definition of explanatory hypothesis

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Implement Sci

Understanding contexts: how explanatory theories can help

Frank davidoff.

1 Lexington, USA

2 Geisel School of Medicine, Dartmouth College, Hanover, NH 03755 USA

Associated Data

Not applicable, because this analysis involves no original research data.

To rethink the nature and roles of context in ways that help improvers implement effective, sustained improvement interventions in healthcare quality and safety.

Critical analysis of existing concepts of context; synthesis of those concepts into a framework for the construction of explanatory theories of human environments, including healthcare systems.

Data sources

Published literature in improvement science, as well as in social, organization, and management sciences. Relevant content was sought by iteratively building searches from reference lists in relevant documents.

Scientific thought is represented in both causal and explanatory theories. Explanatory theories are multi-variable constructs used to make sense of complex events and situations; they include basic operating principles of explanation, most importantly: transferring new meaning to complex and confusing phenomena; separating out individual components of an event or situation; unifying the components into a coherent construct (model); and adapting that construct to fit its intended uses. Contexts of human activities can be usefully represented as explanatory theories of peoples’ environments; they are valuable to the extent they can be translated into practical changes in behaviors.

Healthcare systems are among the most complex human environments known. Although no single explanatory theory adequately represents those environments, multiple mature theories of human action, taken together, can usually make sense of them. Current mature theories of context include static models , universal-plus-variable models , activity theory and related models , and the FITT framework (Fit between Individuals, Tasks, and Technologies). Explanatory theories represent contexts most effectively when they include basic explanatory principles.

Conclusions

Healthcare systems can usefully be represented in explanatory theories. Improvement interventions in healthcare quality and safety are most likely to bring about intended and sustained changes when improvers use explanatory theories to align interventions with the host systems into which they are being introduced.

Introduction

Human contexts—defined in this commentary primarily as the meaning of human environments to the people who live and work in them—are major determinants of the effectiveness and generalizability of interventions to improve healthcare quality and safety [ 1 , 2 ]. Despite the importance of contex, much about it remains obscure, as do the specific mechanisms by which local contexts affect the implementation of improvement interventions. As a consequence, context is still sometimes vaguely referred to in scholarly work as “All those things in the situation which are relevant to meaning in some sense, but which I haven’t identified.”([ 2 ] p. 6).

Context plays an important role in both improvement science and implementation science; limited understanding of context therefore limits understanding of both the fundamental principles of improvement and the actions that put improvements into practice. Achieving deep understanding of context is a challenge that has baffled serious improvers, researchers, and scholars for years [ 2 ]. This difficulty [ 3 ] suggests that multiple complementary explanatory theories might prove more useful than any single theory in understanding both context in general and specific local contexts.

This commentary is intended as a complement to the SQUIRE guidelines for publication of work in quality improvement [ 4 ]. It explores the premise that explanatory theories of human environments can help improvers work flexibly from first principles rather than rigid formulas, and, as is true for good theories generally [ 6 ], can provide improvers with explicit reasons why particular interventions are likely to be effective in specific systems; it examines the nature of explanatory theories and the basic principles of explanation, considers the contributions of those principles to mature (i.e., fully-developed, refined) explanatory theories of complex human environments, and considers the nature of the data needed in constructing explanatory theories of local environments, and the methods used for gathering the requisite data. The commentary proposes, finally, that it is both appropriate and useful early in the planning of an improvement program, to create an explanatory theory of the local healthcare environment into which planned intervention is to be introduced, then use that theory in linking the intervention with that environment. The commentary also encourages improvers to reconsider and revise the initial explanatory theories from time to time as more is learned about the local environment during the improvement process.

Explanatory theories

Scientific thought is built primarily around two complementary mental constructs [ 5 ]: causal and explanatory theories. Explanatory theories are created to help to understand complex, confusing events and situations; they often also serve as sources of testable causal theories of events and situations.

Although explanatory theories are sometimes thought to play a less central role in science than causal theories [ 5 – 7 ], many explanatory theories— including the theory of evolution, the periodic table of the elements, and the structure of DNA—have proven uniquely helpful in understanding important phenomena in natural sciences. Political science is built largely around explanatory theories [ 7 ]; process flow diagrams and Pareto charts are among the explanatory theories that help understand events and situations in improvement science [ 8 ].

The concepts in this commentary were developed from the published literature in improvement science as well as the social, behavioral, organizational, improvement, and management sciences. Sources that proved especially important include Bate et al. [ 2 ] on the dynamic properties of context, Squires et al. [ 3 ] on the construction of explanatory theories, Braithwaite et al. [ 9 ] and Greenhalgh [ 10 ] on complexity, Nardi [ 11 ] and Greenhalgh et al. [ 12 ] on theories of human action, Vandenbroucke [ 5 ] and Clarke and Primo [ 7 ] on explanatory theory, and Pitt [ 13 ] on the fundamentals of explanation. Literature searches were built out from reference citations in these and related publications.

The author’s experience as editor of a major clinical journal ( Annals of Internal Medicine ), and as publications editor at the Institute for Healthcare Improvement (Cambridge, MA), also helped in constructing this commentary. Discussions in the improvement science development group at the Health Foundation (London, UK) and in the Standards for Quality Improvement Reporting Excellence (SQUIRE) leadership group [ 4 ] also contributed importantly to this effort.

The complexity and dynamism of human environments

The most salient properties of human environments are arguably their complexity [ 9 , 10 ] and their dynamic nature [ 2 ]. This commentary rests on the concept of “complex systems” summarized in Table  1 .

Distinctive properties of complex systems (adapted from references [ 9 , 10 , 14 ])

The degrees of complexity in human systems are usefully characterized in the following schema [ 14 ], in which the cooking of a specific dish is represented as simple . Challenges at this basic level are usually managed successfully by following explicit, straightforward recipes or protocols.

By comparison, sending a rocket to the moon is complicated for multiple psychological, social, and technical reasons. Successful management of complicated challenges often requires the use of dedicated management tools such as checklists (mainly to overcome the limitations of human memory) and protocols that map out contingency-dependent decision points (mainly to avoid oversimplification).

Finally, the challenge of raising a child can be seen as  complex , largely because it involves such a large number of variables, many of them poorly defined, which often leads to unpredictable outcomes, e.g., when the experience of raising one child successfully is of little use in raising the next.

Principles of explanation (sense-making)

Although a human event or situation can sometimes be explained adequately in terms of causal mechanisms, the inherent complexity and dynamic nature of events and situations usually requires explanations that go beyond causality and include descriptive explanatory principles [ 5 , 6 , 10 , 13 – 16 ]. Most importantly, those principles include transferring new meaning to the event or situation, establishing its familiarity and internal logic, separating out its individual components, unifying its components into a coherent mental construct or “gestalt”, and adapting the explanation to fit its intended uses.

Transferring (sharing) meaning

The classic human system for transferring or sharing meaning is, of course, language [ 17 ]: witness the substantial loss or distortion of its meaning that results when a word or phrase is taken out of context, and conversely the greater precision of a literature search that uses search terms embedded in linguistic contexts, when contrasted with a search that uses search terms lacking such embellishment [ 18 , 19 ]. (Salmon proposes that the transfer of information, energy or causal inference between processes is more meaningful than transfer between events [ 16 ].)

Familiarity

Familiarity, by itself, is neither necessary nor sufficient to make sense of an event or situation. But familiarity is nonetheless an important component of explanation, because a sense of familiarity provides a sense of understanding ([ 20 ], p. 52). Metaphor is often the chosen mechanism for transferring meaning from familiar things to those that are less familiar, a property that prompted Aristotle to comment that it is metaphor that most produces knowledge. The psychologist Julian Jaynes has argued that metaphor is not a “mere extra trick of language” but is rather “the very constitutive ground of language,” and that “it is by metaphor that language grows” ([ 20 ], pp. 48-9).

Explanation in natural sciences is usually considered adequate when its logic is clear, as when statement of a general law (a “regularity”) is coupled with statement of a specific antecedent condition. In physics, for example, a statement such as “All wave phenomena of a certain type satisfy the law of refraction, and light is a wave of that type” is accepted as a logical construct that meaningfully explains the refraction of light ([ 13 ] p. 10]).

Separating out and unifying components

By themselves, the individual components of an event or situation ordinarily have little if any inherent meaning. But the construct that results when those components are brought together to make a coherent whole (usually as a narrative, map, model, or mathematical expression) is uniquely helpful in making sense of that event or situation [ 4 , 21 , 22 ]. Important new meanings can emerge as well—often unexpectedly—from the resulting construct. For these reasons, some philosophers of science consider unification of a phenomenon’s individual components into a coherent whole as the main principle by which explanation renders a phenomenon understandable [ 4 , 5 , 21 , 22 ].

The sharing of meaning among a phenomenon’s individual components finds expression in catch-phrases such as the jigsaw puzzle effect , and “The whole is greater than the sum of its parts.” On a more grand scale, the theory of evolution is said to acquire its explanatory power when “an apparently modest allegiance to mere fact gathering” abruptly crystallizes into a “whole world view” [ 23 ].

Details of the mental process through which unification creates explanations unfortunately remain obscure. And curiously, even a highly coherent construct of an event or situation does not necessarily help understand whether its components are truly independent, whether the interactions among them are uni-directional or recursive, and which components (if any) are most important in determining its overall behavior. Moreover, craftspeople such as watchmakers and car mechanics understand that success in their work depends on their ability to separate out the components of the complex systems they are called on to assemble or repair (disaggregate them) at least as much as on their ability to understand how the components contribute collectively to an event or situation’s overall behavior (unify them). At least in theory, the explanatory principles of disaggregation and unification appear to contradict each other, but in practice, the two principles are often complementary. In managing a human system, for example, a leader’s ability to unify various groups’ individual modes of decision-making can complement his or her ability to distinguish those modes from one another [ 24 ].

Adapting explanations

Explanatory theories are arguably successful to the extent people can translate them into practical implementation behavior—e.g., manage the environments in which they live and work or predict the likelihood that a specific event will happen in the future ([ 16 ] p. 77). Not surprisingly, therefore, the explanatory theories people develop on their own to manage their personal environments differ substantially from the ones they develop collectively to further the work of the organizations in which they work. For similar reasons, personal and organization-related explanatory theories differ from those that outside researchers create to understand these various environments.

Personal contexts

Peoples’ intense, universal need to give meaning to “the brutal aboriginal flux” of their lived experience [ 1 ] suggests that humans can be defined as “reason-giving animals” [ 25 ]. They begin creating explanatory theories of their personal environments at an extremely early age [ 26 ], then extend and refine those theories as they and their personal environments change over time. Personal explanatory theories are usually implicit and poorly articulated; they can also be distorted, incomplete, or inappropriate since they frequently lack independent reality testing.

Organizational and professional contexts

Workers in organizations are called on to create explanatory models that make sense of the internal structure and function of those organizations, as well as of the external environments in which their organizations are embedded. Weick et al. describe this work as a creative, collaborative undertaking that involves “language, talk, and communication” and is “ongoing, subtle, swift, social, and easily taken for granted” [ 1 , 27 , 28 ]. Early in this sense-making process, workers in an organization “bracket” information (i.e., identify items they see as especially relevant to their particular situation), then name (label) those items, which stabilizes the streaming of their experience [ 1 ].

The way people in organizations envision events and situations also immediately begins their social and administrative work of organizing, because bracketing and labeling events predisposes them to find common ground and provides them with a set of cognitive categories, plus a typology of potential actions. (Bracketing central venous catheter infection and labeling it as primarily a social rather than a biological problem [ 29 ] played an important role in shaping an intervention that successfully lowers the infection rate [ 30 ].) Workers then use such newly defined contextual elements as they literally talk their organization-related explanatory theories into existence [ 1 ].

The sense-making process described above closely resembles the one that professionals in applied disciplines, together with their clients, use to make sense of the problem situations they are called on to manage ([ 31 ], pp. 267–83). More specifically, medical professionals will recognize its resemblance to the process by which they and their patients formulate the essential explanatory theories they know as diagnoses .

Mature explanatory theories of human environments

People initially sketch out rough explanatory theories of environments which usually involve basic principles of explanation, then subsequently broaden and refine these nascent constructs into more mature theories. Important examples of such mature explanatory theories include static theories , universal-plus-variable theories , activity theory and related general theories of human action , and the FITT framework (Fit between Individuals, Task, and Technology).

Static theories

Several research groups have developed explanatory theories of outstanding healthcare systems by selecting the components they judge to be most closely associated with certain systems’ ability to deliver exceptionally safe, high-quality care [ 32 – 36 ], then assembling those components into structured models. (A recent international effort is engaged in constructing a new and more meaningful theory of this type [ 3 ]).

The individual components identified in these theory-building exercises—buildings, equipment, leadership, geographic location, teaching status, financial and intellectual resources, and the like—are quite heterogeneous and the resulting constructs often pay little attention to functional relationships among the components or to the ways in which the process of care plays out over time for individual patients. Metaphorically speaking, then, explanatory theories such as these describe the anatomy of exceptional healthcare environments, but not their physiology ; that is, they are static , which could account for the limited ability of this type of explanatory theory to explain variation in the effectiveness of improvement interventions across different healthcare systems.

Universal-plus-variable models

Working from detailed on-site observations in high-performing healthcare systems, Bate et al. [ 2 , 37 ] have constructed a generalized explanatory theory of such systems. Their experience is reflected in their comment that “although research has provided an abundance of data on key success factors in QI efforts, very little was previously known about how these combine and interact with each other in the improvement process over time.” They comment further that the context of a healthcare system is “a process; dynamic, fluid, and constantly moving, not lumpen, material, or static,” and that “it is the dynamic and ongoing interaction between [the domains of an environment] rather than any one of them individually or independently, that accounts for the effectiveness of a QI intervention,” as well as for “the striking variation between similar QI interventions in different places” ([ 2 ]p. 11).

These investigators then refine and sharpen the focus of their emerging explanatory theory by postulating that a healthcare system’s ability to deliver outstanding care lies in the combination of the two major components— universals and variables —that characterize an organization’s local situation. More specifically, they identify the challenges inherent in several distinct areas—physical/technological, emotional, educational, cultural and political, and structural—as the universals in all healthcare organizations; they also characterize the actions that individual workers and groups take in response to those challenges as differing both within and across organizations to the point where those actions and the possible combinations among them can be assumed to be “practically innumerable” ([ 37 ], p. 168), i.e., they are the variables .

The resulting universal-plus-variable explanatory theory of human contexts gains plausibility from its affinity with other established cognitive systems in which people represent the complex meanings that matter to them. The best known and arguably most important of such systems is of course language [ 17 ]; people produce language by embedding differing strings of individual words (the variables) in a relatively small number of stable grammatical structures (the universals). They then use the resulting construct to create a virtually unlimited number of statements that are meaningful to others, even though many of those statements have not been seen or heard previously.

Music provides another illuminating example of a meaningful universal-plus-variable explanatory theory [ 39 ]. Composers in each musical tradition embed differing arrays of tones (variables) in a limited set of stable, widely recognized harmonic constructs (universals). One critic has elegantly captured this explanatory theory of music (or at least of Western music) in his pithy comment that “Mozart used the same B-flat as everyone else.”

Activity theory and related models of human action

The universal-plus-variable explanatory theory of contexts also resonates with several earlier mature explanatory theories of human action, including Activity Theory and related models [ 11 ]. Some of these action theories are now seen as especially useful in understanding the interaction between people and computer systems [ 12 ]. In these theories, it is precisely the ongoing bi-directional interaction between static human environments and the dynamic needs, interests, and experiences people bring to encounters with those environments that creates most of the contexts (meanings) of human life. For example, context is understood as follows in Activity Theory as an overarching, albeit secondary, consideration: “[W]hat takes place in an activity system composed of object, actions, and operations, is the context… [C]ontext is not an outer container or shell inside of which people behave in certain ways.” Context in these theories is thus “both internal to people…and at the same time, external to people” [ 11 ], i.e., as an integrated whole. This unifying perspective invalidates “simplistic explanations that divide internal and external, and schemes that see context as external to people.”

The FITT framework (Fit between Individuals, Tasks, and Technologies)

Developed largely to explain the adoption of information and communication technologies (IT) [ 40 ], the FITT framework clearly distinguishes an organization’s established and widely recognized tasks and technologies from its workers’ shifting dynamic behaviors [ 5 , 12 , 40 , 41 ] (Table  2 ), and in that respect, it resembles other universal-plus-variable explanatory theories of human activity.

Use of explanatory principles in constructing an electronic decision-support system to improve postoperative care (adapted from references [ 5 , 6 , 13 , 41 ])

As noted elsewhere [ 6 ], the FITT framework has been used to guide the successful implementation of an innovative electronic order system for post-operative surgical care [ 41 ]. Researchers in that study explicitly used the FITT framework to help them interweave their new electronic system with the healthcare environment in which they implemented it.

The nature of data needed to construct explanatory theories of healthcare environments

Adequate understanding of human environments requires that explanatory theories take the enormous complexity of those environments appropriately into account. Although complexity of this magnitude can be a cause for despair among improvers and researchers, the statistician George Box’s pungent comment that “All theories are wrong, but some are illuminating and useful” offers reassurance that creating explanatory theories of human environments, including healthcare systems, is likely on balance to be worth the effort.

Data used to create meaningful explanatory theories of human environments

Creating explanatory theories of human environments that help implement successful improvement interventions apparently requires open-ended, multi-level data on working relationships in organizations [ 1 , 9 – 11 , 29 , 31 , 36 – 38 , 41 – 48 ]. Research groups are now laboring to clarify the essential nature of such data (Table  3 ), while also obtaining insights into effective techniques for collecting and analyzing those data (Table  4 ).

Characteristics of data that contribute meaningfully to explanatory theories of human environments (adapted from references [ 9 , 42 – 48 ])

Methods for collecting and analyzing data that help to plan, implement, and evaluate the impact of improvement interventions (adapted from references [ 31 , 42 , 47 , 48 ])

It is important to note in this connection that improvement interventions reach their full potential more successfully when their implementation builds on the complexity of the systems they intend to change than when they underestimate or ignore that complexity [ 9 ]. Even documenting that a healthcare system has “a long way to go” to achieve specific solutions within each of the six universal challenge area (in contrast to being either “some way there” or “already there”) can help improvers pinpoint current gaps and opportunities in that system’s quality and safety, and facilitate productive discussions on their future improvement efforts (Cf. Codebook for Quality Improvement Practice, for example) ([ 37 ], p., 177).

In like fashion, answering a question regarding organizational complexity (e.g., “How did this practice miss a diagnosis?”) can be more effective in changing system performance than obtaining answering a narrowly focused question such as “How did an individual practitioner miss a diagnosis?”) [ 42 – 48 ].

Traditional scientific methods will undoubtedly continue over time to help understand human environments, including environments that are as complex and dynamic as healthcare systems. At the same time, the difficulty of understanding those environments in the concepts and language of sciences suggests that explanatory theories of those environments will be more meaningful when they include contributions from the arts and humanities.

An important, and intriguing, painting by the Belgian surrealist René Magritte hints at the potential of such an ecumenical approach. In this work, Magritte apparently tries to represent the complex, emotionally freighted world of tobacco use by juxtaposing the image of a tobacco pipe with a written comment: “Ceci n’est pas une pipe” (“This is not a pipe”). The resulting cognitive dissonance suggests the artist’s intent is to increase the painting’s impact by cautioning his viewers that “This is only the image of a pipe, not the actual object; don’t confuse the two,” and encouraging them not to mistake the part for the whole (a pipe is, after all, only one small part of tobacco smoking).

But he does not stop there: in his effort to jolt viewers toward even deeper and more precise awareness of tobacco use, Magritte resorts to a particularly unorthodox representation of the pernicious habit, when he flatly asserts that “a pipe actually isn’t a pipe,” his surrogate for a paradoxical characterization of tobacco use in terms of what it is not . Examples of this startling apophatic (i.e., reverse) way to represent complex, confusing realities are now appearing in the literature of improvement science, as in “wake-up calls” telling us that  neither a checklist of infection control measures [ 49 ] nor a surgical safety checklist  [ 50 ], by itself, is an improvement intervention (the unstated subtext being that successful, sustained improvement absolutely requires explicit, extensive coordination, and tight linkage, between the intervention and the environment in which it is being implemented).

In articulating her explanatory theory of the world of falconry , the scholar and writer Helen Macdonald also turns, as follows, to this paradoxical, inverse way of understanding the deeper meaning of a complex human environment [ 51 ]:

“[T]here is a world of things out there – rocks and trees and grass and all the things that crawl and run and fly. They are all things in themselves, but we make them sensible to us by giving them meanings that shore up our own views of the world. In my time [living with and training my goshawk] Mabel I’ve learned how you feel more human once you have known, even in your imagination, what it is likely to be not”.

This commentary considers evidence that reinforces the crucial reality that the healthcare systems in which improvement programs take place—or, more specifically, the values and character of those systems—are at least as important in improving care as the specifics of the improvement interventions themselves. This obvious but often underappreciated reality environmental feature argues strongly for the development of sophisticated, nuanced understanding of those environments early in the implementation of improvement programs, and consistent application of that understanding during the improvement process. Realistically, understanding a human environment—especially one as complex and dynamic as a healthcare system—is an arduous, demanding undertaking, which further underscores the value of building a basic set of context-related initiatives into the implementation of any sizeable healthcare improvement program. These initiatives might include the following:

• As early as possible in planning the program, create an explanatory theory of the host environment that incorporates the basic principles of explanation, especially unification of the environment’s major components;
• If possible, involve social scientists, as well as professionals from humanities (e.g., creative writers, reporters, historians, graphic artists and the like) in the development of that explanatory theory;
• Use that explanatory theory in coordinating and linking the intervention with the host environment;
• Explore the use of established mature explanatory theories, individually or collectively, in making sense of the local host environment;
• Assess the relative importance of the environment’s major components as determinants of its nature and behavior; its successes and failures;
• From time to time, review the most current version of the explanatory theory and revise it if necessary as more is learned about the host environment and about the interaction between environment and intervention
• To avoid creating jitter and instability in the program, resist unnecessary tinkering with the makeup and application of the explanatory theory;
• Make explicit efforts to assure that all members of the improvement team are familiar with the major components of the host environment, and understand how those components fit/work together;
• Adapt the focus, comprehensiveness, organization, and level of detail of the explanatory theory of the host environment, to make it as useful as possible for its most important users.

Acknowledgements

The author gratefully acknowledges useful comments of Paul Batalden, Trisha Greenhalgh, Mary Dixon-Woods, Lucian Leape, Tom Sheridan, Cyrus Hopkins, and Judith Singer on earlier versions of this article.

Acronym for Standards for Quality Improvement Reporting Excellence.

No funding was received for this work.

Availability of data and materials

Author’s contributions.

The author gathered all the reference material, drafted the initial versions of the paper and all subsequent revisions, and takes responsibility for the entire content of the article.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Competing interests.

The author declares no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

What is explanatory research?

Last updated

12 June 2023

Reviewed by

Miroslav Damyanov

The search for knowledge and understanding never stops in the field of research. Researchers are always finding new techniques to help analyze and make sense of the world. Explanatory research is one such technique. It provides a new perspective on various areas of study.

So, what exactly is explanatory research? This article will provide an in-depth overview of everything you need to know about explanatory research and its purpose. You’ll also get to know the different types of explanatory research and how they’re conducted.

Analyze explanatory research

Get a deeper understanding of your explanatory research when you analyze it in Dovetail

• Explanatory research: definition

Explanatory research is a technique used to gain a deeper understanding of the underlying reasons for, causes of, and relationships behind a particular phenomenon that has yet to be extensively studied.

Researchers use this method to understand why and how a particular phenomenon occurs the way it does. Since there is limited information regarding the phenomenon being studied, it’s up to the researcher to develop fresh ideas and collect more data.

The results and conclusions drawn from explanatory research give researchers a deeper understanding and help predict future occurrences.

• Descriptive research vs. explanatory research

Descriptive research aims to define or summarize an event or population without explaining why it exists. It focuses on acquiring and conveying facts.

On the other hand, explanatory research aims to explain why a phenomenon occurs by working to understand the causes and correlations between variables.

Unlike descriptive research, which focuses on providing descriptions and characteristics of a given phenomenon, explanatory research goes a step further to explain different mechanisms and the reasons behind them. Explanatory research is never concerned with producing new knowledge or solving problems. Instead, it aims to explain why and how something happens.

• Exploratory research vs. explanatory research

Explanatory research explains why specific phenomena function as they do. Meanwhile, exploratory research examines and investigates an issue that is not clearly defined. Both methods are crucial for problem analysis.

Researchers use exploratory research at the outset to discover new ideas, concepts, and opportunities. Once exploratory research has identified a potential area of interest or problem, researchers employ explanatory research to delve further into the specific subject matter.

Researchers employ the explanatory research technique when they want to explain why and how something occurs in a certain way. Researchers who employ this approach usually have an outcome in mind, and carrying it out is their top priority.

• When to use explanatory research

Explanatory research may be helpful in the following situations:

When testing a theoretical model: explanatory research can help researchers develop a theory. It can provide sufficient evidence to validate or refine existing theories based on the available data.

When establishing causality: this research method can determine the cause-and-effect relationships between study variables and determine which variable influences the predicted outcome most. Explanatory research explores all the factors that lead to a certain outcome or phenomenon.

When making informed decisions: the results and conclusions drawn from explanatory research can provide a basis for informed decision-making. It can be helpful in different industries and sectors. For example, entrepreneurs in the business sector can use explanatory research to implement informed marketing strategies to increase sales and generate more revenue.

When addressing research gaps: a research gap is an unresolved problem or unanswered question due to inadequate research in that space. Researchers can use explanatory research to gather information about a certain phenomenon and fill research gaps. It also enables researchers to answer previously unanswered questions and explain different mechanisms that haven’t yet been studied.

When conducting program evaluation: researchers can also use the technique to determine the effectiveness of a particular program and identify all the factors that are likely to contribute to its success or failure.

• Types of explanatory research

Here are the different types of explanatory research:

Case study research: this method involves the in-depth analysis of a given individual, company, organization, or event. It allows researchers to study individuals or organizations that have faced the same situation. This way, they can determine what worked for them and what didn’t.

Experimental research: this involves manipulating independent variables and observing how they affect dependent variables. This method allows researchers to establish a cause-and-effect relationship between different variables.

Quasi-experimental research: this type of research is quite similar to experimental research, but it lacks complete control over variables. It’s best suited to situations where manipulating certain variables is difficult or impossible.

Correlational research: this involves identifying underlying relationships between two or more variables without manipulating them. It determines the strength and direction of the relationship between different variables.

Historical research: this method involves studying past events to gain a better understanding of their causes and effects. It’s mostly used in fields like history and sociology.

Survey research: this type of explanatory research involves collecting data using a set of structured questionnaires or interviews given to a representative sample of participants. It helps researchers gather information about individuals’ attitudes, opinions, and behaviors toward certain phenomena.

Observational research: this involves directly observing and recording people in their natural setting, like the home, the office, or a shop. By studying their actions, needs, and challenges, researchers can gain valuable insights into their behavior, preferences, and pain points. This results in explanatory conclusions.

• How to conduct explanatory research

Take the following steps when conducting explanatory research:

Develop the research question

The first step is to familiarize yourself with the topic you’re interested in and clearly articulate your specific goals. This will help you define the research question you want to answer or the problem you want to solve. Doing this will guide your research and ensure you collect the right data.

Formulate a hypothesis

The next step is to formulate a hypothesis that will address your expectations. Some researchers find that literature material has already covered their topic in the past. If this is the case with you, you can use such material as the main foundation of your hypothesis. However, if it doesn’t exist, you must formulate a hypothesis based on your own instincts or literature material on closely related topics.

Select the research type

Choose an appropriate research type based on your research questions, available resources, and timeline. Consider the level of control you need over the variables.

Next, design and develop instruments such as surveys, interview guides, or observation guidelines to gather relevant data.

Collect the data

Collecting data involves implementing the research instruments and gathering information from a representative sample of your target audience. Ensure proper data collection protocol, ethical considerations , and appropriate documentation for the data you collect.

Analyze the data

Once you have collected the data you need for your research, you’ll need to organize, code, and interpret it.

Use appropriate analytical methods, such as statistical analysis or thematic coding , to uncover patterns, relationships, and explanations that address your research goals and questions. You may have to suggest or conduct further research based on the results to elaborate on certain areas.

Communicate the results

Finally, communicate your results to relevant stakeholders , such as team members, clients, or other involved partners. Present your insights clearly and concisely through reports, slides, or visualizations. Provide actionable recommendations and avenues for future research.

• Examples of explanatory research

Here are some real-life examples of explanatory research:

Understanding what causes high crime rates in big cities

Law enforcement organizations use explanatory research to pinpoint what causes high crime rates in particular cities. They gather information about various influencing factors, such as gang involvement, drug misuse, family structures, and firearm availability.

They then use regression analysis to examine the data further to understand the factors contributing to the high crime rates.

Factors that influence students’ academic performance

Educators and stakeholders in the Department of Education use questionnaires and interviews to gather data on factors that affect academic performance. These factors include parental engagement, learning styles, motivation, teaching quality, and peer pressure.

The data is used to ascertain how these variables affect students’ academic performance.

Examining what causes economic disparity in certain areas

Researchers use correlational and experimental research approaches to gather information on variables like education levels, household income, and employment rates. They use the information to examine the causes of economic disparity in certain regions.

• Advantages of explanatory research

Here are some of the benefits you can expect from explanatory research:

Deeper understanding : the technique helps fill research gaps in previous studies by explaining the reasons, causes, and relationships behind particular behaviors or phenomena.

Competitive edge: by understanding the underlying factors that drive customer satisfaction and behavior, companies can create more engaging products and desirable services.

Predictable capabilities: it helps researchers and teams make predictions regarding certain phenomena like user behavior or future iterations of product features.

Informed decision-making: explanatory research generates insights that can help individuals make informed decisions in various sectors.

• Disadvantages of explanatory research

Explanatory research is a great approach for better understanding various phenomena, but it has some limitations.

It’s time-consuming: explanatory research can be a time-consuming process, requiring careful planning, data collection, analysis, and interpretation. The technique might extend your timeline.

It’s resource intensive: explanatory research often requires a significant allocation of resources, including financial, human, and technological. This could pose challenges for organizations with limited budgets or constraints.

You have limited control over real-world factors: this type of research often takes place in controlled environments. Researchers may find this limits their ability to capture real-world complexities and variables that influence a particular behavior or phenomenon.

Depth and breadth are difficult to balance : explanatory research mainly focuses on a narrow hypothesis, which can limit the scope of the research and prevent researchers from understanding a problem more broadly.

Should you be using a customer insights hub?

Do you want to discover previous research faster?

Do you share your research findings with others?

Do you analyze research data?

Start for free today, add your research, and get to key insights faster

Editor’s picks

Last updated: 11 January 2024

Last updated: 15 January 2024

Last updated: 17 January 2024

Last updated: 12 May 2023

Last updated: 30 April 2024

Last updated: 18 May 2023

Last updated: 25 November 2023

Last updated: 13 May 2024

Latest articles

Related topics, .css-je19u9{-webkit-align-items:flex-end;-webkit-box-align:flex-end;-ms-flex-align:flex-end;align-items:flex-end;display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-flex-direction:row;-ms-flex-direction:row;flex-direction:row;-webkit-box-flex-wrap:wrap;-webkit-flex-wrap:wrap;-ms-flex-wrap:wrap;flex-wrap:wrap;-webkit-box-pack:center;-ms-flex-pack:center;-webkit-justify-content:center;justify-content:center;row-gap:0;text-align:center;max-width:671px;}@media (max-width: 1079px){.css-je19u9{max-width:400px;}.css-je19u9>span{white-space:pre;}}@media (max-width: 799px){.css-je19u9{max-width:400px;}.css-je19u9>span{white-space:pre;}} decide what to .css-1kiodld{max-height:56px;display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}@media (max-width: 1079px){.css-1kiodld{display:none;}} build next, decide what to build next.

Users report unexpectedly high data usage, especially during streaming sessions.

Users find it hard to navigate from the home page to relevant playlists in the app.

It would be great to have a sleep timer feature, especially for bedtime listening.

I need better filters to find the songs or artists I’m looking for.

Log in or sign up

Get started for free

• Table of Contents
• Random Entry
• Chronological
• Editorial Information
• About the SEP
• Editorial Board
• How to Cite the SEP
• Special Characters
• Advanced Tools
• Support the SEP
• PDFs for SEP Friends
• Make a Donation
• SEPIA for Libraries
• Entry Contents

Bibliography

Academic tools.

• Friends PDF Preview
• Author and Citation Info
• Back to Top

Scientific Explanation

Issues concerning scientific explanation have been a focus of philosophical attention from Pre-Socratic times through the modern period. However, modern discussion really begins with the development of the Deductive-Nomological ( DN ) model. This model has had many advocates (including Popper 1959, Braithwaite 1953, Gardiner, 1959, Nagel 1961) but unquestionably the most detailed and influential statement is due to Carl Hempel (1942, 1965a, and Hempel & Oppenheim 1948). These papers and the reaction to them have structured subsequent discussion concerning scientific explanation to an extraordinary degree. After some general remarks by way of background and orientation ( Section 1 ), this entry describes the DN model and its extensions, and then turns to some well-known objections ( Section 2 ). It next describes a variety of subsequent attempts to develop alternative models of explanation, including Wesley Salmon’s Statistical Relevance ( Section 3 ) and Causal Mechanical ( Section 4 ) models, Unificationist models due to Michael Friedman and Philip Kitcher ( Section 5 ), and Pragmatic theories found in the work of van Fraassen ( Section 6 ). Section 7 provides a summary and discusses directions for future work. This article thus discusses treatments of scientific explanation up to the end of the twentieth century.

1. Background and Introduction

2.1 the basic idea, 2.2 the role of laws in the dn model, 2.3 inductive statistical explanation, 2.4 motivation for the dn model: nomic expectability and a regularity account of causation, 2.5 explanatory understanding and nomic expectability: counterexamples to sufficiency, 3.1 the basic idea, 3.2 the sr model and low probability events, 3.3 what do statistical theories explain what sorts of examples are accounts of statistical explanation intended to capture, 3.4 causation and statistical relevance relationships, 4.1 the basic idea, 4.2 the cm model and explanatory relevance, 4.3 the cm model and complex systems, 4.4 more recent developments, 5.1 the basic idea, 5.2 illustrations of the unificationist model, 5.3 the illustrations criticized, 6.1 introduction, 6.2 constructive empiricism and the pragmatic theory of explanation, 7.1 the role of causation, 7.2 a single model of explanation, other internet resources, related entries.

As will become apparent, “scientific explanation” is a topic that raises a number of interrelated issues. Some background orientation will be useful before turning to the details of competing models. A presupposition of most recent discussion has been that science sometimes provides explanations (rather than something that falls short of explanation—e.g., “mere description”) and that the task of a “theory” or “model” of scientific explanation is to characterize the structure of such explanations. It is thus assumed that there is (at some suitably abstract and general level of description) a single kind or form of explanation that is “scientific”. In fact, the notion of “scientific explanation” suggests at least two contrasts—first, a contrast between those “explanations” that are characteristic of “science” and those explanations that are not, and, second, a contrast between “explanation” and something else. However, with respect to the first contrast, much of the recent philosophical literature assumes that there is substantial continuity between explanations found in science and some forms of explanation found in ordinary, non-scientific contexts. It is further assumed that it is the task of a theory of explanation to capture what is common to both scientific and some ordinary, non-scientific forms of explanation. These assumptions help to explain (what may otherwise strike the reader as curious) why, as this entry will illustrate, discussions of scientific explanation often move back and forth between examples drawn from bona-fide science (e.g., explanations of the trajectories of the planets that appeal to Newtonian mechanics) and more homey examples (e.g., the tipping over of inkwells).

With respect to the second contrast, most models of explanation assume that it is possible for a set of claims to be true, accurate, supported by evidence, and so on and yet unexplanatory. For example, all of the accounts of scientific explanation described below would agree that an account of the appearance of a particular species of bird of the sort found in a bird guidebook is, however accurate, not an explanation of anything of interest to biologists (such as the development, characteristic features, or behavior of that species). Instead, such an account is “merely descriptive”. However, different models of explanation provide different accounts of what the contrast between the explanatory and merely descriptive consists in.

A related point is that, while most theorists of scientific explanation have proposed models that are intended to cover at least some cases of explanation that we would not think of as part of science, they have nonetheless assumed some implicit restriction on the kinds of explanation they have sought to reconstruct. It has often been noted that the word “explanation” is used in a wide variety of ways in ordinary English—we speak of explaining the meaning of a word, explaining how to bake a pie, explaining why one made a certain decision (where this is to offer a justification) and so on. Although the various models discussed below have sometimes been criticized for their failure to capture all of these forms of “explanation” (see, e.g., Scriven 1959), it is clear that they were never intended to do this. Instead, their intended explicandum is, very roughly, explanations of why things happen, where the “things” in question can be either particular events or something more general—e.g., regularities or repeatable patterns in nature. Paradigms of this sort of explanation include: the explanation for the advance in the perihelion of mercury provided by General Relativity, the explanation of the extinction of the dinosaurs in terms of the impact of a large asteroid at the end of the Cretaceous period, the explanation provided by the police for why a traffic accident occurred (the driver was speeding and there was ice on the road), and the standard explanation provided in economics textbooks for why monopolies will, in comparison with firms in perfectly competitive markets, raise prices and reduce output.

Finally, a few words about the broader epistemological and methodological background to the models described below. Many philosophers think of concepts like “explanation”, “law”, “cause”, and “support for counterfactuals” as part of an interrelated family of concepts that are “modal” in character. For familiar “empiricist” reasons, Hempel and many other early defenders of the DN model regarded these concepts as not well understood, at least prior to analysis. It was assumed that it would be “circular” to explain one concept from this family in terms of others from the same family and that they must instead be explicated in terms of other concepts from outside the modal family—concepts that more obviously satisfied (what were taken to be) empiricist standards of intelligibility and testability. For example, in Hempel’s version of the DN model, the notion of a “law” plays a key role in explicating the concept of “explanation”, and his assumption is that laws are just regularities that meet certain further conditions that are also acceptable to empiricists. As we shall see, these empiricist standards (and an accompanying unwillingness to employ modal concepts as primitives) have continued to play a central role in the models of explanation developed subsequent to the DN model.

A related issue has to do with whether all scientific explanations are causal and if not, what distinguishes causal from non-causal explanations. Hempel recognized both causal and non-causal forms of explanation but held that both were captured by the DN model –in his view, causal explanations are simply DN explanations that cite causal laws (which he regarded as a proper subset of all laws). Many but not all of the accounts discussed below in effect assume that many of the problems with the DN model can be traced to its commitment to an inadequate account of causation; thus that getting clearer about causal notions would lead to more adequate accounts of explanation. By contrast, a substantial amount of recent discussion of explanation has moved away from this focus on causation and instead explores the possibility of non-causal forms of explanation. [ 1 ]

Suggested Readings : Salmon (1989) is a superb critical survey of all the models of scientific explanation discussed in this entry. Kitcher and Salmon (1989), Pitt (1988), and Ruben (1993) are anthologies that contain a number of influential articles.

2. The DN Model

According to the Deductive-Nomological Model, a scientific explanation consists of two major “constituents”: an explanandum , which is a sentence “describing the phenomenon to be explained” and an explanans , “the class of those sentences which are adduced to account for the phenomenon” (Hempel & Oppenheim 1948 [1965: 247]). For the explanans to successfully explain the explanandum several conditions must be met. First, “the explanandum must be a logical consequence of the explanans” and “the sentences constituting the explanans must be true” (Hempel 1948 [1965: 248]). That is, the explanation should take the form of a sound deductive argument in which the explanandum follows as a conclusion from the premises in the explanans. This is the “deductive” component of the model. Second, the explanans must contain at least one “law of nature” and this must be an essential premise in the derivation in the sense that the derivation of the explanandum would not be valid if this premise were removed. This is the “nomological” component of the model—“nomological” being a philosophical term of art which, suppressing some niceties, means (roughly) “lawful”. In its most general formulation, the DN model is meant to apply both to the explanation of “general regularities” or “laws” such as (to use Hempel and Oppenheim’s examples) why light conforms to the law of refraction and also to the explanation of particular events, conceived as occurring at a particular time and place, such as the bent appearance of the partially submerged oars of a rowboat on a particular occasion of viewing. As an additional illustration of a DN explanation of a particular event, consider a derivation of the position of Mars at some future time from Newton’s laws of motion, the Newtonian inverse square law governing gravity, and information about the mass of the sun, the mass of Mars and the present position and velocity of each. In this derivation the various Newtonian laws figure as essential premises and they are used, in conjunction with appropriate information about initial conditions (the masses of Mars and the sun and so on), to derive the explanandum (the future position of Mars) via a deductively valid argument. The DN criteria are thus satisfied.

The notion of a sound deductive argument is (arguably) relatively clear (or at least something that can be regarded as antecedently understood from the point of view of characterizing scientific explanation). But what about the other major component of the DN model—that of a law of nature? The basic intuition that guides the DN model goes something like this: Within the class of true generalizations, we may distinguish between those that are only “accidentally true” and those that are “laws”. To use Hempel’s examples, the generalization

is, if true, only accidentally so. In contrast,

is a law. Thus, according to the DN model, the latter generalization can be used, in conjunction with information that some particular sample of gas has been heated under constant pressure, to explain why it has expanded. By contrast, the former generalization (1) in conjunction with the information that a particular person n is a member of the 1964 Greensbury school board, cannot be used to explain why n is bald.

While this example may seem clear enough, what exactly is it that distinguishes true accidental generalizations from laws? This has been the subject of a great deal of philosophical discussion, most of which must be beyond the scope of this entry. [ 2 ] For reasons explained in Section 1 , Hempel assumes that an adequate account must explain the notion of law in terms of notions that lie outside the modal family. [ 3 ] He considers (1965b) a number of familiar proposals having this character [ 4 ] and finds them all wanting, remarking that the problem of characterizing the notion of law has proved “highly recalcitrant” (1965b: 338). It seems fair to say, however, that his underlying assumption is that, at bottom, laws are just exceptionless generalizations describing regularities that meet certain additional distinguishing conditions that he is not at present able to formulate. In subsequent decades, a variety of criteria for lawhood have been proposed. Of these the so-called best systems analysis (Lewis 1973) is probably the most popular, but no single account has won general acceptance. Finding an adequate characterization of lawhood is thus an ongoing issue for the DN model.

One point at which this issue is particularly pressing concerns the explanatory status of the so-called special sciences—biology, psychology, economics and so on. These sciences are full of generalizations that appear to play an explanatory role and yet fail to satisfy many of the standard criteria for lawfulness. For example, although Mendel’s law of segregation (M) (which states that in sexually reproducing organisms each of the two alternative forms (alleles) of a gene specifying a trait at a locus in a given organism has 0.5 probability of ending up in a gamete) is widely used in models in evolutionary biology, it has a number of exceptions, such as meiotic drive. A similar point holds for the principles of rational choice theory (such as the generalization that preferences are transitive) which figure centrally in economics. Other widely used generalizations in the special sciences have very narrow scope in comparison with paradigmatic laws, hold only over restricted spatio-temporal regions, and lack explicit theoretical integration.

There is considerable disagreement over whether such generalizations are laws. Some philosophers (e.g., Woodward 2000) suggest that such generalizations satisfy too few of the standard criteria to count as laws but can nevertheless figure in explanations; if so, it apparently follows that we must abandon the DN requirement that all explanations must appeal to laws. Others (e. g., Mitchell 1997), emphasizing different criteria for lawfulness, conclude instead that generalizations like (M) are laws and hence no threat to the requirement that explanations must invoke laws. In the absence of a more principled account of laws, it is hard to evaluate these competing claims and hence hard to assess the implications of the DN model for the special sciences. At the very least, providing such an account is an important item of unfinished business for advocates of the DN model.

The DN model is meant to capture explanation via deduction from deterministic laws and this raises the obvious question of the explanatory status of statistical laws. Do such laws explain at all and if so, what do they explain, and under what conditions? Hempel (1965b) distinguishes two varieties of statistical explanation. The first of these, deductive-statistical ( DS ) explanation, involves the deduction of “a narrower statistical uniformity” from a more general set of premises, at least one of which involves a more general statistical law. Since DS explanation involves deduction of the explanandum from a law, it conforms to the same general pattern as the DN explanation of regularities. However, in addition to DS explanation, Hempel also recognizes a distinctive sort of statistical explanation, which he calls inductive-statistical or IS explanation, involving the subsumption of individual events (like the recovery of a particular person from streptococcus infection) under (what he regards as) statistical laws (such as a law specifying the probability of recovery, given that penicillin has been taken).

While the explanandum of a DN or DS explanation can be deduced from the explanans, one cannot deduce that some particular individual, John Jones, has recovered from the above statistical law and the information that he has taken penicillin. At most what can be deduced from this information is that recovery is more or less probable. In IS explanation, the relation between explanans and explanandum is, in Hempel’s words, “inductive,” rather than deductive—hence the name inductive-statistical explanation. The details of Hempel’s account are complex, but the underlying idea is roughly this: an IS explanation will be good or successful to the extent that its explanans confers high probability on its explanandum outcome.

Thus if it is a statistical law that the probability of recovery from streptococcus, given that one has taken penicillin, is high, and Jones has taken penicillin and recovered, this information can be used to provide an IS explanation of Jones’s recovery. However if the probability of recovery is low (e.g., less than 0.5), given that Jones has taken penicillin, then, even if Jones recovers, we cannot use this information to provide an IS explanation of his recovery.

Why suppose that all (or even some) explanations have a DN or IS structure? There are two ideas which play a central motivating role in Hempel’s (1965b) discussion. The first connects the information provided by a DN argument with a certain conception of what it is to achieve understanding of why something happens—it appeals to an idea about the object or point of giving an explanation. Hempel writes

… a DN explanation answers the question “ Why did the explanandum-phenomenon occur?” by showing that the phenomenon resulted from certain particular circumstances, specified in \(C_1,C_2,\ldots,C_k\), in accordance with the laws \(L_1,L_2,\ldots,L_{\gamma}\). By pointing this out, the argument shows that, given the particular circumstances and the laws in question, the occurrence of the phenomenon was to be expected ; and it is in this sense that the explanation enables us to understand why the phenomenon occurred. (1965b: 337, italics in original)

One can think of IS explanation as involving a natural generalization of this idea. While an IS explanation does not show that the explanandum-phenomenon was to be expected with certainty, it does the next best thing: it shows that the explanandum-phenomenon is at least to be expected with high probability and in this way provides understanding. Stated more generally, both the DN and IS models, share the common idea that, as Salmon (1989) puts it,

the essence of scientific explanation can be described as nomic expectability —that is expectability on the basis of lawful connections. (1989: 57)

The second main motivation for the DN/IS model has to do with the role of causal claims in scientific explanation. There is considerable disagreement among philosophers about whether all explanations in science and in ordinary life are causal and also disagreement about what the distinction (if any) between causal and non-causal explanations consists in. Nonetheless, virtually everyone, including Hempel, agrees that many scientific explanations cite information about causes. However, Hempel, along with most other early advocates of the DN model, is unwilling to take the notion of causation as primitive in the theory of explanation—that is, he was unwilling to simply say that X figures in an explanation of Y if and only if X causes Y . Instead, adherents of the DN model have generally looked for an account of causation that satisfies the empiricist requirements described in Section 1 . In particular, advocates of the DN model have generally accepted a broadly Humean or regularity theory of causation, according to which (very roughly) all causal claims imply the existence of some corresponding regularity (a “law”) linking cause to effect. This is then taken to show that all causal explanations “imply,” perhaps only “implicitly,” that such a law/regularity exists and hence that laws are “involved” in all such explanations, just as the DN model claims.

To illustrate this line of argument, consider

(3) is a so-called singular causal explanation, advanced by Michael Scriven (1962) as a counterexample to the claim that the DN model describes necessary conditions for successful explanation. According to Scriven, (3) explains the tipping over of the inkwell even though no law or generalization figures explicitly in (3) and (3) appears to consist of a single sentence, rather than a deductive argument. Hempel’s response (1965b: 360ff) is that the occurrence of “caused” in (3) should not be left unanalyzed or taken as explanatory just as it stands. Instead (3) should be understood as “implicitly” or “tacitly” claiming there is a “law” or regularity linking knee impacts to tipping over of inkwells. According to Hempel, it is the implicit claim that some such law holds that “distinguishes” (3) from “a mere sequential narrative” in which the spilling is said to follow the impact but without any claim of causal connection—a narrative that (Hempel thinks) would clearly not be explanatory. This linking law is the nomological premise in the DN argument that, according to Hempel, is “implicitly” asserted by (3) .

The basic idea is thus that a proper explication of the role of causal claims in explanation leads via a Humean or regularity theory of causation, to the conclusion that, at least ideally, explanations should satisfy the DN/IS model. Let us call this line of argument the “hidden structure” argument in recognition of the role it assigns to a hidden (or at least non-explicit) DN structure that is claimed to be associated with (3) .

At this point a comment is in order regarding a feature of this proposal that may seem puzzling. The boundaries of the category “scientific explanation” are far from clear, but while (3) is arguably an explanation, it is not what one usually thinks of as “science”—instead it is a claim from “ordinary life” or “common sense”. This raises the question of why adherents of the DN/IS model don’t simply respond to the alleged counterexample (3) by denying that it is an instance of the category “scientific explanation”—that is, by claiming that the DN/IS model is not an attempt to reconstruct the structure of explanations like (3) but is rather only meant to apply to explanations that are properly regarded as “scientific”. The fact that this response is not often adopted by advocates of the DN model is an indication of the extent to which, as noted in Section 1 , it is implicitly assumed in most discussions of scientific explanation that there are important similarities or continuities in structure between explanations like (3) and explanations that are more obviously scientific and that these similarities that should be captured by some common account that applies to both. Indeed, it is a striking feature not just of Hempel (1965b) but of many other treatments of scientific explanation that much of the discussion in fact focuses on “ordinary life” singular causal explanations similar to (3) , the tacit assumption being that conclusions about the structure of such explanations have fairly direct implications for understanding explanation in science.

As explained above, examples like (3) are potential counterexamples to the claim that the DN model provides necessary conditions for explanation. There are also a number of well-known counterexamples to the claim that the DN model provides sufficient conditions for successful scientific explanation. Here are two illustrations.

Explanatory Asymmetries . There are many cases in which a derivation of an explanandum E from a law L and initial conditions I seems explanatory but a “backward” derivation of I from E and the same law L does not seem explanatory, even though the latter, like the former, appears to meet the criteria for successful DN explanation. For example, one can derive the length s of the shadow cast by a flagpole from the height h of the pole and the angle θ of the sun above the horizon and laws about the rectilinear propagation of light. This derivation meets the DN criteria and seems explanatory. On the other hand, the following derivation from the same laws also meets the DN criteria but does not seem explanatory:

Examples like this suggest that at least some explanations possess directional or asymmetric features to which the DN model is insensitive.

Explanatory Irrelevancies . A derivation can satisfy the DN criteria and yet be a defective explanation because it contains irrelevancies besides those associated with the directional features of explanation. Consider an example due to Wesley Salmon (1971a: 34):

It is arguable that ( L ) meets the criteria for lawfulness imposed by Hempel and many other writers. (If one wants to deny that ( L ) is a law one needs some principled, generally accepted basis for this judgment and, as explained above, it is unclear what this basis is.) Moreover, (5) is certainly a sound deductive argument in which ( L ) occurs as an essential premise. Nonetheless, most people judge that ( L ) and ( K ) are no explanation of ( E ). There are many other similar illustrations. For example (Kyburg 1965), it is presumably a law (or at least an exceptionless, counterfactual supporting generalization) that all samples of table salt that have been hexed by being touched with the wand of a witch dissolve when placed in water. One may use this generalization as a premise in a DN derivation which has as its conclusion that some particular hexed sample of salt has dissolved in water. But again the hexing is irrelevant to the dissolving and such a derivation is no explanation.

One obvious diagnosis of the difficulties posed by examples like (4) and (5) focuses on the role of causation in explanation. According to this analysis, to explain an outcome we must cite its causes and (4) and (5) fail to do this. As Salmon (1989a: 47) puts it,

a flagpole of a certain height causes a shadow of a given length and thereby explains the length of the shadow.

By contrast,

the shadow does not cause the flagpole and consequently cannot explain its height.

Similarly, taking birth control pills does not cause Jones’ failure to get pregnant and this is why (5) fails to be an acceptable explanation. On this analysis, what (4) and (5) show is that a derivation can satisfy the DN criteria and yet fail to identify the causes of an explanandum—when this happens the derivation will fail to be explanatory.

As explained above, advocates of the DN model would not regard this diagnosis as very illuminating, unless accompanied by some account of causation that does not simply take this notion as primitive. (Salmon in fact provides such an account, which we will consider in Section 4 .) We should note, however, that an apparent lesson of (4) and (5) is that the regularity account of causation favored by DN theorists is at best incomplete: the occurrence of c , e , and the existence of some regularity or law linking them (or x ’s having property P and x ’s having property Q and some law linking these) is not a sufficient condition for the truth of the claim that c caused or x ’s having P is causally or explanatorily relevant to x ’s having Q . More generally, if the counterexamples (4) and (5) are accepted, it follows that the DN model fails to state sufficient conditions for explanation. Explaining an outcome isn’t just a matter of showing that it is nomically expectable.

There are two possible reactions one might have to this observation. One is that the idea that explanation is a matter of nomic expectability is correct as far as it goes, but that something more is required as well. According to this assessment, the DN/IS model does state a necessary condition for successful explanation and, moreover, a condition that is a non-redundant part of a set of conditions that are jointly sufficient for explanation. However, some other, independent feature, X (which will account for the directional features of explanation and insure the kind of explanatory relevance that is apparently missing in the birth control example) must be added to the DN model to achieve a successful account of explanation. The idea is thus that Nomic Expectability + X = Explanation. Something like this idea is endorsed, by the unificationist models of explanation developed by Friedman (1974) and Kitcher (1989), which are discussed in Section 5 below.

A second, more radical possible conclusion is that the DN account of the goal or rationale of explanation is mistaken in some much more fundamental way and that the DN model does not even state necessary conditions for successful explanation. As noted above, unless the hidden structure argument is accepted, this conclusion is strongly suggested by examples like (3) (“The impact of my knee caused the tipping over of the inkwell”) which appear to involve explanation without the explicit citing of a law or a deductive structure.

Suggested Readings . The most authoritative and comprehensive statement of the DN and IS models is probably Hempel (1965b). This is reprinted in Hempel 1965a, along with a number of other papers that touch on various aspects of the problem of scientific explanation. In addition to the references cited in this section, Salmon (1989: 46ff.) describes a number of well-known counterexamples to the DN/IS models and discusses their significance.

3. The SR Model

Much of the subsequent literature on explanation has been motivated by attempts to capture the features of causal or explanatory relevance that appear to be left out of examples like (4) and (5) , typically within the empiricist constraints described above. Wesley Salmon’s statistical relevance (or SR ) model (Salmon 1971a) is a very influential attempt to capture these features in terms of the notion of statistical relevance or conditional dependence relationships. Given some class or population \(A\), an attribute \(C\) will be statistically relevant to another attribute \(B\) if and only if \(P(B\pmid A.C) \ne P(B\pmid A)\)—that is, if and only if the probability of \(B\) conditional on \(A\) and \(C\) is different from the probability of \(B\) conditional on \(A\) alone. The intuition underlying the SR model is that statistically relevant properties (or information about statistically relevant relationships) are explanatory and statistically irrelevant properties are not. In other words, the notion of a property making a difference for an explanandum is unpacked in terms of statistical relevance relationships.

As an illustration, suppose that in the birth control pills example (5) the original population T includes both sexes. Then

assuming that not all women in the population take birth control pills. In other words, if you are a male in this population, taking birth control pills is statistically irrelevant to whether you become pregnant, while if you are a female it is relevant. Thus taking birth control pills is explanatorily irrelevant to pregnancy among males but not among females.

To characterize the SR model more precisely we need the notion of a homogenous partition. A homogenous partition of \(A\) is a set of subclasses or cells \(C_i\) of \(A\) that are mutually exclusive and exhaustive, where \(P(B\pmid A.C_i) \ne P(B\pmid A.C_j)\) for all \(C_i \ne C_j\) and where no further statistically relevant partition of any of the cells \(A\), \(C_i\) can be made with respect to \(B\)—that is, there are no additional attributes \(D_k\) in \(A\) such that

On the SR model, an explanation of why some member x of the class characterized by attribute \(A\) has attribute \(B\) consists of the following information:

• The prior probability of \(B\) within \(A\): \(P(B\pmid A) = p\).
• A homogeneous partition of A with respect to B , (\(A.C_1,\ldots ,A.C_n)\), together with the probability of B within each cell of the partition: \(P(B\pmid A.C_i) = p_i\) and
• The cell of the partition to which x belongs.

To employ one of Salmon’s examples, suppose we want to construct an SR explanation of why x who has a strep infection = S , recovers quickly = Q . Let \(T(-T)\) according to whether x is (is not) treated with penicillin, and \(R(-R)\) = according to whether x has a penicillin-resistant strain. Assume for the sake of argument that no other factors are relevant to quick recovery. There are four possible combinations of these properties: \(T.R,\) \(-T.R,\) \(T.{-R},\) \({-T}.{-R},\) but let us assume that

That is, the probability of quick recovery, given that one has strep, is the same for those who have the resistant strain regardless of whether or not they are treated and also the same for those who have not been treated. By contrast, the probability of recovery is different (presumably greater) among those with strep who have been treated and do not have the resistant strain.

In this case

is a homogenous partition of S with respect to Q . The SR explanation of x ’s recovery will consist of a statement of the probability of quick recovery among all those with strep ((i) above), a statement of the probability of recovery in each of the two cells of the above partition ((ii) above), and the cell to which x belongs, which is \(S.T.R\) ((iii) above). Intuitively, the idea is that this information tells us about the relevance of each of the possible combinations of the properties T and R to quick recovery among those with strep and is explanatory for just this reason.

The SR model has a number of distinctive features that have generated substantial discussion. First, note that according to the SR model, and in contrast to the DN/IS model, an explanation is not an argument—either in the sense of a deductively valid argument in which the explanandum follows as a conclusion from the explanans or in the sense of a so-called inductive argument in which the explanandum follows with high probability from the explanans, as in the case of IS explanation. Instead, an explanation is an assembly of information that is statistically relevant to an explanandum. Salmon argues (and takes the birth control example (5) to illustrate) that the criteria that a good argument must satisfy (e.g., criteria that insure deductive soundness or some inductive analogue) are simply different from those a good explanation must satisfy. Among other things, as Salmon puts it, “irrelevancies [are] harmless in arguments but fatal in explanations” (1989: 102). As explained above, in associating successful explanation with the provision of information about statistical relevance relationships, the SR model attempts to accommodate this observation.

A second, closely related point is that the SR model departs from the IS model in abandoning the idea that a statistical explanation of an outcome must provide information from which it follows the outcome occurred with high probability. As the reader may check, the statement of the SR model above imposes no such high probability requirement; instead, even very unlikely outcomes will be explained as long as the criteria for SR explanation are met. Suppose that, in the above example, the probability of quick recovery from strep, given treatment and the presence of a non-resistant strain, is rather low (e.g., 0.2). Nonetheless, if the criteria (i)–(iii) above—a homogeneous partition with correct probability values for each cell in the partition—are satisfied, we may use this information to explain why x , who had a non-resistant strain of strep and was treated, recovered quickly. Indeed, according to the SR model, we may explain why some x which is A is B , even if the conditional probability of B given A and the cell \(C_i\) to which x belongs \((p_i = P(B\pmid A.C_i))\) is less than the prior probability \((p = P(B\pmid A))\) of B in A . For example, if the prior probability of quick recovery among all those with any form of strep is 0.5 and the probability of quick recovery of those with a resistant strain who are untreated is 0.1, we may nonetheless explain why y , who meets these last conditions \(({-T}.R),\) recovered quickly (assuming he did) by citing the cell to which he belongs, the probability of recovery given that he falls in this cell, and the other sort of information described above. More generally, what matters on the SR model is not whether the value of the probability of the explanandum-outcome is high or low (or even high or low in comparison with its prior probability) but rather whether the putative explanans cites all and only statistically relevant factors and whether the probabilities it invokes are correct. One consequence of this, which Salmon endorses while acknowledging that many will regard it as unintuitive, is that on the SR model, the same explanans E may explain both an explanandum M and explananda that are inconsistent with M , such as \(-M\). For example, the same explanans will explain both why a subject with strep and certain other properties (e.g., T and \(-R\)) recovers quickly, if he does, and also why he does not recover if he does not. By contrast, on the DN or IS models, if E explains M, E cannot also explain \(-M\).

This judgment that, contrary to the IS model, the value that a candidate explanans assigns to an explanandum-outcome should not matter for the goodness of the explanation, is motivated as follows: When an outcome is the result of a genuinely indeterministic process we understand both high probability and low probability outcomes (the latter of which of course will sometimes occur) equally well: in both cases, once an SR model has been constructed, there are no additional factors that distinguish the two outcomes.

Stepping back from the issues (such as the status of the high probability requirement) that have dominated discussions of statistical explanation, there are several more general issues that deserve mention. One is that these models have been applied to a range of examples that seem prima-facie to be quite different, including quantum mechanical examples (e.g., radioactive decay) but also more ordinary examples such as recovery from disease and (to use an example of Salmon’s) causes of juvenile delinquency. Radioactive decay is a process that is usually taken to be irreducibly indeterministic and hence is the sort of thing that Salmon’s objective homogeneity requirement is designed to capture. By contrast, although the evidence for many models of juvenile delinquency comes from population level statistics these models do not assume that delinquency is the outcome of an irreducibly indeterministic process (and the models themselves are very far from satisfying an objective homogeneity requirement). Indeed, taken literally, standard causal models assume the opposite: the assumed equations are deterministic with the stochastic element supplied by an “error term”. This raises the question of whether it is sensible to look for a single model that captures all of these examples.

A second, more radical assessment focuses on the question of whether it is appropriate to think of the sorts of statistical theories and hypotheses on which Hempel and Salmon discuss as explaining individual events or outcomes at all. For example, why not instead take quantum mechanics to explain (i) the probabilities with which individual outcomes like decay events occur but not (ii) those individual outcomes themselves? If we adopt (i) the relationship between a quantum mechanical model and such explananda will be deductive and thus subsumable under whatever model of deductive explanation we favor. This raises the question of what additional work is accomplished by models of statistical explanation of either the IS or SR sort.

Putting aside the issues raised in the previous section, the SR model embodies several generic assumptions of ongoing philosophical interest. In particular the model assumes that (i) explanations must cite causal relationships and that (ii) causal relationships are fully captured by statistical relevance (or conditional dependence and independence) relationships. While (i) is a matter of current controversy, (ii) is clearly false. As a substantial body of work [ 5 ] has made clear, causal relationships are greatly underdetermined by statistical relevance relationships, even given additional assumptions. For example, a structure in which B is a common cause of the joint effects A and C implies (assuming the Causal Markov assumption which is the appropriate generalization of Salmon’s assumptions connecting causation and probability) implies the same statistical relevance relations as a chain structure in which A causes B which causes C .

Selected Readings . Salmon (1971a) provides a detailed statement and defense of the SR model. This essay, as well as papers by Jeffrey (1969) and Greeno (1970) which defend views broadly similar to the SR model, are collected in Salmon (1971b). Additional discussion of the model as well as a more recent characterization of “objective homogeneity” can be found in Salmon (1984). Cartwright (1979) contains some influential criticisms of the SR model. Theorems specifying the precise extent of the underdetermination of causal claims by evidence about statistical relevance relationships can be found in Spirtes, Glymour and Scheines (1993 [2000: chapter 4]). For additional discussion of “screening off” and the principle of the common cause, see the entry on Reichenbach’s Principle of the Common Cause .

4. The Causal Mechanical Model

In more recent work (especially, Salmon 1984) Salmon abandoned the attempt to characterize explanation or causal relationships in purely statistical terms. Instead, he developed a new account which he called the Causal Mechanical ( CM ) model of explanation—an account which is similar in both content and spirit to so-called causal process theories of causation of the sort defended by philosophers like Philip Dowe (2000). We may think of the CM model as an attempt to capture the “something more” involved in causal and explanatory relationships over and above facts about statistical relevance, again while attempting to remain within a broadly Humean framework.

The CM model employs several central ideas. A causal process is a physical process, like the movement of a baseball through space, that is characterized by the ability to transmit a mark in a continuous way. (“Continuous” generally, although perhaps not always, means “spatio-temporally continuous”.) Intuitively, a mark is some local modification to the structure of a process—for example, a scuff on the surface of a baseball or a dent an automobile fender. A process is capable of transmitting a mark if, once the mark is introduced at one spatio-temporal location, it will persist to other spatio-temporal locations even in the absence of any further interaction. In this sense the baseball will transmit the scuff mark from one location to another. Similarly, a moving automobile is a causal process because a mark in the form of a dent in a fender will be transmitted by this process from one spatio-temporal location to another. Causal processes contrast with pseudo-processes which lack the ability to transmit marks. An example is the shadow of a moving physical object. The intuitive idea is that, if we try to mark the shadow by modifying its shape at one point (for example, by altering a light source or introducing a second occluding object), this modification will not persist unless we continually intervene to maintain it as the shadow occupies successive spatio-temporal positions. In other words, the modification will not be transmitted by the structure of the shadow itself, as it would in the case of a genuine causal process.

We should note for future reference that, as characterized by Salmon, the ability to transmit a mark is clearly a counterfactual notion, in several senses. To begin with, a process may be a causal process even if it does not in fact transmit any mark, as long as it is true that if it were appropriately marked, it would transmit the mark. Moreover, the notion of marking itself involves a counterfactual contrast—a contrast between how a process behaves when marked and how it would behave if left unmarked. Although Salmon, like Hempel, has always been suspicious of counterfactuals, his view at the time that he first introduced the CM model was that the counterfactuals involved in the characterization of mark transmission were relatively unproblematic, in part because they seemed experimentally testable in a fairly direct way. Nonetheless the reliance of the CM model, as originally formulated, on counterfactuals shows that it does not completely satisfy the Humean strictures described above. In subsequent work, described in Section 4.4 below, Salmon attempted to construct a version of the CM model that completely avoids reliance on counterfactuals.

The other major element in Salmon’s model is the notion of a causal interaction . A causal interaction involves a spatio-temporal intersection between two causal processes which modifies the structure of both—each process comes to have features it would not have had in the absence of the interaction. A collision between two cars that dents both is a paradigmatic causal interaction.

According to the CM model, an explanation of some event E will trace the causal processes and interactions leading up to E (Salmon calls this the etiological aspect of the explanation), or at least some portion of these, as well as describing the processes and interactions that make up the event itself (the constitutive aspect of explanation). In this way, the explanation shows how E “fit[s] into a causal nexus”(1984: 9). For example, when two billiard balls collide (event E ), the trajectory of each of the balls is a causal process (as shown by the fact that if the balls were scratched, such marks would persist) and their collision is a causal interaction. Explaining E will involve both tracing these trajectories and noting that E involves an interaction.

As the billiard example illustrates, the CM model takes as its paradigms of causal interaction examples such as collisions in which there is “action by contact” and no spatio-temporal gaps in the transmission of causal influence. There is little doubt that explanations in which there are no such gaps (no “action at a distance”) often strike us as particularly satisfying. However, as Christopher Hitchcock shows in an illuminating paper (Hitchcock 1995), even here the CM model leaves out something important. Consider the usual elementary textbook “scientific explanation” of the motion of the balls in the above example following their collision. This explanation proceeds by deriving that motion from information about their masses and velocity before the collision, the assumption that the collision is perfectly elastic, and the law of the conservation of linear momentum. We usually think of the information conveyed by this derivation as showing that it is the mass and velocity of the balls, rather than, say, the scratches on their surface or chalk that may be transmitted by the cue stick that is explanatorily relevant to their subsequent motion. However, it is hard to see what in the CM model allows us to pick out the linear momentum of the balls, as opposed to these other features, as explanatorily relevant. Part of the difficulty is that to express such relatively fine-grained judgments of explanatory relevance (that it is linear momentum rather than chalk marks that matters) we need to talk about relationships between properties or magnitudes and it is not clear how to express such judgments purely in terms of facts about causal processes and interactions. Both the linear momentum and the chalk mark communicated to the cue ball by the cue stick are marks transmitted by the spatio-temporally continuous causal process consisting of the motion of the cue ball.

Ironically, as Hitchcock goes on to note, a similar observation may be made about the birth control pills example (5) originally devised by Salmon to illustrate the failure of the DN model to capture the notion of explanatory relevance. Spatio-temporally continuous causal processes that transmit marks as well as causal interactions are at work when male Mr. Jones ingests birth control pills—the pills dissolve, components enter his bloodstream, are metabolized or processed in some way, and so on. Similarly, spatio-temporally continuous causal processes (albeit different processes) are at work when female Ms. Jones takes birth control pills. However, the pills are irrelevant to Mr. Jones’s non-pregnancy, and relevant to Ms. Jones’s non-pregnancy. Again, it looks as though the relevance or irrelevance of the birth control pills to Mr. or Ms. Jones’s failure to become pregnant cannot be captured just by asking whether the processes leading up to these outcomes are causal processes in Salmon’s sense.

A more general way of putting the problem revealed by these examples is that those features of a process P in virtue of which it qualifies as a causal process (ability to transmit mark M ) may not be the features of P that are causally or explanatorily relevant to the outcome E that we want to explain ( M may be irrelevant to E with some other property R of P being the property which is causally relevant to E ). So while mark transmission may well be a criterion that correctly distinguishes between causal processes and pseudo-processes , it does not, as it stands, provide the resources for distinguishing those features or properties of a causal process that are causally or explanatorily relevant to an outcome and those features that are irrelevant.

A second set of worries has to do with the application of the CM model to systems which depart in various respects from simple physical paradigms such as the collision described above. There are a number of examples of such systems. First, there are theories like Newtonian gravitational theory which involve “action at a distance” in a physically interesting sense. Second, there are a number of examples from the literature on causation that do not involve physically interesting forms of action at a distance but which arguably involve causal interactions without intervening spatio-temporally continuous processes or transfer of energy and momentum from cause to effect. These include cases of causation by omission and causation by “double prevention” or “disconnection.” [ 6 ] In all these cases, a literal application of the CM model seems to yield the judgment that no explanation has been provided—that Newtonian gravitational theory is unexplanatory and so on. Many philosophers have been reluctant to accept this assessment.

Yet another class of examples that raise problems for the CM model involves putative explanations of the behavior of complex or “higher level” systems—explanations that do not explicitly cite spatio-temporally continuous causal processes involving transfer of energy and momentum, even though we may think that such processes are at work at a more “underlying” level. Most explanations in disciplines like biology, psychology and economics fall under this description, as do a number of straightforwardly physical explanations.

As an illustration, suppose that a mole of gas is confined to a container of volume \(V_1,\) at pressure \(P_1,\) and temperature \(T_1\). The gas is then allowed to expand isothermally into a larger container of volume \(V_2\). One standard way of explaining the behavior of the gas—its rate of diffusion and its subsequent equilibrium pressure \(P_2\)—appeals to the generalizations of phenomenological thermodynamics—e.g., the ideal gas law, Graham’s law of diffusion, and so on. Salmon appears to regard putative explanations based on at least the first of these generalizations as not explanatory because they do not trace continuous causal processes—he thinks of the individual molecules as causal processes but not the gas as a whole. [ 7 ] However, it is plainly impossible to trace the causal processes and interactions represented by each of the \(6 \times 10^{23}\) molecules making up the gas and the successive interactions (collisions) it undergoes with every other molecule. Even the usual statistical mechanical treatment, which Salmon presumably would regard as explanatory, does not attempt to do this. Instead, it makes certain general assumptions about the distribution of molecular velocities and the forces involved in molecular collisions and then uses these, in conjunction with the laws of mechanics, to derive and solve a differential equation (the Boltzmann transport equation) describing the overall behavior of the gas. This treatment abstracts radically from the details of the causal processes involving particular individual molecules and instead focuses on identifying higher level variables that aggregate over many individual causal processes and that figure in general patterns that govern the behavior of the gas.

This example raises a number of questions. Just what does the CM model require in the case of complex systems in which we cannot trace individual causal processes, at least at a fine-grained level? How exactly does the causal mechanical model avoid the (disastrous) conclusion that any successful explanation of the behavior of the gas must trace the trajectories of individual molecules? Does the statistical mechanical explanation described above successfully trace causal processes and interactions or specify a causal mechanism in the sense demanded by the CM model, and if so, what exactly does tracing causal processes and interactions involve or amount to in connection with such a system? A fully adequate development of the CM model needs to address such questions.

There is another aspect of this example that is worthy of comment. Suppose that a particular sample of gas expands in a way that meets the conditions described above and that it is somehow possible to provide an account that traces each of the individual molecular trajectories of its component molecules. Such an account would nonetheless leave out information that seems explanatorily relevant. This information has to do with the fact that there are a large number of alternative trajectories besides the actual trajectories followed by the component molecules on this particular occasion that would lead to the same final pressure \(P_2\) which is what we want to explain. Modal information of this sort—both about what would happen if the molecules followed different trajectories consistent with the initial conditions \(P_1\), \(V_1\) and \(T_1\) and what would happen if they instead followed initial conditions consistent trajectories consistent with different macroscopic initial conditions is better captured by a treatment in terms of upper level thermodynamic variables.

In more recent work (e.g., Salmon 1994), prompted in part by a desire to avoid certain counterexamples advanced by Philip Kitcher (1989) to his characterization of mark transmission, Salmon attempted to fashion a theory of causal explanation that completely avoids any appeal to counterfactuals. In this new theory which is influenced by the conserved process theory of causation of Dowe (2000), Salmon defined a causal process as a process that transmits a non-zero amount of a conserved quantity at each moment in its history. Conserved quantities are quantities so characterized in physics—linear momentum, angular momentum, charge, and so on. A causal interaction is an intersection of world lines associated with causal processes involving exchange of a conserved quantity. Finally, a process transmits a conserved quantity from A to B if it possesses that quantity at every stage without any interactions that involve an exchange of that quantity in the half-open interval \((A,B]\). [ 8 ]

One may doubt that this new theory really avoids reliance on counterfactuals, but an even more fundamental difficulty is that it still does not adequately deal with the problem of causal or explanatory relevance described above. That is, we still face the problem that the feature that makes a process causal (transmission of some conserved quantity or other) may tell us little about which features of the process are causally or explanatorily relevant to the outcome we want to explain. For example, a moving billiard ball will transmit many conserved quantities (linear momentum, angular momentum, charge etc.) and many of these may be exchanged during a collision with another ball. What is it that entitles us to single out the linear momentum of the balls, rather than these other conserved quantities as the property that is causally relevant to their subsequent motion? In cases in which there appear to be no conservation laws governing the explanatorily relevant property (i.e., cases in which the explanatorily relevant variables are not conserved quantities) this difficulty seems even more acute. Properties like “having ingested birth control pills,” “being pregnant”, or “being a sample of hexed salt” do not themselves figure in conservation laws. While one may say that both birth control pills and hexed salt are causal processes because both consist, at some underlying level, of processes that unambiguously involve the transmission of conserved quantities like mass and charge, this observation does not by itself tell us what, if anything, about these underlying processes is relevant to pregnancy or dissolution in water.

In a more recent paper (Salmon 1997), Salmon conceded this point. He agreed that the notion of a causal process cannot by itself capture the notion of causal and explanatory relevance. He suggested, however, that this notion can be adequately captured by appealing to the notion of a causal process and information about statistical relevance relationships (that is, information about conditional and unconditional (in)dependence relationships), with the latter capturing the element of causal or explanatory dependence that was missing from his previous account:

I would now say that (1) statistical relevance relations, in the absence of information about connecting causal processes, lack explanatory import and that (2) connecting causal processes, in the absence of statistical relevance relations, also lack explanatory import. (1997: 476)

This suggestion is not developed in any detail in Salmon’s paper, and it is not easy to see how it can be made to work. We noted above that statistical relevance relationships often greatly underdetermine the causal relationships among a set of variables. What reason is there to suppose that appealing to the notion of a causal process, in Salmon’s sense, will always or even usually remove this indeterminacy? We also noted that the notion of a causal process cannot capture fine-grained notions of relevance between properties, that there can be causal relevance between properties instances of which (at least at the level of description at which they are characterized) are not linked by spatio-temporally continuous or transference of conserved quantities, and that properties can be so linked without being causally relevant (recall the chalk mark that is transmitted from one billiard ball to another). As long as it is possible (and why should it not be?) for different causal claims to imply the same facts about statistical relevance relationships and for these claims to differ in ways that cannot be fully cashed out in terms of Salmon’s notions of causal processes and interactions, this new proposal will fail as well.

Selected Readings: Salmon (1984) provides a detailed statement of the Causal Mechanical model, as originally formulated. Salmon (1994 and 1997) provide a restatement of the model and respond to criticisms. For discussion and criticism of the CM model, see Kitcher (1989, especially pages 461ff), Woodward (1989), and Hitchcock (1995).

5. A Unificationist Account of Explanation

In unificationist accounts of explanation developed by philosophers, scientific explanation is a matter of providing a unified account of a range of different phenomena. [ 9 ] This idea is unquestionably intuitively appealing. Successful unification may exhibit connections or relationships between phenomena previously thought to be unrelated and this seems to be something that we expect good explanations to do. Moreover, theory unification has clearly played an important role in science. Paradigmatic examples include Newton’s unification of terrestrial and celestial theories of motion and Maxwell’s unification of electricity and magnetism. The key question, however, is whether (and which) intuitive notions of unification can be made more precise in a way that allows us to recover the features that we think that good explanations should possess.

Michael Friedman (1974) is an important early attempt to do this. Friedman’s formulation of the unificationist idea was subsequently shown to suffer from various technical problems (Kitcher 1976) and subsequent development of the unificationist treatment of explanation has been most closely associated with Philip Kitcher (especially Kitcher 1989).

Let us begin by introducing some of Kitcher’s technical vocabulary. A schematic sentence is a sentence in which some of the nonlogical vocabulary has been replaced by dummy letters. To use Kitcher’s examples, the sentence “Organisms homozygous for the sickling allele develop sickle cell anemia” is associated with a number of schematic sentences including “Organisms homozygous for A develop P ” and “For all X if X is O and A then X is P ”. Filling instructions are directions that specify how to fill in the dummy letters in schematic sentences. For example, filling instructions might tell us to replace A with the name of an allele and P with the name of a phenotypic trait in the first of the above schematic sentences. Schematic arguments are sequences of schematic sentences. Classifications describe which sentences in schematic arguments are premises and conclusions and what rules of inference are used. An argument pattern is an ordered triple consisting of a schematic argument, a set or sets of filling instructions, one for each term of the schematic argument, and a classification of the schematic argument. The more restrictions an argument pattern imposes on the arguments that instantiate it, the more stringent it is said to be.

Roughly speaking, Kitcher’s guiding idea is that explanation is a matter of deriving descriptions of many different phenomena by using as few and as stringent argument patterns as possible over and over again-the fewer the patterns used, the more stringent they are, and the greater the range of different conclusions derived, the more unified our explanations. Kitcher summarizes this view as follows:

Science advances our understanding of nature by showing us how to derive descriptions of many phenomena, using the same pattern of derivation again and again, and in demonstrating this, it teaches us how to reduce the number of facts we have to accept as ultimate. (Kitcher 1989: 432)

Kitcher does not propose a completely general theory of how the various considerations he describes—number of conclusions, number of patterns and stringency of patterns—are to be traded-off against one another, but does suggest that it often will be clear enough what these considerations imply about the evaluation of particular candidate explanations. His basic strategy is to attempt to show that the derivations we regard as good or acceptable explanations are instances of patterns that, taken together, score better according to the criteria just described than the patterns instantiated by what we regard as defective explanations. Following Kitcher, let us define the explanatory store \(E(K)\) as the set of argument patterns that maximally unifies K , the set of beliefs accepted at a particular time in science. Showing that a particular derivation is a good or acceptable explanation is then a matter of showing that it belongs to the explanatory store.

As an illustration, consider Kitcher’s treatment of the problem of explanatory asymmetries (recall Section 2.5 ). Our present explanatory practices—call these P —are committed to the idea that derivations of a flagpole’s height from the length of its shadow are not explanatory. Kitcher compares P with an alternative systemization in which such derivations are regarded as explanatory. According to Kitcher, P includes the use of a single “origin and development” ( OD ) pattern of explanation, according to which the dimensions of objects-artifacts, mountains, stars, organisms etc. are traced to “the conditions under which the object originated and the modifications it has subsequently undergone” (1989: 485). Now consider the consequences of adding to P an additional pattern S (the shadow pattern) which permits the derivation of the dimensions of objects from facts about their shadows. Since the OD pattern already permits the derivation of all facts about the dimensions of objects, the addition of the shadow pattern S to P will increase the number of argument patterns in P and will not allow us to derive any new conclusions. On the other hand, if we were to drop OD from P and replace it with the shadow pattern, we would have no net change in the number of patterns in P , but would be able to derive far fewer conclusions than we would with OD , since many objects do not have shadows (or enough shadows) from which to derive all of their dimensions. Thus OD belongs to the explanatory store, and the shadow pattern does not. Kitcher’s treatment of other familiar problem cases is similar. For example he claims that explanations that contain irrelevancies (such as Salmon’s birth control pills case are less unifying than competing explanations that do not contain such irrelevancies.

Kitcher acknowledges that there is nothing in the unificationist account per se that requires that all explanation be deductive: “there is no bar in principle to the use of non-deductive arguments in the systemization of our beliefs”. Nonetheless, “the task of comparing the unifying power of different systemizations looks even more formidable if non-deductive arguments are considered” and in part for this reason Kitcher endorses the view that “in a certain sense, all explanation is deductive” (1989: 448).

What is the role of causation on this account? Kitcher claims that “the ‘because’ of causation is always derivative from the ‘because’ of explanation.” (1989: 477). That is, our causal judgments simply reflect the explanatory relationships that fall out of our (or our intellectual ancestors’ ) attempts to construct unified theories of nature. There is no independent causal order over and above this which our explanations must capture. Like many other philosophers, Kitcher takes very seriously (even if in the end he perhaps does not fully endorse) standard empiricist or Humean worries about the epistemic accessibility and intelligibility of causal claims. Taking causal, counterfactual or other notions belonging to the same family as primitive in the theory of explanation is problematic. Kitcher believes that it is a virtue of his theory that it does not do this. Instead, Kitcher proposes to begin with the notion of explanatory unification, characterized in terms of constraints on deductive systemizations, where these constraints can be specified in a quite general way that is independent of causal or counterfactual notions, and then show how the causal claims we accept derive from our efforts at unification.

As remarked at the beginning of this section, the idea that explanation is connected in some way to unification is intuitively appealing. Nonetheless Kitcher’s particular way of cashing out this connection has been subject to criticism. In connection with Kitcher’s treatment of explanatory asymmetries, consider, following Barnes (1992), a time-symmetric theory like Newtonian mechanics, applied to a closed system like the solar system. Call derivations of the state of motion of planets at some future time t from information about their present positions (at time \(t_0\)), masses, and velocities, the forces incident on them at \(t_0\), and the laws of mechanics predictive . Now contrast such derivations with retrodictive derivations in which the present motions of the planets are derived from information about their future velocities and positions at t , the forces operative at t , and so on. It looks as though there will be just as many retrodictive derivations as predictive derivations, and each will require premises of exactly the same general sort—information about positions, velocities, masses, etc. and the same laws. Thus the pattern or patterns instantiated by the retrodictive derivations look(s) exactly as unified as the pattern or patterns associated with the predictive derivations. However, we ordinarily think of the predictive derivations and not the retrodictive derivations as explanatory and the present state of the planets as the cause of their future state and not vice-versa.

One possible response to this second example is to bite the bullet and to argue that from the point of view of fundamental physics, there really is no difference in the explanatory import of the retrodictive and predictive derivations, and that it is a virtue, not a defect, of the unificationist approach that it reproduces this judgment. Whatever might be said in favor of this response, it is not Kitcher’s. His claim is that our ordinary judgments about causal asymmetries can be derived from the unificationist account. The example just described casts doubt on this claim. More generally, it casts doubt on Kitcher’s contention that one can begin with the notion of explanatory unification, understood in a way that does not presuppose causal notions, and use it to derive the content of causal judgments.

Selected Readings : The most detailed statement of Kitcher’s position can be found in Kitcher (1989). Salmon (1989: 94ff.) contains a critical discussion of Friedman’s version of the unificationist account of explanation but ends by advocating a “rapprochement” between unificationist approaches and Salmon’s own causal mechanical model. Woodward (2003) contains additional criticisms of Kitcher’s version of unificationism.

6. Pragmatic Theories of Explanation

Despite their many differences, the accounts of Hempel (focusing now on just the DN rather than the IS model), Salmon, Kitcher, and others discussed above, largely share a common overall conception of what the project of constructing a theory of explanation should involve and (to a considerable extent) what criteria such a theory should satisfy if it is to be successful. “Pragmatic” theories of explanation depart from this consensus in important respects. Let us say that a theory of explanation contains “pragmatic” elements if (i) according to the theory, those elements require irreducible reference to facts about the interests, beliefs or other features of the psychology of those providing or receiving the explanation and/or (ii) irreducible reference to the “context” in which the explanation occurs. (For what this means, see below.) Although the writers discussed above agree that pragmatic elements play some role in the activity of giving and receiving explanations, they assume that there is a non-pragmatic core to the notion of explanation, which it is the central task of a theory of explanation to capture. That is, it is assumed that this core notion can be specified without reference to psychological features of explainers or their audiences and with reference to non-contextual features that are sufficiently general, abstract and “structural,” in the sense that they hold across a range of explanations with different contents and across a range of different contexts. Relations like deductive entailment or statistical relevance are examples of candidates for such “structural” relationships. In addition, these writers see the goal of a theory of explanation, as capturing the notion of a correct explanation, as opposed to the notion of an explanation’s being considered explanatory by a particular audience or not, a matter which presumably depends on such considerations as whether the audience understands the terms in which the explanation is framed. (In this sense, “correctness” requires (at least) that the explanans be true or “well-confirmed” and that the explanans stands in the right relationship to the explanandum.) Finally, as noted in the Introduction to this entry, writers in this tradition have not had the goal of capturing all the various ways in which the word “explanation” is used in ordinary English. They have instead focused on a much more restricted class of examples in which what is of interest is (something like) explaining “why” some outcome or general phenomenon occurred, as opposed to explaining, e.g., the meaning of a word or how to solve a differential equation. The motivation for this restriction is simply the judgment that an interesting and non-trivial theory is more likely to emerge if it is restricted in scope in this way. For ease of reference, let us call this the “traditional” conception of the task of a theory of explanation.

Some or all of these assumptions and goals are rejected in pragmatic accounts of explanation. Early contributors to this approach include Michael Scriven (e.g., 1962) and Sylvan Bromberger (e.g., 1966), with more systematic statements, due to van Fraassen (1980) and Achinstein (1983) appearing in the 1980s. Since it is not always clear just what the points of disagreement are between pragmatic and traditional accounts, some orienting remarks about this will be useful before turning to details. Defenders of pragmatic approaches to explanation typically stress the point that whether provision of a certain body of information to some audience produces understanding or is illuminating for that audience depends on the background knowledge of the audience members and on other factors having to do with the local context. For example, an explanation of the deflection of starlight by the sun that appeals to the field equations of General Relativity may be highly illuminating to a trained physicist but unintelligible to a layperson because of their background. Factors of this sort are grouped together as “pragmatic” and their influence is taken to illustrate at least one way in which pragmatic considerations enter into the notion of explanation.

Taken in itself, the observation just described seems completely uncontroversial and not in conflict with traditional approaches to explanation. Indeed, as remarked above writers like Hempel and Salmon explicitly agree that explanation has a pragmatic dimension in the sense just described—in fact, Hempel invokes the role of pragmatic factors at a number of points to address prima-facie counterexamples to the DN model. This suggests that, often at least, what is distinctive about pragmatic approaches to explanation is not just the bare idea that explanation has a “pragmatic dimension” but rather the much stronger claim that the traditional project of constructing a model of explanation pursued by Hempel and others has so far been unsuccessful ( and perhaps is bound to be unsuccessful) and that this is so because pragmatic or contextual factors play a central and ineliminable role in explanation in a way that resists incorporation into models of the traditional sort. On this view, much of what is distinctive about pragmatic accounts is their opposition to traditional accounts and their diagnosis of why such accounts fail—they fail because they omit pragmatic or contextual elements. It will be important to keep this point in mind in what follows because there is a certain tendency among advocates of pragmatic theories to argue as though the superiority of their approach is established simply by the observation that explanation has a pragmatic dimension; instead it seems more appropriate to think that the real issue is whether traditional approaches are inadequate in principle because of their neglect of the pragmatic dimension of explanation.

A second issue concerns an important ambiguity in the notion of “pragmatic”. On one natural understanding of this notion, a pragmatic consideration is one that has to do with utility or usefulness in the service of some goal connected to human interests, where these interests are in some relevant sense “practical”. Call this notion “pragmatic 1 ”. On this construal, Hempel’s DN model might be correctly characterized as a pragmatic 1 theory since it links explanatory information closely to the provision of information that is useful for purposes of prediction and prediction certainly qualifies as a pragmatic goal. For similar reasons, Woodward’s (2003) theory of explanation might also be counted as a pragmatic 1 theory since it connects explanation with the provision of information that is useful for manipulation and control—unquestionably useful goals. As these examples suggest, models of explanation that aspire to traditional goals can be pragmatic 1 theories.

In the context of theories of explanation, however, the label “pragmatic” is usually intended to suggest a somewhat different set of associations. In particular, as noted above, “pragmatic” is typically used to characterize considerations having to do with facts about the psychology (interests, beliefs etc.) of those involved in providing or receiving explanations and/or to characterize considerations involving the local context, often with the suggestion that both sets of considerations may vary in complex and idiosyncratic ways that resist incorporation into the sort of general theory sought by traditional models. [ 10 ] Call this set of associations “pragmatic 2 ”. Neither Hempel’s nor Woodward’s theory is pragmatic 2 . In particular, as the example of the DN model illustrates, the fact that a theory is pragmatic 1 in the sense that it appeals to facts about goals generally shared by human beings (such as prediction) to help to motivate a model of explanation does not preclude attempting to construct models of explanation satisfying traditional goals and does not require commitment to the idea that explanation must be understood as a pragmatic 2 notion. We need to be careful to distinguish these two different ways of thinking about the “pragmatic” dimension of explanation.

Finally, as emphasized above, a concern with the pragmatics of explanation naturally connects with an interest in the “psychology” of explanation and this in turn suggests the relevance of empirical studies of sorts of information that various subjects (ordinary folks, scientists) find explanatory, treat as providing “understanding”, the distinctions subjects make among explanations and so on. Although there is a growing literature in this area, the most prominent philosophical advocates of pragmatic approaches to explanation have so far tended not to make use of it. In this connection, it is worth pointing out that this psychological literature goes well beyond the truisms found in philosophical discussion about different people finding different sorts of information explanatory depending on their interests. In particular, psychologists have been very interested in exploring general features or structural patterns present in information that various subjects find explanatory. For example, Lombrozo (2010) finds evidence that subjects prefer explanations that appeal to relationships that are relatively stable (in the sense of continuing to hold across changing circumstances) [ 11 ] and Lien and Cheng (2000) present evidence that in cases in which the explanandum \(E\) has a single candidate cause \(C\), subjects prefer levels of explanation/causal description that maximize \(\Delta p = \Pr(E\pmid C) - \Pr(E \pmid \text{not-}C)\).

Notice that in both cases these are relationships or patterns of the sort that traditional accounts of explanation attempt to capture. As these examples bring out, there is no necessary incompatibility between the project of trying to formulate an account of explanation that satisfies traditional goals and an interest in the psychology of explanation. It may be that subjects find certain sorts of information explanatory or understanding-producing because certain structural features of the sort that traditional accounts attempt to characterize are present in that information—indeed this is what the Lombrozo and Lien and Cheng papers suggest. Thus, we should distinguish the project of investigating the empirical psychology of explanation (which can be pursued with a variety of different commitments about how to best theorize about explanation) from the more specific claim that the characterization of what it is for an explanatory relationship to hold between explanans and explanandum must be given in “psychologistic” terms, in the sense that this requires irreducible reference to psychological facts about particular audiences such as the vagaries of what they happen to be interested in.

One of the most influential recent pragmatic accounts of explanation is associated with constructive empiricism. [ 12 ] This is the thesis, defended by Bas van Fraassen in his 1980 book, The Scientific Image , that the aim of science (or at least “pure” science) is the construction of theories that are “empirically adequate” (that is, that yield a true or correct description of observables) and not, as scientific realists suppose, theories that aim to tell literally true stories about unobservables. Relatedly, “acceptance” of a theory involves only the belief that it is empirically adequate (van Fraassen 1980: 12). van Fraassen’s account of explanation, which is laid out in several articles and, most fully, in Chapter Six of his book, is meant to fit with this overall conception of science: it is a conception according to which explanation per se is not an epistemic aim of “pure” science (empirical adequacy is the only such aim), but rather a “pragmatic” virtue, having to do with the “application” of science. (Note that the application of science is arguably a matter of pragmatics 1 . However, the idea that explanation has to do with the application of science is used to motivate the adoption of a pragmatic 2 theory of explanation. We thus have an elision of the two notions of “pragmatic” distinguished above.) According to van Fraassen, because explanation is a merely pragmatic virtue, a concern with explanation is not something that can require scientists to move beyond belief in the empirical adequacy of their theories to belief in the literal truth of claims about unobservable entities.

According to van Fraassen, explanations are answers to questions and getting clear about the logic of questions is central to constructing a theory of explanation. Questions can take many different forms, but when the question of interest is a “why” question, explanatory queries will typically take the following form: a query about why some explanandum \(P_k\) rather than any one of the members of a contrast class \(X\) (a set of possible alternatives to \(P_k\)) obtained. In addition, some “relevance relation” \(R\) is assumed by the question. An answer \(A\) to this question will take the form “\(P_k\) in contrast to (the rest of) \(X\) because \(A\), where \(A\) bears the relevance relation \(R\) to \([P_k,X]\)”. To use van Fraassen’s example, consider “Why is this conductor warped?” Depending on the context, the intended contrast might have to with, e.g., why this particular conductor is warped in contrast to some other conductor that is unwarped. Alternatively, it might have to do with why this particular conductor is warped now when it was previously unwarped. The relevance relation \(R\) similarly depends on the context and the information which the questioner is interested in obtaining. For example, \(R\) might involve causal information (the question might be a request for what caused the warping) but it also might have to do with information about function, if the context was one in which it is assumed that the shape of the conductor plays some functional role in a power station which the questioner wants to know about. Thus “context” enters into the explanation both by playing a role in specifying the contrast class \(X\) and the relevance relation \(R\). van Fraassen describes various rules for the “evaluation” of answers. For example, \(P_k\) and \(A\) must be true, the other members of the contrast class must not be true, \(A\) must “favor” (raise the conditional probability of) \(P_k\) against alternatives, and \(A\) must compare favorably with other answers to the same question, a condition which itself has several aspects including, for example, whether \(A\) favors the topic more than these other answers and whether \(A\) is screened off by other answers. However, he also makes it clear (as the example above suggests) that a variety of different relevance relations may be appropriate depending on context and that the evaluation of answers also depends on context. Moreover, he explicitly denies that there is anything distinctive about the category of scientific explanation that has to do with its structure or form—instead, a scientific explanation is simply an explanation that makes use of information that is (or at least, is treated as) grounded in a “scientific” theory.

Van Fraassen sums up his view of explanation (and gestures at his grounds for rejecting traditional objectivist approaches) as follows:

The discussion of explanation went wrong at the very beginning when explanation was conceived of as a relation like description: a relation between a theory and a fact. Really, it is a three-term relation between theory, fact, and context. No wonder that no single relation between theory and fact ever managed to fit more than a few examples! Being an explanation is essentially relative for an explanation is an answer … it is evaluated vis-à-vis a question, which is a request for information. But exactly… what is requested differs from context to context. (1980: 156)

Van Fraassen begins his chapter on explanation with a brief story that provides a good point of entry into how he intends his account to work. Recall from Section 2.5 that a well-known counterexample to the DN model involves the claim that one can explain the length S of the shadow cast by a flagpole in terms of the height H of the flagpole but that (supposedly) one cannot explain H in terms of S , despite the fact that one can construct a DN derivation from S to H . This is commonly taken to show that the DN model has left out some factor having to do with the directional or asymmetric features of explanation—e.g., perhaps an asymmetry in the relation between cause and effect that ought to be incorporated into one’s model of explanation. In van Fraassen’s story, a straightforward causal explanation of the usual sort of S in terms of H (although the object in question is a tower rather than a flagpole) is first offered. Then a second explanation, according to which the height of the tower is “explained” by the fact that it was designed to cast a shadow of a certain length is advanced. Presumably the moral we are to draw is that as the context and perhaps the relevance relation R are varied, both

are acceptable (legitimate, appropriate etc.) explanations. Moreover, since these variations in context and relevance relation turn on variations in what is of interest to the explainer and his audience, we are further encouraged to conclude that explanatory asymmetries have their source in psychological facts about people’s interests and background beliefs, rather than in, say, some asymmetry that exits in nature independently of these. Pragmatists about explanation think that a similar conclusion holds for other features of the explanatory relevance relation that philosophers have tried to characterize in terms of traditional models of explanation.

One obvious response to this claim, made by several critics (e.g., Kitcher & Salmon 1987: 317), is that the example does not really involve a case in which, depending on context, H causally explains S and S causally explains H . Instead, although H does causally explain S , it is (something like) the desire for a shadow of length S (rather than S itself) that explains (or at least causally explains) the height (or the choice of height) for the tower. Or, if one prefers, in the latter case we are given something like a functional explanation (but not a causal explanation) for the height of the tower, in the sense that we are told what the intended function of that choice of height is. On either of these diagnoses, this will not be a case in which whether H provides a causal explanation of S or whether instead S provides a causal explanation of H shifts depending on factors having to do with the interests of the speaker or audience or other contextual factors. If so, the story about the tower does not show that the asymmetry present in the flagpole example must be accounted for in terms of pragmatic factors. It may be accounted for in some other way. In fact, although discussion must be beyond the scope of this entry, a number of possible candidates for such a non-pragmatic account of causal asymmetries have been proposed, both in philosophy and outside of it (for example, in the machine learning literature [ 13 ] ).

A much more general criticism has been advanced against van Fraassen’s version of a pragmatic theory by Kitcher and Salmon (1987). Basically, their complaint is that the relevance relation R in van Fraassen’s account is completely unconstrained, with the (in their view, obviously unacceptable) consequence that for any pair of true propositions P and A , answer A is relevant to P via some relevance relation and thus “explains” P . For example, according to Salmon and Kitcher, we might define a relationship of “astral influence” \(R^*\), meeting van Fraassen’s criteria for being a relevance relation, such that the time t of a person’s death is explained in terms of \(R^*\) and the position of various heavenly bodies at t . Here it may seem that van Fraassen has a ready response. As noted above, on van Fraassen’s view, background knowledge and, in the case of scientific explanation, current scientific knowledge, helps to determine which are the acceptable relevance relations and acceptable answers to the questions posed in requests for explanation—such knowledge and the expectations that go along with it are part of the relevant context when one asks for an explanation of time of death. Obviously, astral influence is not an acceptable or legitimate relevance relation according to modern science—hence appeal to such a relation is not countenanced as explanatory by van Fraassen’s theory. More generally it might be argued that available scientific knowledge will provide constraints on the relevance relations and answers that exclude the “anything goes” worry raised by Salmon and Kitcher—at least insofar as the context is one in which a “scientific explanation” is sought.

While this response may seem plausible enough as far as it goes, it does bring out the extent to which much of the work of distinguishing the explanatory from the non-explanatory in van Fraassen’s account comes from a very general appeal to what is accepted as legitimate background information in current science. Put differently, this raises the worry that once one moves beyond van Fraassen’s formal machinery concerning questions and answers (which van Fraassen himself acknowledges is relatively unconstraining), one is left with an account according to which a scientific explanation is simply any explanation employing claims from current science and a currently scientifically approved relevance relation. Even if otherwise unexceptionable, this proposal is, if not exactly trivial, at least rather deflationary—it provides much less than many have hoped for from a theory of explanation. In particular, in cases (of which there are many examples) in which there is an ongoing argument or dispute in some area of science not about whether some proposed theory or model is true but rather about whether it explains some phenomenon, it is not easy to see how the proposal even purports to provide guidance. On the other hand, the obvious rejoinder that might be made on van Fraassen’s behalf is that no more ambitious treatment that would satisfy the expectations associated with more traditional accounts of explanation (including a demarcation of candidate explanations into those that are “correct” and “incorrect”) is possible—a theory like van Fraassen’s is as good as it gets. If there is no defensible theory of explanation embodying a non-trivially constraining relevance relation, it cannot be a good criticism of van Fraassen’s theory that he fails to provide this.

A final point that is suggested by van Fraassen’s theory is this: In considering pragmatic theories, it matters a great deal exactly where the “pragmatic” elements are claimed to enter into the account of explanation. One point at which such considerations seem clearly to enter is in the selection or characterization of what an audience wants explained. This is reflected in van Fraassen’s theory in the choice of a \(P_k\) and an associated contrast class \(X\). Obviously, whether we are looking for an explanation of why, say, this particular conductor is now bent when it was previously straight or whether instead we want to know why this conductor is bent while some other conductor is straight is a matter that depends on our interests. However, this particular sort of “interest relativity” (and associated phenomena having to do with the role of contrastive focus in the characterization of explananda, which really just serve to specify more exactly which particular explananda we want explained) seems something that can be readily acknowledged by traditional theories. [ 14 ] After all, it is not a threat to the DN or other models with similar traditional aspirations that one audience may be interested in an explanation of the photoelectric effect but not the deflection of starlight by the sun and another audience may have the opposite interests. What would be a threat to the DN and similar models would be an argument that once some explanandum E is fully specified, whether explanans M explains E (that is, whether there is an explanatory relation between M and E ) is itself “interest-relative”. It is natural to interpret van Fraassen as making this latter claim, both in connection with explanatory asymmetries and more generally.

Selected Readings . van Fraassen (1980, especially Chapter Six) and Achinstein (1983) are classic statements of pragmatic approaches to explanation. These pragmatic accounts are discussed and criticized in Salmon (1989). van Fraassen’s account is also discussed in Kitcher and Salmon (1987). De Regt and Dieks (2005) is a recent defense of what the authors describe as a “contextual” account of scientific understanding and which engages with some of the themes in the “pragmatics of explanation” literature.

7. Conclusions, Open Issues, and Future Directions

What can we conclude from this recounting of some of the more prominent recent attempts to construct models of scientific explanation? What important issues remain open and what are the most promising directions for future work? Of course, any effort at stock-taking will reflect a particular point of view, but with this caveat in mind, several observations seem plausible, even if not completely uncontroversial.

One issue concerns the role of causal information in scientific explanation. All of the traditional models considered above attempt to capture causal explanations, although some attempt to capture non-causal explanations as well. It is a natural thought (endorsed by many) that many of the difficulties faced by the models described above derive at least in part from their reliance on inadequate treatments of causation. [ 15 ] The problems of explanatory asymmetries and explanatory irrelevance described in Section 2.5 seem to show that the holding of a law between C and E is not sufficient for C to cause E ; hence not a sufficient condition for C to figure in an explanation of E . If the argument of section 3.3 is correct, a fundamental problem with the SR model is that statistical relevance information is insufficient to fully capture causal information in the sense that different causal structures can be consistent with the same information about statistical relevance relationships. Similarly, the CM model faces the difficulty that information about causal processes and interactions is also insufficient to fully capture causal relevance relations and that there is a range of cases in which causal relationships hold between C and E (and hence in which C figures in an explanation of E ) although there is no connecting causal process between C and E . Finally, a fundamental problem with unificationist models is that the content of our causal judgments does not seem to fall out of our efforts at unification, at least when unification is understood along the lines advocated by Kitcher. For example, as discussed above, considerations having to do with unification do not by themselves explain why it is appropriate to explain effects in terms of their causes rather than vice-versa.

These observations suggest that insofar as we are interested in causal forms of scientific explanation progress may require more attention to the notion of causation and a more thorough-going integration of discussions of explanation with the burgeoning literature on causation, both within and outside of philosophy. [ 16 ] A number of steps in this direction have been taken. (cf. Woodward 2003).

Does this mean that a focus on causation should entirely replace the project of developing models of explanation or that philosophers should stop talking about explanation and instead talk just about causation? Despite the apparent centrality of causation to many explanations, it is arguable that completely subsuming the latter into the former loses connections with some important issues. For one thing, causal claims themselves seem to vary greatly in the extent to which they are explanatorily deep or illuminating. Causal claims found in Newtonian mechanics seem deeper or more satisfying from the point of view of explanation than causal claims of “the rock broke the window” variety. It is usually supposed that such differences are connected to other features—for example to how general, stable, coherent with background knowledge a causal claim is. However, notions like “generality” are vague and not all forms of generality seem to be connected to explanatory goodness. So even if one focuses only on causal explanation, there remains the important project of trying to understand better what sorts of distinctions among causal claims matter for goodness in explanation. To the extent this is so, the kinds of concerns that have animated traditional treatments of explanation don’t seem to be entirely subsumable into standard accounts of causation, which have tended to focus largely on the project of distinguishing causal from non-causal relationships rather than on the features that make causal relationships “good” for purposes of explanation.

Another important question has to do with whether there are forms of why-explanation that are non-causal. If so, how important are these are in science and what is their structure? Hempel seems to have thought of causal explanations simply as those DN explanations that appeal to causal laws which he regarded as a proper subset of all laws. Thus on his view, causal and non-causal explanations share a common structure. Kitcher’s unificationist model was also intended to apply to both causal and non-causal explanations such as unifying argument patterns in linguistics. More recently, there has been a great upsurge of interest in whether there are non-causal forms of explanation, with some claiming they are ubiquitous in science (e.g., Lange 2017, Reutlinger & Saatsi 2018). If there are such explanations, this raises the issue of what distinguishes them from causal explanations and whether there is some overarching theory that subsumes both causal and non-causal explanations.

As noted above, one way in which the attempt to develop a single general model of explanation might fail is that we might conclude that there are causal and non-causal forms of explanation that have little in common. But even putting this possibility aside, another possibility is that explanation differs across different areas of science in a way that precludes the development of a single, general model. It is, after all, uncontroversial that explanatory practice—what is accepted as an explanation, how explanatory goals interact with others, what sort of explanatory information is thought to be achievable, discoverable, testable etc.—varies in significant ways across different disciplines. Nonetheless, all of the models of explanation surveyed above are “universalist” in aspiration—they claim that a single, “one size” model of explanation fits all areas of inquiry in so far as these have a legitimate claim to explain. Although the extreme position that explanation in biology or history has nothing interesting in common with explanation in physics seems unappealing (and in any case has attracted little support), it seems reasonable to expect that more effort will be devoted in the future to developing models of explanation that are more sensitive to disciplinary differences. Ideally, such models would reveal commonalities across disciplines but they should also enable us to see why explanatory practice varies as it does across different disciplines and the significance of such variation. For example, as noted above, biologists, in contrast to physicists, often describe their explanatory goals as the discovery of mechanisms rather than the discovery of laws. Although it is conceivable that this difference is purely terminological, it is also worth exploring the possibility that there is a distinctive story to be told about what a mechanism is, as this notion is understood by biologists, and how information about mechanisms contributes to explanation.

A closely related point is that at least some of the models described above impose requirements on explanation that may be satisfiable in some domains of inquiry but are either unachievable (in any practically interesting sense) in other domains or, to the extent that they may be achievable, bear no discernible relationship to generally accepted goals of inquiry in those domains. For example, we noted above that many scientists and philosophers hold that there are few if any laws to be discovered in biology and the social and behavioral sciences. If so, models of explanation that assign a central role to laws may not be very illuminating regarding how explanation works in these disciplines. As another example, even if we suppose that the partition into objectively homogeneous reference classes recommended by the SR model is an achievable goal in connection with certain quantum mechanical phenomena, it may be that (as suggested above) it is simply not a goal that can be achieved in a non-trivial way in economics and sociology, disciplines in which causal inference from statistics also figures prominently. In such disciplines, it may be that additional statistically relevant partitions of any population or subpopulation of interest will virtually always be possible, so that the activity of finding such partitions is limited only by the costs of gathering additional information. A similar assessment may hold for most applications of the CM model to the social sciences.

• Achinstein, Peter, 1983, The Nature of Explanation , New York: Oxford University Press.
• Barnes, Eric, 1992, “Explanatory Unification and the Problem of Asymmetry”, Philosophy of Science , 59(4): 558–571. doi:10.1086/289695
• Braithwaite, R. B., 1953, Scientific Explanation: A Study of the Function of Theory, Probability and Law in Science , Cambridge: Cambridge University Press.
• Bromberger, Sylvain, 1966, “Why-Questions”, in Mind and Cosmos: Essays in Contemporary Science and Philosophy , Robert G. Colodny, (ed), Pittsburgh, PA: University of Pittsburgh Press, 86–111.
• Cartwright, Nancy, 1979, “Causal Laws and Effective Strategies”, Noûs , 13(4): 419–437. doi:10.2307/2215337
• –––, 1983, How the Laws of Physics Lie , Oxford: Clarendon Press.
• De Regt, Henk W. and Dennis Dieks, 2005, “A Contextual Approach to Scientific Understanding”, Synthese , 144(1): 137–170. doi:10.1007/s11229-005-5000-4
• Dowe, Phil, 2000, Physical Causation , Cambridge: Cambridge University Press. doi:10.1017/CBO9780511570650
• Earman, John, 1986, A Primer on Determinism , Dordrecht: Reidel.
• Friedman, Michael, 1974, “Explanation and Scientific Understanding”, The Journal of Philosophy , 71(1): 5–19. doi:10.2307/2024924
• Gardiner, Patrick L., 1959, The Nature of Historical Explanation , Oxford: Oxford University Press.
• Goodman, Nelson, 1955, Fact, Fiction, and Forecast , Cambridge, MA: Harvard University Press.
• Greeno, James G., 1970, “Evaluation of Statistical Hypotheses Using Information Transmitted”, Philosophy of Science , 37(2): 279–294. Reprinted in Salmon, 1971b: 89–104. doi:10.1086/288301
• Hall, Ned, 2004, “Two Concepts of Causation”, in Causation and Counterfactuals , John Collins, Ned Hall, and L. A. Paul (eds), Cambridge: MIT Press, pp. 225–276.
• Hausman, Daniel M., 1998, Causal Asymmetries , Cambridge: Cambridge University Press. doi:10.1017/CBO9780511663710
• Hempel, Carl G., 1965a, Aspects of Scientific Explanation and Other Essays in the Philosophy of Science , New York: Free Press.
• –––, 1965b, “Aspects of Scientific Explanation”, in Hempel 1965a: 331–496.
• –––, 1942 [1965], “The Function of General Laws in History”, Journal of Philosophy , 39(2): 35–48; reprinted with slight modifications in Hempel 1965a: 231–244. doi:10.2307/2017635
• Hempel, Carl G. and Paul Oppenheim, 1948 [1965], “Studies in the Logic of Explanation”, Philosophy of Science , 15(2): 135–175. Reprinted in Hempel 1965a: 245–290. doi:10.1086/286983
• Janzing, Dominik, Joris Mooij, Kun Zhang, Jan Lemeire, Jakob Zscheischler, Povilas Daniušis, Bastian Steudel, and Bernhard Schölkopf, 2012, “Information-Geometric Approach to Inferring Causal Directions”, Artificial Intelligence , 182–183: 1–31. doi:10.1016/j.artint.2012.01.002
• Jeffrey, Richard C., 1969, “Statistical Explanation vs. Statistical Inference”, in Essays in Honor of Carl G. Hempel , Nicholas Rescher (ed.), Dordrecht: D. Reidel, 104–113. Reprinted in Salmon 1971b: 19–28. doi:10.1007/978-94-017-1466-2_6
• Hitchcock, Christopher Read, 1995, “Discussion: Salmon on Explanatory Relevance”, Philosophy of Science , 62(2): 304–320. doi:10.1086/289858
• Kitcher, Philip, 1976, “Explanation, Conjunction, and Unification”, The Journal of Philosophy , 73(8): 207–212. doi:10.2307/2025559
• –––, 1989, “Explanatory Unification and the Causal Structure of the World”, in Kitcher and Salmon 1989: 410–505.
• Kitcher, Philip and Wesley Salmon, 1987, “Van Fraassen on Explanation”, The Journal of Philosophy , 84(6): 315–330. doi:10.2307/2026782
• ––– (eds), 1989, Scientific Explanation (Minnesota Studies in the Philosophy of Science, Volume 13), Minneapolis, MN: University of Minnesota Press.
• Kyburg, Henry E., 1965, “Discussion: Salmon’s Paper”, Philosophy of Science , 32(2): 147–151. doi:10.1086/288034
• Lange, M., 2017, Because Without Cause: Non-causal Explanations in Science and Mathematics , Oxford: Oxford University Press.
• Lewis, David K., 1973, Counterfactuals , Cambridge, MA: Harvard University Press.
• –––, 1986, Philosophical Papers (Volume II), Oxford: Oxford University Press. doi:10.1093/0195036468.001.0001
• –––, 2000, “Causation as Influence”, The Journal of Philosophy , 97(4): 182–197. doi:10.2307/2678389
• Lien, Yunnwen and Patricia W. Cheng, 2000, “Distinguishing Genuine from Spurious Causes: A Coherence Hypothesis”, Cognitive Psychology , 40(2): 87–137. doi:10.1006/cogp.1999.0724
• Lombrozo, Tania, 2010, “Causal–Explanatory Pluralism: How Intentions, Functions, and Mechanisms Influence Causal Ascriptions”, Cognitive Psychology , 61(4): 303–332. doi:10.1016/j.cogpsych.2010.05.002
• Mitchell, Sandra D., 1997, “Pragmatic Laws”, Philosophy of Science , 64(Supplement PSA 1996): S468–S479. doi:10.1086/392623
• Nagel, Ernest, 1961, The Structure of Science: Problems in the Logic of Scientific Explanation , New York: Harcourt, Brace and World.
• Pearl, Judea, 2000, Causality: Models, Reasoning and Inference , Cambridge: Cambridge University.
• Pitt, Joseph C. (ed.), 1988, Theories of Explanation , New York: Oxford University Press.
• Popper, Karl, 1959, The Logic of Scientific Discovery , London: Hutchinson.
• Reutlinger, Alexander and Juha Saatsi (eds.), 2018, Explanation Beyond Causation; Philosophical Perspectives on Non-Causal Explanations , Oxford: Oxford University Press. doi:10.1093/oso/9780198777946.001.0001
• Ruben, David-Hillel (ed.), 1993, Explanation (Oxford Readings in Philosophy), Oxford: Oxford University Press.
• Salmon, Wesley C., 1971a, “Statistical Explanation”, in Salmon 1971b: 29–87.
• ––– (ed.), 1971b, Statistical Explanation and Statistical Relevance , Pittsburgh, PA: University of Pittsburgh Press.
• –––, 1984, Scientific Explanation and the Causal Structure of the World , Princeton, NJ: Princeton University Press.
• –––, 1989, “Four Decades of Scientific Explanation”, in Kitcher and Salmon 1989: 3–219. Reprinted as a separate monograph, Minneapolis, MN: University of Minnesota Press, 1989. Page numbers are from the monograph.
• –––, 1994, “Causality without Counterfactuals”, Philosophy of Science , 61(2): 297–312. doi:10.1086/289801
• –––, 1997, “Causality and Explanation: A Reply to Two Critiques”, Philosophy of Science , 64(3): 461–477. doi:10.1086/392561
• Schaffer, Jonathan, 2000, “Causation by Disconnection”, Philosophy of Science , 67(2): 285–300. doi:10.1086/392776
• Scriven, Michael, 1959, “Truisms as the Grounds of Historical Explanations”, in Theories of History: Readings from Classical and Contemporary Sources , Patrick Gardiner (ed.), Glencoe, IL: The Free Press, 443–475.
• –––, 1962, “Explanations, Predictions, and Laws”, in Scientific Explanation, Space, and Time (Minnesota Studies in the Philosophy of Science: Vol. 3), Herbert Feigl and Grover Maxwell (eds), Minneapolis: University of Minnesota Press, 170–230.
• Spirtes, Peter, Clark Glymour, and Richard Scheines, 1993 [2000], Causation, Prediction and Search , New York: Springer-Verlag. Second Edition, Cambridge, MA: MIT Press, 2000.
• van Fraassen, Bas. C., 1980, The Scientific Image , Oxford: Oxford University Press. doi:10.1093/0198244274.001.0001
• –––, 1989, Laws and Symmetry , Oxford: Oxford University Press. doi:10.1093/0198248601.001.0001
• Whewell, William, 1840, The Philosophy of the Inductive Sciences, Founded upon their History , two volumes, London: John W. Parker.
• Woodward, James, 1989, “The Causal/Mechanical Model of Explanation”, in Kitcher and Salmon 1989: 357–383.
• –––, 2000, “Explanation and Invariance in the Special Sciences”, The British Journal for the Philosophy of Science , 51(2): 197–254. doi:10.1093/bjps/51.2.197
• –––, 2002, “What Is a Mechanism?: A Counterfactual Account”, Philosophy of Science , 69: S366–S377.
• –––, 2003, Making Things Happen: A Theory of Causal Explanation , Oxford: Oxford University Press. doi:10.1093/0195155270.001.0001
• –––, 2006, “Sensitive and Insensitive Causation”, The Philosophical Review , 115(1): 1–50. doi:10.1215/00318108-2005-001

How to cite this entry . Preview the PDF version of this entry at the Friends of the SEP Society . Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO). Enhanced bibliography for this entry at PhilPapers , with links to its database.

• “ Theories of Explanation ”, by G. Randolph Mayes (CSU/Sacramento), in the Internet Encyclopedia of Philosophy (edited by J. Fieser)

causation: counterfactual theories of | causation: probabilistic | causation: the metaphysics of | laws of nature | Salmon, Wesley

Copyright © 2021 by James Woodward < jfw @ pitt . edu > Lauren Ross < rossl @ uci . edu >

• Accessibility

Support SEP

Mirror sites.

View this site from another server:

• Info about mirror sites

The Stanford Encyclopedia of Philosophy is copyright © 2023 by The Metaphysics Research Lab , Department of Philosophy, Stanford University

Library of Congress Catalog Data: ISSN 1095-5054

The Plagiarism Checker Online For Your Academic Work

Start Plagiarism Check

Editing & Proofreading for Your Research Paper

Get it proofread now

Online Printing & Binding with Free Express Delivery

Configure binding now

• Academic essay overview
• The writing process
• Structuring academic essays
• Types of academic essays
• Academic writing overview
• Sentence structure
• Academic writing process
• Improving your academic writing
• Titles and headings
• APA style overview
• APA citation & referencing
• APA structure & sections
• Citation & referencing
• Structure and sections
• APA examples overview
• Commonly used citations
• Other examples
• British English vs. American English
• Chicago style overview
• Chicago citation & referencing
• Chicago structure & sections
• Chicago style examples
• Citing sources overview
• Citation format
• Citation examples
• College essay overview
• Application
• How to write a college essay
• Types of college essays
• Commonly confused words
• Definitions
• Dissertation overview
• Dissertation structure & sections
• Dissertation writing process
• Graduate school overview
• Application & admission
• Study abroad
• Master degree
• Harvard referencing overview
• Language rules overview
• Grammatical rules & structures
• Parts of speech
• Punctuation
• Methodology overview
• Analyzing data
• Experiments
• Observations
• Inductive vs. Deductive
• Qualitative vs. Quantitative
• Types of validity
• Types of reliability
• Sampling methods
• Theories & Concepts
• Types of research studies
• Types of variables
• MLA style overview
• MLA examples
• MLA citation & referencing
• MLA structure & sections
• Plagiarism overview
• Plagiarism checker
• Types of plagiarism
• Printing production overview
• Research bias overview
• Types of research bias
• Example sections
• Types of research papers
• Research process overview
• Problem statement
• Research proposal
• Research topic
• Statistics overview
• Levels of measurment
• Frequency distribution
• Measures of central tendency
• Measures of variability
• Hypothesis testing
• Parameters & test statistics
• Types of distributions
• Correlation
• Effect size
• Hypothesis testing assumptions
• Types of ANOVAs
• Types of chi-square
• Statistical data
• Statistical models
• Spelling mistakes
• Tips overview
• Academic writing tips
• Dissertation tips
• Sources tips
• Working with sources overview
• Evaluating sources
• Finding sources
• Including sources
• Types of sources

Your Step to Success

Plagiarism Check within 10min

Printing & Binding with 3D Live Preview

Explanatory Research – Guide with Definition & Examples

How do you like this article cancel reply.

Save my name, email, and website in this browser for the next time I comment.

Explanatory research, a vital part of research methodology , is dedicated to providing a deep understanding of a phenomenon through the explanation of causal relationships among variables. Unlike exploratory research that seeks to generate new insights or ideas, explanatory research dives deeper to identify why and how certain situations occur. This methodology is often employed when there is a clear understanding of the problem but the reasons behind it remain obscure, thereby necessitating a comprehensive explanation.

Inhaltsverzeichnis

• 1 Explanatory Research – In a Nutshell
• 2 Definition: Explanatory Research
• 3 The usage of explanatory research
• 4 Explanatory research questions
• 5 Explanatory research: Data collection
• 6 Explanatory research: Data analysis
• 7 The 5 Steps of explanatory research with examples
• 8 Explanatory vs. exploratory research
• 9 Advantages vs. disadvantages

Explanatory Research – In a Nutshell

• Explanatory research is a cornerstone of other research.
• Without an explanatory study, your future research will be incomplete and inefficient.
• This research improves survey and study design and reduces unintended bias.

Definition: Explanatory Research

Explanatory research is a study method that investigates the causes of a phenomenon when only limited data is presented. It can help you better grasp a topic, determine why a phenomenon is happening, and forecast future events.

This research can be described as a “cause and effect” model, researching previously unexplored patterns and trends in current data. Consequently, it is sometimes considered a sort of causal research .

The usage of explanatory research

Explanatory research investigates how or why something happens. Therefore, this type of research is one of the first steps in the research process , serving as a beginning point for future work. Your topic may have data, but the causal relationship you’re interested in may not.

This research helps evaluate patterns and generate hypotheses for future work. An explanatory study can help you comprehend a variable’s relationship. However, don’t expect conclusive outcomes.

Explanatory research questions

This research answers “why” and “how” inquiries, resulting in a better knowledge of a previously unsolved topic or clarification for relevant future research.

• Why do bilingual individuals exhibit more risky behavior than monolingual individuals during commercial negotiations?
• How does a child’s capacity to resist gratification predict their future success?
• Why are adolescents more prone to litter in highly littered areas than in clean areas?

Explanatory research: Data collection

After deciding on your research subject, you have numerous alternatives for research and data collection methods.

The following are some of the most prevalent research methods:

• Literature reviews
• Interviews and focus groups
• Pilot studies

Explanatory research: Data analysis

Ensure that your explanatory research is conducted appropriately and that your analysis is causal and not merely correlative.

Correlated variables are merely linked: when one changes, so does the other . There is no direct or indirect causal relationship.

Causation means independent variable changes cause dependent variable changes. The link between variables is direct.

The requirements for causal evidence are:

• Temporal : Cause must precede effect.
• Variation : Independent and dependent variable intervention must be systematic.
• Non-spurious : Be sure no mitigating factors or third hidden variables contradict your results.

The 5 Steps of explanatory research with examples

The data collection approach determines your explanatory research design. In most circumstances, you’ll utilize an experiment to test causality. The steps are illustrated in the following.

Step 1 of explanatory research: Research question

The initial stage in the research is familiarizing yourself with the topic of interest to formulate a research question.

Suppose you are interested in adult language retention rates.

You’ve examined language retention in adoptees. People who learned a foreign language as infants had an easier time learning it again than those who weren’t exposed.

You want to know how language exposure affects long-term retention. You’re designing an experiment to answer this question: How does early language exposure affect language retention in adoptees?

Step 2 of explanatory research: Hypothesis

Next, set your expectations. In some circumstances, you can use relevant literature to build your hypothesis. In other cases, the topic isn’t well-studied; therefore, you must create your theory based on instincts or literature on distant themes.

You hypothesize that individuals exposed to a language in infancy for a shorter duration will be less likely to retain features of this language than adults exposed for a longer time.

You express your predictions in terms of the null (H 0 ) and alternative (H 1 ) hypotheses:

• H 0 : Infancy language exposure does not affect language retention in adopted adults.
• H 1 : Exposure to a language in infancy improves language retention in adult adoptees.

Step 3 of explanatory research: Methodology and data collection

Next, choose your data collecting and data analysis methodologies and document them. After meticulously planning your research, you can begin data collection.

To test a causal relationship, you run an experiment. You gather a group of adults adopted from Colombia and raised in the U.S.

You compare:

• 0-6-month-old Colombian adoptees.
• 6-12 month-old Colombian adoptees
• 12-18-month-old Colombian adoptees.
• Unexposed monolingual adults.

Using a three-stage research design, you administer two tests of their Spanish language skills during the study:

• Pre-test : Several language proficiency tests are administered to identify group variations before instruction.
• Intervention : You deliver eight hours of Spanish lessons to each group.
• Post-test : After the intervention, you administer multiple language proficiency tests to determine whether there are any differences between the groups.

Step 4 of explanatory research: Analysis and results

After data collection, assess and report results.

After experimenting, you examine the data and observe that:

• The pre-exposed adults demonstrated more excellent Spanish language skills than individuals who were not pre-exposed. The post-test reveals an even more significant disparity.
• Adults adopted between 12 and 18 months had higher Spanish competence than those adopted between 0 and 6 months or 6 and 12 months, but there was no difference between the latter two groups.

For significance, use a mixed ANOVA . ANOVA indicates that pre-test differences aren’t significant, while post-test differences are.

You report your findings following the criteria of your chosen citation style between the groups.

Step 5 of explanatory research: Interpretation and recommendation

Try to explain unexpected results as you interpret them. In most circumstances, you’ll need to provide recommendations for future research.

Your findings were per your expectations. Adopted individuals who were pre-exposed to a language in infancy for a longer time have preserved more of this knowledge than people who weren’t pre-exposed.

After the intervention, this difference becomes large.

You decide to do more research and suggest some topics:

• Replicate the study with a larger sample
• Study other mother tongues (e.g., Korean, Lingala, Arabic)
• Study other linguistic features, like accent nativeness.

Explanatory vs. exploratory research

Explanatory and exploratory research are often confused. Remember, exploratory research establishes the framework for explanatory research.

Many exploratory research inquiries begin with “what.” They are intended to guide future studies and typically lack definite conclusions. The research is frequently employed as the initial step in the research process to assist you in refining your study topic and ideas.

Explanatory research questions begin with “why” or “how.” They assist you in understanding why and how something happens.

Advantages vs. disadvantages

As with any other study methodology, this research involves trade-offs: while it offers a unique set of benefits, it also has major drawbacks.

What is explanatory research?

An explanatory study investigates how or why something happens with limited information. It helps you understand a topic.

Is explanatory research quantitative or qualitative?

The explanatory research model is a quantitative strategy used to examine a hypothesis by gathering evidence that either supports or contradicts it.

When should I use explanatory research?

Explanatory research aims to explain a phenomenon. Consequently, this form of research is frequently one of the initial steps of the research process, acting as a springboard for subsequent analysis.

We use cookies on our website. Some of them are essential, while others help us to improve this website and your experience.

• External Media

Individual Privacy Preferences

Cookie Details Privacy Policy Imprint

Here you will find an overview of all cookies used. You can give your consent to whole categories or display further information and select certain cookies.

Accept all Save

Essential cookies enable basic functions and are necessary for the proper function of the website.

Show Cookie Information Hide Cookie Information

Statistics cookies collect information anonymously. This information helps us to understand how our visitors use our website.

Content from video platforms and social media platforms is blocked by default. If External Media cookies are accepted, access to those contents no longer requires manual consent.

Privacy Policy Imprint

What Is A Research (Scientific) Hypothesis? A plain-language explainer + examples

By:  Derek Jansen (MBA)  | Reviewed By: Dr Eunice Rautenbach | June 2020

If you’re new to the world of research, or it’s your first time writing a dissertation or thesis, you’re probably noticing that the words “research hypothesis” and “scientific hypothesis” are used quite a bit, and you’re wondering what they mean in a research context .

“Hypothesis” is one of those words that people use loosely, thinking they understand what it means. However, it has a very specific meaning within academic research. So, it’s important to understand the exact meaning before you start hypothesizing. 

Research Hypothesis 101

• What is a hypothesis ?
• What is a research hypothesis (scientific hypothesis)?
• Requirements for a research hypothesis
• Definition of a research hypothesis
• The null hypothesis

What is a hypothesis?

Let’s start with the general definition of a hypothesis (not a research hypothesis or scientific hypothesis), according to the Cambridge Dictionary:

Hypothesis: an idea or explanation for something that is based on known facts but has not yet been proved.

In other words, it’s a statement that provides an explanation for why or how something works, based on facts (or some reasonable assumptions), but that has not yet been specifically tested . For example, a hypothesis might look something like this:

Hypothesis: sleep impacts academic performance.

This statement predicts that academic performance will be influenced by the amount and/or quality of sleep a student engages in – sounds reasonable, right? It’s based on reasonable assumptions , underpinned by what we currently know about sleep and health (from the existing literature). So, loosely speaking, we could call it a hypothesis, at least by the dictionary definition.

But that’s not good enough…

Unfortunately, that’s not quite sophisticated enough to describe a research hypothesis (also sometimes called a scientific hypothesis), and it wouldn’t be acceptable in a dissertation, thesis or research paper . In the world of academic research, a statement needs a few more criteria to constitute a true research hypothesis .

What is a research hypothesis?

A research hypothesis (also called a scientific hypothesis) is a statement about the expected outcome of a study (for example, a dissertation or thesis). To constitute a quality hypothesis, the statement needs to have three attributes – specificity , clarity and testability .

Let’s take a look at these more closely.

Need a helping hand?

Hypothesis Essential #1: Specificity & Clarity

A good research hypothesis needs to be extremely clear and articulate about both what’ s being assessed (who or what variables are involved ) and the expected outcome (for example, a difference between groups, a relationship between variables, etc.).

Let’s stick with our sleepy students example and look at how this statement could be more specific and clear.

Hypothesis: Students who sleep at least 8 hours per night will, on average, achieve higher grades in standardised tests than students who sleep less than 8 hours a night.

As you can see, the statement is very specific as it identifies the variables involved (sleep hours and test grades), the parties involved (two groups of students), as well as the predicted relationship type (a positive relationship). There’s no ambiguity or uncertainty about who or what is involved in the statement, and the expected outcome is clear.

Contrast that to the original hypothesis we looked at – “Sleep impacts academic performance” – and you can see the difference. “Sleep” and “academic performance” are both comparatively vague , and there’s no indication of what the expected relationship direction is (more sleep or less sleep). As you can see, specificity and clarity are key.

Hypothesis Essential #2: Testability (Provability)

A statement must be testable to qualify as a research hypothesis. In other words, there needs to be a way to prove (or disprove) the statement. If it’s not testable, it’s not a hypothesis – simple as that.

For example, consider the hypothesis we mentioned earlier:

Hypothesis: Students who sleep at least 8 hours per night will, on average, achieve higher grades in standardised tests than students who sleep less than 8 hours a night.  

We could test this statement by undertaking a quantitative study involving two groups of students, one that gets 8 or more hours of sleep per night for a fixed period, and one that gets less. We could then compare the standardised test results for both groups to see if there’s a statistically significant difference. 

Again, if you compare this to the original hypothesis we looked at – “Sleep impacts academic performance” – you can see that it would be quite difficult to test that statement, primarily because it isn’t specific enough. How much sleep? By who? What type of academic performance?

So, remember the mantra – if you can’t test it, it’s not a hypothesis 🙂

Defining A Research Hypothesis

You’re still with us? Great! Let’s recap and pin down a clear definition of a hypothesis.

A research hypothesis (or scientific hypothesis) is a statement about an expected relationship between variables, or explanation of an occurrence, that is clear, specific and testable.

So, when you write up hypotheses for your dissertation or thesis, make sure that they meet all these criteria. If you do, you’ll not only have rock-solid hypotheses but you’ll also ensure a clear focus for your entire research project.

What about the null hypothesis?

You may have also heard the terms null hypothesis , alternative hypothesis, or H-zero thrown around. At a simple level, the null hypothesis is the counter-proposal to the original hypothesis.

For example, if the hypothesis predicts that there is a relationship between two variables (for example, sleep and academic performance), the null hypothesis would predict that there is no relationship between those variables.

At a more technical level, the null hypothesis proposes that no statistical significance exists in a set of given observations and that any differences are due to chance alone.

And there you have it – hypotheses in a nutshell. 

If you have any questions, be sure to leave a comment below and we’ll do our best to help you. If you need hands-on help developing and testing your hypotheses, consider our private coaching service , where we hold your hand through the research journey.

Psst... there’s more!

This post was based on one of our popular Research Bootcamps . If you're working on a research project, you'll definitely want to check this out ...

You Might Also Like:

16 Comments

Very useful information. I benefit more from getting more information in this regard.

Very great insight,educative and informative. Please give meet deep critics on many research data of public international Law like human rights, environment, natural resources, law of the sea etc

In a book I read a distinction is made between null, research, and alternative hypothesis. As far as I understand, alternative and research hypotheses are the same. Can you please elaborate? Best Afshin

This is a self explanatory, easy going site. I will recommend this to my friends and colleagues.

Very good definition. How can I cite your definition in my thesis? Thank you. Is nul hypothesis compulsory in a research?

It’s a counter-proposal to be proven as a rejection

Please what is the difference between alternate hypothesis and research hypothesis?

It is a very good explanation. However, it limits hypotheses to statistically tasteable ideas. What about for qualitative researches or other researches that involve quantitative data that don’t need statistical tests?

In qualitative research, one typically uses propositions, not hypotheses.

could you please elaborate it more

I’ve benefited greatly from these notes, thank you.

This is very helpful

well articulated ideas are presented here, thank you for being reliable sources of information

Excellent. Thanks for being clear and sound about the research methodology and hypothesis (quantitative research)

I have only a simple question regarding the null hypothesis. – Is the null hypothesis (Ho) known as the reversible hypothesis of the alternative hypothesis (H1? – How to test it in academic research?

this is very important note help me much more

Trackbacks/Pingbacks

• What Is Research Methodology? Simple Definition (With Examples) - Grad Coach - […] Contrasted to this, a quantitative methodology is typically used when the research aims and objectives are confirmatory in nature. For example,…

Submit a Comment Cancel reply

Your email address will not be published. Required fields are marked *

Save my name, email, and website in this browser for the next time I comment.

• Print Friendly

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, automatically generate references for free.

• Knowledge Base
• Methodology
• Explanatory Research | Definition, Guide, & Examples

Explanatory Research | Definition, Guide & Examples

Published on 7 May 2022 by Tegan George and Julia Merkus. Revised on 20 January 2023.

Explanatory research is a research method that explores why something occurs when limited information is available. It can help you increase your understanding of a given topic, ascertain how or why a particular phenomenon is occurring, and predict future occurrences.

Explanatory research can also be explained as a ’cause and effect’ model, investigating patterns and trends in existing data that haven’t been previously investigated. For this reason, it is often considered a type of causal research .

Table of contents

When to use explanatory research, explanatory research questions, explanatory research data collection, explanatory research data analysis, step-by-step example of explanatory research, explanatory vs exploratory research, advantages and disadvantages of exploratory research, frequently asked questions about explanatory research.

Explanatory research is used to investigate how or why a phenomenon takes place. Therefore, this type of research is often one of the first stages in the research process, serving as a jumping-off point for future research. While there is often data available about your topic, it’s possible the particular causal relationship you are interested in has not been robustly studied.

Explanatory research helps you analyse these patterns, formulating hypotheses that can guide future endeavors. If you are seeking a more complete understanding of a relationship between variables, explanatory research is a great place to start. However, keep in mind that it will likely not yield conclusive results.

You analysed their final grades and noticed that the students who take your course in the first semester always obtain higher grades than students who take the same course in the second semester.

Prevent plagiarism, run a free check.

Explanatory research answers ‘why’ and ‘what’ questions, leading to an improved understanding of a previously unresolved problem or providing clarity for related future research initiatives.

Here are a few examples:

• Why do undergraduate students obtain higher average grades in the first semester than in the second semester?
• How does marital status affect labour market participation?
• Why do multilingual individuals show more risky behaviour during business negotiations than monolingual individuals?
• How does a child’s ability to delay immediate gratification predict success later in life?
• Why are teenagers more likely to litter in a highly littered area than in a clean area?

After choosing your research question, there is a variety of options for research and data collection methods to choose from.

A few of the most common research methods include:

• Literature reviews
• Interviews and focus groups
• Pilot studies
• Observations
• Experiments

The method you choose depends on several factors, including your timeline, your budget, and the structure of your question.

If there is already a body of research on your topic, a literature review is a great place to start. If you are interested in opinions and behaviour, consider an interview or focus group format. If you have more time or funding available, an experiment or pilot study may be a good fit for you.

In order to ensure you are conducting your explanatory research correctly, be sure your analysis is definitively causal in nature, and not just correlated.

Always remember the phrase ‘correlation doesn’t imply causation’. Correlated variables are merely associated with one another: when one variable changes, so does the other. However, this isn’t necessarily due to a direct or indirect causal link.

Causation means that changes in the independent variable bring about changes in the dependent variable. In other words, there is a direct cause-and-effect relationship between variables.

Causal evidence must meet three criteria:

• Temporal : What you define as the ’cause’ must precede what you define as the ‘effect’.
• Variation : Intervention must be systematic between your independent variable and dependent variable.
• Non-spurious : Be careful that there are no mitigating factors or hidden third variables that confound your results.

Correlation doesn’t imply causation, but causation always implies correlation. In order to get conclusive causal results, you’ll need to conduct a full experimental design .

Your explanatory research design depends on the research method you choose to collect your data . In most cases, you’ll use an experiment to investigate potential causal relationships. We’ll walk you through the steps using an example.

Step 1: Develop the research question

The first step in conducting explanatory research is getting familiar with the topic you’re interested in, so that you can develop a research question .

Let’s say you’re interested in language retention rates in adults.

You are interested in finding out how the duration of exposure to language influences language retention ability later in life.

Step 2: Formulate a hypothesis

The next step is to address your expectations. In some cases, there is literature available on your subject or on a closely related topic that you can use as a foundation for your hypothesis . In other cases, the topic isn’t well studied, and you’ll have to develop your hypothesis based on your instincts or on existing literature on more distant topics.

• H 0 : The duration of exposure to a language in infancy does not influence language retention in adults who were adopted from abroad as children.
• H 1 : The duration of exposure to a language in infancy has a positive effect on language retention in adults who were adopted from abroad as children.

Step 3: Design your methodology and collect your data

Next, decide what data collection and data analysis methods you will use and write them up. After carefully designing your research, you can begin to collect your data.

• Adults who were adopted from Colombia between 0 and 6 months of age
• Adults who were adopted from Colombia between 6 and 12 months of age
• Adults who were adopted from Colombia between 12 and 18 months of age
• Monolingual adults who have not been exposed to a different language

During the study, you test their Spanish language proficiency twice in a research design that has three stages:

• Pretest : You conduct several language proficiency tests to establish any differences between groups pre-intervention.
• Intervention : You provide all groups with 8 hours of Spanish class.
• Posttest : You again conduct several language proficiency tests to establish any differences between groups post-intervention.

You made sure to control for any confounding variables , such as age, gender, and proficiency in other languages.

Step 4: Analyse your data and report results

After data collection is complete, proceed to analyse your data and report the results.

• The pre-exposed adults showed higher language proficiency in Spanish than those who had not been pre-exposed. The difference is even greater for the posttest.
• The adults who were adopted between 12 and 18 months of age had a higher Spanish language proficiency level than those who were adopted between 0 and 6 months or 6 and 12 months of age, but there was no difference found between the latter two groups.

To determine whether these differences are significant, you conduct a mixed ANOVA. The ANOVA shows that all differences are not significant for the pretest, but they are significant for the posttest.

Step 5: Interpret your results and provide suggestions for future research

As you interpret the results, try to come up with explanations for the results that you did not expect. In most cases, you want to provide suggestions for future research.

However, this difference is only significant after the intervention (the Spanish class).

You decide it’s worth it to further research the matter, and propose a few additional research ideas:

• Replicate the study with a larger sample
• Replicate the study for other maternal languages (e.g., Korean, Lingala, Arabic)
• Replicate the study for other language aspects, such as nativeness of the accent

It can be easy to confuse explanatory research with exploratory research. If you’re in doubt about the relationship between exploratory and explanatory research, just remember that exploratory research lays the groundwork for later explanatory research.

Exploratory research questions often begin with ‘what’. They are designed to guide future research and do not usually have conclusive results. Exploratory research is often utilised as a first step in your research process, to help you focus your research question and fine-tune your hypotheses.

Explanatory research questions often start with ‘why’ or ‘how’. They help you study why and how a previously studied phenomenon takes place.

Like any other research design , exploratory research has its trade-offs: while it provides a unique set of benefits, it also has significant downsides:

• It gives more meaning to previous research. It helps fill in the gaps in existing analyses and provides information on the reasons behind phenomena.
• It is very flexible and often replicable, since the internal validity tends to be high when done correctly.
• As you can often use secondary research, explanatory research is often very cost- and time-effective, allowing you to utilise pre-existing resources to guide your research before committing to heavier analyses.

Disadvantages

• While explanatory research does help you solidify your theories and hypotheses, it usually lacks conclusive results.
• Results can be biased or inadmissible to a larger body of work and are not generally externally valid . You will likely have to conduct more robust (often quantitative ) research later to bolster any possible findings gleaned from explanatory research.
• Coincidences can be mistaken for causal relationships , and it can sometimes be challenging to ascertain which is the causal variable and which is the effect.

Explanatory research is a research method used to investigate how or why something occurs when only a small amount of information is available pertaining to that topic. It can help you increase your understanding of a given topic.

Explanatory research is used to investigate how or why a phenomenon occurs. Therefore, this type of research is often one of the first stages in the research process , serving as a jumping-off point for future research.

Exploratory research explores the main aspects of a new or barely researched question.

Explanatory research explains the causes and effects of an already widely researched question.

Quantitative research deals with numbers and statistics, while qualitative research deals with words and meanings.

Quantitative methods allow you to test a hypothesis by systematically collecting and analysing data, while qualitative methods allow you to explore ideas and experiences in depth.

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the ‘Cite this Scribbr article’ button to automatically add the citation to our free Reference Generator.

George, T. & Merkus, J. (2023, January 20). Explanatory Research | Definition, Guide & Examples. Scribbr. Retrieved 14 May 2024, from https://www.scribbr.co.uk/research-methods/explanatory-research-design/

Is this article helpful?

Tegan George

Other students also liked, exploratory research | definition, guide, & examples, descriptive research design | definition, methods & examples, a quick guide to experimental design | 5 steps & examples.

• Privacy Policy

Home » What is a Hypothesis – Types, Examples and Writing Guide

What is a Hypothesis – Types, Examples and Writing Guide

Table of Contents

Definition:

Hypothesis is an educated guess or proposed explanation for a phenomenon, based on some initial observations or data. It is a tentative statement that can be tested and potentially proven or disproven through further investigation and experimentation.

Hypothesis is often used in scientific research to guide the design of experiments and the collection and analysis of data. It is an essential element of the scientific method, as it allows researchers to make predictions about the outcome of their experiments and to test those predictions to determine their accuracy.

Types of Hypothesis

Types of Hypothesis are as follows:

Research Hypothesis

A research hypothesis is a statement that predicts a relationship between variables. It is usually formulated as a specific statement that can be tested through research, and it is often used in scientific research to guide the design of experiments.

Null Hypothesis

The null hypothesis is a statement that assumes there is no significant difference or relationship between variables. It is often used as a starting point for testing the research hypothesis, and if the results of the study reject the null hypothesis, it suggests that there is a significant difference or relationship between variables.

Alternative Hypothesis

An alternative hypothesis is a statement that assumes there is a significant difference or relationship between variables. It is often used as an alternative to the null hypothesis and is tested against the null hypothesis to determine which statement is more accurate.

Directional Hypothesis

A directional hypothesis is a statement that predicts the direction of the relationship between variables. For example, a researcher might predict that increasing the amount of exercise will result in a decrease in body weight.

Non-directional Hypothesis

A non-directional hypothesis is a statement that predicts the relationship between variables but does not specify the direction. For example, a researcher might predict that there is a relationship between the amount of exercise and body weight, but they do not specify whether increasing or decreasing exercise will affect body weight.

Statistical Hypothesis

A statistical hypothesis is a statement that assumes a particular statistical model or distribution for the data. It is often used in statistical analysis to test the significance of a particular result.

Composite Hypothesis

A composite hypothesis is a statement that assumes more than one condition or outcome. It can be divided into several sub-hypotheses, each of which represents a different possible outcome.

Empirical Hypothesis

An empirical hypothesis is a statement that is based on observed phenomena or data. It is often used in scientific research to develop theories or models that explain the observed phenomena.

Simple Hypothesis

A simple hypothesis is a statement that assumes only one outcome or condition. It is often used in scientific research to test a single variable or factor.

Complex Hypothesis

A complex hypothesis is a statement that assumes multiple outcomes or conditions. It is often used in scientific research to test the effects of multiple variables or factors on a particular outcome.

Applications of Hypothesis

Hypotheses are used in various fields to guide research and make predictions about the outcomes of experiments or observations. Here are some examples of how hypotheses are applied in different fields:

• Science : In scientific research, hypotheses are used to test the validity of theories and models that explain natural phenomena. For example, a hypothesis might be formulated to test the effects of a particular variable on a natural system, such as the effects of climate change on an ecosystem.
• Medicine : In medical research, hypotheses are used to test the effectiveness of treatments and therapies for specific conditions. For example, a hypothesis might be formulated to test the effects of a new drug on a particular disease.
• Psychology : In psychology, hypotheses are used to test theories and models of human behavior and cognition. For example, a hypothesis might be formulated to test the effects of a particular stimulus on the brain or behavior.
• Sociology : In sociology, hypotheses are used to test theories and models of social phenomena, such as the effects of social structures or institutions on human behavior. For example, a hypothesis might be formulated to test the effects of income inequality on crime rates.
• Business : In business research, hypotheses are used to test the validity of theories and models that explain business phenomena, such as consumer behavior or market trends. For example, a hypothesis might be formulated to test the effects of a new marketing campaign on consumer buying behavior.
• Engineering : In engineering, hypotheses are used to test the effectiveness of new technologies or designs. For example, a hypothesis might be formulated to test the efficiency of a new solar panel design.

How to write a Hypothesis

Here are the steps to follow when writing a hypothesis:

Identify the Research Question

The first step is to identify the research question that you want to answer through your study. This question should be clear, specific, and focused. It should be something that can be investigated empirically and that has some relevance or significance in the field.

Conduct a Literature Review

Before writing your hypothesis, it’s essential to conduct a thorough literature review to understand what is already known about the topic. This will help you to identify the research gap and formulate a hypothesis that builds on existing knowledge.

Determine the Variables

The next step is to identify the variables involved in the research question. A variable is any characteristic or factor that can vary or change. There are two types of variables: independent and dependent. The independent variable is the one that is manipulated or changed by the researcher, while the dependent variable is the one that is measured or observed as a result of the independent variable.

Formulate the Hypothesis

Based on the research question and the variables involved, you can now formulate your hypothesis. A hypothesis should be a clear and concise statement that predicts the relationship between the variables. It should be testable through empirical research and based on existing theory or evidence.

Write the Null Hypothesis

The null hypothesis is the opposite of the alternative hypothesis, which is the hypothesis that you are testing. The null hypothesis states that there is no significant difference or relationship between the variables. It is important to write the null hypothesis because it allows you to compare your results with what would be expected by chance.

Refine the Hypothesis

After formulating the hypothesis, it’s important to refine it and make it more precise. This may involve clarifying the variables, specifying the direction of the relationship, or making the hypothesis more testable.

Examples of Hypothesis

Here are a few examples of hypotheses in different fields:

• Psychology : “Increased exposure to violent video games leads to increased aggressive behavior in adolescents.”
• Biology : “Higher levels of carbon dioxide in the atmosphere will lead to increased plant growth.”
• Sociology : “Individuals who grow up in households with higher socioeconomic status will have higher levels of education and income as adults.”
• Education : “Implementing a new teaching method will result in higher student achievement scores.”
• Marketing : “Customers who receive a personalized email will be more likely to make a purchase than those who receive a generic email.”
• Physics : “An increase in temperature will cause an increase in the volume of a gas, assuming all other variables remain constant.”
• Medicine : “Consuming a diet high in saturated fats will increase the risk of developing heart disease.”

Purpose of Hypothesis

The purpose of a hypothesis is to provide a testable explanation for an observed phenomenon or a prediction of a future outcome based on existing knowledge or theories. A hypothesis is an essential part of the scientific method and helps to guide the research process by providing a clear focus for investigation. It enables scientists to design experiments or studies to gather evidence and data that can support or refute the proposed explanation or prediction.

The formulation of a hypothesis is based on existing knowledge, observations, and theories, and it should be specific, testable, and falsifiable. A specific hypothesis helps to define the research question, which is important in the research process as it guides the selection of an appropriate research design and methodology. Testability of the hypothesis means that it can be proven or disproven through empirical data collection and analysis. Falsifiability means that the hypothesis should be formulated in such a way that it can be proven wrong if it is incorrect.

In addition to guiding the research process, the testing of hypotheses can lead to new discoveries and advancements in scientific knowledge. When a hypothesis is supported by the data, it can be used to develop new theories or models to explain the observed phenomenon. When a hypothesis is not supported by the data, it can help to refine existing theories or prompt the development of new hypotheses to explain the phenomenon.

When to use Hypothesis

Here are some common situations in which hypotheses are used:

• In scientific research , hypotheses are used to guide the design of experiments and to help researchers make predictions about the outcomes of those experiments.
• In social science research , hypotheses are used to test theories about human behavior, social relationships, and other phenomena.
• I n business , hypotheses can be used to guide decisions about marketing, product development, and other areas. For example, a hypothesis might be that a new product will sell well in a particular market, and this hypothesis can be tested through market research.

Characteristics of Hypothesis

Here are some common characteristics of a hypothesis:

• Testable : A hypothesis must be able to be tested through observation or experimentation. This means that it must be possible to collect data that will either support or refute the hypothesis.
• Falsifiable : A hypothesis must be able to be proven false if it is not supported by the data. If a hypothesis cannot be falsified, then it is not a scientific hypothesis.
• Clear and concise : A hypothesis should be stated in a clear and concise manner so that it can be easily understood and tested.
• Based on existing knowledge : A hypothesis should be based on existing knowledge and research in the field. It should not be based on personal beliefs or opinions.
• Specific : A hypothesis should be specific in terms of the variables being tested and the predicted outcome. This will help to ensure that the research is focused and well-designed.
• Tentative: A hypothesis is a tentative statement or assumption that requires further testing and evidence to be confirmed or refuted. It is not a final conclusion or assertion.
• Relevant : A hypothesis should be relevant to the research question or problem being studied. It should address a gap in knowledge or provide a new perspective on the issue.

Advantages of Hypothesis

Hypotheses have several advantages in scientific research and experimentation:

• Guides research: A hypothesis provides a clear and specific direction for research. It helps to focus the research question, select appropriate methods and variables, and interpret the results.
• Predictive powe r: A hypothesis makes predictions about the outcome of research, which can be tested through experimentation. This allows researchers to evaluate the validity of the hypothesis and make new discoveries.
• Facilitates communication: A hypothesis provides a common language and framework for scientists to communicate with one another about their research. This helps to facilitate the exchange of ideas and promotes collaboration.
• Efficient use of resources: A hypothesis helps researchers to use their time, resources, and funding efficiently by directing them towards specific research questions and methods that are most likely to yield results.
• Provides a basis for further research: A hypothesis that is supported by data provides a basis for further research and exploration. It can lead to new hypotheses, theories, and discoveries.
• Increases objectivity: A hypothesis can help to increase objectivity in research by providing a clear and specific framework for testing and interpreting results. This can reduce bias and increase the reliability of research findings.

Limitations of Hypothesis

Some Limitations of the Hypothesis are as follows:

• Limited to observable phenomena: Hypotheses are limited to observable phenomena and cannot account for unobservable or intangible factors. This means that some research questions may not be amenable to hypothesis testing.
• May be inaccurate or incomplete: Hypotheses are based on existing knowledge and research, which may be incomplete or inaccurate. This can lead to flawed hypotheses and erroneous conclusions.
• May be biased: Hypotheses may be biased by the researcher’s own beliefs, values, or assumptions. This can lead to selective interpretation of data and a lack of objectivity in research.
• Cannot prove causation: A hypothesis can only show a correlation between variables, but it cannot prove causation. This requires further experimentation and analysis.
• Limited to specific contexts: Hypotheses are limited to specific contexts and may not be generalizable to other situations or populations. This means that results may not be applicable in other contexts or may require further testing.
• May be affected by chance : Hypotheses may be affected by chance or random variation, which can obscure or distort the true relationship between variables.

About the author

Muhammad Hassan

Researcher, Academic Writer, Web developer

You may also like

Data Collection – Methods Types and Examples

Delimitations in Research – Types, Examples and...

Research Process – Steps, Examples and Tips

Research Design – Types, Methods and Examples

Institutional Review Board – Application Sample...

Evaluating Research – Process, Examples and...

Scientific Hypothesis, Model, Theory, and Law

Understanding the Difference Between Basic Scientific Terms

Hero Images / Getty Images

• Chemical Laws
• Periodic Table
• Projects & Experiments
• Scientific Method
• Biochemistry
• Physical Chemistry
• Medical Chemistry
• Chemistry In Everyday Life
• Famous Chemists
• Activities for Kids
• Abbreviations & Acronyms
• Weather & Climate
• Ph.D., Biomedical Sciences, University of Tennessee at Knoxville
• B.A., Physics and Mathematics, Hastings College

Words have precise meanings in science. For example, "theory," "law," and "hypothesis" don't all mean the same thing. Outside of science, you might say something is "just a theory," meaning it's a supposition that may or may not be true. In science, however, a theory is an explanation that generally is accepted to be true. Here's a closer look at these important, commonly misused terms.

A hypothesis is an educated guess, based on observation. It's a prediction of cause and effect. Usually, a hypothesis can be supported or refuted through experimentation or more observation. A hypothesis can be disproven but not proven to be true.

Example: If you see no difference in the cleaning ability of various laundry detergents, you might hypothesize that cleaning effectiveness is not affected by which detergent you use. This hypothesis can be disproven if you observe a stain is removed by one detergent and not another. On the other hand, you cannot prove the hypothesis. Even if you never see a difference in the cleanliness of your clothes after trying 1,000 detergents, there might be one more you haven't tried that could be different.

Scientists often construct models to help explain complex concepts. These can be physical models like a model volcano or atom  or conceptual models like predictive weather algorithms. A model doesn't contain all the details of the real deal, but it should include observations known to be valid.

Example: The  Bohr model shows electrons orbiting the atomic nucleus, much the same way as the way planets revolve around the sun. In reality, the movement of electrons is complicated but the model makes it clear that protons and neutrons form a nucleus and electrons tend to move around outside the nucleus.

A scientific theory summarizes a hypothesis or group of hypotheses that have been supported with repeated testing. A theory is valid as long as there is no evidence to dispute it. Therefore, theories can be disproven. Basically, if evidence accumulates to support a hypothesis, then the hypothesis can become accepted as a good explanation of a phenomenon. One definition of a theory is to say that it's an accepted hypothesis.

Example: It is known that on June 30, 1908, in Tunguska, Siberia, there was an explosion equivalent to the detonation of about 15 million tons of TNT. Many hypotheses have been proposed for what caused the explosion. It was theorized that the explosion was caused by a natural extraterrestrial phenomenon , and was not caused by man. Is this theory a fact? No. The event is a recorded fact. Is this theory, generally accepted to be true, based on evidence to-date? Yes. Can this theory be shown to be false and be discarded? Yes.

A scientific law generalizes a body of observations. At the time it's made, no exceptions have been found to a law. Scientific laws explain things but they do not describe them. One way to tell a law and a theory apart is to ask if the description gives you the means to explain "why." The word "law" is used less and less in science, as many laws are only true under limited circumstances.

Example: Consider Newton's Law of Gravity . Newton could use this law to predict the behavior of a dropped object but he couldn't explain why it happened.

As you can see, there is no "proof" or absolute "truth" in science. The closest we get are facts, which are indisputable observations. Note, however, if you define proof as arriving at a logical conclusion, based on the evidence, then there is "proof" in science. Some work under the definition that to prove something implies it can never be wrong, which is different. If you're asked to define the terms hypothesis, theory, and law, keep in mind the definitions of proof and of these words can vary slightly depending on the scientific discipline. What's important is to realize they don't all mean the same thing and cannot be used interchangeably.

• Theory Definition in Science
• Hypothesis, Model, Theory, and Law
• What Is a Scientific or Natural Law?
• Scientific Hypothesis Examples
• The Continental Drift Theory: Revolutionary and Significant
• What 'Fail to Reject' Means in a Hypothesis Test
• What Is a Hypothesis? (Science)
• Hypothesis Definition (Science)
• Definition of a Hypothesis
• Processual Archaeology
• The Basics of Physics in Scientific Study
• What Is the Difference Between Hard and Soft Science?
• Tips on Winning the Debate on Evolution
• Geological Thinking: Method of Multiple Working Hypotheses
• 5 Common Misconceptions About Evolution
• Deductive Versus Inductive Reasoning

• Science Notes Posts
• Contact Science Notes
• Todd Helmenstine Biography
• Anne Helmenstine Biography
• Free Printable Periodic Tables (PDF and PNG)
• Periodic Table Wallpapers
• Interactive Periodic Table
• Periodic Table Posters
• How to Grow Crystals
• Chemistry Projects
• Fire and Flames Projects
• Holiday Science
• Chemistry Problems With Answers
• Physics Problems
• Unit Conversion Example Problems
• Chemistry Worksheets
• Biology Worksheets
• Periodic Table Worksheets
• Physical Science Worksheets
• Science Lab Worksheets
• My Amazon Books

Scientific Theory Definition and Examples

A scientific theory is a well-established explanation of some aspect of the natural world. Theories come from scientific data and multiple experiments. While it is not possible to prove a theory, a single contrary result using the scientific method can disprove it. In other words, a theory is testable and falsifiable.

Examples of Scientific Theories

There are many scientific theory in different disciplines:

• Astronomy : theory of stellar nucleosynthesis , theory of stellar evolution
• Biology : cell theory, theory of evolution, germ theory, dual inheritance theory
• Chemistry : atomic theory, Bronsted Lowry acid-base theory , kinetic molecular theory of gases , Lewis acid-base theory , molecular theory, valence bond theory
• Geology : climate change theory, plate tectonics theory
• Physics : Big Bang theory, perturbation theory, theory of relativity, quantum field theory

Criteria for a Theory

In order for an explanation of the natural world to be a theory, it meets certain criteria:

• A theory is falsifiable. At some point, a theory withstands testing and experimentation using the scientific method.
• A theory is supported by lots of independent evidence.
• A theory explains existing experimental results and predicts outcomes of new experiments at least as well as other theories.

Difference Between a Scientific Theory and Theory

Usually, a scientific theory is just called a theory. However, a theory in science means something different from the way most people use the word. For example, if frogs rain down from the sky, a person might observe the frogs and say, “I have a theory about why that happened.” While that theory might be an explanation, it is not based on multiple observations and experiments. It might not be testable and falsifiable. It’s not a scientific theory (although it could eventually become one).

Value of Disproven Theories

Even though some theories are incorrect, they often retain value.

For example, Arrhenius acid-base theory does not explain the behavior of chemicals lacking hydrogen that behave as acids. The Bronsted Lowry and Lewis theories do a better job of explaining this behavior. Yet, the Arrhenius theory predicts the behavior of most acids and is easier for people to understand.

Another example is the theory of Newtonian mechanics. The theory of relativity is much more inclusive than Newtonian mechanics, which breaks down in certain frames of reference or at speeds close to the speed of light . But, Newtonian mechanics is much simpler to understand and its equations apply to everyday behavior.

Difference Between a Scientific Theory and a Scientific Law

The scientific method leads to the formulation of both scientific theories and laws . Both theories and laws are falsifiable. Both theories and laws help with making predictions about the natural world. However, there is a key difference.

A theory explains why or how something works, while a law describes what happens without explaining it. Often, you see laws written in the form of equations or formulas.

Theories and laws are related, but theories never become laws or vice versa.

Theory vs Hypothesis

A hypothesis is a proposition that is tested via an experiment. A theory results from many, many tested hypotheses.

Theory vs Fact

Theories depend on facts, but the two words mean different things. A fact is an irrefutable piece of evidence or data. Facts never change. A theory, on the other hand, may be modified or disproven.

Difference Between a Theory and a Model

Both theories and models allow a scientist to form a hypothesis and make predictions about future outcomes. However, a theory both describes and explains, while a model only describes. For example, a model of the solar system shows the arrangement of planets and asteroids in a plane around the Sun, but it does not explain how or why they got into their positions.

• Frigg, Roman (2006). “ Scientific Representation and the Semantic View of Theories .”  Theoria . 55 (2): 183–206. 
• Halvorson, Hans (2012). “What Scientific Theories Could Not Be.”  Philosophy of Science . 79 (2): 183–206. doi: 10.1086/664745
• McComas, William F. (December 30, 2013).  The Language of Science Education: An Expanded Glossary of Key Terms and Concepts in Science Teaching and Learning . Springer Science & Business Media. ISBN 978-94-6209-497-0.
• National Academy of Sciences (US) (1999). Science and Creationism: A View from the National Academy of Sciences (2nd ed.). National Academies Press. doi: 10.17226/6024  ISBN 978-0-309-06406-4. 
• Suppe, Frederick (1998). “Understanding Scientific Theories: An Assessment of Developments, 1969–1998.”  Philosophy of Science . 67: S102–S115. doi: 10.1086/392812

Related Posts

Understanding contexts: how explanatory theories can help

Affiliations.

• 1 , Lexington, USA. [email protected].
• 2 Geisel School of Medicine, Dartmouth College, Hanover, NH, 03755, USA. [email protected].
• PMID: 30841932
• PMCID: PMC6404339
• DOI: 10.1186/s13012-019-0872-8

Objective: To rethink the nature and roles of context in ways that help improvers implement effective, sustained improvement interventions in healthcare quality and safety.

Design: Critical analysis of existing concepts of context; synthesis of those concepts into a framework for the construction of explanatory theories of human environments, including healthcare systems.

Data sources: Published literature in improvement science, as well as in social, organization, and management sciences. Relevant content was sought by iteratively building searches from reference lists in relevant documents.

Results: Scientific thought is represented in both causal and explanatory theories. Explanatory theories are multi-variable constructs used to make sense of complex events and situations; they include basic operating principles of explanation, most importantly: transferring new meaning to complex and confusing phenomena; separating out individual components of an event or situation; unifying the components into a coherent construct (model); and adapting that construct to fit its intended uses. Contexts of human activities can be usefully represented as explanatory theories of peoples' environments; they are valuable to the extent they can be translated into practical changes in behaviors. Healthcare systems are among the most complex human environments known. Although no single explanatory theory adequately represents those environments, multiple mature theories of human action, taken together, can usually make sense of them. Current mature theories of context include static models, universal-plus-variable models, activity theory and related models, and the FITT framework (Fit between Individuals, Tasks, and Technologies). Explanatory theories represent contexts most effectively when they include basic explanatory principles.

Conclusions: Healthcare systems can usefully be represented in explanatory theories. Improvement interventions in healthcare quality and safety are most likely to bring about intended and sustained changes when improvers use explanatory theories to align interventions with the host systems into which they are being introduced.

• Cooperative Behavior
• Delivery of Health Care / standards*
• Human Activities
• Models, Theoretical
• Organizational Innovation
• Patient Safety / standards*
• Quality Improvement / standards*

• Methodology
• Open access
• Published: 06 March 2019

Understanding contexts: how explanatory theories can help

• Frank Davidoff   ORCID: orcid.org/0000-0001-7924-4005 1 , 2  

Implementation Science volume  14 , Article number:  23 ( 2019 ) Cite this article

25k Accesses

32 Citations

39 Altmetric

Metrics details

To rethink the nature and roles of context in ways that help improvers implement effective, sustained improvement interventions in healthcare quality and safety.

Critical analysis of existing concepts of context; synthesis of those concepts into a framework for the construction of explanatory theories of human environments, including healthcare systems.

Data sources

Published literature in improvement science, as well as in social, organization, and management sciences. Relevant content was sought by iteratively building searches from reference lists in relevant documents.

Scientific thought is represented in both causal and explanatory theories. Explanatory theories are multi-variable constructs used to make sense of complex events and situations; they include basic operating principles of explanation, most importantly: transferring new meaning to complex and confusing phenomena; separating out individual components of an event or situation; unifying the components into a coherent construct (model); and adapting that construct to fit its intended uses. Contexts of human activities can be usefully represented as explanatory theories of peoples’ environments; they are valuable to the extent they can be translated into practical changes in behaviors.

Healthcare systems are among the most complex human environments known. Although no single explanatory theory adequately represents those environments, multiple mature theories of human action, taken together, can usually make sense of them. Current mature theories of context include static models , universal-plus-variable models , activity theory and related models , and the FITT framework (Fit between Individuals, Tasks, and Technologies). Explanatory theories represent contexts most effectively when they include basic explanatory principles.

Conclusions

Healthcare systems can usefully be represented in explanatory theories. Improvement interventions in healthcare quality and safety are most likely to bring about intended and sustained changes when improvers use explanatory theories to align interventions with the host systems into which they are being introduced.

Peer Review reports

Introduction

Human contexts—defined in this commentary primarily as the meaning of human environments to the people who live and work in them—are major determinants of the effectiveness and generalizability of interventions to improve healthcare quality and safety [ 1 , 2 ]. Despite the importance of contex, much about it remains obscure, as do the specific mechanisms by which local contexts affect the implementation of improvement interventions. As a consequence, context is still sometimes vaguely referred to in scholarly work as “All those things in the situation which are relevant to meaning in some sense, but which I haven’t identified.”([ 2 ] p. 6).

Context plays an important role in both improvement science and implementation science; limited understanding of context therefore limits understanding of both the fundamental principles of improvement and the actions that put improvements into practice. Achieving deep understanding of context is a challenge that has baffled serious improvers, researchers, and scholars for years [ 2 ]. This difficulty [ 3 ] suggests that multiple complementary explanatory theories might prove more useful than any single theory in understanding both context in general and specific local contexts.

This commentary is intended as a complement to the SQUIRE guidelines for publication of work in quality improvement [ 4 ]. It explores the premise that explanatory theories of human environments can help improvers work flexibly from first principles rather than rigid formulas, and, as is true for good theories generally [ 6 ], can provide improvers with explicit reasons why particular interventions are likely to be effective in specific systems; it examines the nature of explanatory theories and the basic principles of explanation, considers the contributions of those principles to mature (i.e., fully-developed, refined) explanatory theories of complex human environments, and considers the nature of the data needed in constructing explanatory theories of local environments, and the methods used for gathering the requisite data. The commentary proposes, finally, that it is both appropriate and useful early in the planning of an improvement program, to create an explanatory theory of the local healthcare environment into which planned intervention is to be introduced, then use that theory in linking the intervention with that environment. The commentary also encourages improvers to reconsider and revise the initial explanatory theories from time to time as more is learned about the local environment during the improvement process.

Explanatory theories

Scientific thought is built primarily around two complementary mental constructs [ 5 ]: causal and explanatory theories. Explanatory theories are created to help to understand complex, confusing events and situations; they often also serve as sources of testable causal theories of events and situations.

Although explanatory theories are sometimes thought to play a less central role in science than causal theories [ 5 , 6 , 7 ], many explanatory theories— including the theory of evolution, the periodic table of the elements, and the structure of DNA—have proven uniquely helpful in understanding important phenomena in natural sciences. Political science is built largely around explanatory theories [ 7 ]; process flow diagrams and Pareto charts are among the explanatory theories that help understand events and situations in improvement science [ 8 ].

The concepts in this commentary were developed from the published literature in improvement science as well as the social, behavioral, organizational, improvement, and management sciences. Sources that proved especially important include Bate et al. [ 2 ] on the dynamic properties of context, Squires et al. [ 3 ] on the construction of explanatory theories, Braithwaite et al. [ 9 ] and Greenhalgh [ 10 ] on complexity, Nardi [ 11 ] and Greenhalgh et al. [ 12 ] on theories of human action, Vandenbroucke [ 5 ] and Clarke and Primo [ 7 ] on explanatory theory, and Pitt [ 13 ] on the fundamentals of explanation. Literature searches were built out from reference citations in these and related publications.

The author’s experience as editor of a major clinical journal ( Annals of Internal Medicine ), and as publications editor at the Institute for Healthcare Improvement (Cambridge, MA), also helped in constructing this commentary. Discussions in the improvement science development group at the Health Foundation (London, UK) and in the Standards for Quality Improvement Reporting Excellence (SQUIRE) leadership group [ 4 ] also contributed importantly to this effort.

The complexity and dynamism of human environments

The most salient properties of human environments are arguably their complexity [ 9 , 10 ] and their dynamic nature [ 2 ]. This commentary rests on the concept of “complex systems” summarized in Table  1 .

The degrees of complexity in human systems are usefully characterized in the following schema [ 14 ], in which the cooking of a specific dish is represented as simple . Challenges at this basic level are usually managed successfully by following explicit, straightforward recipes or protocols.

By comparison, sending a rocket to the moon is complicated for multiple psychological, social, and technical reasons. Successful management of complicated challenges often requires the use of dedicated management tools such as checklists (mainly to overcome the limitations of human memory) and protocols that map out contingency-dependent decision points (mainly to avoid oversimplification).

Finally, the challenge of raising a child can be seen as  complex , largely because it involves such a large number of variables, many of them poorly defined, which often leads to unpredictable outcomes, e.g., when the experience of raising one child successfully is of little use in raising the next.

Principles of explanation (sense-making)

Although a human event or situation can sometimes be explained adequately in terms of causal mechanisms, the inherent complexity and dynamic nature of events and situations usually requires explanations that go beyond causality and include descriptive explanatory principles [ 5 , 6 , 10 , 13 , 14 , 15 , 16 ]. Most importantly, those principles include transferring new meaning to the event or situation, establishing its familiarity and internal logic, separating out its individual components, unifying its components into a coherent mental construct or “gestalt”, and adapting the explanation to fit its intended uses.

Transferring (sharing) meaning

The classic human system for transferring or sharing meaning is, of course, language [ 17 ]: witness the substantial loss or distortion of its meaning that results when a word or phrase is taken out of context, and conversely the greater precision of a literature search that uses search terms embedded in linguistic contexts, when contrasted with a search that uses search terms lacking such embellishment [ 18 , 19 ]. (Salmon proposes that the transfer of information, energy or causal inference between processes is more meaningful than transfer between events [ 16 ].)

Familiarity

Familiarity, by itself, is neither necessary nor sufficient to make sense of an event or situation. But familiarity is nonetheless an important component of explanation, because a sense of familiarity provides a sense of understanding ([ 20 ], p. 52). Metaphor is often the chosen mechanism for transferring meaning from familiar things to those that are less familiar, a property that prompted Aristotle to comment that it is metaphor that most produces knowledge. The psychologist Julian Jaynes has argued that metaphor is not a “mere extra trick of language” but is rather “the very constitutive ground of language,” and that “it is by metaphor that language grows” ([ 20 ], pp. 48-9).

Explanation in natural sciences is usually considered adequate when its logic is clear, as when statement of a general law (a “regularity”) is coupled with statement of a specific antecedent condition. In physics, for example, a statement such as “All wave phenomena of a certain type satisfy the law of refraction, and light is a wave of that type” is accepted as a logical construct that meaningfully explains the refraction of light ([ 13 ] p. 10]).

Separating out and unifying components

By themselves, the individual components of an event or situation ordinarily have little if any inherent meaning. But the construct that results when those components are brought together to make a coherent whole (usually as a narrative, map, model, or mathematical expression) is uniquely helpful in making sense of that event or situation [ 4 , 21 , 22 ]. Important new meanings can emerge as well—often unexpectedly—from the resulting construct. For these reasons, some philosophers of science consider unification of a phenomenon’s individual components into a coherent whole as the main principle by which explanation renders a phenomenon understandable [ 4 , 5 , 21 , 22 ].

The sharing of meaning among a phenomenon’s individual components finds expression in catch-phrases such as the jigsaw puzzle effect , and “The whole is greater than the sum of its parts.” On a more grand scale, the theory of evolution is said to acquire its explanatory power when “an apparently modest allegiance to mere fact gathering” abruptly crystallizes into a “whole world view” [ 23 ].

Details of the mental process through which unification creates explanations unfortunately remain obscure. And curiously, even a highly coherent construct of an event or situation does not necessarily help understand whether its components are truly independent, whether the interactions among them are uni-directional or recursive, and which components (if any) are most important in determining its overall behavior. Moreover, craftspeople such as watchmakers and car mechanics understand that success in their work depends on their ability to separate out the components of the complex systems they are called on to assemble or repair (disaggregate them) at least as much as on their ability to understand how the components contribute collectively to an event or situation’s overall behavior (unify them). At least in theory, the explanatory principles of disaggregation and unification appear to contradict each other, but in practice, the two principles are often complementary. In managing a human system, for example, a leader’s ability to unify various groups’ individual modes of decision-making can complement his or her ability to distinguish those modes from one another [ 24 ].

Adapting explanations

Explanatory theories are arguably successful to the extent people can translate them into practical implementation behavior—e.g., manage the environments in which they live and work or predict the likelihood that a specific event will happen in the future ([ 16 ] p. 77). Not surprisingly, therefore, the explanatory theories people develop on their own to manage their personal environments differ substantially from the ones they develop collectively to further the work of the organizations in which they work. For similar reasons, personal and organization-related explanatory theories differ from those that outside researchers create to understand these various environments.

Personal contexts

Peoples’ intense, universal need to give meaning to “the brutal aboriginal flux” of their lived experience [ 1 ] suggests that humans can be defined as “reason-giving animals” [ 25 ]. They begin creating explanatory theories of their personal environments at an extremely early age [ 26 ], then extend and refine those theories as they and their personal environments change over time. Personal explanatory theories are usually implicit and poorly articulated; they can also be distorted, incomplete, or inappropriate since they frequently lack independent reality testing.

Organizational and professional contexts

Workers in organizations are called on to create explanatory models that make sense of the internal structure and function of those organizations, as well as of the external environments in which their organizations are embedded. Weick et al. describe this work as a creative, collaborative undertaking that involves “language, talk, and communication” and is “ongoing, subtle, swift, social, and easily taken for granted” [ 1 , 27 , 28 ]. Early in this sense-making process, workers in an organization “bracket” information (i.e., identify items they see as especially relevant to their particular situation), then name (label) those items, which stabilizes the streaming of their experience [ 1 ].

The way people in organizations envision events and situations also immediately begins their social and administrative work of organizing, because bracketing and labeling events predisposes them to find common ground and provides them with a set of cognitive categories, plus a typology of potential actions. (Bracketing central venous catheter infection and labeling it as primarily a social rather than a biological problem [ 29 ] played an important role in shaping an intervention that successfully lowers the infection rate [ 30 ].) Workers then use such newly defined contextual elements as they literally talk their organization-related explanatory theories into existence [ 1 ].

The sense-making process described above closely resembles the one that professionals in applied disciplines, together with their clients, use to make sense of the problem situations they are called on to manage ([ 31 ], pp. 267–83). More specifically, medical professionals will recognize its resemblance to the process by which they and their patients formulate the essential explanatory theories they know as diagnoses .

Mature explanatory theories of human environments

People initially sketch out rough explanatory theories of environments which usually involve basic principles of explanation, then subsequently broaden and refine these nascent constructs into more mature theories. Important examples of such mature explanatory theories include static theories , universal-plus-variable theories , activity theory and related general theories of human action , and the FITT framework (Fit between Individuals, Task, and Technology).

Static theories

Several research groups have developed explanatory theories of outstanding healthcare systems by selecting the components they judge to be most closely associated with certain systems’ ability to deliver exceptionally safe, high-quality care [ 32 , 33 , 34 , 35 , 36 ], then assembling those components into structured models. (A recent international effort is engaged in constructing a new and more meaningful theory of this type [ 3 ]).

The individual components identified in these theory-building exercises—buildings, equipment, leadership, geographic location, teaching status, financial and intellectual resources, and the like—are quite heterogeneous and the resulting constructs often pay little attention to functional relationships among the components or to the ways in which the process of care plays out over time for individual patients. Metaphorically speaking, then, explanatory theories such as these describe the anatomy of exceptional healthcare environments, but not their physiology ; that is, they are static , which could account for the limited ability of this type of explanatory theory to explain variation in the effectiveness of improvement interventions across different healthcare systems.

Universal-plus-variable models

Working from detailed on-site observations in high-performing healthcare systems, Bate et al. [ 2 , 37 ] have constructed a generalized explanatory theory of such systems. Their experience is reflected in their comment that “although research has provided an abundance of data on key success factors in QI efforts, very little was previously known about how these combine and interact with each other in the improvement process over time.” They comment further that the context of a healthcare system is “a process; dynamic, fluid, and constantly moving, not lumpen, material, or static,” and that “it is the dynamic and ongoing interaction between [the domains of an environment] rather than any one of them individually or independently, that accounts for the effectiveness of a QI intervention,” as well as for “the striking variation between similar QI interventions in different places” ([ 2 ]p. 11).

These investigators then refine and sharpen the focus of their emerging explanatory theory by postulating that a healthcare system’s ability to deliver outstanding care lies in the combination of the two major components— universals and variables —that characterize an organization’s local situation. More specifically, they identify the challenges inherent in several distinct areas—physical/technological, emotional, educational, cultural and political, and structural—as the universals in all healthcare organizations; they also characterize the actions that individual workers and groups take in response to those challenges as differing both within and across organizations to the point where those actions and the possible combinations among them can be assumed to be “practically innumerable” ([ 37 ], p. 168), i.e., they are the variables .

The resulting universal-plus-variable explanatory theory of human contexts gains plausibility from its affinity with other established cognitive systems in which people represent the complex meanings that matter to them. The best known and arguably most important of such systems is of course language [ 17 ]; people produce language by embedding differing strings of individual words (the variables) in a relatively small number of stable grammatical structures (the universals). They then use the resulting construct to create a virtually unlimited number of statements that are meaningful to others, even though many of those statements have not been seen or heard previously.

Music provides another illuminating example of a meaningful universal-plus-variable explanatory theory [ 39 ]. Composers in each musical tradition embed differing arrays of tones (variables) in a limited set of stable, widely recognized harmonic constructs (universals). One critic has elegantly captured this explanatory theory of music (or at least of Western music) in his pithy comment that “Mozart used the same B-flat as everyone else.”

Activity theory and related models of human action

The universal-plus-variable explanatory theory of contexts also resonates with several earlier mature explanatory theories of human action, including Activity Theory and related models [ 11 ]. Some of these action theories are now seen as especially useful in understanding the interaction between people and computer systems [ 12 ]. In these theories, it is precisely the ongoing bi-directional interaction between static human environments and the dynamic needs, interests, and experiences people bring to encounters with those environments that creates most of the contexts (meanings) of human life. For example, context is understood as follows in Activity Theory as an overarching, albeit secondary, consideration: “[W]hat takes place in an activity system composed of object, actions, and operations, is the context… [C]ontext is not an outer container or shell inside of which people behave in certain ways.” Context in these theories is thus “both internal to people…and at the same time, external to people” [ 11 ], i.e., as an integrated whole. This unifying perspective invalidates “simplistic explanations that divide internal and external, and schemes that see context as external to people.”

The FITT framework (Fit between Individuals, Tasks, and Technologies)

Developed largely to explain the adoption of information and communication technologies (IT) [ 40 ], the FITT framework clearly distinguishes an organization’s established and widely recognized tasks and technologies from its workers’ shifting dynamic behaviors [ 5 , 12 , 40 , 41 ] (Table  2 ), and in that respect, it resembles other universal-plus-variable explanatory theories of human activity.

As noted elsewhere [ 6 ], the FITT framework has been used to guide the successful implementation of an innovative electronic order system for post-operative surgical care [ 41 ]. Researchers in that study explicitly used the FITT framework to help them interweave their new electronic system with the healthcare environment in which they implemented it.

The nature of data needed to construct explanatory theories of healthcare environments

Adequate understanding of human environments requires that explanatory theories take the enormous complexity of those environments appropriately into account. Although complexity of this magnitude can be a cause for despair among improvers and researchers, the statistician George Box’s pungent comment that “All theories are wrong, but some are illuminating and useful” offers reassurance that creating explanatory theories of human environments, including healthcare systems, is likely on balance to be worth the effort.

Data used to create meaningful explanatory theories of human environments

Creating explanatory theories of human environments that help implement successful improvement interventions apparently requires open-ended, multi-level data on working relationships in organizations [ 1 , 9 , 10 , 11 , 29 , 31 , 36 , 37 , 38 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 ]. Research groups are now laboring to clarify the essential nature of such data (Table  3 ), while also obtaining insights into effective techniques for collecting and analyzing those data (Table  4 ).

It is important to note in this connection that improvement interventions reach their full potential more successfully when their implementation builds on the complexity of the systems they intend to change than when they underestimate or ignore that complexity [ 9 ]. Even documenting that a healthcare system has “a long way to go” to achieve specific solutions within each of the six universal challenge area (in contrast to being either “some way there” or “already there”) can help improvers pinpoint current gaps and opportunities in that system’s quality and safety, and facilitate productive discussions on their future improvement efforts (Cf. Codebook for Quality Improvement Practice, for example) ([ 37 ], p., 177).

In like fashion, answering a question regarding organizational complexity (e.g., “How did this practice miss a diagnosis?”) can be more effective in changing system performance than obtaining answering a narrowly focused question such as “How did an individual practitioner miss a diagnosis?”) [ 42 , 43 , 44 , 45 , 46 , 47 , 48 ].

Traditional scientific methods will undoubtedly continue over time to help understand human environments, including environments that are as complex and dynamic as healthcare systems. At the same time, the difficulty of understanding those environments in the concepts and language of sciences suggests that explanatory theories of those environments will be more meaningful when they include contributions from the arts and humanities.

An important, and intriguing, painting by the Belgian surrealist René Magritte hints at the potential of such an ecumenical approach. In this work, Magritte apparently tries to represent the complex, emotionally freighted world of tobacco use by juxtaposing the image of a tobacco pipe with a written comment: “Ceci n’est pas une pipe” (“This is not a pipe”). The resulting cognitive dissonance suggests the artist’s intent is to increase the painting’s impact by cautioning his viewers that “This is only the image of a pipe, not the actual object; don’t confuse the two,” and encouraging them not to mistake the part for the whole (a pipe is, after all, only one small part of tobacco smoking).

But he does not stop there: in his effort to jolt viewers toward even deeper and more precise awareness of tobacco use, Magritte resorts to a particularly unorthodox representation of the pernicious habit, when he flatly asserts that “a pipe actually isn’t a pipe,” his surrogate for a paradoxical characterization of tobacco use in terms of what it is not . Examples of this startling apophatic (i.e., reverse) way to represent complex, confusing realities are now appearing in the literature of improvement science, as in “wake-up calls” telling us that  neither a checklist of infection control measures [ 49 ] nor a surgical safety checklist  [ 50 ], by itself, is an improvement intervention (the unstated subtext being that successful, sustained improvement absolutely requires explicit, extensive coordination, and tight linkage, between the intervention and the environment in which it is being implemented).

In articulating her explanatory theory of the world of falconry , the scholar and writer Helen Macdonald also turns, as follows, to this paradoxical, inverse way of understanding the deeper meaning of a complex human environment [ 51 ]:

“[T]here is a world of things out there – rocks and trees and grass and all the things that crawl and run and fly. They are all things in themselves, but we make them sensible to us by giving them meanings that shore up our own views of the world. In my time [living with and training my goshawk] Mabel I’ve learned how you feel more human once you have known, even in your imagination, what it is likely to be not”.

This commentary considers evidence that reinforces the crucial reality that the healthcare systems in which improvement programs take place—or, more specifically, the values and character of those systems—are at least as important in improving care as the specifics of the improvement interventions themselves. This obvious but often underappreciated reality environmental feature argues strongly for the development of sophisticated, nuanced understanding of those environments early in the implementation of improvement programs, and consistent application of that understanding during the improvement process. Realistically, understanding a human environment—especially one as complex and dynamic as a healthcare system—is an arduous, demanding undertaking, which further underscores the value of building a basic set of context-related initiatives into the implementation of any sizeable healthcare improvement program. These initiatives might include the following:

As early as possible in planning the program, create an explanatory theory of the host environment that incorporates the basic principles of explanation, especially unification of the environment’s major components;

If possible, involve social scientists, as well as professionals from humanities (e.g., creative writers, reporters, historians, graphic artists and the like) in the development of that explanatory theory;

Use that explanatory theory in coordinating and linking the intervention with the host environment;

Explore the use of established mature explanatory theories, individually or collectively, in making sense of the local host environment;

Assess the relative importance of the environment’s major components as determinants of its nature and behavior; its successes and failures;

From time to time, review the most current version of the explanatory theory and revise it if necessary as more is learned about the host environment and about the interaction between environment and intervention

To avoid creating jitter and instability in the program, resist unnecessary tinkering with the makeup and application of the explanatory theory;

Make explicit efforts to assure that all members of the improvement team are familiar with the major components of the host environment, and understand how those components fit/work together;

Adapt the focus, comprehensiveness, organization, and level of detail of the explanatory theory of the host environment, to make it as useful as possible for its most important users.

Weick KE, Sutcliffe KM, Obstfeld D. Organizing and the process of sense-making. Organization Sci. 2005;16:409–21.

Article   Google Scholar  

Bate P. Context is everything. In: Perspectives on context. London: The Health Foundation; 2014. p. 1–29.

Google Scholar  

Squires JE, Graham ID, Hutchinson AM, Michie S, Francis, JJ, Sales A, et al. Identifying the domains of context important to implementation science: a study protocol. Implement Sci. 2015;10:135. Doi: 10:1186/s13012–015-0325-y.

Ogrinc G, Davies L, Goodman D, et al. SQUIRE 2.0 Standards for Quality Improvement Reporting Excellence: revised publication guidelines from a detailed consensus process. BMJ Qual Saf. 2015. https://doi.org/10.1136/bmjqs-2015-004411 .

Vandenbroucke JP. Observational research, randomized trials, and two views of medical science. PLoS Med. 2008;5:c67. https://doi.org/10.1371/journal.pmed.0050067 .

Davidoff F, Dixon-Woods M, Leviton L, Michie S. Demystifying theory and its use in improvement. BMJ Qual Saf 2015;24:228–238. doi 1136/bmjqs-2014-003627.

Clarke KA, Primo DM. A model discipline. Political science and the logic of representations. New York: Oxford University Press; 2012.

Berwick DM, Godfrey AB, Roessner J. Curing healthcare: new strategies for quality improvement. San Francisco: Jossey-Bass; 2003.

Braithwaite, J., Churruca, K., Ellis, LA, et al. Complexity science in healthcare –aspirations, approaches, applications and accomplishments: a white paper. Australian Institute of Health Innovation, Macquarie University: Sydney; 2017; https://www.researchgate.net/publications/319643112 . Accessed 1–23-19.

Greenhalgh T. Systems. In: How to implement evidence-based medicine. Hoboken: John Wiley & Sons; 2018.

Nardi BA. Studying context: A comparison of activity theory, situated action models, and distributed cognition. In Activity Theory and Human-Computer Interaction. Cambridge: MIT Press; 1995:69-102.

Greenhalgh T, Stones R. Theorising big IT programs in healthcare: Strong structuration theory meets actor-network theory. Soc Sci Med; 2010;70:1285–1294. doi: 10.10.16/j.socscimed.2009.12.034.

Pitt JC. Theories of explanation. New York: Oxford University Press; 1988.

Glouberman S, Zimmerman B. Complicated and complex systems: what would successful reform of Medicare look like? Toronto: Commission on the Future of Health Care in Canada; 2002. Accessed 8–24-18

Hempel CG, Oppenheim P. Studies in the logic of explanation. In: Pitt JC, editor. Theories of explanation. New York: Oxford University Press; 1988. p. 9–50.

Salmon WC. Statistical explanation and causality. In: Pitt JC, editor. Theories of explanation. New York: Oxford University Press; 1988. p. 75–118.

Radford A. Transformational grammar. Cambridge (UK): Cambridge University Press; 1988.

Book   Google Scholar  

Purcell GP, Shortliffe EH. Contextual models of clinical publications for enhancing retrieval from full-text databases. Proc Annu Symp Comput Appl Med Care. 1995:851–7.

Moskovitch P, Martins SB, Behiri E, et al. A comparative evaluation of full-text, concept-based, and context-sensitive search. J Am Med Inform Assoc. 2007;14(2):164–74 Epub 2007 Jan 9.

Jaynes J. The origin of consciousness in the breakdown of the bicameral mind. Boston: Houghton Mifflin; 1990.

Kitcher P. Explanatory Unification, In Pitt, reference 11, pp. 167–87 .

Sheridan TB. Modeling human-system interaction. Hoboken: John Wiley & Sons; 2017.

Gopnik, A. Rewriting nature. Charles Darwin, natural novelist. The New Yorker, October 23, 2006.

Wears RL, Barsky CL, Perry DJ. Human Factors in Organizational Design and Management. In: Broberg O, Fallentin N, Hasle P, et al., editors. Control modes in care delivery organizations: Nordic Ergonomics Society Annual Conference, XI; 2014. p. 957–6.

Tilly C. Why? What happens when people give reasons…and why. Princeton University Press: Princeton; 2006.

Schulz L. Infants explore the unexpected. Science. 2015;348:42–3.

Article   CAS   Google Scholar  

Weick KE. Making sense of the organization. The impermanent organization. Volume 2. Chichester, West Sussex (UK): John Wiley & Sons; 2009.

Daft RI, Weick KE. Toward a model of organizations as interpretation systems. Acad Manag Rev. 1984;9:284–95.

Dixon-Woods M, Bosk CL, Aveling EL, et al. Explaining Michigan: developing an ex post theory of an improvement program. Milbank Q. 2011;89:167–205. https://doi.org/10.1111/j.1468-0009.2011.00625.x . https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3142336/ . Accessed 12 Mar 2018.

Pronovost P, Needham D, Berenholtz S, et al. An intervention to decrease catheter-related bloodstream infections in the ICU. N Engl J Med. 2006;355:2725–32. https://doi.org/10.1056/NEJMoa061115 .

Article   CAS   PubMed   Google Scholar  

Schön DA. The reflective practitioner. Aldershot, Hants, UK: Ashgate; 1991.

Bradley EH, Curry LA, Cherlin E, et al. Hospital strategies for reducing risk-stratified mortality rates in acute myocardial infarction. Ann Intern Med. 2012;156:618–26. https://doi.org/10.7326/0003-4819-156-9-201205010-00003 .

Article   PubMed   PubMed Central   Google Scholar  

Kaplan HC, Provost LP, Froehle CM, et al. The model for understanding success in quality (MUSIQ): building a theory of context in healthcare quality improvement. BMJ Qual Saf. 2012;21:13–20. https://doi.org/10.1136/bmjqs-2011-000010 .

Article   PubMed   Google Scholar  

Damschroder LJ, Aron DC, Keith RE, et al. Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science. Implement Sci. 2009;4:50. https://doi.org/10.1186/1748-5908-4-50 accessed 12 Mar 2018.

Taylor SL, Dy S, Foy R, et al. What context features might be important determinants of the effectiveness of patient safety practice interventions? BMJ Qual Saf. 2011;20:611–7. https://doi.org/10.1136/bmjqs.2010.049379 Epub 2011 May 26.

Gustafson DH, Sainfort F, Eichler M, et al. Developing and testing a model to predict outcomes of organizational change. Health Serv Res. 2003;38:751–76.

Bate P, Mendel P, Robert G. Organizing for quality. New York: Radcliffe Publishing; 2008.

Greenhalgh T, Stones R. Theorizing big IT programs in healthcare: strong structuration theory meets actor-network theory. Soc Sci Med . 2010;70:1285–1294. doi 0.10.16/j.socscimed. 2009.12.034.

Bernstein L. The unanswered question: whither music? Cambridge, MA: Harvard University Press; 1981.

Ammenwerth E, Iller C, Mahler C. IT-adoption and the interactions of task, technology, and individuals: a fit framework and a case study. BMC Med Inform Decis Mak. 2006;6:3.

Lesselroth BJ, Yang J, McComachie J, et al. Addressing the sociotechnical drivers of quality improvement: a case study of post-operative DVT prophylaxis computerized decision support. BMJ Qual Saf. 2011;20:381–9. https://doi.org/10.1136/bmjqs.2010.042689 Epub 2011 Jan 5.

Hawe P, Reilly T. Ecological theory in practice. Illustrations from a community-based intervention to promote the health of recent mothers. Prev Sci. 2005;6:227–36. https://doi.org/10.1007/s11121-005-0009-z .

Lanham HJ, McDaniel RR, Crabtree BF, Miller WL, Stange KC, Talka AF, et al. How improving practice relationships among clinicians and nonclinicians can improve quality in primary care. Jt Comm J Qual Patient Saf 2009;35:457–456; PMCID:2928073.

Leykum LK, Pugh J, Lawrence V, et al. Organizational interventions employing principles of complexity science have improved outcomes for patients with type II diabetes. Implementation Sci. 2007;2(28). https://doi.org/10.1186/1748-5908-2-28 .

Leykum LK, Parchman M, Pugh J, et al. The importance of organizational characteristics in patients with chronic disease: a systematic review of congestive heart failure. Implementation Sci. 2010;5:66. https://doi.org/10.1186/1748-5908-5-66 .

Hawe P, Shiell A, Riley T. Theorizing events in systems. Am J Community Psychol. 2009;43:267–76. https://doi.org/10.1007/s10464-009-9229-9 .

Leykum LK, Lanham JH, Provost SM, et al. Improving outcomes of hospitalized patients: the physician relationships, improvising, and sense making protocol. Implementation Sci. 2014;9:171. https://doi.org/10.1186/s13012-014-0171-3 .

McAllister C, Leykum LK, Lanham H, et al. Relationships within impatient physician housestaff teams and their association with hospitalized patient outcomes. J Hosp Medicine. 2014;9:764–71. https://doi.org/10.1002/jhm.2274 Epub 2014 Oct 30.

Bosk CL, Dixon-Woods M, Goeschel CA, Pronovost PJ. Reality check for checklists. Lancet 2009;374:444–445. PMID:19681190.

Aveling EL, McCulloch P, Dixon-Woods M. A qualitative study comparing experiences of the surgical safety checklist in hospitals in high-income and low-income countries. BMJ Open. 2013:e003039. https://doi.org/10.1136/bmjopen-2013-003039 .

Macdonald H. H is for hawk. New York: Grove Press; 2014.

Download references

Acknowledgements

The author gratefully acknowledges useful comments of Paul Batalden, Trisha Greenhalgh, Mary Dixon-Woods, Lucian Leape, Tom Sheridan, Cyrus Hopkins, and Judith Singer on earlier versions of this article.

Acronym for Standards for Quality Improvement Reporting Excellence.

No funding was received for this work.

Availability of data and materials

Not applicable, because this analysis involves no original research data.

Author information

Authors and affiliations.

Lexington, USA

Frank Davidoff

Geisel School of Medicine, Dartmouth College, Hanover, NH, 03755, USA

You can also search for this author in PubMed   Google Scholar

Contributions

The author gathered all the reference material, drafted the initial versions of the paper and all subsequent revisions, and takes responsibility for the entire content of the article.

Corresponding author

Correspondence to Frank Davidoff .

Ethics declarations

Ethics approval and consent to participate.

Not applicable.

Consent for publication

Competing interests.

The author declares no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License ( http://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Cite this article.

Davidoff, F. Understanding contexts: how explanatory theories can help. Implementation Sci 14 , 23 (2019). https://doi.org/10.1186/s13012-019-0872-8

Download citation

Received : 25 July 2018

Accepted : 18 February 2019

Published : 06 March 2019

DOI : https://doi.org/10.1186/s13012-019-0872-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Implementation Science

ISSN: 1748-5908

• Submission enquiries: Access here and click Contact Us
• General enquiries: [email protected]

What is a scientific hypothesis?

It's the initial building block in the scientific method.

Hypothesis basics

What makes a hypothesis testable.

• Types of hypotheses
• Hypothesis versus theory

Additional resources

Bibliography.

A scientific hypothesis is a tentative, testable explanation for a phenomenon in the natural world. It's the initial building block in the scientific method . Many describe it as an "educated guess" based on prior knowledge and observation. While this is true, a hypothesis is more informed than a guess. While an "educated guess" suggests a random prediction based on a person's expertise, developing a hypothesis requires active observation and background research. 

The basic idea of a hypothesis is that there is no predetermined outcome. For a solution to be termed a scientific hypothesis, it has to be an idea that can be supported or refuted through carefully crafted experimentation or observation. This concept, called falsifiability and testability, was advanced in the mid-20th century by Austrian-British philosopher Karl Popper in his famous book "The Logic of Scientific Discovery" (Routledge, 1959).

A key function of a hypothesis is to derive predictions about the results of future experiments and then perform those experiments to see whether they support the predictions.

A hypothesis is usually written in the form of an if-then statement, which gives a possibility (if) and explains what may happen because of the possibility (then). The statement could also include "may," according to California State University, Bakersfield .

Here are some examples of hypothesis statements:

• If garlic repels fleas, then a dog that is given garlic every day will not get fleas.
• If sugar causes cavities, then people who eat a lot of candy may be more prone to cavities.
• If ultraviolet light can damage the eyes, then maybe this light can cause blindness.

A useful hypothesis should be testable and falsifiable. That means that it should be possible to prove it wrong. A theory that can't be proved wrong is nonscientific, according to Karl Popper's 1963 book " Conjectures and Refutations ."

An example of an untestable statement is, "Dogs are better than cats." That's because the definition of "better" is vague and subjective. However, an untestable statement can be reworded to make it testable. For example, the previous statement could be changed to this: "Owning a dog is associated with higher levels of physical fitness than owning a cat." With this statement, the researcher can take measures of physical fitness from dog and cat owners and compare the two.

Types of scientific hypotheses

In an experiment, researchers generally state their hypotheses in two ways. The null hypothesis predicts that there will be no relationship between the variables tested, or no difference between the experimental groups. The alternative hypothesis predicts the opposite: that there will be a difference between the experimental groups. This is usually the hypothesis scientists are most interested in, according to the University of Miami .

For example, a null hypothesis might state, "There will be no difference in the rate of muscle growth between people who take a protein supplement and people who don't." The alternative hypothesis would state, "There will be a difference in the rate of muscle growth between people who take a protein supplement and people who don't."

If the results of the experiment show a relationship between the variables, then the null hypothesis has been rejected in favor of the alternative hypothesis, according to the book " Research Methods in Psychology " (​​BCcampus, 2015). 

There are other ways to describe an alternative hypothesis. The alternative hypothesis above does not specify a direction of the effect, only that there will be a difference between the two groups. That type of prediction is called a two-tailed hypothesis. If a hypothesis specifies a certain direction — for example, that people who take a protein supplement will gain more muscle than people who don't — it is called a one-tailed hypothesis, according to William M. K. Trochim , a professor of Policy Analysis and Management at Cornell University.

Sometimes, errors take place during an experiment. These errors can happen in one of two ways. A type I error is when the null hypothesis is rejected when it is true. This is also known as a false positive. A type II error occurs when the null hypothesis is not rejected when it is false. This is also known as a false negative, according to the University of California, Berkeley . 

A hypothesis can be rejected or modified, but it can never be proved correct 100% of the time. For example, a scientist can form a hypothesis stating that if a certain type of tomato has a gene for red pigment, that type of tomato will be red. During research, the scientist then finds that each tomato of this type is red. Though the findings confirm the hypothesis, there may be a tomato of that type somewhere in the world that isn't red. Thus, the hypothesis is true, but it may not be true 100% of the time.

Scientific theory vs. scientific hypothesis

The best hypotheses are simple. They deal with a relatively narrow set of phenomena. But theories are broader; they generally combine multiple hypotheses into a general explanation for a wide range of phenomena, according to the University of California, Berkeley . For example, a hypothesis might state, "If animals adapt to suit their environments, then birds that live on islands with lots of seeds to eat will have differently shaped beaks than birds that live on islands with lots of insects to eat." After testing many hypotheses like these, Charles Darwin formulated an overarching theory: the theory of evolution by natural selection.

"Theories are the ways that we make sense of what we observe in the natural world," Tanner said. "Theories are structures of ideas that explain and interpret facts." 

• Read more about writing a hypothesis, from the American Medical Writers Association.
• Find out why a hypothesis isn't always necessary in science, from The American Biology Teacher.
• Learn about null and alternative hypotheses, from Prof. Essa on YouTube .

Encyclopedia Britannica. Scientific Hypothesis. Jan. 13, 2022. https://www.britannica.com/science/scientific-hypothesis

Karl Popper, "The Logic of Scientific Discovery," Routledge, 1959.

California State University, Bakersfield, "Formatting a testable hypothesis." https://www.csub.edu/~ddodenhoff/Bio100/Bio100sp04/formattingahypothesis.htm  

Karl Popper, "Conjectures and Refutations," Routledge, 1963.

Price, P., Jhangiani, R., & Chiang, I., "Research Methods of Psychology — 2nd Canadian Edition," BCcampus, 2015.‌

University of Miami, "The Scientific Method" http://www.bio.miami.edu/dana/161/evolution/161app1_scimethod.pdf  

William M.K. Trochim, "Research Methods Knowledge Base," https://conjointly.com/kb/hypotheses-explained/  

University of California, Berkeley, "Multiple Hypothesis Testing and False Discovery Rate" https://www.stat.berkeley.edu/~hhuang/STAT141/Lecture-FDR.pdf  

University of California, Berkeley, "Science at multiple levels" https://undsci.berkeley.edu/article/0_0_0/howscienceworks_19

Sign up for the Live Science daily newsletter now

Get the world’s most fascinating discoveries delivered straight to your inbox.

'The most critically harmful fungi to humans': How the rise of C. auris was inevitable

Odd earthquake swarm in Central Europe hints at magma bubbling below the surface

James Webb telescope detects 1-of-a-kind atmosphere around 'Hell Planet' in distant star system

Most Popular

• 2 32 of the most colorful birds on Earth
• 3 Can mirrors facing each other create infinite reflections?
• 4 See stunning reconstruction of ancient Egyptian mummy that languished at an Australian high school for a century
• 5 New invention transforms any smartphone or TV display into a holographic projector
• 2 'The most critically harmful fungi to humans': How the rise of C. auris was inevitable
• 3 32 of the most colorful birds on Earth
• 4 Roman-era skeletons buried in embrace, on top of a horse, weren't lovers, DNA analysis shows
• 5 Stone with 1,600-year-old Irish inscription found in English garden

Species as Explanatory Hypotheses: Refinements and Implications

• Regular Article
• Published: 18 February 2009
• Volume 57 , pages 201–248, ( 2009 )

Cite this article

• Kirk Fitzhugh 1  

465 Accesses

31 Citations

Explore all metrics

The formal definition of species as explanatory hypotheses presented by Fitzhugh (Marine Biol 26:155–165, 2005a , b ) is emended. A species is an explanatory account of the occurrences of the same character(s) among gonochoristic or cross-fertilizing hermaphroditic individuals by way of character origin and subsequent fixation during tokogeny. In addition to species, biological systematics also employs hypotheses that are ontogenetic, tokogenetic, intraspecific, and phylogenetic, each of which provides explanatory hypotheses for distinctly different classes of causal questions. It is suggested that species hypotheses can not be applied to organisms with obligate asexual, parthenogenetic, and self-fertilizing modes of reproduction. Hypotheses explaining shared characters among such organisms are, instead, strictly phylogenetic. Several implications of this emended definition are examined, especially the relations between species, intraspecific, and phylogenetic hypotheses, as well as the limitations of species names to be applied to temporally different characters within populations.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Similar content being viewed by others

On species concepts, species delimitation criteria, taxonomy committees, and biases: a response to Lima (2022a)

The use and limitations of null-model-based hypothesis testing

Informative ecological models without ecological forces

Throughout this paper, I intentionally avoid use the more common phrase, ‘shared similarities among members of two or more species .’ As I (Fitzhugh 2005a , 2006b , 2008b ) deny that species have the status of class constructs, individuals, or natural kinds, in lieu of being explanatory hypotheses, it would be entirely incorrect to connote some sort of membership relation between organisms and species.

Throughout this paper, formal names of species will be indicated using uninomials, rather than the binomial form required by the International Codes of Nomenclature (e.g., ICZN 1999 ). As was pointed out by Fitzhugh ( 2008b ), and in this paper, since hypotheses in the categories of species and genus are inferred from entirely different sets of premises applying different theories, the two classes of hypotheses stand separate from one another. The implication is that while the binomial requirement allows for monotypic genera and other taxa, those constructs are, by definition, empirically vacuous as they do not refer to any phylogenetic hypotheses. The solution is to acknowledge that names of species hypotheses should stand on their own, distinct from phylogenetic (supraspecific) hypotheses.

Achinstein P (1970) Inference to scientific laws. In: Stuewer RH (ed) Volume V: historical and philosophical perspectives of science. Minnesota studies in the philosophy of science. University of Minnesota Press, Minneapolis, pp 87–111

Google Scholar  

Atocha A (2006) Abductive reasoning: logical investigations into discovery and explanation. Springer, Dordrecht

Ben-Menahem Y (1990) The inference to the best explanation. Erkenntnis 33:319–344. doi: 10.1007/BF00717590

Article   Google Scholar  

Boyd R (1999) Homeostasis, species, and higher taxa. In: Wilson RA (ed) Species: new interdisciplinary essays. MIT Press, Cambridge, pp 141–185

Caplan AL (1980) Have species become déclassé? In: Asquith PD, Giere RN (eds) PSA 1980, vol 1. Philosophy of Science Association, East Lansing, pp 71–82

Carnap R (1951) Logical foundations of probability. University of Chicago Press, Illinois

Cohan FM (2002) What are bacterial species? Annu Rev Microbiol 56:457–487. doi: 10.1146/annurev.micro.56.012302.160634

Curd MV (1980) The logic of discovery: an analysis of three approaches. In: Nickles T (ed) Scientific discovery, logic and rationality. D. Reidel Publishing Company, Dordrecht, pp 201–219

de Queiroz K (1998) The general lineage concept of species, species criteria, and the process of speciation: a conceptual unification and terminological recommendations. In: Howard DJ, Berlocher SH (eds) Endless forms: species and speciation. Oxford University Press, New York, pp 57–75

de Queiroz K (1999) The general lineage concept of species and the defining properties of the species category. In: Wilson RA (ed) Species: new interdisciplinary essays. MIT Press, Cambridge, pp 49–89

Douven I (2002) Testing inference to the best explanation. Synthese 130:355–377. doi: 10.1023/A:1014859910339

Fann KT (1970) Peirce’s theory of abduction. Martinus Nijhoff, The Hague

Fitzhugh K (2005a) The inferential basis of species hypotheses: the solution to defining the term ‘species’. Mar Ecol (Berl) 26:155–165. doi: 10.1111/j.1439-0485.2005.00058.x

Fitzhugh K (2005b) Les bases philosophiques de l’inférence phylogénétique: une vue d’ensemble. Biosystema 24:83–105

Fitzhugh K (2006a) The abduction of phylogenetic hypotheses. Zootaxa 1145:1–110

Fitzhugh K (2006b) DNA barcoding: an instance of technology-driven science? Bioscience 56:462–463. doi: 10.1641/0006-3568(2006)56[462:DBAIOT]2.0.CO;2

Fitzhugh K (2006c) The philosophical basis of character coding for the inference of phylogenetic hypotheses. Zool Scr 35:261–286. doi: 10.1111/j.1463-6409.2006.00229.x

Fitzhugh K (2006d) The ‘requirement of total evidence’ and its role in phylogenetic systematics. Biol Philos 21:309–351. doi: 10.1007/s10539-005-7325-2

Fitzhugh K (2008a) Fact, theory, test and evolution. Zool Scr 37:109–113. doi: 10.1111/j.1463-6409.2008.00338.x

Fitzhugh K (2008b) Abductive inference: implications for ‘Linnean’ and ‘phylogenetic’ approaches for representing biological systematization. Evol Biol 35:52–82. doi: 10.1007/s11692-008-9015-x

Fitzhugh K (2008c) Clarifying the role of character loss in phylogenetic inference. Zool Scr 37:561–569. doi: 10.1111/j.1463-6409.2008.00338.x

Ghiselin MT (1974) A radical solution to the species problem. Syst Zool 23:536–544. doi: 10.2307/2412471

Ghiselin MT (1997) Metaphysics and the origin of species. State University of New York Press, Albany

Godfrey-Smith P (2003) Theory and reality: an introduction to the philosophy of science. University of Chicago Press, Illinois

Griffiths PE (1999) Squaring the circle: natural kinds with historical essences. In: Wilson RA (ed) Species: new interdisciplinary essays. MIT Press, Cambridge, pp 209–228

Hacking I (2001) An introduction to probability and inductive logic. Cambridge University Press, New York

Hanson NR (1958) Patterns of discovery: an inquiry into the conceptual foundations of science. Cambridge University Press, New York

Harman G (1965) The inference to the best explanation. Philos Rev 74:88–95. doi: 10.2307/2183532

Hempel CG (1965) Aspects of scientific explanation and other essays in the philosophy of science. The Free Press, New York

Hennig W (1966) Phylogenetic systematics. University of Illinois Press, Urbana

Hull DL (1976) Are species really individuals? Syst Zool 25:174–191. doi: 10.2307/2412744

Hull DL (1978) A matter of individuality. Philos Sci 45:335–360. doi: 10.1086/288811

Hull DL (1997) The ideal species concept—and why we can’t get it. In: Claridge MF, Dawah HA, Wilson MR (eds) Species: the units of biodiversity. Chapman & Hall, New York, pp 357–380

International Commission on Zoological Nomenclature (1999) International code of zoological nomenclature. The International Trust for Zoological Nomenclature, London

Josephson JR, Josephson SG (eds) (1994) Abductive inference: computation, philosophy, technology. Cambridge University Press, New York

Keller RA, Boyd RN, Wheeler QD (2003) The illogical basis of phylogenetic nomenclature. Bot Rev 69:93–110. doi: 10.1663/0006-8101(2003)069[0093:TIBOPN]2.0.CO;2

Lipton P (1993) Inference to the best explanation. Routledge, New York

Magnani L (2001) Abduction, reason, and science: processes of discovery and explanation. Kluwer Academic, New York

Mayden RL (1997) A hierarchy of species concepts: the denouement in the saga of the species problem. In: Claridge MF, Dawah HA, Wilson MR (eds) Species: the units of biodiversity. Chapman & Hall, New York, pp 381–424

Nickles T (1980) Introductory essay: scientific discovery and the future of philosophy of science. In: Nickles T (ed) Scientific discovery, logic and rationality. D. Reidel Publishing Company, Dordrecht, pp 1–59

Peirce CS (1878) Illustrations of the logic of science. Sixth paper. Deduction, induction, and hypothesis. Pop Sci Mon 13:470–482

Peirce CS (1931–1935) Collected papers of Charles Sanders Peirce. In: Hartshorne C, Weiss P, Burks A (eds) vol 1–6. Harvard University Press, Cambridge

Peirce CS (1958) Collected papers of Charles Sanders Peirce. In: Burks A (ed) vol 7–8. Harvard University Press, Cambridge

Psillos S (2002) Simply the best: a case for abduction. In: Kakas AC, Sadri F (eds) Computational logic: logic programming and beyond. Springer, New York, pp 605–625

Chapter   Google Scholar  

Reilly FE (1970) Charles Peirce’s theory of scientific method. Fordham University Press, New York

Rieppel O (2006) The PhyloCode: a critical discussion of its theoretical foundation. Cladistics 22:186–197. doi: 10.1111/j.1096-0031.2006.00097.x

Rieppel O (2007) Species: kinds of individuals or individuals of a kind. Cladistics 23:373–384. doi: 10.1111/j.1096-0031.2007.00152.x

Salmon WC (1984) Scientific explanation and the causal structure of the world. Princeton University Press, Princeton

Salmon WC (1989) Four decades of scientific explanation. In: Kitcher P, Salmon WC (eds) Volume XIII, scientific explanation. Minnesota studies in the philosophy of science. University of Minnesota Press, Minneapolis, pp 3–219

Sober E (1986) Explanatory presupposition. Australas J Philos 64:143–149. doi: 10.1080/00048408612342351

Sober E (1994) From a biological point of view: essays in evolutionary biology. Cambridge University Press, New York

Stamos DN (2003) The species problem: biological species, ontology, and the metaphysics of biology. Lexington Books, New York

Thagard P (1988) Computational philosophy of science. The MIT Press, Cambridge

Van Fraassen BC (1990) The scientific image. Clarendon Press, Oxford

Walton D (2004) Abductive reasoning. The University of Alabama Press, Tuscaloosa

Wilkins J (2007) The dimensions, modes and definitions of species and speciation. Biol Philos 22:247–266. doi: 10.1007/s10539-006-9043-9

Wilson RA (1999) Realism, essence, and kind: resuscitating species essentialism? In: Wilson RA (ed) Species: new interdisciplinary essays. MIT Press, Cambridge, pp 187–207

Download references

Acknowledgments

Sincere thanks are extended to Drs. Francisco Vergara-Silva and Rasmus G. Winther for inviting me to participate in their symposium, “Systematics, Darwinism, and the Philosophy of Science,” held at the Instituto de Investigaciones Filosóficas, Universidad Nacional Autónoma de México (UNAM), Mexico City. I am also very grateful to Drs. Brian Brown and Jody Martin for commenting on an earlier draft of this paper.

Author information

Authors and affiliations.

Research & Collections Branch, Natural History Museum of Los Angeles County, 900 Exposition Boulevard, Los Angeles, CA, 90007, USA

Kirk Fitzhugh

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Kirk Fitzhugh .

Rights and permissions

Reprints and permissions

About this article

Fitzhugh, K. Species as Explanatory Hypotheses: Refinements and Implications. Acta Biotheor 57 , 201–248 (2009). https://doi.org/10.1007/s10441-009-9071-3

Download citation

Received : 25 November 2008

Accepted : 26 January 2009

Published : 18 February 2009

Issue Date : June 2009

DOI : https://doi.org/10.1007/s10441-009-9071-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Species concepts
• Systematics
• Find a journal
• Publish with us
• Track your research