research studies on journalism

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 10 July 2023

The old-new epistemology of digital journalism: how algorithms and filter bubbles are (re)creating modern metanarratives

• Luca Serafini 1  

Humanities and Social Sciences Communications volume  10 , Article number:  395 ( 2023 ) Cite this article

1833 Accesses

1 Citations

4 Altmetric

Metrics details

• Cultural and media studies

In journalism studies, the advent of the World Wide Web and the rise of online journalism are generally associated with going beyond the objective, normative paradigm associated with the principles of philosophical and scientific modernity towards a postmodern paradigm centred on subjectivity and relativism. This article offers an alternative reading of the epistemology of online journalism: the fragmentation of audiences into homophilic networks, the formation of ideological bubbles, and the growing polarisation caused by algorithms make the contents circulating online a reintroduction of modernity’s metanarratives. These metanarratives in no way correspond to the principles typical of postmodernism, such as the equivalence of interpretations and openness to dialogue. Journalistic content also comes under this charge: although it conveys narratives that are subjective, they are perceived as absolute truths inside the information bubbles in which they circulate. This phenomenon is caused by “information platformization” processes. Based on these premises, a new definition of online journalism is proposed: rather than “postmodern”, it can be better understood as a fulfilment of the foundational principles of modernism, but in a subjective form.

Similar content being viewed by others

Ecology of the digital world of Wikipedia

How do we study misogyny in the digital age? A systematic literature review using a computational linguistic approach

Online searches to evaluate misinformation can increase its perceived veracity

Introduction.

Epistemology has proven to be a fertile field for journalism studies. Over the years, the discipline has developed in multiple directions and across varying levels of analysis. There have been studies on the epistemology of specific kinds of journalism, such as investigative journalism (Ettema and Glasser, 1985 ); and on the journalism conveyed by specific media, such as television (Ekström, 2002 ). With the advent of digital media, a new field defined as “Epistemologies of Digital Journalism” (Ekström and Westlund, 2019 ) emerged. A longitudinal study on digital journalism – that is, journalism in which editorial content is distributed and consumed through a digital medium – has shown how epistemology represents one of the most significant areas of research carried out in recent years (Steensen et al., 2019 ). Within this specific field of study, changes in the languages and practices of journalism brought about by the rise of digital media have been viewed by some scholars as the manifestation of a “postmodern turn” (Gade, 2011 ; Bogaerts and Carpenter, 2013 ; Wahl-Jorgensen, 2017 ). Underpinning this interpretation is a more general association between postmodernism and digital media that also applies to information when it is conveyed primarily by online media.

This article argues, however, that this interpretation is not entirely adequate to describe the mechanisms of online information production and consumption over the last 10 to 15 years. In this article, we will therefore attempt to define a new epistemological paradigm. We will do so by developing a conceptual argument, and not through the presentation of empirical findings. The aim is in fact to operate a terminological and conceptual redefinition of phenomena that, in theoretical studies on journalism, have so far mostly been framed within different paradigms. Starting from this theoretical and epistemological redefinition, the aim is for a path of enquiry to be opened for future empirical studies on journalism that frame the data collected within the new paradigm proposed here. In our current scenario, news narratives are increasingly conditioned by algorithm-driven processes of selective exposure and by a consequent formation of ideological and cultural bubbles. Within this framework, the thesis of this paper is that journalistic narratives increasingly take the form of metanarratives typical of the philosophical and scientific discourses of modernity, since they aim to provide their polarised audiences with a meaning that is absolute, complete, and impermeable to factual denial. This phenomenon is antithetical to some of the main assumptions of postmodernism: that all interpretations are equivalent, that there exists an openness to dialogue, and that texts and narratives possess a plurality of meanings. Consequently, in the algorithmic web of ideological bubbles, contemporary journalism can better be defined as a reintroduction and subjective form of fulfilment of the paradigm underpinning the modern normative model. Although subjectivism is the characteristic feature of online news narratives, this does not lead to the assumption that all interpretations are legitimate in the absence of an absolute truth; rather, it results in a clash between narratives, each of which seeks to prevail as an absolute truth within its respective information bubble. Due to information “platformization” mechanisms, journalism reinforces these tendencies, and it too is absorbed into the more generalised reintroduction of an epistemological model that is much closer to modernity than postmodernity.

The argument developed here refers to the way in which digital technologies have influenced journalistic practices in the global North, especially in Europe and the United States, and is thus limited to this social and geographical context. The clarification is important to avoid what Mabweazara ( 2014 : 2) described as a tendency on the part of many scholars to “seek explanatory frameworks in the uneven distribution and use of technological resources between the economically developed North and the poor South”. With regard to the countries of the global South and especially African countries, we will only mention some aspects related to the relationships between elements such as digital literacy, the frequently limited access of audiences to the products of digital journalism, and even the pre-digital polarisation of society with algorithmic news dissemination processes. It is hoped that these suggestions will enable future research focused, for example, on African countries to explore in more detail the links between the epistemological paradigm proposed here and what has already been termed an “African digital journalism epistemology” (Mabweazara, 2014 ).

From modern to postmodern journalism

The concept of “postmodern journalism” refers, by antithesis, to the “modern” paradigm that preceded it. In this sense, what happened in journalism is considered as a particular manifestation of a more general phenomenon, namely, the transition in the cultural and philosophical sphere from modernism to postmodernism. According to Jean-François Lyotard’s well-known thesis, postmodernism can be defined as scepticism toward the metanarratives that structured the modern philosophical discourse (Lyotard, 1984 ): the systems of thought that claimed to provide overarching explanations, such as the Enlightenment, Idealism, and Marxism. All these philosophical movements and systems contained unitary principles, the bases of which made it possible to encompass the meaning of reality (Reason, Spirit, the laws of materialism). Postmodernism signalled the winding down of these grand narratives and, simultaneously, of the emancipatory projects that philosophical systems, understood as universalising forms of knowledge, bear with them. For Lyotard, in the pre-industrial age the grand narratives that ensured the existence and preservation of a social order belonged to the realm of myth; with modernity, a new set of narratives arose, whose cornerstone was the scientific rationality professed by Enlightenment thinkers. As Isaiah Berlin ( 1980 : 1–32) explains, the Enlightenment proclaimed the autonomy of reason and the natural sciences, while at the same time rejecting the authority and tradition of all forms of transcendental and nonrational knowledge. For Lyotard, with postmodernism the modernist faith in reason lost its self-evident character: many discourses that were modelled on the scientific rationality promoted by the Enlightenment went into crisis. These discourses had taken the form of metanarratives founded on objectivity, universality, and certain knowledge; they were rational and uncontaminated by anything subjective or transcendent to reality.

Journalism was among the discourses that entered into crisis with the postmodern turn. Between the nineteenth century and the first half of the twentieth century, its normative model had been forged from the Enlightenment spirit, the idea that objective methods and rational procedures could be used to describe reality (Schudson, 2018 : 29). The paradigm of objectivity is thus generally viewed as a direct application of scientific modernism to journalism (Schudson, 1990 ; Durham, 1998 ). This model of objective journalism took shape in the United States during the 1830s and was developed more fully beginning in the 1920s, when, partly due to the need to make journalism a fully-fledged professional activity, it was codified through specific procedures and rules. As Walter Lippmann ( 1922 ) noted at the time, this need was motivated in part by a desire to distinguish journalism from the activities of public relations: the latter were focused on persuasion, while the former aimed at the objective reporting of facts. This need for distinction turned objective journalism into a defensive, strategic ritual (Tuchman, 1972 ), put into practice to avoid the influence exerted by professional communicators. But, as Nerone ( 2013 ) points out, journalism is also an -ism, and as such shares the characteristics of a belief system. Consequently, objectivity was not embraced as a journalistic norm purely as a defensive paradigm but also as the expression of a metanarrative: from this perspective, journalism could and should report factual truths. Consensus regarding the rules of the profession was based on objectivity understood in terms of an ideology (Schudson, 2001 : 151). Any challenge to journalism’s claim to present itself as a “bearer of truth” was therefore seen as a challenge to its normative paradigm, and the concept of impartiality and modernist assumptions regarding the role of journalism as a bulwark of democracy fused into its professional mission (McNair, 2012 ; Schudson, 2008 ). For a long time, the principles of modern scientific rationality, which hinged on a description and study of reality uncontaminated by subjective opinions, ensured that emotions were excluded from the reporting of facts (Richard and Rees, 2011; Peters, 2011 : 298). The paradigm of objectivity was thus borrowed from the principles of modern scientific rationality, because “objectivity relies on the modern perception of a textual message – one that is rigid and permanent – it rejects the idea that message reception is a dialogical site with varying possibilities of meaning” (Soffer, 2009 : 474). But a monological conception of this sort “is therefore associated with a single world view that […] sees the world as an object of deduction” (Soffer, 2009 : 477); and this monological voice goes hand-in-hand with the modern scientific perception (Shotter, 1997 : 26).

The fact that journalism’s age of professionalisation, which took place in tandem with the rise of the normative paradigm of objectivity, was defined as “high modernism” in connection with the processes of the 1920s, and “high modernity” in connection with what occurred in the 1950s and 60 s, when objectivity was identified by journalists as both an ideal and a daily practice (Hallin, 1994 ), is therefore not a terminological coincidence but rather the sign of a significant analogy.

Starting in the 1960s and 70 s, however, something different took place: the normative paradigm of journalism came under increasing scrutiny. The historical period when this occurred is no coincidence: these were the years when the rationalist tradition inspired by Descartes was also being critiqued by philosophers, clearing the path to an “affective turn” that accompanied the progressive rise of postmodernism (La Caze and Lloyd, 2011 ). As in philosophical postmodernism, which rejects any scientistic belief in the possibility of an objective account of reality, in journalism, too, a view began to gain ground that the normative model does nothing but reinforce “official” versions of reality propagated by power, a power that holds the means to impose its own symbolic representation of the facts (Jukes, 2020 : 28). Postmodernism does indeed imply the notion that the symbols used to describe reality are nothing but symbols, expressions of subjective choice that, as such, are incapable of describing reality as it truly is (Baudrillard, 1984 : 159–164; Rorty, 1989 ). This also applies, of course, to the linguistic symbols through which journalism reports on reality.

New journalistic models of reporting on reality that emerged during this historical period were to some extent a sign of the times: New Journalism, for example, was defined, significantly, as a “signpost to the postmodern,” because the subjective and narrative form it expressed hewed more closely to the demands of postmodern society (Basu, 2010 ). As Schudson ( 2018 ) points out, during the 1960s and 70 s a more analytical and in many cases interpretive account of reality took hold among journalists – a model he defines as “objectivity 2.0”, marking a first break with the modernist model of “objectivity 1.0”. For all these reasons, the changes that started in the 1960s and 70s and developed more fully in the subsequent decades “could be seen to represent a “postmodern turn” in journalism insofar as they challenge the conventional “grand narratives”, certainties and rationalities that underpin the profession and its practices” (Wahl-Jorgensen, 2017 : 98).

It is important to note that this questioning of the “grand narratives” and the concurrent “postmodern turn” are characterised above all by two processes: a shift from (presumed) objective reporting to subjective reporting; and, consequently, a shift from the disembodied rationality typical of the modern scientific paradigm to an increasingly greater presence of emotions in journalistic texts (Jukes, 2020 ). The postmodern turn of the 1960s and 70 s, which, as we have seen, led to postmodern philosophical assumptions being applied to journalism as well, is generally summarised by Nietzsche’s well-known aphorism (Nietzsche, 1967 [1885–1887], aphorism 481): “facts is precisely what there is not, only interpretations”. Nietzsche’s aphorism became the emblem of the postmodern paradigm because it clarifies the shift that took place under it, from presumed objectivity to subjectivism. According to the prevailing view in journalism studies, the main components of the postmodern turn of the 1960s and 70 s just described came about because of new technologies and the way they reshaped journalism. As Gade ( 2011 : 126) remarks, “the digital age unleashed already present postmodern forces, creating a networked, interactive, and consumer-oriented era that destroyed the stability of the mass media”.

This transition, which was already visible in the 1960s and 70 s, was brought to maturity primarily by the hybrid media system, in which different types of older and newer media form a system that evolves through mutual interactions (Chadwick, 2013 ), and the affordances Footnote 1 of digital technologies. In line with postmodern assumptions on the loss of boundaries between high culture and pop culture, and the de-differentiation between cultural and social spheres (Lash, 1990 ), online journalism causes a contamination between traditional and digital media, between actors of information (created as much as by users as by professionals), between communicative models (broadcast and conversational), and, above all, between content types (with a progressive mingling of hard and soft news). The typically modern concept of boundary work (Gieryn, 1983 ), understood as what allows a clearly demarcated line to be drawn between what is and is not journalism (Carlson, 2015 ), has faded. The result is a genuine epistemological rupture, definable according to (Wahl-Jorgensen, 2017 : 106) as postmodern in all respects: “The changes occasioned by technological transformations could be understood as a postmodern form of journalism because they have destabilised conventional: (a) physical, stylistic and genre distinctions; (b) differentiations between amateur and professional content; and (c) distinctions around the truth value of objective versus emotional content”.

These factors, often interconnected, have caused a slippage toward an increasingly subjective recounting of facts, leading journalism to gradually lose its self-representation as a bearer of truth and opening the way for perspectivism and relativism. As for the contamination between information actors, the rise of citizen journalism Footnote 2 – of the participatory news consumer (who not only consumes but also produces news) and a general “editorial society” (Hartley, 2000 ), in which citizens act as journalists and no longer trust in journalism as an open system – has led to news production that is increasingly contaminated by subjective points of view. The postmodern turn must therefore be understood as the rise of a biographical society, in which life stories are everywhere (Plummer, 2001 : 78). As noted by Papacharissi ( 2015 ), the incorporation of user-generated content in the news means that it becomes simultaneously more subjective and more emotional: an affective news stream is generated, replete with emotions, opinions, and subjective experiences. In social media especially, the news has become almost indistinguishable from conversation about the news.

Beckett and Deuze ( 2016 ) have found a trend towards an increasingly personalised and emotionalised journalism in the age of networked news, with growing use of the first-person in writing. Accelerated by digital media, definable in all respects as postmodern media, we are thus moving toward a confessional, subjective journalism (Coward, 2013 ). The crisis in the objective reporting of reality and the very concept of truth brought to light by postmodernism have caused, to use the words of Bogaerts and Carpenter ( 2013 : 69-70), “a new truth-claim in journalism, turning from claims based on objectivity to those based on authenticity”. The concept of truth has thus become subjective on all counts, losing any reference to universality or to a shared certainty. All the convergence processes just mentioned, typically postmodern insofar as they supersede the (modern) division between professional, normative journalism and its “other”, have induced the rise of subjectivism, perspectivism, and relativism: if the confines of journalism have eroded and there is no more truth, then all that can be told are different stories from different perspectives. This process is seen on all counts as the expression of a postmodern turn: all opinions become legitimate, and references to facts become increasingly remote in this “cacophony of mediated voices” (Silverstone, 2007 : 1–24). For all these reasons, the end of modern metanarratives in the network society means that journalism has also been transformed into a subjective, perspectival account of reality that some scholars define as “postmodern”.

The subjective fulfilment of modernity: Heidegger and Nietzsche

As mentioned earlier, Nietzsche’s aphorism – “facts is precisely what there is not, only interpretations” – is considered the emblem of the postmodern paradigm. Nevertheless, the epistemology of digital media, in the form it has taken for the last 10 to 15 years at least, can be viewed more appropriately as a fulfilment of the modern paradigm in a subjective form. This is also pertinent to online journalism, as we shall see in the last section. To understand what is meant by “a fulfilment of the modern paradigm in a subjective form”, we can turn to one of the better-known interpretations of Nietzsche’s thought, provided by Martin Heidegger in his courses between 1936 and 1946. In the third and fourth volumes of Nietzsche (Heidegger, 1991 [1961]), The will to power as knowledge and as metaphysics and Nihilism , Heidegger explains why Nietzschean philosophy should be understood as a fulfilment of metaphysical thought and scientific rationalism. Heidegger views Nietzsche as the thinker who carried out “the fulfilment of the metaphysics that began with Plato” (Heidegger, 1991 [1961]: 261). Nietzsche preserves the Platonic distinction between a true world and an apparent world but reverses it (the sensible world takes the place of the supersensible or transcendent world). Important to our enquiry is Heidegger’s idea that the fulfilment of metaphysical thought derives from Nietzsche’s assumption that being is will to power, which, as such, rests on nothing but itself: it has no foundation, precisely because the distinction between a true world and an apparent world has fallen into decline. As a result, there is no transcendent principle based on which one can establish what is true or false: being as will to power is a radical form of perspectivism and subjectivism. In Heidegger’s reading, however, this is what makes Nietzsche’s work a form of metaphysical thought that leads to the triumph of scientific rationality. Nietzsche does indeed conceive of the world as material available to the will to power, as that which can be forged by humans as they please. All this is summarised in Nietzsche’s aphorism, recalled by Heidegger: "To ‘humanise’ the world, that is, to feel ourselves more and more masters within it" (Heidegger, 1991 [1961]: 614). Eliminating any transcendental foundation of truth opens the way to a different form of metaphysical thought: it is flipped over from an “objective” to a “subjective” metaphysics. Although subjective, it remains a form of metaphysics but one intended as a fulfilment of scientific rationality.

Taking Heidegger’s approach and applying it to the aphorism “facts is precisely what there is not, only interpretations” casts a different light on Nietzsche’s words: no longer do they appear as the foundation of postmodernism but, rather, as expressive of the modern paradigm – a paradigm tied to scientific rationality and the search for truth but flipped over into subjectivity. The following sections explain why the epistemology of the online world is more in line with the modern paradigm than the postmodern one, and how this also affects online journalism.

The algorithmic web as a reintroduction of modern metanarratives

Several assumptions of the postmodern paradigm are not easily reconciled with the form that online public discussion has taken in recent years, also challenging the association between digital media and postmodernism. Selective content exposure and the consequent formation of homophilic networks caused by the functioning of algorithms on the leading platforms generate a communicative exchange whose foundations are quite different from those of postmodern thought.

The first assumption of postmodernism that is in strong contradiction with the phenomena just mentioned concerns the virtual equivalence of subjective points of view, which is a corollary of the collapse of all metaphysics and all modern rationalistic projects. The idea that the ending of metanarratives opens the way to a plurality of narratives, micronarratives, and differences in points of view appears both in philosophical currents belonging to “weak thought” (Vattimo, Rovatti 2013 [1983]), issuing directly from postmodernism, and in the work of Lyotard himself ( 1984 ). The basic assumption of this opening to difference is the acceptance of all diversity. Since no one discourse can set itself up as hegemonic and absolute (because any claim to truth is a manifestation of power and leads to authoritarianism), postmodern subjectivism takes the form of “charity”. No one can establish which point of view is true: as a result, public debate starts from an acceptance of all outlooks, leading to dialogue and, ultimately, mutual solidarity. In the linguistic games that Lyotard speaks of, no discourse is privileged over others: knowledge emerges as the acceptance of a plurality of discourses.

These points are also pertinent to the more general association between postmodernism and new media, i.e., the mass media developed after the emergence of information technology and using digital technologies. In giving life to the network society, new media pluralise public discourse in such a way that, theoretically, it can no longer be subjected to any form of domination: it is freed, in other words, from all metanarratives. Nietzsche’s aphorism, the emblem of postmodernism, postulates reality’s reduction to interpretation. But as Maddalena and Gili ( 2020 : 66) recently noted, “when there are no longer any criteria to evaluate the validity of different discourses […] every idea and interpretation is equally legitimate”. The premise of modernism was a monological voice that described reality by objectifying it, in which a text had only one possible meaning. With postmodernism, the closed meaning of the modernist text is replaced by a dialogical reality, in which every message is open to a plurality of meanings and points of view on the world: this also applies to journalistic texts (Lähteenmäki, 1998 ; Soffer, 2009 ). The idea of a plural dialogue, which assumes that all points of view are equivalent, is in blatant contradiction, however, with the fragmentation of online public debate, the selective content exposure caused by algorithms, and the ensuing polarisation and radicalisation of opinions. The concepts of filter bubbles – i.e., the customised information ecosystem created by algorithms in which users are exposed to information that supports what they already believe and like (Pariser, 2011 ), and echo chambers, closed and homogeneous virtual environments in which divergent views have no place and subjects always hear the echo of their own voice and opinions (Sunstein, 2001 ) – have found their way into academic discussion, entering into the lexicon of journalism and common speech. Naturally, these concepts are not accepted uncritically by the scholarly community. Over the years, some studies have objected to a lack of empirical evidence, arguing that the findings do not support the thesis of greater polarisation online than offline (Fletcher and Nielsen, 2017 ; Bruns, 2019 ).

It must nevertheless be noted that recent empirical studies give greater credence to the idea that pre-existing opinions on the web are reinforced due to selective content exposure (especially the news), to such an extent that the plurality of public debate is put at risk, fostering the radicalisation of opinions (Claussen et al., 2019 ; Levy, 2021 ). In a comparative analysis of these and other studies, Aral ( 2020 : 250) recently argued that, despite conflicting theses that have alternated over recent years, “evidence from multiple experimental studies shows that the machine’s recommendation algorithms create filter bubbles of polarised content consumption”.

This is why the narratives that form in online “tribes” (including journalistic ones, as we shall see in the final section of this article) are more and more similar to the grand récits of modernity than to the open narratives of postmodernity. In concrete terms, this refers to the fact that within homophilic networks, the complexity of reality is reduced to simple, all-encompassing schemas, in which every facet of reality serves to confirm pre-established opinions. These are the same principles that Lyotard identified as the basis of modernity’s metanarratives, such as Marxism, Idealism, and the Enlightenment, which sought to explain reality through unitary principles (Reason, Spirit, the laws of materialism). These were the principles that gave reality an absolute, overarching meaning, and were somehow able to explain every phenomenon. As Hannah Arendt ( 1951 ) observes, the concept of ideology should be understood etymologically as the “logic of the idea”: as an attitude that compresses the infinite variety of reality into an absolute logical schema and satisfies the desire for meaning. In a nutshell, precisely because of its ultrarational basis, the ideological attitude is impervious to factual denial. This is exactly what led to Karl Popper’s observation (2002 [1963]: 45–46) on the Marxist theory of history and Freudian psychoanalysis:

“I felt that these other three theories, though posing as sciences, had in fact more in common with primitive myths than with science […]. These theories appeared to be able to explain practically everything that happened within the fields to which they referred. […] the world was full of verifications of the theory. Whatever happened always confirmed it […] and unbelievers were clearly people who did not want to see the manifest truth. […] A Marxist could not open a newspaper without finding on every page confirming evidence for his interpretation of history”.

Like the philosophical systems developed from the principles of modern science (think of the Cartesian Method), the algorithmic epistemology works on a completely deductive and falsely empirical basis to make each successive piece of data confirm the initial thesis. This mechanism is one of the most studied and analysed dynamics belonging to the world of online communities, and it relates to what media sociology defines as "confirmation bias" (Zhao et al., 2020 ) and the "backfire effect" (Nyhan and Reifler, 2010 ; Jarman, 2016 ). Through studies conducted on numerous Facebook pages and groups, Quattrociocchi and Vicini ( 2016 , 48-50), among others, have pointed out that communities of users who are grouped together and strongly polarised on a specific position tend to take as true news that is scarcely credible, provided it is consistent with their reference narrative. This process has to do with confirmation bias, namely, a mechanism that leads individuals to consider true only information that falls within their belief system. To make matters worse, empirical studies on Facebook groups show that debunking operations are not only unsuccessful, but they also tend to reinforce the very belief system they seek to discredit, thus causing a backfire effect. For users who are convinced, for instance, of the truthfulness of a conspiracy theory, the more they are exposed to information showing the fallacy of their position, the more they will strengthen their initial belief. They do this either by ignoring and discarding information that contradicts their theory, or by taking it as devious attempts to conceal the truth. The backfire effect thus consists in a paradoxical reinforcement of the polarisation of individuals and groups who are exposed to factual denials of the ideas they believe in: guided by the tribal emotionality produced in them by selective exposure to certain content, instead of reflecting on their ideas and reconsidering them, these individuals and groups will increasingly reinforce their identity and their sense of group belonging.

The narratives of online communities, which also absorb journalistic narratives, as we shall see, thus seem to draw on the grand narratives of modern thought. Instead of moving toward a postmodernist relativism and perspectivism, the web of algorithmic engineering leads toward a subjective fulfilment of modernity: just as the metanarratives of modernity claimed to provide an objective and truthful explanation of all aspects of reality, so in online metanarratives each individual group believes its own worldview to be true, absolute, and capable of explaining all things. Contrasting narratives are not accepted in the name of a fundamental relativism, as postmodernist assumptions would have it; instead, they are rejected as false and often bitterly opposed, in line with the dynamics of radicalisation and polarisation described earlier.

Although, as specified, the epistemological paradigm presented here refers to the global North countries, a few examples concerning the nations of the global South can be given to highlight some possible effects of algorithmic news dissemination processes on societies that already present strong levels of polarisation and, sometimes, low levels of digital literacy. In Myanmar, for example, Facebook has admitted to playing a role (described as “decisive” by a UN report) in fomenting hatred against the Rohingya Muslim minority. Low digital literacy, in Myanmar, meant that users lacked the tools to respond critically and reflectively to the proliferation of disparaging and inflammatory posts against the Rohingya, which Facebook soon made viral (Osnos, 2018 ). In another low digitally literate country like the Philippines, on the other hand, Facebook-induced polarisation dynamics and the proliferation of fake news fostered Duterte’s legitimisation of his own autocratic and repressive power, as highlighted by a BuzzFeed investigation (Alba, 2018 ). And again, a BBC investigation highlighted how, in an already heavily “tribalised” nation marked by ethnic-religious conflicts like Nigeria, users’ hyper-emotional responses to content that went viral on social media (and often manipulated) contributed to massacres of Christians by Muslims and vice versa (Adegoke, 2018 ). This scenario of further balkanisation, in nations of the Global South, may thus be influenced by variables such as “digital literacy and competencies, limited access to information and exposure to various kinds of self-sorting online groups. Thus, not everyone shares fake news with the intention to cause harm” (Mare et al., 2019 : 6). It follows from what has been said so far that the subjectivism of the algorithmic web cannot be defined as postmodern but rather as a subjective reversal of modernity: it does not lead to dialogue between different but equally legitimate opinions but to an antagonism between narratives that are indeed subjective but at the same time claim to be true and absolute (like those of modernity, which founded this same claim on universality rather than subjectivity). Heidegger’s interpretation of Nietzsche thus serves particularly well to explain the turn that subjectivism has taken in the algorithmic web. As regards Nietzsche’s aphorism that celebrates the end of facts and the triumph of interpretations, Heidegger’s idea allows us to locate it in a different paradigm than that of postmodernism, seeing it instead as a fulfilment of the metaphysics of subjectivity, and to clarify the idea that the universal assumptions of modernity’s metanarratives have been reversed into subjectivity.

As Gade ( 2011 : 114) reminds us, "to postmodernists, science is but one discourse, and it includes all the biases of any discourse: it imposes a set of processes and rules on how to see and define the world, and these processes shape thinking in ways that obscure seeing the world as it actually is". What happens in online communities through the cognitive dynamics of confirmation bias and the backfire effect represents a similar shutting out of reality, which is now encapsulated in a preconstructed narrative and system of meaning within a single group and constantly reinforced by selective exposure to content (including news content).

The equivalence of interpretations is not associated with the algorithmic web and the narratives formed there; similarly, the postmodern assumption of the plural and never-definitive meaning of a text also does not correspond to what is found online, particularly on social networking sites. To meet a social network’s need for meaning, the texts circulating on the web often present a full, absolute, easily and immediately comprehensible meaning capable of generating emotional reactions (in the form of likes, shares, and so on), thereby winning the battle for users’ attention. Taking as an example the headlines of news articles as they are conveyed on social media immediately calls up the phenomenon of “sharebaiting”, in which users are prompted to share content based on the headline alone, by clicking on an article without reading it through. This happens precisely because of the emotional charge and fullness of meaning expressed by the wording of online headlines, which are instrumental in making them go instantly viral. A study by Columbia University (reported in Dewey, 2016 ) showed that almost 60 per cent of the links shared on social media had never been opened by users, confirming the effectiveness of these strategies: the headlines express an immediate, complete sense of meaning for the social networks for whom the information content is intended.

Clearly, there is no plural meaning here, no polysemy, no multi-voice dialogue, as the principles of postmodernism would have it. Rather, the headlines on social media are constructed to be short and concise, with a suggestive character, to reduce the complexity of the narrative (and reality) and fit into a pre-established schema of meaning. In this case too, then, the web of algorithms, filter bubbles, and echo chambers draws on a paradigm that is much closer to the grand narratives of modernity than to the dialogic plurality of postmodernity. Subjectivism, as in Heidegger’s interpretation of Nietzsche, should therefore be understood within a rationalist, Enlightenment epistemology, one that is modern in all respects, in which individual narratives maintain their pretension to express an absolute truth impervious to factual denial, and to convey it in online social networks. It remains to be seen how these observations on headlines also apply to journalistic narratives.

Information platformization and journalistic (meta)narratives

With the massive transfer of information onto the web, journalism has progressively absorbed the principles and imperatives that guide the functioning of online platforms. The shift toward a subjective form of modernism, rather than toward postmodernism, brought about by the algorithmic web can, therefore, also be applied to journalistic narratives, insofar as it is the online platforms and their algorithms that influence both the production of information by news organisations and the consumption of information by users. In other words, there is a "platformization" of information, which makes journalistic narratives subject to the same principles that govern the circulation of other content on the web. Information progressively loses its public value and is transformed into a “private value”, especially on the web and in social media, in line with the commercial rationale and monetisation of interactions that guide the functioning of platforms. As the consumption of information is increasingly influenced by selective exposure to content (due to algorithmic filters), “information bubbles” are generated that also influence how journalistic texts are composed: to maximise interactions and revenues, these too must attract interest within those same bubbles.

First, as is well known, for many years now the consumption of information via social media has increased disproportionately (Newman et al., 2022 ). Social media have become the new “infomediaries” (Rebillard and Smyrnaios, 2010 ), the true mediators of information. This implies, however, that information consumption is strongly conditioned by networks of individual users, who read what their contacts have shared on their walls. Social media have been referred to as “secondary gatekeepers” (Shoemaker and Vos, 2009 ): also included within this category is “audience-based gatekeeping” (Nielsen, 2017 : 90), that is, the contacts in a user’s social network that determine their information consumption. Clearly, the network rationale already steers information consumption to form “information bubbles”, since news consumption takes place within networks of users who share the same mindset, political orientation, and worldview. Moreover, in serving this type of audience, social media and platforms in general follow the rationale of spreadability (Jenkins et al., 2013 ): in other words, they have to offer information content that maximises interactions and generates actions such as likes, comments, and shares. These aspects in particular reinforce the tendency to form ideological bubbles (Klinger and Svensson, 2018 ), since the user network is offered content in line with the ideological orientation of its members (specifically with the goal of making it simpler to generate interactions and revenues).

Indeed, news organisations increasingly produce content based on its estimated potential circulation (Anderson, 2011 ). These estimates hinge on the datafication process (van Dijck 2014 ), whereby metrics on information consumption (such as trending topics) dictate the topics to be covered. However, if information consumption is already highly polarised and ideologised, news organisations will be obliged to produce content in line with their users’ attitudes. Given these conditions, journalistic narratives can hardly be constructed as narratives that are open to a plurality of meanings, dialogical, and designed to elicit reflective responses from users. Quite the opposite: there will be a tendency to propose content that is instantly understandable, has an immediate emotional impact, responds to the ideological orientation of the information bubble for which it is intended, and consequently satisfies the need (typical of any bubble) for an absolute meaning, defined once and for all.

As noted earlier, this is primarily due to the ‘platformization’ of information, that is, the result of news outlets absorbing the principles (and values) of online platforms, as well as the fact that these platforms are gradually taking over the role of information and news providers. A process of “disaggregation” (Carr, 2008 : 153) is taking place, by which news organisations no longer act as news gateways and are replaced by search engines, news aggregators, and social networks. Within this framework, the audience is no longer an information audience: it is composed, rather, of platform users. These processes mean that “as platformization continues to penetrate more sectors of society, the distinction between private and public is increasingly glossed over as an irrelevant societal classification” (van Dijck et al., 2018 : 30). Under these circumstances, journalism also tends to lose a great deal of its public values (its role in democratic systems, separation from power, comprehensive coverage so that everyone has a voice, and so on). A shift takes place "from a model that primarily revolves around editorial autonomy to one based on datafied user interests and activities.” But "user data are never a neutral reflection of user interests but always shaped by the techno-commercial strategies of platforms" (van Dijck et al., 2018 : 57). The data, which forms a basis for setting the editorial line (dictating which news items are chosen and the language used to cover them), are influenced by the bubble-forming algorithms: consequently, they tend to produce news “for bubbles”, since this is what generates the most interactions and revenues. News organisations are pushed towards communicative models that privilege private, subjective, and often ideological meanings, specifically due to the commercial rationale of the platform to which they are subject.

The equivalence proposed in this article between the algorithmic web, ideological bubbles, and the grand récits of modernity is thus applicable to journalistic narratives as well: they absorb the guiding principles of the platforms, which drive the production process as much as they do news consumption. Here again, subjective, ‘private’ journalism directed toward a platform audience cannot be qualified as ‘postmodern’, since in both its choice of topics and language it is a journalism that aims to create antagonisms rather than dialogue, to provide absolute rather than open narratives, to satisfy the desire for meaning of online communities. Emotional, captivating headlines, the choice of highly divisive, polarising topics, all of this succeeds much better (due to the platform affordances) in capturing the interest of users, thus also catering to their desires and generating greater revenues.

This is not postmodern journalism, therefore: it is modern journalism reversed into a subjective form. Even the narratives conveyed by news organisations in information bubbles are in fact metanarratives, endowed (for that one particular bubble) with an absolute, all-inclusive meaning, capable of encompassing all facets of reality and “explaining everything”. From universal, objective metanarratives we pass to subjective metanarratives. It is indeed true that with the postmodern turn everything became interpretation; however, inside each ideological bubble, each person’s narrative is considered an absolute truth. Having lost its public function of providing information and dialogue, and having acquired the rationale of a platform aimed at maximising revenue (through the interactions of polarised audiences), journalism tends to reinforce this process.

As for the countries of the Global South, which have been hinted at during this work, it should be noted how the production and especially the consumption of news is often influenced by the cost of accessing information. Indeed, it has been noted how, especially in African countries, many media outlets lock their content behind paywalls: this limits access to information for substantial portions of the population (Mare et al., 2019 ). The consequence of this limited access to a multiplicity of information sources is that people often mistake “the popularity or virality of a shared piece of information as an indication of its veracity” (Chakrabarti et al., 2018 : 44). Again, there are therefore mechanisms of news platformization that, in the absence of full access to information by the population, can generate or exacerbate some of the effects described in relation to the context of Global North countries.

Conclusions

This article offers an alternative conceptual framework to the interpretation that views online journalism as a shift from a modern to a postmodern paradigm. It begins by reconstructing the main theories that explain the link between the modern paradigm of journalism and the assumptions of scientific modernity and shows how the journalistic ideal of objectivity is often equated with a metanarrative like those that structured the philosophical and scientific discourses of modernity. It then reconstructs theories showing how journalism has shifted, along with the gradual rise of the web, from the paradigm of objectivity to a form of marked subjectivism. This trend has been reinforced by platform affordances, the phenomena of media convergence, and a contamination between information actors, communication models, and types of content. This subjective stance is generally associated with the postmodern turn and with Nietzsche’s aphorism that "facts is precisely what there is not, only interpretations". It is argued, however, that Nietzsche’s aphorism and the postmodern paradigm, in general, are inadequate to define the turn taken in the production and consumption of online content (including information), which is strongly influenced by platform algorithms and the formation of ideological bubbles. The narratives that emerge within highly homogenous and polarised online social networks are actually the metanarratives of modernity: like the systems of thought that claimed to explain every aspect of reality, narratives that can “explain everything” are favoured within bubbles, because they provide an absolute meaning never contradicted by the data of reality. Confirmation bias and the backfire effect explain the imperviousness of these bubbles to evidence that refutes pre-constituted narratives. This clashes with several assumptions of postmodernism: the equivalence of interpretations, an openness to other people’s points of view, dialogue, and a plurality of textual meanings. On the contrary, online texts tend to assume a monolithic meaning, one that is absolute and defined once and for all.

All these factors lead to the hypothesis that the algorithmic web represents a fulfilment of the universal metanarratives of modernity but flipped over into a subjective form. For this reason, starting from Nietzsche’s aphorism, Heidegger’s interpretation of Nietzsche is proposed as most suited to capture the characteristics of this subjective fulfilment of modernity. Heidegger saw Nietzsche’s thought as a reversal of an objective metaphysics into a subjective metaphysics, making it a fulfilment of scientific and Enlightenment rationalism (rather than the forerunner of postmodern thought, as Nietzsche would be considered from the 1970s on).

Lastly, this algorithmic, polarised logic, which revolves around the phenomena of filter bubbles and echo chambers and reintroduces the principles of modern metanarratives, also applies to journalistic narratives. Indeed, information makes up part of the content that is increasingly dependent on the operating mechanisms of online platforms, which, through their affordances, impose a highly “privatised”, subjectivist and ideological model of news production and consumption. Journalistic texts and narratives thus succumb to a fate similar to that of other content circulating on the web. For this reason, within the framework of algorithms and ideological information bubbles, journalism does indeed express an epistemological model that revolves around subjectivity, but it cannot be defined as “postmodern”: rather, it is a subjective reversal of the metanarratives typical of modern journalism.

There are some limitations to this study. Firstly, the article’s intention is to make a theoretical argument in order to propose a new epistemological paradigm for online journalism. As such, it makes a theoretical synthesis that cannot take into account all the practices of digital journalism. Secondly, the proposed epistemological paradigm is limited to the countries of the Global North. However, some elements referring to the countries of the Global South, particularly those in Africa, have been presented, which may allow for comparative reflection. This may lead future studies to analyse how the epistemological paradigm proposed here can be integrated with epistemological models more directly referring to journalism in the countries of the Global South.

Affordances are the properties that a technological object possesses and which in fact already suggest a use for the object itself.

Journalism by non-professionals who disseminate information using social media, blogs and websites.

Adegoke, Y (2018) Like, Share, Kill. Nigerian police say false information on Facebook is killing people. BBC World, November 13. Available at: https://www.bbc.co.uk/news/resources/idt-sh/nigeria_fake_news

Alba, D (2018) How Duterte Used Facebook To Fuel the Philippine Drug War. BuzzFeed, September 4. Available at: https://www.buzzfeed.com/daveyalba/facebook-philippines-dutertes-drug-war

Anderson, K (2011) Live blogging evolved: context and curation not just collection. In: Strange Attractor. Available at: https://charman-anderson.com/tag/curation/

Aral, S (2020) The Hype Machine: How Social Media Disrupts Our Elections, Our Economy and Our Health – and How We Must Adapt. London: HarperCollins

Arendt, H (1951) The Origins of Totalitarianism. New York: Schocken Books

Basu, M (2010) ‘New Journalism’, Subjectivity and Postmodern News. Proof, Set 2. Available at: https://www.academia.edu/5822142/_New_Journalism_Subjectivity_and_Postmodern_News

Baudrillard, J (1984) Simulacra and Simulation. Translated by Glaser SF. Ann Arbor: The University of Michigan Press

Beckett C, Deuze M (2016) On the role of emotion in the future of journalism. Social Media + Society 2(3):205630511666239

Article   Google Scholar  

Berlin, I (1980) Against the Current: Essays in the History of Ideas. New York: Viking

Bogaerts, J, Carpenter, N (2013) The postmodern challenge to journalism. In: Peters C, Broersma M (eds) Rethinking Journalism. Trust and participation in a transformed news landscape. Abingdon-New York: Routledge, 60–72

Bruns, A (2019) Are Filter Bubbles Real? Cambridge: Polity Press

Carlson, M (2015) Introduction: The Many Boundaries of Journalism. In: Carlson M, Lewis SC (eds) Boundaries of Journalism. Professionalism, Practices and Participation. London: Routledge, 1–18

Carr, N (2008) The Big Switch: Rewiring the World, from Edison to Google. New York: W.W. Norton

Chadwick, A (2013) The Hybrid Media System: Politics and Power. New York: Oxford University Press

Chakrabarti, S, Rooney, C, Kweon, M (2018) Verification, Duty, Credibility: Fake News and Ordinary Citizens in Kenya and Nigeria. London: BBC News. Available at: https://downloads.bbc.co.uk/mediacentre/bbc-fake-news-research-paper-nigeria-kenya.pdf

Claussen, J, Peukert, C, Sen, A (2019) The Editor vs. the Algorithm: Returns to Data and Externalities in Online News. CESifo Working Paper Series 8012. Available at: https://ideas.repec.org/p/ces/ceswps/_8012.html

Coward, R (2013) Speaking Personally: The rise of subjective and confessional journalism. Basingstoke, UK: Palgrave Macmillan

Dewey, C (2016) 6 in 10 of You Will Share This Link Without Reading It, A New Depressing Study Says. Washington Post, 16 June. Available at: https://www.washingtonpost.com/news/the-intersect/wp/2016/06/16/six-in-10-of-you-will-share-this-link-without-reading-it-according-to-a-new-and-depressing-study/

Durham MG (1998) On the relevance of standpoint epistemology to the practice of journalism: the case for ‘strong objectivity’. Commun Theory 8(2):117–140

Ekström M (2002) Epistemologies of TV journalism: a theoretical framework. Journalism 3(3):259–282

Ekström M, Westlund O (2019) The dislocation of news journalism: a conceptual framework for the study of epistemologies of digital journalism. Med Commun 7(1):259–270

Ettema JS, Glasser TL (1985) On the epistemology of investigative journalism. Communication 8:183–206

Google Scholar  

Fletcher R, Nielsen RK (2017) Are news audiences increasingly fragmented? A cross-national analysis of cross-platform news audience fragmentation and duplication. J Commun 67(4):476–498

Gade, PJ (2011) Postmodernism, Uncertainty, and Journalism. In: Lowrey, W, Gade PJ (eds) Changing The News: The Forces Shaping Journalism in Uncertain Times. Routledge: New York, 77–96

Gieryn TF (1983) Boundary-work and the Demarcation of Science from Non-science: Strains and Interests in Professional Ideologies of Scientists. Am Sociol Rev 48(6):781–795

Hallin, D (1994) We Keep America on the Top of the World. Routledge: New York

Hartley J (2000) Communicative democracy in a redactional society: the future of journalism studies. J Theory Pract Crit 1(1):39–48

Heidegger, M (1991 [1961]) Nietzsche, Volumes Three and Four: III: The Will To Power as Knowledge and as Metaphysics. IV : Nihilism. Edited by DF Krell. San Francisco: Harper Collins

Jarman JW (2016) Influence of political affiliation and criticism on the effectiveness of political fact checking. Commun Res Rep 33(1):9–15

Jenkins, H, Ford, S, Green, J (2013) Spreadable Media: Creating Value and Meaning in a Networked Culture. New York University Press: New York

Jukes, S (2020) Journalism and Emotion. Sage: London

Klinger U, Svensson J (2018) The end of media logics? On algorithms and agency. New Media Soc 17(8):4653–4670

Lähteenmäki M (1998) On meaning and understanding: a dialogical approach. Dialogism 1:74–91

La Caze, M, Lloyd, HM (2011) Editor’s Introduction: Philosophy and the Affective Turn. Parrhesia 13:1–13

Lash, S (1990) Sociology of Postmodernism. Routledge: London-New York

Levy R (2021) Social media, news consumption, and polarization: evidence from a field experiment. Am Econ Rev 111(3):831–870

Lippmann, W (1922) Public Opinion. Brace & Co, New York: Harcourt

Lyotard, J-F (1984 [1979]) The Postmodern Condition: A Report on Knowledge. Translated by Bennington G and Massumi B. Minneapolis: University of Minnesota Press

Mabweazara HM (2014) Introduction ‘Digital technologies and the evolving African newsroom’: towards an African digital journalism epistemology. Digit Journal 2(1):2–11

Maddalena, G, Gili, G (2020) The History and Theory of Post-Truth Communication. Palgrave Macmillan: London

Mare A, Mabweazara HM, Moyo D (2019) “Fake News” and cyber-propaganda in sub-saharan Africa: recentering the research agenda. Afr Journal Stud 40(4):1–12

McNair, B (2012) Journalism and Democracy: An evaluation of the political public sphere. Routledge: London

Nerone J (2013) The historical roots of the normative model of journalism. Journalism 14(4):446–458

Newman, N, Fletcher, R, Levy, D, Robertson, CT, Nielsen, RK (2022) Reuters Institute Digital News Report 2022. Oxford: Reuters Institute for the Study of Journalism. Available at: https://reutersinstitute.politics.ox.ac.uk/digital-news-report/2022

Nielsen, RK (2017) News media, search engines and social networking sites as varieties of online gatekeeepers. In: Peters C, Broersma M (eds) Rethinking Journalism Again. Routledge: Abingdon-New York, 81–96

Nietzsche, F (1967 [1885-1887]) The Will to Power. Random House: New York

Nyhan B, Reifler J (2010) When corrections fail: the persistence of political misperceptions. Polit Behav 32(2):303–330

Osnos, E (2018) Can Mark Zuckerberg Fix Facebook Before It Breaks Democracy? New Yorker, September 17. Available at: https://www.newyorker.com/magazine/2018/09/17/can-mark-zuckerberg-fix-facebook-before-it-breaks-democracy

Papacharissi Z (2015) Affective publics and structure of storytelling: sentiment, events and mediality. Inf Commun Soc 19(3):307–324

Pariser, E (2011) The Filter Bubble. Penguin Press: New York

Peters C (2011) Emotion aside or emotional side? Crafting an ‘experience on involvement’ in the news. Journalism 12(3):297–316

Plummer, K (2001) Documents of life 2: An invitation to a critical humanism. SAGE: London

Popper, K (2002 [1963]) Conjectures and Refutations: The Growth of Scientific Knowledge. Routledge: London

Quattrociocchi, W, Vicini, A (2016) Misinformation. Guida alla società dell’informazione e della credulità. Milano: FrancoAngeli

Rebillard F, Smyrnaios N (2010) Les infomédiaires, au coeur de la filière del l’information d’actualité en ligne: Les cas del Google, Wikio et Paperblog. Réseaux 160:163–194

Rorty, R (1989) Contingency, Irony, and Solidarity. Cambridge University Press: Cambridge, UK

Schudson, M (1990) Origins of the Ideal of Objectivity in the Professions: American Journalism and Law, 1830 to 1940. Garland Publishing: New York

Schudson M (2001) The objectivity norm in American journalism. Journalism 2(2):149–170

Schudson, M (2008) Why Democracies Need an Unlovable Press. Polity Press: Cambridge, UK

Schudson, M (2018) Why Journalism Still Matters . Polity Press: Cambridge, UK

Shoemaker, PJ, Vos, T (2009) Gatekeeping Theory. Routledge: New York

Shotter J (1997) On a different ground: from contest between monologues to dialogical contest. Argumentation 11:95–112

Silverstone, R (2007) Media and Morality: On the Rise of the Mediapolis. Polity Press: Cambridge, UK

Soffer O (2009) The competing ideals of objectivity and dialogue in American journalism. Journalism 10(4):473–491

Steensen S, Grøndahl Larsen AM, Hågvar YB, Fonn BK (2019) What does digital journalism studies look like? Digit Journal 7(3):320–342

Sunstein, C (2001) Republic.com. Princeton University Press: Princeton, NJ

Tuchman G (1972) Objectivity as strategic ritual: an examination of newsmen’s notions of objectivity. Am J Sociol 77:660–679

van Dijck J (2014) Datafication, dataism and dataveillance: big data between scientific paradigm and ideology. Surveill Soc 12(2):197–208

van Dijck, J, Poell, T, de Waal, M (2018) The Platform Society: Public Values in a Connective World. Oxford University Press: New York

Vattimo, G, Rovatti, PA (2013 [1983]) Weak Thought. Suny Press: New York

Wahl-Jorgensen, K (2017) Is there a ‘postmodern turn’ in journalism? In: Peters C, Broersma M (eds) Rethinking Journalism Again. Routledge: Abingdon-New York, 98–112

Zhao H, Fu S, Chen X (2020) Promoting users’ intention to share online health articles on social media: The role of confirmation bias. Inf Process Manag 57(6):102354. https://doi.org/10.1016/j.ipm.2020.102354

Article   PubMed   PubMed Central   Google Scholar  

Download references

Author information

Authors and affiliations.

University of Naples Federico II, Naples, Italy

Luca Serafini

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Luca Serafini .

Ethics declarations

Competing interests.

The author declares no competing interests.

Additional information

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 licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as 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 images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Serafini, L. The old-new epistemology of digital journalism: how algorithms and filter bubbles are (re)creating modern metanarratives. Humanit Soc Sci Commun 10 , 395 (2023). https://doi.org/10.1057/s41599-023-01905-6

Download citation

Received : 14 March 2023

Accepted : 26 June 2023

Published : 10 July 2023

DOI : https://doi.org/10.1057/s41599-023-01905-6

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

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Research in Journalism, Media and Culture

Main navigation.

Faculty in Journalism, Culture and Communication analyze emerging and enduring forms of public communication and the institutional and economic conditions that sustain them. They employ a range of research methods, including ethnography, textual and historical analysis, and political economic approaches to media industries. As digital technologies have transformed mediated practices, the faculty has opened a series of new areas of inquiry, including computational journalism, the study of algorithms in institutions, and the cultural history of Silicon Valley.

Angèle Christin  is an associate professor. She is interested in fields and organizations where algorithms and ‘big data’ analytics transform professional values, expertise, and work practices. In her dissertation, she analyzed the growing importance of audience metrics in web journalism in the United States and France. Drawing on ethnographic methods, she examined how American and French journalists make sense of traffic numbers in different ways, which in turn has distinct effects on the production of news in the two countries. In a new project, she studies the construction, institutionalization, and reception of analytics and predictive algorithms in the U.S. criminal justice system.

Ted Glasser  is an emeritus professor.  His teaching and research focus on media practices and performance, with emphasis on questions of press responsibility and accountability. His books include  Normative Theories of the Media: Journalism in Democratic Societies,  written with Clifford Christians, Denis McQuail, Kaarle Nordenstreng, and Robert White, which in 2010 won the Frank Luther Mott-Kappa Tau Alpha award for best research-based book on journalism/mass communication and was one of three finalists for the Association for Education in Journalism and Mass Communication’s Tankard Book Award;  The Idea of Public Journalism , an edited collection of essays, recently translated into Chinese;  Custodians of Conscience: Investigative Journalism and Public Virtue , written with James S. Ettema, which won the Society of Professional Journalists’ award for best research on journalism, the Bart Richards Award for Media Criticism, and the Frank Luther Mott-Kappa Tau Alpha award for the best research-based book on journalism/mass communication;  Public Opinion and the Communication of Consent , edited with Charles T. Salmon; and  Media Freedom and Accountability , edited with Everette E. Dennis and Donald M. Gillmor.  His research, commentaries and book reviews have appeared in a variety of publications, including the  Journal of Communication, Journalism & Mass Communication Quarterly, Critical Studies in Mass Communication, Journalism Studies, Policy Sciences, Journal of American History, Quill, Nieman Reports  and  The New York Times Book Review .

James T. Hamilton  is a professor and the director of the Journalism Program. His work on the economics of news focuses on the market failures involved in the production of public affairs coverage and the generation of investigative reporting. Through research in the emerging area of computational journalism, he is exploring how to lower the cost of discovering stories about the operation of political institutions.

Xiaochang Li  is an assistant professor. She is broadly interested in the history of informatics, computation, and related data practices. Drawing upon media history, history of science, and STS, her work is concerned with how information technologies shape the production and circulation of knowledge and the relationship between technical practices and social worlds. Her current research examines the history of speech recognition and natural language processing and how the pursuit of language influenced the development of AI, Machine Learning, and contemporary algorithmic culture. Her work also touches on sound studies and the history of acoustics and she has previously worked on topics concerning transnational media audiences and digital content circulation.

Fred Turner  is a professor and cultural historian of media and media technologies. Trained in both Communication and Science and Technology Studies, he has long been interested in how media and American culture have shaped one another over time. His most recent work has focused on the rise of American technocracy since World War II and on the aesthetic and ideological manifestations of that rise in the digital era. Before earning his Ph.D., Turner worked as a journalist for ten years. He continues to write for newspapers and magazines and strongly supports researchers who seek to have a public impact with their work.

Like all Communication faculty, the members of the Journalism, Communication and Culture group routinely collaborate with colleagues from around the campus. The group enjoys particularly strong collaborations with sociologists, historians, art historians, and computer scientists.

Faculty — Journalism, Media and Culture

Doctoral Students — Journalism, Media and Culture

Selected Graduates

• Sanna Ali, Ph.D. 2023. AI Policy Analyst, Stanford Cyber Policy Center and RegLab
• Jeff Nagy, Ph.D. 2023. Assistant Professor, Communication and Media Studies, York University
• Anna Gibson, Ph.D. 2022. Postdoctoral Fellow, MIT Comparative Media Studies/Writing
• Jihye Lee , Ph.D. 2022. Assistant Professor, School of Advertising and Public Relations, UT Austin
• Andreas Katsanevas, Ph.D. 2020. Technology Policy Researcher, Meta
• Sheng Zou,  Ph.D. 2020. Assistant Professor, School of Communication, Hong Kong Baptist University
• Christine Larson , Ph.D. 2017. Assistant Professor, Journalism, University of Colorado, Boulder
• Yeon Joo , Ph.D. 2014. Associate Professor, Department of Digital Media, Myongji University, Seoul
• Morgan G. Ames , Ph.D. 2013. Assistant Adjunct Professor, School of Information, University of California, Berkeley
• Seeta Pena Gangdaharan , Ph.D. 2012. Assistant Professor, Department of Media and Communications, London School of Economics
• Mike Ananny , Ph.D. 2011, Associate Professor, Annenberg School for Communication and Journalism, University of Southern California
• Daniel Kreiss , Ph.D. 2010, Associate Professor, School of Media and Journalism, University of North Carolina, Chapel Hill
• John Kim , Ph.D. 2009, Associate Professor, Media and Cultural Studies, Macalaster College
• Erica Robles-Anderson , Ph.D. 2009, Associate Professor, Media, Culture, and Communication, New York University
• Isabel Awad , Ph.D. 2007, Associate Professor, Department of Media and Communication, Erasmus University (Netherlands)
• Cherian George , Ph.D. 2003, Professor, Associate Dean, School of Communication, Hong Kong Baptist University
• Francis Lapfung Lee , Ph.D. 2003, Professor, School of Journalism and Communication, City University of Hong Kong

Journalism in Practice Studies: A Systematic Review

• First Online: 22 February 2024

Cite this chapter

• Thouraya Snoussi 4 &
• Nourah Ahmed Al-Hooti   ORCID: orcid.org/0000-0002-3663-0130 5  

Part of the book series: Springer Handbooks of Political Science and International Relations ((SHPPSIR))

146 Accesses

This chapter deals with the literature related to journalism in practice studies (JPS) during the years from 2003 to 2023 in the hope of recognizing the usefulness of these studies for the society, especially for journalism practitioners. The findings of the systematic review highlight the growing interest in JPS research along with the increased recognition of its perspectives over the past 20 years. The analysis provided us with many studies that were based on different disciplines and dealt with a variety of fields, such as journalism ethics and professionalism, newsroom cultures and practices, and technological shifts in journalism. Furthermore, the sample studied shows that JPS has focused on qualitative studies, including textual and narrative analyses. For this, the authors believe that the JPS literature considered in this chapter provides insights for journalist practitioners into role development, changing audience behaviors, and technological advances, allowing them to adapt and stay informed. As advances in technology continue to shape the field, JPS is keeping an eye on digital transformation, especially artificial intelligence, and looks for ways to leverage it in the practice of journalism. Thus, future research will be devoted to exploring the ever-evolving landscape of journalism and its multifaceted dimensions.

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

Access this chapter

• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Available as EPUB and PDF
• Durable hardcover edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Allan, S. (2018). What is journalism studies for? Journalism Studies, 19 (7), 972–974.

Google Scholar  

Allan, S., & Thorsen, E. (2009). Citizen journalism: Global perspectives . Peter Lang.

Anderson, C. W. (2013). Journalism: From tradition to practice . Routledge.

Barendt, E., & Lustgarten, L. (2005). Freedom of speech and the press . Oxford University Press.

Barnhurst, K. G., & Nerone, J. (2001). The formative years of American journalism. Journalism Studies, 2 (2), 261–279.

Boczkowski, P. J., & Anderson, C. W. (2017). Remaking the news: Essays on the future of journalism scholarship in the digital age . MIT Press.

Book   Google Scholar  

Boczkowski, P. J., & Papacharissi, Z. (Eds.). (2018). Trump and the media . MIT Press.

Bourdieu, P. (1993). The field of cultural production: Essays on art and literature . Columbia University Press.

Boyd, D. (2014). It’s complicated: The social lives of networked teens . Yale University Press.

Bruns, A., & Burgess, J. E. (2012). Researching news discussion on Twitter: New methodologies. Journalism Studies, 13 (5–6), 801–814.

Article   Google Scholar  

Bruns, A., & Highfield, T. (2017). The quantified backchannel: Digital traces on conference Twitter. Digital Journalism, 5 (6), 682–700.

Bull, A. (2018). Slow journalism and the digital disruption of news. Digital Journalism, 6 (3), 352–368. https://doi.org/10.1080/21670811.2017.1343815

Campbell, W. J., & Jamieson, K. H. (1990). The press effect: Politicians, journalists, and the stories that shape the political world . Oxford University Press.

Couldry, N., & Hepp, A. (2017). The mediated construction of reality. Media, Culture and Society, 39 (1), 5–8.

Dahlgren, P. (2018). Journalism and the public sphere. Journalism, 19 (11), 1405–1409.

Deuze, M. (2005). What is journalism? Professional identity and ideology of journalists reconsidered. Journalism, 6 (4), 442–464.

Deuze, M. (2007). Media work and the reshaping of journalism. Journalism, 8 (2), 121–140.

Deuze, M., & Witschge, T. (2018). Beyond journalism: Theorizing the transformation of journalism. Journalism Studies, 19 (11), 1537–1552.

Entman, R. M. (2012). Scandal and silence: Media responses to presidential misconduct . Wiley-Blackwell.

Entman, R. M., & Matthes, J. (2018). Media framing effects: Conceptualizing and measuring issue salience. In The Oxford handbook of political communication theories (pp. 1–18). Oxford University Press.

Franklin, B. (2009). Journalism studies at the crossroads: Challenges and opportunities in the digital age. Journalism, 10 (1), 5–7.

Franklin, B. (2016). Researching news production worldwide: Some reflections on the craft of comparative journalism studies. Journalism Studies, 17 (6), 667–680.

Franklin, B., & Murphy, D. (Eds.). (2020). Journalism in practice: Ethical and professional issues . Routledge.

Friedland, L. A., Hoveyda, A. R., & Lee, B. Y. L. (2006). Intellectual property law and interactive media: Free speech vs fair use? Journal of Interactive Advertising, 6 (1), 73–84.

Fulton, J., & Scott, P. (2022). Time well spent: ABC journalists reflect on slow journalism and the remote communities project. Journalism Practice, 16 (8), 1597–1613.

Galison, P. (1997). Image and logic: A material culture of microphysics . University of Chicago Press.

Galison, P., & Thompson, E. A. (Eds.). (1999). The architecture of science . MIT Press.

Garnett, J., & Lewis, D. (2009). Defamation Law and Social Media Communications in Australia: An Overview Journal of Media Law, 1 (2), 269–286.

Gieryn, T. F. (1983). Boundary-work and the demarcation of science from non-science: Strains and interests in professional ideologies of scientists. American Sociological Review, 48 (6), 781–795.

Gillmor, D. (2006). We the media: Grassroots journalism by the people, for the people . O’Reilly Media.

Ginosar, A., & Reich, Z. (2020). Obsessive-activist journalists: A new model of journalism? Journalism Practice, 16 (4), 1–21. https://doi.org/10.1080/17512786.2020.1816488

Hallin, D. C., & Mancini, P. (2004). Comparing media systems: Three models of media and politics . Cambridge University Press.

Hanusch, F. (2018). Journalism as a practice profession. In The handbook of journalism studies (2nd ed., pp. 189–205). Routledge.

Hanusch, F., & Tandoc, E. C., Jr. (2019). Journalism Studies Today Journalism Studies . https://doi.org/10.1080/1461670X.2019

Herman, E. S., & Chomsky, N. (1988). Manufacturing consent: The political economy of the mass media . Pantheon Books.

Hermida A., Fletcher F., Korell D., Logan D., & Glasgow J. (2012). The Tweeted Times: Journalism Skills in the Information Age Journal of Education for Journalism and Mass Communication .

Hirst, M., & Patching, R. (2018). Slow journalism: Exploring an emerging news paradigm. In J. Bennett & N. Strange (Eds.), Digital journalism: Theory, practice and research (pp. 201–214). Routledge.

Holton, A. (2016). The elements of slow Journalism: A case study in the long-form feature. In A. Williams & M. Mair (Eds.), Handbook of journalism studies (pp. 287–301). Routledge.

Kovach, B., & Rosenstiel, T. (2010). The elements of journalism: What newspeople should know and the public should expect . Three Rivers Press.

Kovach, B., & Rosenstiel, T. (2014). The elements of journalism: What newspeople should know and the public should expect . Three Rivers Press.

Krippendorff, K. (2018). Content analysis: An introduction to its methodology . Sage.

Lewis, S. C., & Usher, N. (2013). Introduction: Citizen journalism and democracy. In S. C. Lewis & N. Usher (Eds.), Citizen journalism: Global perspectives (pp. 1–18). Peter Lang Publishing.

Lewis, S. C., & Molyneux, L. (2018). A decade of research on social media and journalism: Assumptions, blind spots, and a way forward. Media and communication, 6 (4), 11–23.

Marmot, M. (2005). Social determinants of health inequalities. The Lancet, 365 (9464), 1099–1104.

Maras, S. (2013). Objectivity in journalism . John Wiley & Sons.

McChesney, R. W., & Nichols, J. (2010). The death and life of American journalism: The media revolution that will begin the world again . Nation Books.

McQuail, D. (2013). Media policy: Premature obsequies? In Media power, professionals and policies (pp. 19–34). Routledge.

McQuail, D., & Windahl, S. (1993). Communication models for the study of mass communication . Longman.

Merriam-Webster. (2021). Journalism . In Merriam-Webster.com dictionary. Retrieved April 6, 2021, from https://www.merriam-webster.com/dictionary/journalism

Merrill, J. C., & Fisher, D. R. (2016). Media history: Society culture technology . Routledge.

Merritt, D., & Milligan, J. (2006). A history of American journalism . Routledge.

Nelkin, D. (1995). Selling science: How the press covers science and technology . W.H. Freeman.

Palmer, J. (2013). Environmental journalism . Routledge.

PRISMA. (2020). PRISMA 2020 explanation and elaboration: Updated guidance and exemplars for reporting systematic reviews. PLoS Medicine, 17 (3), e1003293. https://doi.org/10.1371/journal.pmed.1003293

Rogers, E. M. (2003). Diffusion of innovations (5th ed.). Free Press.

Schudson, M. (1978). Discovering the news: A social history of American newspapers . Basic Books.

Schudson, M. (1995). The power of news . Harvard University Press.

Schudson, M. (2005). Four approaches to the sociology of news. Journalism Studies, 6 (2), 115–129.

Schultz, B., Sheffer, M. L., & Sporting News. (2011). Sports journalism and society . Routledge.

Singer, J. B. (2010). Journalism ethics amid structural change: A study of international codes. Journal of Mass Media Ethics, 25 (4), 261–279.

Singer, J. B. (2018). User-generated visibility: Secondary gatekeeping in a shared media space. Digital Journalism, 6 (2), 146–164.

Singer, J. B., Hermida, A., Domingo, D., Heinonen, A., Paulussen, S., Quandt, T., et al. (2011). Participatory journalism . Malden, MA: John Wiley & Sons.

Society of Professional Journalists. (2014). SPJ code of ethics . Retrieved from https://www.spj.org/ethicscode.asp

Stake, R. E. (2005). Qualitative case studies. In N. K. Denzin & Y. S. Lincoln (Eds.), The Sage handbook of qualitative research (3rd ed., pp. 443–466). Sage.

Tuchman, G. (1978). The news net. Social Research , 253–276.

Wahl-Jorgensen, K. (2009). Journalistic epistemology: Exploring the tensions between journalistic professionalism and academic research. Journalism, 10 (3), 299–319.

Ward, S. J. A. (2014). Ethics and the media: An introduction . Cambridge University Press.

Ward, S. (2015). Radical media ethics: A global approach . Wiley-Blackwell.

Wardle, C., & Derakhshan, H. (2017). Information disorder: Toward an interdisciplinary framework for research and policy making . Council of Europe Report.

Weaver, D. H., & Wilhoit, G. C. (1996). The American journalist in the 1990s: U.S. news people at the end of an era . Lawrence Erlbaum Associates.

Weaver, D. H., & Willnat, L. (2016). Changes in US journalism: How do journalists think about social media? Journalism Practice, 10 (7), 844–855.

Download references

Declaration of Conflicting Interests

The authors declared no conflicts.

Disclosure Statement

The authors reported no potential conflict of interest.

The authors received no financial support for the research.

Author information

Authors and affiliations.

Department of Mass Communication, College of Communication, University of Sharjah, Sharjah, UAE

Thouraya Snoussi

Public Relations Department, College of Communication, University of Sharjah, Sharjah, UAE

Nourah Ahmed Al-Hooti

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Thouraya Snoussi .

Editor information

Editors and affiliations.

College of Communication, University of Sharjah, Sharjah, United Arab Emirates

Leon Barkho

Jairo Alfonso Lugo-Ocando

School of International Communications, University of Nottingham Ningbo China, Ningbo, China

Sadia Jamil

Rights and permissions

Reprints and permissions

Copyright information

© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this chapter

Snoussi, T., Al-Hooti, N.A. (2024). Journalism in Practice Studies: A Systematic Review. In: Barkho, L., Lugo-Ocando, J.A., Jamil, S. (eds) Handbook of Applied Journalism. Springer Handbooks of Political Science and International Relations. Springer, Cham. https://doi.org/10.1007/978-3-031-48739-2_9

Download citation

DOI : https://doi.org/10.1007/978-3-031-48739-2_9

Published : 22 February 2024

Publisher Name : Springer, Cham

Print ISBN : 978-3-031-48738-5

Online ISBN : 978-3-031-48739-2

eBook Packages : Political Science and International Studies Political Science and International Studies (R0)

Share this chapter

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

• Publish with us

Policies and ethics

• Find a journal
• Track your research

Numbers, Facts and Trends Shaping Your World

Read our research on:

Full Topic List

Regions & Countries

• Publications
• Our Methods
• Short Reads
• Tools & Resources

Read Our Research On:

• Journalists Sense Turmoil in Their Industry Amid Continued Passion for Their Work

77% would choose their career all over again, though 57% are highly concerned about future restrictions on press freedom

Table of contents.

• ‘Struggling’ and ‘chaotic’ are the words journalists most often use to describe their industry
• Journalists say the industry’s best function is getting the news out, but many say it fails at getting the story right
• Mixed reports of salaries and newsroom growth in the past year
• Most journalists are not in a union, but many say they would join
• About one-in-ten journalists have experienced harassment inside their organization, most often along political lines
• Journalists largely say COVID-19 has changed their organizations forever
• Journalists see false and made-up news as a big problem and don’t have much confidence in how the industry handles it
• Most journalists think it is important to report on the false statements of public figures
• Vast majority of journalists are against requiring a license to be a journalist
• Most journalists who were harassed by someone outside their organization in the past year say it happened on social media
• Journalists sense that Americans don’t have high trust in the news media overall, but think their own audience trusts them
• Many journalists feel connected to their audiences; the public doesn’t feel the same level of connection
• Journalists express confidence in their sources but rarely interview them in person
• Journalists less likely than the public to say each side of a story always deserves equal coverage
• Journalists have mixed views on whether they can keep their personal views out of their reporting
• About half of journalists took part in workplace diversity training in the past year; more likely among Black, Hispanic and Asian journalists
• Journalists who say their news organization has a left-leaning audience are more likely to engage with diversity issues, see diversity shortfalls
• Freelancers far less engaged with issues of diversity at news organizations
• Younger journalists less satisfied with their organization’s diversity efforts
• Oldest journalists most concerned about possible press restrictions, but also most satisfied in their jobs
• Online journalists most likely to say their organization is expanding
• TV journalists most likely to want a license requirement
• Radio and podcast journalists feel most connected to their audience
• TV journalists are more likely to say their newsroom is diverse – and to take part in diversity trainings
• Acknowledgments
• Appendix: Detailed tables of the composition of the journalist survey sample
• Survey of journalists
• Advisory board
• Open-ended questions asked in the survey of journalists
• February survey of U.S. adults: American Trends Panel survey methodology
• March survey of U.S. adults: American Trends Panel survey methodology

Much of Pew Research Center’s earlier research on the U.S. news environment has focused on the public’s news consumption habits and views toward the news media. This major new undertaking was designed to capture the other side of the equation, asking U.S.-based journalists to provide their own perspective on the industry they work in.

The main source of data for this study is a Pew Research Center survey of 11,889 U.S.-based journalists who are currently working in the news industry and said that they report, edit or create original news stories in their current job. The survey was conducted online between Feb. 16 and March 17, 2022. See Appendix for a detailed demographic profile of the journalists who completed the survey.

Because there is no readily available list of all U.S. journalists, Center researchers relied on commercial databases of journalists based in the U.S. as well as supplemental lists of news organizations to create a broad and diverse sample of over 160,000 journalists from as many types of outlets and areas of reporting as possible. Although it is impossible to be certain every segment of the journalism profession in the U.S. is covered by the sample, the use of multiple databases and supplemental lists ensured that journalists from a variety of different reporting areas, news platform types, as well as outlet sizes and types – such as those who work for organizations that are intended to primarily reach a particular demographic group – were represented.

Propensity weighting was used to ensure that the responses of the 11,889 respondents aligned with the full sample of over 160,000 journalists with respect to job titles, media outlet type, freelance status and geographic location.

See topline for the questions asked in the survey. For more information on the development of the sample of journalists or the survey weighting, please see the methodology .

Another major goal of this study was to see how the views and attitudes of journalists compare with those of the American public. To accomplish this, the report also draws upon two surveys of U.S. adults – one of 9,388 adults conducted Feb. 7-13, 2022, and another of 10,441 adults conducted March 7-13, 2022. All respondents who took part in these two surveys are members of the Center’s American Trends Panel , an online survey panel that is recruited through national, random sampling of residential addresses. Recruiting our panelists by phone or mail ensures that nearly all U.S. adults have a chance of selection. This gives us confidence that any sample can represent the whole population (see our Methods 101 explainer on random sampling). To further ensure that each survey reflects a balanced cross-section of the nation, the data is weighted to match the U.S. adult population by gender, race, ethnicity, partisan affiliation, education and other categories.

See topline for the questions asked in the two surveys of U.S. adults, and the methodology .

This is the latest report in Pew Research Center’s ongoing investigation of the state of news, information and journalism in the digital age, a research program funded by The Pew Charitable Trusts, with generous support from the John S. and James L. Knight Foundation.

From the economic upheaval of the digital age to the rise of political polarization and the COVID-19 pandemic , journalism in America has been in a state of turmoil for decades . While U.S. journalists recognize the many challenges facing their industry, they continue to express a high degree of satisfaction and fulfillment in their jobs, according to an extensive new Pew Research Center survey of nearly 12,000 working U.S.-based journalists.

Seven-in-ten journalists surveyed say they are “very” or “somewhat” satisfied with their job, and an identical share say they often feel excited about their work. Even larger majorities say they are either “extremely” or “very” proud of their work – and that if they had to do it all over again, they would still pursue a career in the news industry. About half of journalists say their job has a positive impact on their emotional well-being, higher than the 34% who say it is bad for their emotional well-being.

At the same time, however, journalists recognize serious challenges in the news media more broadly. Indeed, when asked to describe their industry in a single word, nearly three-quarters of journalists surveyed (72%) use a word with negative connotations, with the most common responses being words that relate to “struggling” and “chaos.” Other, far less common negative words include “biased” and “partisan,” as well as “difficult” and “stressful.” (See Chapter 1 for more detailed figures and the methodology for more details about the question asked.)

The survey of 11,889 U.S. journalists, conducted Feb. 16-March 17, 2022, identified several specific areas of concern for journalists, including the future of press freedom, widespread misinformation, political polarization and the impact of social media.

More than half of journalists surveyed (57%) say they are “extremely” or “very” concerned about the prospect of press restrictions being imposed in the United States. And about seven-in-ten journalists (71%) say made-up news and information is a very big problem for the country, higher than the 50% of U.S. adults who say the same. At the same time, four-in-ten journalists say that news organizations are generally doing a bad job managing or correcting misinformation.

A large majority of journalists say they come across misinformation at least sometimes when they are working on a story, and while most say they are confident in their ability to recognize it, about a quarter of reporting journalists (26%) say they have unknowingly reported on a story that was later found to contain false information.

How to report on false statements has become a vexing question for journalists amid a turbulent political climate. The survey asked journalists what they think is the best approach to coverage when a public figure makes a false statement. By two-to-one, journalists are more likely to say the best approach is to “report on the statement because it is important for the public to know about” (64%) rather than to “not report on the statement because it gives attention to the falsehoods and the public figure” (32%).

Still, there is no consensus that opposing views always warrant equal coverage. What historically may have been considered a standard norm of journalism (and even a requirement for broadcast stations in their election coverage ) seems, in today’s political environment, to be facing a reevaluation as heated debate ensues around the issue of “bothsidesism” – whether news outlets should be committed to always giving equal attention to all sides of an issue.

A little over half of journalists surveyed (55%) say that in reporting the news, every side does not always deserve equal coverage, greater than the share who say journalists should always strive to give every side equal coverage (44%). 

On the other hand, journalists express wide support for another long-standing norm of journalism: keeping their own views out of their reporting. Roughly eight-in-ten journalists surveyed (82%) say journalists should do this, although there is far less consensus over whether journalists meet this standard. Just over half (55%) think journalists are largely able to keep their views out of their reporting, while 43% say journalists are often unable to.

Some of journalists’ views – such as whether every side deserves equal coverage – are connected to the ideological composition of their audiences. Journalists were asked about the political leanings of the audience at the organization where they work (or the main one they work for if they work for more than one), and roughly half say that their audience leans predominantly to the left (32%) or right (20%). An additional third say their organization has a more politically mixed audience, while 13% are unsure.

Even as they recognize audience leanings, journalists express deep concerns over political sorting in news consumption habits, with three-quarters of those surveyed saying it is a major problem when people with the same political views get their news from the same news organizations. The American public, however, appears much less worried: Roughly four-in-ten U.S. adults (39%) say this a major problem. To be able to make comparisons between journalists’ views and those of the public on certain key issues, the Center conducted two separate surveys around the same time as the journalist survey, posing some of the same questions to roughly 10,000 U.S. adults who are part of the Center’s  American Trends Panel . (Read more about those two surveys in the methodology section .)

Maintaining widespread journalistic credibility in a polarized climate can seem like an impossible task – and many journalists seem to recognize that. While three-quarters of journalists say that journalists largely agree on the basic facts of the news – even if they report on them in different ways – about half of journalists surveyed (52%) say it is not possible to report news that “nearly everyone finds accurate.” An even greater share of the U.S. public overall (62%) says it is not possible to report news that is universally accepted as accurate.

Journalists and the public far apart on their assessment of today’s news industry 

The survey’s results show that journalists recognize that the public views them and their work with deep skepticism . When asked what one word they think the public would use to describe the news industry these days, journalists overwhelmingly give negative responses, with many predicting that the public would describe the news media as “inaccurate,” “untrustworthy,” “biased” or “partisan.” (Read the methodology for more detail.)

Moreover, just 14% of journalists surveyed say they think the U.S. public has a great deal or fair amount of trust in the information it gets from news organizations these days. Most believe that Americans as a whole have some trust (44%) or little to no trust (42%).

When a similar question was posed to the general public, 29% of U.S. adults say they have at least a fair amount of trust in the information they get from news outlets, while 27% say they have some trust and 44% have little to none.

This disconnect between journalists and the public also comes through when each group is asked about the job news organizations are doing with five core functions of journalism: covering the most important stories of the day, reporting the news accurately, serving as a watchdog over elected leaders, giving voice to the underrepresented, and managing or correcting misinformation.

In all five areas, journalists give far more positive assessments than the general public of the work news organizations are doing. And on four of the five items, Americans on the whole are significantly more likely to say the news media is doing a bad job than a good job. For example, while 65% of journalists say news organizations do a very or somewhat good job reporting the news accurately, 35% of the public agrees, while 43% of U.S. adults say journalists do a bad job of this.

Similarly, while nearly half of journalists (46%) say they feel extremely or very connected with their audiences, only about a quarter of the public (26%) feels that connection with their main news organizations.

Journalists see social media as both a blessing and a curse

These days, many journalists connect with audiences through social media, and they see this as a double-edged sword. On the one hand, 94% of journalists surveyed report using social media in their work to some degree, and these journalists see a number of ways that social media helps them do their jobs. For example, among journalists who use social media for their work, 87% say it has a very or somewhat positive impact on promoting news stories, and 79% say it helps them connect with their audience and find sources for their news stories.

At the same time, however, two-thirds of all journalists surveyed (67%) say social media has a very or somewhat negative impact on the state of journalism as a whole. Just 18% say social media has a positive impact on the news industry, while 14% say it has neither a positive nor negative impact.

Additionally, roughly four-in-ten journalists (42%) say they have experienced job-related harassment or threats by someone outside their own organization in the past year, and within this group, a large majority (78%) say that harassment came through social media at least once. That means that one-third of all journalists surveyed report being harassed on social media in the last 12 months.

Journalists offer mixed views about diversity and inclusion in the newsroom

As issues of diversity and inclusion in the workspace gain heightened attention around the country , journalists offer mixed views of how their organizations are doing at diversifying staff. Although two-thirds (67%) say their organization has achieved sufficient gender diversity, about half as many – 32% – say it has reached sufficient racial and ethnic diversity. And fewer than half of respondents (42%) characterize addressing issues around diversity and inclusion as a major priority for their newsroom.

Roughly two-thirds of journalists say their organization generally treats everyone fairly regardless of age, gender, or race and ethnicity. But these figures are not quite as high among certain groups. For example, Black, Hispanic and Asian journalists are less likely than White journalists to say that their organization treats everyone fairly based on race and ethnicity. Read Chapter 6 for more details. (The appendix provides a detailed demographic profile of the journalists who completed the survey.)

Younger journalists give lowest grades on newsroom diversity, older journalists more worried about the future of press freedom

In a number of areas, there is a large amount of agreement across different groups of journalists. Still, some differences do emerge – particularly by age.

The youngest journalists (ages 18 to 29) are least likely to say their organization has enough diversity in the newsroom in a number of areas. For instance, nearly seven-in-ten (68%) say there is not enough racial and ethnic diversity, compared with 37% of journalists ages 65 and older who say the same. Younger journalists also engage with this issue more: About half of journalists under 50 say they discuss their organization’s diversity with colleagues at least several times a month, far higher than the 30% of those 65 and older who say this.

The oldest group of journalists, meanwhile, tend to feel more fulfilled by their job. Three-quarters of journalists 65 and older say their job has a very or somewhat positive impact on their emotional well-being, substantially higher than the 29% of journalists under 30 who say the same. Older journalists, meanwhile, are more worried about the future of press freedom – 68% say they are extremely or very concerned about possible restrictions on press freedom in the country, compared with 42% of those ages 18 to 29. (Read Chapter 7 for more details.)

Other key findings from the survey include:

• The survey sought to gauge the financial standing of journalists and the economics of the organizations they work for. The findings could be read with optimism or concern – depending on one’s vantage point. About four-in-ten journalists surveyed (41%) say they received a salary increase in the past year. The greatest portion – 50% – say their salary has stayed the same, while far fewer (7%) experienced a pay cut. Looking at news organizations more broadly, journalists are somewhat more likely to say their news organization has been expanding (30%) in the past year than to say it’s been cutting back (22%), with absence of change again being most common (46%). All in all, many journalists – 42% – say they are at least somewhat concerned about their job security, though far fewer express the highest level of concern.
• Despite concerns about the spread of misinformation and reporters inserting their own views into stories, about three-quarters of journalists (74%) say that people should be allowed to practice journalism without needing a license, while one-quarter say people should be required to have a license should in order to practice journalism.
• The vast majority of journalists surveyed say that at least some of the stories they have worked on in the past year had to do with the COVID-19 pandemic. But perhaps even more striking, six-in-ten say that the pandemic has brought either a great deal (26%) or a fair amount (34%) of permanent change to news reporting at their organization.
• Journalists’ opinions also sometimes differ based on the platform of their organization – particularly journalists who work in television. Compared with those in print, audio or online journalism, journalists who work for an organization that originated on TV seem the least happy with their job. About one-third (34%) say the news industry has a very or somewhat positive impact on their emotional well-being, considerably lower than the 54% of those who work online, 52% who work in print and 48% who work in radio or podcasting who say the same. And TV journalists are much more likely to say they were harassed by someone outside their organization in the past year (58%).

Sign up for our weekly newsletter

Fresh data delivery Saturday mornings

Sign up for The Briefing

Weekly updates on the world of news & information

• Free Speech & Press
• Freedom of the Press
• Journalists
• Media Industry
• Media Layoffs & Employment
• Media Polarization
• Misinformation
• Politics & Media
• Social Media & the News
• Trust in Media

Americans’ Views of Technology Companies

Most americans say a free press is highly important to society, ­most americans favor restrictions on false information, violent content online, freedom of speech and lgbt rights: americans’ views of issues in supreme court case, most u.s. journalists are concerned about press freedoms, most popular, report materials.

• What do journalists think the news industry does best and worst?

1615 L St. NW, Suite 800 Washington, DC 20036 USA (+1) 202-419-4300 | Main (+1) 202-857-8562 | Fax (+1) 202-419-4372 |  Media Inquiries

Research Topics

• Email Newsletters

ABOUT PEW RESEARCH CENTER  Pew Research Center is a nonpartisan fact tank that informs the public about the issues, attitudes and trends shaping the world. It conducts public opinion polling, demographic research, media content analysis and other empirical social science research. Pew Research Center does not take policy positions. It is a subsidiary of  The Pew Charitable Trusts .

© 2024 Pew Research Center

RUNC: Lightning Czabovsky

This article was originally posted on endeavors.unc.edu . photo by megan mendenhall..

Lightning Czabovsky is an associate professor in the UNC Hussman School of Journalism and Media. He studies the intersection of strategic communication, audience analysis and diverse audiences.

Q: How did you discover your specific field of study?

A: I was working in Los Angeles examining box office returns. Tyler Perry’s “Diary of a Mad Black Woman” opened at number-one, beating much higher budgeted studio films released in twice the number of theaters. As a gay kid that viewed the world differently, seeing a film no one in the industry was even talking about until it succeeded, I was awestruck. Even though I was from a different background, seeing how the macro-environment missed diverse realities really struck a chord. Ever since, I’ve been focused on how media measures miss or don’t understand marginalized, micro or niche audiences.

Q: Academics are problem-solvers. Describe a research challenge you’ve faced and how you overcame it.

A: As someone who primarily focuses on marginalized or niche audiences in media, it’s often hard to operationalize what that means. I have done studies about the marketing materials of LGBTQ+ content, but the question often is: How do we even define or measure that? I’ve learned that in research you have to use existing models or tools to speak to those in your scientific community. But I encourage others to be themselves and make their own tools or measures, especially if the existing science minimizes the nuances of people.

Q: Describe your research in five words.

A: PR measures should be nuanced.

Q: Who or what inspires you? Why?

A: My trees in my backyard. They’ve been here longer than I have — and may remain long after my time here. They’re built by mother nature to be used in their climate dynamic. They’re tall and lanky to blow in the wind when a hurricane comes. They’re a great lesson in both serenity and survival. My husband. We have complete opposite backgrounds, yet we both value work, hard work, kindness and people. Life is tough. But life can be kind. And he always reminds me of the things beyond academia.

Q: If you could pursue any other career, what would it be and why?

A: I wouldn’t pick another career, really. I love my students and my field. It makes my life quite valuable. But, I suppose, if forced, I’ve always loved natural sciences. So, I would be a volcanologist, seismologist, astronomer or meteorologist — or something similar. The combination of science and saving people’s lives is where my soul is. In crises, we’re all in it together, regardless of our backgrounds. I just wish we operated that way every day. My career lets me blend my interest in people and science as a communication person who can understand the science and then try to explain it to those in need.

Research UNCovered delves into the lives of UNC researchers from all disciplines and career levels, showcasing not only their research prowess but personal experiences in academia and beyond.

Try searching for

• Misinformation
• Subscriptions
• Fact-checking

In this piece

What does the public in six countries think of generative ai in news.

An electronic screen displaying Japan's Nikkei share average. Credit: Reuters / Issei Kato

DOI: 10.60625/risj-4zb8-cg87

Executive summary.

Based on an online survey focused on understanding if and how people use generative artificial intelligence (AI), and what they think about its application in journalism and other areas of work and life across six countries (Argentina, Denmark, France, Japan, the UK, and the USA), we present the following findings.

Findings on the public’s use of generative AI

ChatGPT is by far the most widely recognised generative AI product – around 50% of the online population in the six countries surveyed have heard of it. It is also by far the most widely used generative AI tool in the six countries surveyed. That being said, frequent use of ChatGPT is rare, with just 1% using it on a daily basis in Japan, rising to 2% in France and the UK, and 7% in the USA. Many of those who say they have used generative AI have used it just once or twice, and it is yet to become part of people’s routine internet use.

In more detail, we find:

• While there is widespread awareness of generative AI overall, a sizable minority of the public – between 20% and 30% of the online population in the six countries surveyed – have not heard of any of the most popular AI tools.
• In terms of use, ChatGPT is by far the most widely used generative AI tool in the six countries surveyed, two or three times more widespread than the next most widely used products, Google Gemini and Microsoft Copilot.
• Younger people are much more likely to use generative AI products on a regular basis. Averaging across all six countries, 56% of 18–24s say they have used ChatGPT at least once, compared to 16% of those aged 55 and over.
• Roughly equal proportions across six countries say that they have used generative AI for getting information (24%) as creating various kinds of media, including text but also audio, code, images, and video (28%).
• Just 5% across the six countries covered say that they have used generative AI to get the latest news.

Findings on public opinion about the use of generative AI in different sectors

Most of the public expect generative AI to have a large impact on virtually every sector of society in the next five years, ranging from 51% expecting a large impact on political parties to 66% for news media and 66% for science. But, there is significant variation in whether people expect different sectors to use AI responsibly – ranging from around half trusting scientists and healthcare professionals to do so, to less than one-third trusting social media companies, politicians, and news media to use generative AI responsibly.

• Expectations around the impact of generative AI in the coming years are broadly similar across age, gender, and education, except for expectations around what impact generative AI will have for ordinary people – younger respondents are much more likely to expect a large impact in their own lives than older people are.
• Asked if they think that generative AI will make their life better or worse, a plurality in four of the six countries covered answered ‘better’, but many have no strong views, and a significant minority believe it will make their life worse. People’s expectations when asked whether generative AI will make society better or worse are generally more pessimistic.
• Asked whether generative AI will make different sectors better or worse, there is considerable optimism around science, healthcare, and many daily routine activities, including in the media space and entertainment (where there are 17 percentage points more optimists than pessimists), and considerable pessimism for issues including cost of living, job security, and news (8 percentage points more pessimists than optimists).
• When asked their views on the impact of generative AI, between one-third and half of our respondents opted for middle options or answered ‘don’t know’. While some have clear and strong views, many have not made up their mind.

Findings on public opinion about the use of generative AI in journalism

Asked to assess what they think news produced mostly by AI with some human oversight might mean for the quality of news, people tend to expect it to be less trustworthy and less transparent, but more up to date and (by a large margin) cheaper for publishers to produce. Very few people (8%) think that news produced by AI will be more worth paying for compared to news produced by humans.

• Much of the public think that journalists are currently using generative AI to complete certain tasks, with 43% thinking that they always or often use it for editing spelling and grammar, 29% for writing headlines, and 27% for writing the text of an article.
• Around one-third (32%) of respondents think that human editors check AI outputs to make sure they are correct or of a high standard before publishing them.
• People are generally more comfortable with news produced by human journalists than by AI.
• Although people are generally wary, there is somewhat more comfort with using news produced mostly by AI with some human oversight when it comes to soft news topics like fashion (+7 percentage point difference between comfortable and uncomfortable) and sport (+5) than with ‘hard’ news topics, including international affairs (-21) and, especially, politics (-33).
• Asked whether news that has been produced mostly by AI with some human oversight should be labelled as such, the vast majority of respondents want at least some disclosure or labelling. Only 5% of our respondents say none of the use cases we listed need to be disclosed.
• There is less consensus on what uses should be disclosed or labelled. Around one-third think ‘editing the spelling and grammar of an article’ (32%) and ‘writing a headline’ (35%) should be disclosed, rising to around half for ‘writing the text of an article’ (47%) and ‘data analysis’ (47%).
• Again, when asked their views on generative AI in journalism, between a third and half of our respondents opted for neutral middle options or answered ‘don’t know’, reflecting a large degree of uncertainty and/or recognition of complexity.

Introduction

The public launch of OpenAI’s ChatGPT in November 2022 and subsequent developments have spawned huge interest in generative AI. Both the underlying technologies and the range of applications and products involving at least some generative AI have developed rapidly (though unevenly), especially since the publication in 2017 of the breakthrough ‘transformers’ paper (Vaswani et al. 2017) that helped spur new advances in what foundation models and Large Language Models (LLMs) can do.

These developments have attracted much important scholarly attention, ranging from computer scientists and engineers trying to improve the tools involved, to scholars testing their performance against quantitative or qualitative benchmarks, to lawyers considering their legal implications. Wider work has drawn attention to built-in limitations, issues around the sourcing and quality of training data, and the tendency of these technologies to reproduce and even exacerbate stereotypes and thus reinforce wider social inequalities, as well as the implications of their environmental impact and political economy.

One important area of scholarship has focused on public use and perceptions of AI in general, and generative AI in particular (see, for example, Ada Lovelace Institute 2023; Pew 2023). In this report, we build on this line of work by using online survey data from six countries to document and analyse public attitudes towards generative AI, its application across a range of different sectors in society, and, in greater detail, in journalism and the news media specifically.

We go beyond already published work on countries including the USA (Pew 2023; 2024), Switzerland (Vogler et al. 2023), and Chile (Mellado et al. 2024), both in terms of the questions we cover and specifically in providing a cross-national comparative analysis of six countries that are all relatively privileged, affluent, free, and highly connected, but have very different media systems (Humprecht et al. 2022) and degrees of platformisation of their news media system in particular (Nielsen and Fletcher 2023).

The report focuses on the public because we believe that – in addition to economic, political, and technological factors – public uptake and understanding of generative AI will be among the key factors shaping how these technologies are being developed and are used, and what they, over time, will come to mean for different groups and different societies (Nielsen 2024). There are many powerful interests at play around AI, and much hype – often positive salesmanship, but sometimes wildly pessimistic warnings about possible future risks that might even distract us from already present issues. But there is also a fundamental question of whether and how the public at large will react to the development of this family of products. Will it be like blockchain, virtual reality, and Web3? All promoted with much bombast but little popular uptake so far. Or will it be more like the internet, search, and social media – hyped, yes, but also quickly becoming part of billions of people’s everyday media use.

To advance our understanding of these issues, we rely on data from an online survey focused on understanding if and how people use generative AI, and what they think about its application in journalism and other areas of work and life. In the first part of the report, we present the methodology, then we go on to cover public awareness and use of generative AI, expectations for generative AI’s impact on news and beyond, how people think AI is being used by journalists right now, and how people think about how journalists should use generative AI, before offering a concluding discussion.

As with all survey-based work, we are reliant on people’s own understanding and recall. This means that many responses here will draw on broad conceptions of what AI is and might mean, and that, when it comes to generative AI in particular, people are likely to answer based on their experience of using free-standing products explicitly marketed as being based on generative AI, like ChatGPT. Most respondents will be less likely to be thinking about incidents where they may have come across functionalities that rely in part on generative AI, but do not draw as much attention to it – a version of what is sometimes called ‘invisible AI’ (see, for example, Alm et al. 2020). We are also aware that these data reflect a snapshot of public opinion, which can fluctuate over time.

We hope the analysis and data published here will help advance scholarly analysis by complementing the important work done on the use of AI in news organisations (for example, Beckett and Yaseen 2023; Caswell 2024; Diakopoulos 2019; Diakopoulos et al 2024; Newman 2024; Simon 2024), including its limitations and inequities (see, for example, Broussard 2018, 2023; Bender et al. 2021), and help centre the public as a key part of how generative AI will develop and, over time, potentially impact many different sectors of society, including journalism and the news media.

Methodology

The report is based on a survey conducted by YouGov on behalf of the Reuters Institute for the Study of Journalism (RISJ) at the University of Oxford. The main purpose is to understand if and how people use generative AI, and what they think about its application in journalism and other areas of work and life.

The data were collected by YouGov using an online questionnaire fielded between 28 March and 30 April 2024 in six countries: Argentina, Denmark, France, Japan, the UK, and the USA.

YouGov was responsible for the fieldwork and provision of weighted data and tables only, and RISJ was responsible for the design of the questionnaire and the reporting and interpretation of the results.

Samples in each country were assembled using nationally representative quotas for age group, gender, region, and political leaning. The data were weighted to targets based on census or industry-accepted data for the same variables.

Sample sizes are approximately 2,000 in each country. The use of a non-probability sampling approach means that it is not possible to compute a conventional ‘margin of error’ for individual data points. However, differences of +/- 2 percentage points (pp) or less are very unlikely to be statistically significant and should be interpreted with a very high degree of caution. We typically do not regard differences of +/- 2pp as meaningful, and as a general rule we do not refer to them in the text.

It is important to note that online samples tend to under-represent the opinions and behaviours of people who are not online (typically those who are older, less affluent, and have limited formal education). Moreover, because people usually opt in to online survey panels, they tend to over-represent people who are well educated and socially and politically active.

Some parts of the survey require respondents to recall their past behaviour, which can be flawed or influenced by various biases. Additionally, respondents’ beliefs and attitudes related to generative AI may be influenced by social desirability bias, and when asked about complex socio-technical issues, people will not always be familiar with the terminology experts rely on or understand the terms the same way. We have taken steps to mitigate these potential biases and sources of error by implementing careful questionnaire design and testing.

1. Public awareness and use of generative AI

Most of our respondents have, by now, heard of at least some of the most popular generative AI tools. ChatGPT is by far the most widely recognised of these, with between 41% (Argentina) and 61% (Denmark) saying they’d heard of it.

Other tools, typically those built by incumbent technology companies – such as Google Gemini, Microsoft Copilot, and Snapchat My AI – are some way behind ChatGPT, even with the boost that comes from being associated with a well-known brand. They are, with the exception of Grok from X, each recognised by roughly 15–25% of the public.

Tools built by specialised AI companies, such as Midjourney and Perplexity, currently have little to no brand recognition among the public at large. And there’s little national variation here, even when it comes to brands like Mistral in France; although it is seen by some commentators as a national champion, it clearly hasn’t yet registered with the wider French population.

We should also remember that a sizable minority of the public – between 19% of the online population in Japan and 30% in the UK – have not heard of any of the most popular AI tools (including ChatGPT) despite nearly two years of hype, policy conversations, and extensive media coverage.

While our Digital News Report (Newman et al. 2023) shows that in most countries the news market is dominated by domestic brands that focus on national news, in contrast, the search and social platform space across countries tends to feature the same products from large technology companies such as Google, Meta, and Microsoft. At least for now, it seems like the generative AI space will follow the pattern from the technology sector, rather than the more nationally oriented one of news providers serving distinct markets defined in part by culture, history, and language.

The pattern we see for awareness in Figure 1 extends to use, with ChatGPT by far the most widely used generative AI tool in the six countries surveyed. Use of ChatGPT is roughly two or three times more widespread than the next products, Google Gemini and Microsoft Copilot. What’s also clear from Figure 2 is that, even when it comes to ChatGPT, frequent use is rare, with just 1% using it on a daily basis in Japan, rising to 2% in France and the UK, and 7% in the USA. Many of those who say they have used generative AI have only used it once or twice, and it is yet to become part of people’s routine internet use.

Use of ChatGPT is slightly more common among men and those with higher levels of formal education, but the biggest differences are by age group, with younger people much more likely to have ever used it, and to use it on a regular basis (Figure 3). Averaging across all six countries, 16% of those aged 55 and over say they have used ChatGPT at least once, compared to 56% of 18–24s. But even among this age group infrequent use is the norm, with just over half of users saying they use it monthly or less.

Although people working in many different industries – including news and journalism – are looking for ways of deploying generative AI, people in every country apart from Argentina are slightly more likely to say they are using it in their private life rather than at work or school (Figure 4). If providers of AI products convince more companies and organisations that these tools can deliver great efficiencies and new opportunities this may change, with professional use becoming more widespread and potentially spilling over to people’s personal lives – a dynamic that was part of how the use of personal computers, and later the internet, spread. However, at this stage private use is more widespread.

Averaging across six countries, roughly equal proportions say that they have used generative AI for getting information (24%) as creating media (28%), which as a category includes creating images (9%), audio (3%), video (4%), code (5%), and generating text (Figure 5). When it comes to creating text more specifically, people report using generative AI to write emails (9%) and essays (8%), and for creative writing (e.g. stories and poems) (7%). But it’s also clear that many people who say they have used generative AI for creating media have just been playing around or experimenting (11%) rather than looking to complete a specific real-world task. This is also true when it comes to using generative AI to get information (9%), but people also say they have used it for answering factual questions (11%), advice (10%), generating ideas (9%), and summarisation (8%).

An average of 5% across the six countries say that they have used generative AI to get the latest news, making it less widespread than most of the other uses that were mentioned previously. One reason for this is that the free version of the most widely used generative AI product – ChatGPT – is not yet connected to the web, meaning that it cannot be used for the latest news. Furthermore, our previous research has shown that around half of the most widely used news websites are blocking ChatGPT (Fletcher 2024), and partly as a result, it is rarely able to deliver the latest news from specific outlets (Fletcher et al. 2024).

The figures for using generative AI for news vary by country, from just 2% in the UK and Denmark to 10% in the USA (Figure 6). The 10% figure in the USA is probably partly due to the fact that Google has been trialling Search Generative Experiences (SGE) there for the last year, meaning that people who use Google to search for a news-related topic – something that 23% of Americans do each week (Newman et al. 2023) – may see some generative AI text that attempts to provide an answer. However, given the documented limitations of generative AI when it comes to factual precision, companies like Google may well approach news more cautiously than other types of content and information, and the higher figure in the USA may also simply be because generative AI is more widely used there generally.

Numerous examples have been documented of generative AI giving incorrect answers when asked factual questions, as well as other forms of so-called ‘hallucination’ that result in poor- quality outputs (e.g. Angwin et al. 2024). Although some are quick to point out that it is wrong to expect generative AI to be good at information-based tasks – at least at its current state of development – some parts of the public are experimenting with doing exactly that.

Given the known problems when it comes to reliability and veracity, it is perhaps concerning that our data also show that users seem reasonably content with the performance – most of those (albeit a rather small slice of the online population) who have tried to use generative AI for information-based tasks generally say they trusted the outputs (Figure 7).

In interpreting this, it is important to keep in mind two important caveats.

First, the vast majority of the public has not used generative AI for information-based tasks, so we do not know about their level of trust. Other evidence suggests that trust among the large part of the public that has not used generative AI is low, meaning overall trust levels are likely to be low (Pew 2024).

Second, people are more likely to say that they ‘somewhat trust’ the outputs rather than ‘strongly trust’, which indicates a degree of scepticism – their trust is far from unconditional. However, this may also mean that from the point of view of members of the public who have used the tools, information from generative AI while clearly not perfect is already good enough for many purposes, especially tasks like generating ideas.

When we ask people who have used generative AI to create media whether they think the product they used did it well or badly, we see a very similar picture. Most of those who have tried to use generative AI to create media think that it did it ‘very’ or ‘somewhat’ well, but again, we can only use this data to know what users of the technology think.

The general population’s views on the media outputs may look very different, and while early adopters seem to have some trust in generative AI, and feel these technologies do a somewhat good job for many tasks, it is not certain that everyone will feel the same, even if or when they start using generative AI tools.

2. Expectations for generative AI’s impact on news and beyond

We now move from people’s awareness and use of generative AI products to their expectations around what the development of these technologies will mean. First, we find that most of the public expect generative AI to have a large impact on virtually every sector of society in the next five years (Figure 8). For every sector, there is a smaller number who expect low impact (compared to a large impact), and a significant number of people (roughly between 15% and 20%) who answer ‘don’t know’.

Averaging across six countries, we find that around three-quarters of respondents think generative AI will have a large impact on search and social media companies (72%), while two-thirds (66%) think that it will have a large impact on the news media – strikingly, the same proportion who think it will have a large impact upon the work of scientists (66%). Around half think that generative AI will have a large impact upon national governments (53%) and politicians and political parties (51%).

Interestingly, there are generally fewer people who expect it will have a large impact on ordinary people (48%). Much of the public clearly thinks the impact of generative AI will be mediated by various existing social institutions.

Bearing in mind how different the countries we cover are in many respects, including in terms of how people use and think about news and media (see, for example, Newman et al. 2023), it is striking that we find few cross-country differences in public expectations around the impact of generative AI. There are a few minor exceptions. For example, expectations around impact for politicians and political parties are a bit higher than average in the USA (60% vs 51%) and a bit lower in Japan (44% vs 51%) – but, for the most part, views across countries are broadly similar.

For almost all these sectors, there is little variation across age and gender, and the main difference when it comes to different levels of education is that respondents with lower levels of formal education are more likely to respond with ‘don’t know’, and those with higher levels of education are more likely to expect a large impact. The number who expect a small impact remains broadly stable across levels of education.

The only exception to this relative lack of variation by demographic factors is expectations around what impact generative AI will have for ordinary people. Younger respondents, who, as we have shown in earlier sections, are much more likely to have used generative AI tools, are also much more likely to expect a large impact within the next five years than older people, who often have little or no personal experience of using generative AI (Figure 9).

Expectations around the impact of generative AI, whether large or small, in themselves say nothing about how people think about whether this impact will, on balance, be for better or for worse.

Because generative AI use is highly mediated by institutions, and our data document that much of the public clearly recognise this, a useful additional way to think about expectations is to consider whether members of the public trust different sectors to make responsible use of generative AI.

We find that public trust in different institutions to make responsible use of generative AI is generally quite low (Figure 10). While around half in most of the six counties trust scientists and healthcare professionals to use generative AI responsibly, the figures drop below 40% for most other sectors in most countries. Figures for social media companies are lower than many other sectors, as are those for news media, ranging from 12% in the UK to 30% in Argentina and the USA.

There is more cross-country variation in public trust and distrust in different institutions’ potential use of generative AI, partly in line with broader differences from country to country in terms of trust in institutions.

But there are also some overarching patterns.

First, younger people, while still often sceptical, are for many sectors more likely to say they trust a given institution to use generative AI responsibly, and less likely to express distrust. This tendency is most pronounced in the sectors viewed with greatest scepticism by the public at large, including the government, politicians, and ordinary people, as well as news media, social media, and search engines.

Second, a significant part of the public does not have a firm view on whether they trust or distrust different institutions to make responsible use of generative AI. Varying from sector to sector and from country to country, between roughly one-quarter and half of respondents answer ‘neither trust nor distrust’ or ‘don’t know’ when asked. There is much uncertainty and often limited personal experience; in that sense, the jury is still out.

Leaving aside country differences for a moment and looking at the aggregate across all six countries, we can combine our data on public expectations around the size of the impact that generative AI will have with expectations around whether various sectors will use these technologies responsibly. This will provide an overall picture of how people think about these issues across different social institutions (Figure 11).

If we compare public perceptions relative to the average percentage of respondents who expect a large impact across all sectors (58%, marked by the vertical dashed line in Figure 11) and the average percentage of respondents who distrust actors in a given sector to make responsible use of generative AI (33%, marked by the horizontal dashed line), we can group expectations from sector to sector into four quadrants.

• First, there are those sectors where people expect generative AI to have a relatively large impact, but relatively few expect it will be used irresponsibly (e.g. healthcare and science).
• Second, there are sectors where people expect the impact may not be as great, and relatively fewer fear irresponsible use (e.g. ordinary people and retailers).
• Third, there are sectors where relatively few people expect a large impact, and relatively more people are worried about irresponsible use (e.g. government and political parties).
• Finally, there are sectors where more people expect large impact, and more people fear irresponsible use by the actors involved (e.g. social media and the news media, who are viewed very similarly by the public in this respect).

It is important to keep this quite nuanced and differentiated set of expectations in mind in interpreting people’s general expectations around what impact they think generative AI will have for them personally, as well as for society at large.

Asked if they think that generative AI will make their life better or worse, more than half of our respondents answer ‘neither better nor worse’ or ‘don’t know’, with a plurality in four of the six countries covered answering ‘better’, and a significant minority ‘worse’ (Figure 12). The large number of people with no strong expectations either way is consistent across countries, but the balance between more optimistic responses and more pessimistic ones varies.

People’s expectations when asked whether generative AI will make society better or worse are more pessimistic on average. There are about the same number of optimists, but significantly more pessimists who believe generative AI will make society worse. Expectations around what generative AI might mean for society are more varied across the six countries we cover. In two (France and the UK), there are more who expect it will make society worse than better. In another two (Denmark and the USA), there are as many pessimists as optimists. And in the remaining two (Argentina and Japan) more respondents expect generative AI products will make society better than expect them to make society worse.

Looking more closely at people’s expectations, both in terms of their own life and in terms of society, younger people and people with more formal education also often opt for ‘neither better nor worse’ or ‘don’t know’, but in most countries – Argentina being the exception – they are more likely to answer ‘better’ (Figure 13).

Asked whether they think the use of generative AI will make different areas of life better or worse, again, much of the public is undecided, either opting for ‘neither better nor worse’ or answering ‘don’t know’, underlining that it is still early days.

We now look specifically at the percentage point difference between optimists who expect AI to make things better and pessimists who expect it to make them worse gives a sense of public expectations across different areas (Figure 14). Large parts of the public think generative AI will make science (net ‘better’ of +44 percentage points), healthcare (+36), and many daily routine activities, including transportation (+26), shopping (+22), and entertainment (+17), better, even though there is much less optimism when it comes to core areas of the rule of law, including criminal justice (+1) and more broadly legal rights and due process (-3), and considerable pessimism for some very bread-and-butter issues, including cost of living (-6), equality (-6), and job security (-18).

News and journalism is also an area where, on balance, there is more pessimism than optimism (-8) – a striking contrast to another area involving the media, namely entertainment (+17). But there is a lot of national variation here. In countries that are more optimistic about the potential effects of generative AI, namely Argentina (+19) and Japan (+8), the proportion that think it will make news and journalism better is larger than the proportion that think it will become worse. The UK public are particularly negative about the effect of generative AI on journalism, with a net score of -35. There is a similar lack of consensus across different countries on whether crime and justice, legal rights and due process, cost of living, equality, and job security will be made better or worse.

3. How people think generative AI is being used by journalists right now

Many of the conversations around generative AI and journalism are about what might happen in the future – speculation about what the technology may or may not be able to do one day, and how this will shape the profession as we know it. But it is important to remember that some journalists and news organisations are using generative AI right now, and they have been using some form of AI in the newsroom for several years.

We now focus on how much the public knows about this, what they think journalists currently use generative AI for, and what processes they think news media have in place to ensure quality.

In the survey, we showed respondents a list of journalistic tasks and asked them how often they think journalists perform them ‘using artificial intelligence with some human oversight’. The tasks ranged from behind-the-scenes work like ‘editing the spelling and grammar of an article’ and ‘data analysis’ through to much more audience-facing outputs like ‘writing the text of an article’ and ‘creating a generic image/illustration to accompany the text of an article’.

We specifically asked about doing these ‘using artificial intelligence with some human oversight’ because we know that some newsrooms are already performing at least some tasks in this way, while few are currently doing them entirely using AI without a human in the loop. Even tasks that may seem fanciful to some, like ‘creating an artificial presenter or author’, are not without precedent. In Germany, for example, the popular regional newspaper Express has created a profile for an artificial author called Klara Indernach, 1 which it uses as the byline for its articles created with the help of AI, and several news organisations across the world already use AI-generated artificial presenters for various kinds of video and audio.

Figure 15 shows that a substantial minority of the public believe that journalists already always or often use generative AI to complete a wide range of different tasks. Around 40% believe that journalists often or always use AI for translation (43%), checking spelling and grammar (43%), and data analysis (40%). Around 30% think that journalists often or always use AI for re-versioning – whether it’s rewriting the same article for different people (28%) or turning text into audio or video (30%) – writing headlines (29%), or creating stock images (30%).

In general, the order of the tasks in Figure 15 reflects the fact that people – perhaps correctly – believe that journalists are more likely to employ AI for behind-the-scenes work like spellchecking and translation than they are for more audience-facing outputs. This may be because people understand that some tasks carry a greater reputational risk for journalists, and/or that the technology is simply better at some things than others.

The results may also reveal a degree of cynicism about journalism from some parts of the public. The fact that around a quarter think that journalists always or often use AI to create an image if a real photograph is not available (28%) and 17% think they create an artificial presenter or author may say more about their attitudes towards journalism as an institution than about how they think generative AI is actually being used. However unwelcome they might be – and however wrong they are about how many news media use AI – these perceptions are a social reality, shaping how parts of the public think about the intersection between journalism and AI.

Public perceptions of what journalists and news media already use AI for are quite consistent across different genders and age groups, but there are some differences by country, with respondents in Argentina and the USA a little more likely to believe that AI is used for each of these tasks, and respondents in Denmark and the UK less likely.

Among those news organisations that have decided to implement generative AI for certain tasks, the importance of ‘having a human in the loop’ to oversee processes and check errors is often stressed. Human oversight is nearly always mentioned in public-facing guidelines on the use of AI for editorial work, and journalists themselves mention it frequently (Becker et al. 2024).

Large parts of the public, however, do not think this is happening (Figure 16). Averaging across the six countries, around one-third think that human editors ‘always’ or ‘often’ check AI outputs to make sure they are correct or of a high standard before publishing them. Nearly half think that journalists ‘sometimes’, ‘rarely’, or ‘never’ do this – again, perhaps, reflecting a level of cynicism about the profession among the public, or a tendency to judge the whole profession and industry on the basis of how some parts of it act.

The proportion that think checking is commonplace is lowest in the UK, where only one-third of the population say they ‘trust most news most of the time’ (Newman et al. 2023), but we also see similarly low figures in Denmark, where trust in the news is much higher. The results may, therefore, also partly reflect more than just people’s attitudes towards journalism and the news media.

4. What does the public think about how journalists should use generative AI?

Various forms of AI have long been used to produce news stories by publishers including, for example, Associated Press, Bloomberg, and Reuters. And content produced with newer forms of generative AI has, with mixed results, been published by titles including BuzzFeed, the Los Angeles Times , the Miami Herald , USA Today , and others.

Publishers may be more or less comfortable with how they are using these technologies to produce various kinds of content, but our data suggest that much of the public is not – at least not yet. As we explore in greater detail in our forthcoming 2024 Reuters Institute Digital News Report (Newman et al. 2024), people are generally more comfortable with news produced by human journalists than by AI.

However, averaging across six countries, younger people are significantly more likely to say they are comfortable with using news produced in whole or in part by AI (Figure 17). The USA and Argentina have somewhat higher levels of comfort with news made by generative AI, but there too, much of the public remains sceptical.

We also asked respondents whether they are comfortable or uncomfortable using news produced mostly by AI with some human oversight on a range of different topics. Figure 18 shows the net percentage point difference between those that selected ‘very’ or ‘somewhat’ comfortable and those that selected ‘very’ or ‘somewhat’ uncomfortable (though, as ever, a significant minority selected the ‘neither’ or ‘don’t know’ options). Looking across different topics, there is somewhat more comfort with using news produced mostly by AI with some human oversight when it comes to ‘softer’ news topics, like fashion (+7) and sports (+5), than ‘hard’ news topics including politics (-33) and international affairs (-21).

But in every area, at this point in time, only for a very small number of topics are there more people uncomfortable with relying on AI-generated news than comfortable. As with overall comfort, there is somewhat greater acceptance of the use of AI for generating various kinds of news with at least some human oversight in the USA and Argentina.

Putting aside country differences, there is again a marked difference between our respondents overall and younger respondents. Among respondents overall, there are only three topic areas out of ten where slightly more respondents are comfortable with news made mostly by AI with some human oversight than are uncomfortable with this. Among respondents aged 18 to 24, this rises to six out of ten topic areas.

It is important to remember that much of the public does not have strong views either way, at least at this stage. Between one-quarter and one-third of respondents answer either ‘neither comfortable nor uncomfortable’ or ‘don’t know’ when asked the general questions about comfort with different degrees of reliance on generative AI versus human journalists, and between one-third and half of respondents do the same when asked about generative AI news for specific topics. It is an open question as to how these less clearly formed views will evolve.

One way to assess what the public expects it will mean if and when AI comes to play a greater role in news production is to gauge people’s views on how it will change news, compared to a baseline of news produced entirely by human journalists.

We map this by asking respondents if they think that news produced mostly by AI with some human oversight will differ from what most are used to across a range of different qualities and attributes.

Between one-third and half of our respondents do not have a strong view either way. Focusing on those respondents who do have a view, we can look at the net percentage point difference between how many respondents think AI will make the news somewhat more or much more (e.g. more ‘up to date’ or more ‘transparent’), versus somewhat less or much less, of each, helping to provide an overarching picture of public expectations.

On balance, more respondents expect news produced mostly by AI with some human oversight to be less trustworthy (-17) and less transparent (-8), but more up to date (+22) and – by a large margin – cheaper to make (+33) (Figure 19). There is considerable national variation here, but with the exception of Argentina, the balance of public opinion (net positive or negative) is usually the same for these four attributes. For the others, the balance often varies.

Essentially our data suggest that the public, at this stage, primarily think that the use of AI in news production will help publishers by cutting costs, but identify few, if any, ways in which they expect it to help them – and several key areas where many expect news made with AI to be worse.

In light of this, it makes sense that, when asked if news produced mostly by AI with some human oversight is more or less worth paying for than news produced entirely by a human journalist, an average of 41% across six countries say less worth paying for (Figure 20). Just 8% say they think that news made in this way will be more valuable.

There is some variation here by country and by age, but even among the generally more AI-positive younger respondents aged 18–24, most say either less worth paying for (33%) or about the same (38%). The implications of the spread of generative AI and how it is used by publishers for people’s willingness to pay for news will be interesting to follow going forward, as tensions may well mount between the ‘pivot to pay’ we have seen from many news media in recent years and the views we map here.

Looking across a range of different tasks that journalists and news media might use generative AI for, and in many cases already are using generative AI for, we can again gauge how comfortable the public is by looking at the balance between how many are comfortable with a particular use case and how many are uncomfortable.

As with several of the questions above, about a third have no strong view either way at this stage – but many others do. Across six countries, the balance of public opinion ranges from relatively high levels of comfort with back-end tasks, including editing spelling and grammar (+38), translation (+35), and the making of charts (+28), to widespread net discomfort with synthetic content, including creating an image if a real photo is not available (-13) and artificial presenters and authors (-24) (Figure 21).

When asked if it should be disclosed or labelled as such if news has been produced mostly by AI with some human oversight, only 5% of our respondents say none of the use cases included above need to be disclosed, and the vast majority of respondents say they want some form of disclosure or labelling in at least some cases. Research on the effect of labelling AI-generated news is ongoing, but early results suggest that although labelling may be desired by audiences, it may have a negative effect on trust (Toff and Simon 2023).

There is, however, less consensus on what exactly should be disclosed or labelled, except for somewhat lower expectations around the back-end tasks people are frequently comfortable with AI completing (Figure 22). Averaging across six countries, around half say that ‘creating an image if a real photograph is not available’ (49%), ‘writing the text of an article’ (47%), and ‘data analysis’ (47%) should be labelled as such if generative AI is used. However, this figure drops to around one-third for ‘editing the spelling and grammar of an article’ (32%) and ‘writing a headline’ (35%). Again, variation exists between both countries and demographic groups that are generally more positive about AI.

Based on online surveys of nationally representative samples in six countries, we have, with a particular focus on journalism and news, documented how aware people are of generative AI, how they use it, and their expectations on the magnitude of impact it will have in different sectors – including whether it will be used responsibly.

We find that most of the public are aware of various generative AI products, and that many have used them, especially ChatGPT. But between 19% and 30% of the online population in the six countries surveyed have not heard of any of the most popular generative AI tools, and while many have tried using various of them, only a very small minority are, at this stage, frequent users. Going forward, some use will be driven by people seeking out and using stand-alone generative AI tools such as ChatGPT, but it seems likely that much of it will be driven by a combination of professional adaptation, through products used in the workplace, and the introduction of more generative AI-powered elements into platforms already widely used in people’s private lives, including social media and search engines, as illustrated with the recent announcements of much greater integration of generative AI into Google Search.

When it comes to public expectations around the impact of generative AI and whether these technologies are likely to be used responsibly, we document a differentiated and nuanced picture. First, there are sectors where people expect generative AI will have a greater impact, and relatively fewer people expect it will be used irresponsibly (including healthcare and science). Second, there are sectors where people expect the impact may not be as great, and relatively fewer fear irresponsible use (including from ordinary people and retailers). Third, there are sectors where relatively fewer people expect large impact, and relatively more people are worried about irresponsible use (including government and political parties). Fourth, there are sectors where more people expect large impact, and more people fear irresponsible use by the actors involved (this includes social media and the news media).

Much of the public is still undecided on what the impact of generative AI will be. They are unsure whether, on balance, generative AI will make their own lives and society better or worse. This is understandable, given many are not aware of any of these products, and few have personal experience of using them frequently. Younger people and those with higher levels of formal education – who are also more likely to have used generative AI – are generally more positive.

Expectations around what generative AI might mean for society are more varied across the six countries we cover. In two, there are more who expect it will make society worse than better, in another two, there are as many pessimists as optimists, and in the final two, more respondents expect generative AI products will make society better than expect them to make society worse. These differences may also partly reflect the current situation societies find themselves in, and whether people think AI can fundamentally change the direction of those societies. To some extent we also see this pattern reflected in how people think about AI in news. Across a range of measures, in some countries people are generally more optimistic, but in others more pessimistic.

Looking at journalism and news media more closely, we have found that many believe generative AI is already relatively widely used for many different tasks, but that they are, in most cases, not convinced these uses of AI make news better – they mostly expect it to make it cheaper to produce.

While there is certainly curiosity, openness to new approaches, and some optimism in parts of the public (especially when it comes to the use of these technologies in the health sector and by scientists), generally, the role of generative AI in journalism and news media is seen quite negatively compared to many other sectors – in some ways similar to how much of the public sees social media companies. Basically, we find that the public primarily think that the use of generative AI in news production will help publishers cut costs, but identify few, if any, ways in which they expect it to help them as audiences, and several key areas where many expect news made with AI to be worse.

These views are not solely informed by how people think generative AI will impact journalism in the future. A substantial minority of the public believe that journalists already always or often use generative AI to complete a wide range of different tasks. Some of these are tasks that most are comfortable with, and are within the current capabilities of generative AI, like checking spelling and grammar. But many others are not. More than half of our respondents believe that news media at least sometimes use generative AI to create images if no real photographs are available, and as many believe that news media at least sometimes create artificial authors or presenters. These are forms of use that much of the public are uncomfortable with.

Every individual journalist and every news organisation will need to make their own decisions about which, if any, uses of generative AI they believe are right for them, given their editorial principles and their practical imperatives. Public opinion cannot – and arguably should not – dictate these decisions. But public opinion provides a guide on which uses are likely to influence how people judge the quality of news and their comfort with relying on it, and thus helps, among other things, to identify areas where it is particularly important for journalists and news media to communicate and explain their use of AI to their target audience.

It is still early days, and it remains to be seen how public use and perception of generative AI in general, and its role in journalism and news specifically, will evolve. On many of the questions asking respondents to evaluate AI in different sectors and for different uses, between roughly a quarter and half of respondents pick relatively neutral middle options or answer ‘don’t know’. There is still much uncertainty around what role generative AI should and will have, in different sectors, and for different purposes. And, especially in light of how many have limited personal experience of using these products, it makes sense that much of the public has not made up their minds.

Public debate, opinion commentary, and news coverage will be among the factors influencing how this evolves. So will people’s own experience of using generative AI products, whether for private or professional purposes. Here, it is important to note two things. First, younger respondents generally are much more open to, and in many cases optimistic about, generative AI than respondents overall. Second, despite the many documented limitations and problems with state-of-the-art generative AI products, those respondents who use these tools themselves tend to offer a reasonably positive assessment of how well they work, and how much they trust them. This does not necessarily mean that future adopters will feel the same. But if they do, and use becomes widespread and routine, overall public opinion will change – in some cases perhaps towards a more pessimistic view, but, at least if our data are anything to go by, in a more grounded and cautiously optimistic direction.

• Ada Lovelace Institute and The Alan Turing Institute. 2023. How Do People Feel About AI? A Nationally Representative Survey of Public Attitudes to Artificial Intelligence in Britain
• Alm, C. O., Alvarez, A., Font, J., Liapis, A., Pederson, T., Salo, J. 2020. ‘Invisible AI-driven HCI Systems – When, Why and How’, Proceedings of the 11th Nordic Conference on Human-Computer Interaction: Shaping Experiences, Shaping Society , 1–3. https://doi.org/10.1145/3419249.3420099
• Angwin, J., Nelson, A., Palta, R. 2024. ‘ Seeking Reliable Election Information? Don’t Trust AI ’, Proof News.
• Becker, K. B., Simon, F. M., Crum, C. 2023. ‘Policies in Parallel? A Comparative Study of Journalistic AI Policies in 52 Global News Organisations’, https://doi.org/10.31235/osf.io/c4af9 .
• Beckett, C., Yaseen, M. 2023. ‘ Generating Change: A Global Survey of What News Organisations Are Doing with Artificial Intelligence ’. London: JournalismAI, London School of Economics. .
• Bender, E. M., Gebru, T., McMillan-Major, A., Shmitchell, S. 2021. ‘On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?’, in Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency , 610–23. New York: Association for Computing Machinery. https://doi.org/10.1145/3442188.3445922 .
• Broussard, M. 2018. Artificial Unintelligence: How Computers Misunderstand the World . Reprint edition. Cambridge, MA: The MIT Press.
• Broussard, M. 2023. More than a Glitch: Confronting Race, Gender, and Ability Bias in Tech . Cambridge, MA: The MIT Press.
• Caswell, D. 2024. ‘ AI in Journalism Challenge 2023. ’ London: Open Society Foundations.
• Diakopoulos, N. 2019. Automating the News: How Algorithms Are Rewriting the Media . Cambridge, MA: Harvard University Press.
• Diakopoulos, N., Cools, H., Li, C., Helberger, N., Kung, E., Rinehart, A., Gibbs, L. 2024. ‘ Generative AI in Journalism: The Evolution of Newswork and Ethics in a Generative Information Ecosystem ’. New York: Associated Press. .
• Fletcher, R. 2024. How Many News Websites Block AI Crawlers? Reuters Institute for the Study of Journalism. https://doi.org/10.60625/risj-xm9g-ws87 .
• Fletcher, R., Adami, M., Nielsen, R. K. 2024. ‘ I’m Unable To’: How Generative AI Chatbots Respond When Asked for the Latest News . Reuters Institute for the Study of Journalism. https://doi.org/10.60625/RISJ-HBNY-N953 .
• Humprecht, E., Herrero, L. C., Blassnig, S., Brüggemann, M., Engesser, S. 2022. ‘Media Systems in the Digital Age: An Empirical Comparison of 30 Countries’, Journal of Communication 72(2): 145–64. https://doi.org/10.1093/joc/jqab054 .
• Mellado, C., Cruz, A., Dodds, T. 2024. Inteligencia Artificial y Audiencias en Chile .
• Newman, N. 2024. Journalism, Media, and Technology Trends and Predictions 2024 . Reuters Institute for the Study of Journalism. https://doi.org/10.60625/risj-0s9w-z770 .
• Newman, N., Fletcher, R., Eddy, K., Robertson, C. T., Nielsen, R. K. 2023. Reuters Institute Digital News Report 2023 . Reuters Institute for the Study of Journalism. https://doi.org/10.60625/risj-p6es-hb13 .
• Newman, N., Fletcher, R., Robertson C. T., Ross Arguedas, A. A., Nielsen, R. K. 2024. Reuters Institute Digital News Report 2024 (forthcoming). Reuters Institute for the Study of Journalism.
• Nielsen, R. K. 2024. ‘ How the News Ecosystem Might Look like in the Age of Generative AI ’ .
• Nielsen, R. K., Fletcher, R. 2023. ‘Comparing the Platformization of News Media Systems: A Cross-Country Analysis’, European Journal of Communication 38(5): 484–99. https://doi.org/10.1177/02673231231189043 .
• Pew. 2023. ‘ Growing Public Concern about the Role of Artificial Intelligence in Daily Life ’.
• Pew. 2024. ‘ Americans’ Use of ChatGPT is Ticking Up, but Few Trust its Election Information ’.
• Simon, F. M. 2024. ‘ Artificial Intelligence in the News: How AI Retools, Rationalizes, and Reshapes Journalism and the Public Arena ’, Columbia Journalism Review .
• Toff, B., Simon, F. M. 2023. ‘ Or They Could Just Not Use It?’: The Paradox of AI Disclosure for Audience Trust in News . https://doi.org/10.31235/osf.io/mdvak .
• Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., Polosukhin, I. 2017. Attention Is All You Need. https://doi.org/10.48550/ARXIV.1706.03762 .
• Vogler, D., Eisenegger, M., Fürst, S., Udris, L., Ryffel, Q., Rivière, M., Schäfer, M. S. 2023. Künstliche Intelligenz in der journalistischen Nachrichtenproduktion: Jahrbuch Qualität der Medien Studie 1 / 2023. https://doi.org/10.5167/UZH-238634 .

1 https://www.express.de/autor/klara-indernach-594809

About the authors

Dr Richard Fletcher is Director of Research at the Reuters Institute for the Study of Journalism. He is primarily interested in global trends in digital news consumption, the use of social media by journalists and news organisations, and, more broadly, the relationship between computer-based technologies and journalism.

Professor Rasmus Kleis Nielsen is Director of the Reuters Institute for the Study of Journalism, Professor of Political Communication at the University of Oxford, and served as Editor-in-Chief of the International Journal of Press/Politics from 2015 to 2018. His work focuses on changes in the news media, political communication, and the role of digital technologies in both.

Acknowledgements

We would like to thank Caryhs Innes, Xhoana Beqiri, and the rest of the team at YouGov for their work on fielding the survey. We would also like to thank Felix Simon for his help with the data analysis. We are grateful to the other members of the research team at RISJ for their input on the questionnaire and interpretation of the results, and to Kate Hanneford-Smith, Alex Reid, and Rebecca Edwards for helping to move this project forward and keeping us on track.

Funding acknowledgement

Report published by the Reuters Institute for the Study of Journalism (2024) as part of our work on AI and the Future of News , supported by seed funding from Reuters News and made possible by core funding from the Thomson Reuters Foundation.

At every email we send you'll find original reporting, evidence-based insights, online seminars and readings curated from 100s sources - all in 5 minutes.

• Twice a week
• More than 20,000 people receive it
• Unsubscribe any time

signup block

New study details potential long-term health risks as American girls reach puberty earlier

Amna Nawaz Amna Nawaz

Jackson Hudgins Jackson Hudgins

Leave your feedback

• Copy URL https://www.pbs.org/newshour/show/new-study-details-potential-long-term-health-risks-as-american-girls-reach-puberty-earlier

A new study adds to the evidence that girls in America are reaching puberty earlier with potentially troubling implications for their long-term health. More girls are getting their first periods sooner than previous generations and experiencing longer periods of menstrual irregularity. Amna Nawaz discussed more with Dr. Shruthi Mahalingaiah of the Harvard T.H. Chan School of Public Health.

Read the Full Transcript

Notice: Transcripts are machine and human generated and lightly edited for accuracy. They may contain errors.

Amna Nawaz:

A new study this week adds to the growing body of evidence that girls in America are reaching puberty earlier, with potentially troubling implications for their long-term health. More girls are getting their first periods sooner than previous generations and experiencing longer periods of menstrual irregularity as well.

We're joined now by the senior author of that study, Dr. Shruthi Mahalingaiah, assistant professor at the Harvard T.H. Chan School of Public Health.

Doctor, welcome, and thanks for joining us.

Dr. Shruthi Mahalingaiah, Harvard T.H. Chan School of Public Health: Thank you.

So let me just tick through some highlights of your study real quick here. You found there that girls are now — or, rather, were getting their first period on average at the age of 12.5 between 1950 and 1969.

Now, between 2000 and 2005, that's down to 11.9 years old. You also found the percentage of girls getting their period before the age of 11 almost doubled from about 8.5 to 15.5, and the percentage of girls getting their period before the age of 9 more than doubled.

Doctor, why is this significant and worth paying attention to?

Dr. Shruthi Mahalingaiah:

The age at the first period and time to cycle regularity are two aspects of menstrual health. It is really, really important to understand this trend for a variety of reasons. One of the most important reasons is that both of these characteristics are associated with a variety of long-term health consequences.

And for those with very early age at the first period or menarche and persistently irregular periods, those risks include risk of cardiovascular disease, infertility or fertility problems, mood disorders, and cancers, to name a few.

And the big question here, of course, is, why? Why are we seeing this trend? What do we know about that?

When we evaluated that exact question, we thought about, how could body mass index, particularly around the age at the first period, influence this.

And we found that about half of this trend could be explained by body mass index at menarche. When we looked at maybe what else might be going on, controlling for body mass index, we still noticed a trend towards decreasing age at the first period. And there are a variety of concerns I have around this and other studies have shown, including exposure to environmental factors, like hormone-disrupting chemicals, air pollutants, heavy metals, and exposures to stress and childhood trauma are some important things to consider.

There were also some substantial differences along racial and socioeconomic lines, girls of color, in particular, starting menstruation earlier relative to their white peers.

What explains that?

When we looked at the absolute difference, comparing those of white race ethnicity to other racial minorities, the kind of white girls reduced that age at first period to about six months, and other race ethnic groups had a reduction of almost a year from the first birth cohort to the most recent.

We think about race ethnicity as a surrogate marker for exposures. So, we're thinking about what are the other exposures that might covary with being in a particular race ethnic group, from exposures related to personal care products to stress and racism.

You know, it's worth pointing out that this study was based on self-reported data, right? It was through an app. Over 70,000 women participated.

What should we understand about the limitations of data collected that way and what questions you still have?

There are many important limitations to consider about data collected by self-report. And there are limitations regarding how we collected the data.

Participants had to have an iPhone and download the research app and be comfortable communicating in English. So we do have to interpret all of this information in the context of the population recruited into the study.

However, I think that what we are finding does allow a high-level look into the time trends across a vast time span of birth cohorts. And we are very much interested in promoting awareness of this topic to allow for education of providers and children.

And in terms of future research, looking into further impacts of environmental exposures is something that I'm very interested in.

So, to that goal of raising awareness here, a lot of parents are going to hear this. They're going to be very worried. They're going to wonder what they should do, if anything, to keep their children safe and healthy. What would you say to them?

For parents and caregivers, if you notice early signs of puberty, like breast development, going into the pediatrician for an evaluation is really important.

And then, in terms of what you can do to promote health and wellness around the menstrual cycle, include a diet of whole foods. It's very hard in today's modern world to have a diet of no ultra-processed foods, but try to limit processed foods and fast foods as much as possible and bring in those leafy green vegetables, fruits is really important.

Addressing other aspects of health through behavior, including balancing physical activity and addressing circadian rhythms and sleep health is very important in promoting health in this age group.

That is Dr. Shruthi Mahalingaiah, assistant professor at the Harvard T.H. Chan School of Public Health.

Doctor, thank you so much for joining us. We appreciate your time.

Thank you. OK, take care.

Listen to this Segment

Watch the Full Episode

Amna Nawaz serves as co-anchor of PBS NewsHour.

Support Provided By: Learn more

More Ways to Watch

• PBS iPhone App
• PBS iPad App

Educate your inbox

Subscribe to Here’s the Deal, our politics newsletter for analysis you won’t find anywhere else.

Thank you. Please check your inbox to confirm.

Independent Media demands retraction and apology for defamatory claims in the ICFJ Journalism ‘Big Data Case Study’

I ndependent Media strongly condemns the unethical and biased research practices evident in a recent report on the online harassment faced by journalists working for the Daily Maverick. The International Centre for Journalists (ICFJ) report, titled "The women journalists of South Africa's Daily Maverick: SEXUALIZED, SILENCED AND LABELED SATAN," makes false and defamatory allegations about Independent Media without any substantiation or opportunity for response.

In a blatant violation of basic research ethics, the report, which was ostensibly about the abuse faced by Daily Maverick journalists, deviated from its primary focus and singled out Independent Media for criticism, implicating our company in alleged attacks on journalists. These claims are utterly baseless and untrue. Independent Media has never condoned or engaged in the harassment of journalists. In fact, our own journalists, particularly women, have themselves been victims of such harassment.

Given the well-known hostile relations between Daily Maverick and Independent Media, the critique of Independent Media within a report focusing on Daily Maverick journalists is patently biased. While discussing the broader media environment can provide context, the focus on Independent Media raises serious questions about the impartiality of the researchers and the necessity of including our company in this particular study.

It is deeply disappointing that Prof. Julie Reid, Prof. Julie Posetti, Dr. Diana Maynard, Nabeelah Shabbir, and Don Kevin Hapal have chosen to weaponise this report as a propaganda tool, rather than contributing to a constructive dialogue on the critical issue of journalist abuse. The conduct of the ICFJ and its researchers is highly unethical, particularly the unnecessary invocation of our company's name when it is not directly relevant to the study.

We demand an immediate retraction of all claims made about Independent Media in the report, as well as a public apology to our journalists who have been unfairly maligned by association. The report's authors and affiliated institutions must clarify why Independent Media was inappropriately invoked in a study purportedly focused on a competing media company. They must also disclose any conflicts of interest, funding sources, or political motivations that may have compromised the impartiality of the research.

Independent Media stands in solidarity with all journalists who face harassment and threats for doing their work, something our journalists have experienced too, including death threats and attempts on the lives of some of our reporters.

For the record, Independent Media employs more female reporters and editors than any other media house in South Africa and has been at the forefront of pioneering award-winning gender based violence awareness campaigns, such as the Don’t Look Away Campaign. We have also been recognised for our anti-racism lobbying.

We have spoken up and out against corruption too.

We are not afraid to speak out when the truth matters, such as now. We cannot accept the unethical practice of making grave accusations without evidence or due process.

The ICFJ and the report's authors have failed in their basic duty of care as researchers.

We therefore call on the ICFJ, the research team, and their affiliated institutions to address these lapses transparently and take immediate corrective action. Failure to do so would further undermine the credibility of the study and compound the harm done to Independent Media and our journalists.

Independent Media remains committed to the highest standards of journalism ethics and to the safety and well-being of all journalists. We will continue to advocate for fair and responsible research practices in the study of these vital issues.

While the report is dressed in a cloak of "independence" through the use of researchers and academia, we believe that this "independence" was merely a cover to legitimise the ongoing propaganda attacks against Independent Media.

Independent Media

30 May 2024

• International
• Today’s Paper
• LIVE: LOK SABHA RESULTS 2024
• History of Polls
• Brand Solutions

Does sleep clear more toxins from the brain than when we’re awake? Latest research casts doubt on this theory

Since the brain is an active tissue – with many metabolic and cellular processes happening at any moment – it produces a lot of waste..

There’s no doubt sleep is good for the brain . It allows different parts to regenerate and helps memories stabilise.

When we don’t get enough sleep, this can increase stress levels and exacerbate mental health issues.

Evidence also supports the notion that the brain gets rid of more toxic waste when we’re asleep than when we’re awake. This process is believed to be crucial in getting rid of potentially harmful things such as amyloid, a protein whose build-up in the brain is linked to Alzheimer’s disease.

However, a recent study in mice has come to the opposite conclusion. Its authors suggest that in mice, brain clearance is actually lower during sleep – and that previous findings could also be re-interpreted in this way.

The brain’s cleaning system

Since the brain is an active tissue – with many metabolic and cellular processes happening at any moment – it produces a lot of waste. This waste is removed by our glymphatic system.

Cerebrospinal fluid is a crucial part of the glymphatic system. This fluid surrounds the brain, acting as a liquid cushion that protects it from damage and provides it nourishment, so the brain can function normally.

During the waste removal process, our cerebrospinal fluid helps transfer old and dirty brain fluid – full of toxins, metabolites and proteins – to outside the brain, and welcomes in new fluid.

The waste that has been removed then ends up in the lymphatic system (a part of your immune system), where it’s ultimately eliminated from your body.

The glymphatic system was only discovered in the last decade or so. It was first observed in mice, using dyes injected into their brains to study the movement of fluids there. The existence of the glymphatic system has since been confirmed in humans with the use of MRI scans and contrast dyes.

Based on the results of animal experiments, scientists concluded the glymphatic system is more active at night, during sleep or when under anaesthesia, than during the day.

Other studies have shown this waste removal activity may also vary depending on different conditions – such as sleep position, the type of anaesthetic used, and whether or not the subject’s circadian rhythm was interrupted.

Challenging old interpretations

The recent study used male mice to examine how the movement of brain fluid differed when animals were awake, asleep and anaesthetised. The researchers injected dyes into the animals’ brains to track the flow of fluid through the glymphatic system.

In particular, they examined whether an increase in dye indicated a decrease in fluid movement away from an area, rather than an increase in movement to the area as previous studies had suggested. The former would mean lower clearance via the glymphatic system – and hence less waste being removed.

More dye was found in brain areas after three hours and five hours of being asleep or anaesthetised than when awake. This indicated that less dye, and therefore fluid, was being cleared from the brain when the mouse was asleep or anaesthetised.

Although the findings are interesting, there are a number of limitations with the study’s design. As such, this can’t be considered absolute confirmation that the brain doesn’t flush out as much waste during the night than in the day.

Limitations to this study

First, the study was conducted using mice. The results from animal studies don’t always translate to humans, so it’s difficult to say whether the same will be true for us.

The study also only looked at male mice that were kept awake for a few hours before being allowed to sleep. This may have disturbed their natural sleep-wake rhythm, which could have partially influenced the results.

Studies have shown that interrupted or bad sleep is linked with an increase in stress levels – which in turn lowers brain fluid flow from the glymphatic system.

In contrast, in the first (2013) study that showed more brain toxins were removed during sleep, the mice were observed during their natural sleep time.

Different methods were also used in this study compared with previous ones – including what types of dye were injected and where. Previous studies also used both male and female mice. These differences in study methods could have influenced the results.

The glymphatic system might also behave differently depending on the brain region – with each producing different types of waste when awake or asleep. This may also explain why this study’s results were different from previous ones.

Virtually no studies looking at the glymphatic system and the effects of sleep in mice have examined the contents of the fluid excreted from the brain. So, even if the amount of fluid flowing out of the brain was lower during sleep or anaesthesia, this fluid could still be removing important waste products in different amounts.

A handful of studies have found disturbances in both glymphatic system function and sleep in people with neurological conditions – including Alzheimer’s disease and Parkinson’s.

A study in humans also indicates that more amyloid is found in the brain after even one night of sleep deprivation .

The glymphatic system is important when it comes to how the brain works – but it may very well function differently depending on many factors. We need more research that aims to replicate the latest study’s findings, while also examining the reasons behind its surprising conclusions

• healthy sleep

The Opposition, led by the Congress, is optimistic about a recovery in certain states, particularly in the INDIA bloc. They aim to win 75-80 seats with the support of allies, but their campaign has been geared towards future elections rather than just the Lok Sabha elections. Despite criticism for not aligning with the TMC in West Bengal, the Congress remains hopeful.

More Lifestyle

Buzzing Now

Jun 04: Latest News

• 01 Iran’s acting top diplomat dismisses US-proposed Gaza cease-fire deal in visit to Lebanon
• 02 BJP turning jantantra into mann-tantra, people ready for naya andolan: Akhilesh Yadav
• 03 Greek’s far-right leader to return to prison after judges rescind his early release
• 04 Kylian Mbappe completes transfer to Real Madrid, posts photo with boyhood hero Cristiano Ronaldo: ‘Club of my dream’
• 05 MVA banks on Baramati, Pune with eye on state polls, Mahayuti confident of winning all 4
• Elections 2024
• Political Pulse
• Entertainment
• Movie Review
• Newsletters
• Web Stories

• Tools and Resources
• Customer Services
• Communication and Culture
• Communication and Social Change
• Communication and Technology
• Communication Theory
• Critical/Cultural Studies
• Gender (Gay, Lesbian, Bisexual and Transgender Studies)
• Health and Risk Communication
• Intergroup Communication
• International/Global Communication
• Interpersonal Communication
• Journalism Studies
• Language and Social Interaction
• Mass Communication
• Media and Communication Policy
• Organizational Communication
• Political Communication
• Rhetorical Theory
• Share This Facebook LinkedIn Twitter

Article contents

Journalists’ professional roles and role performance.

• Claudia Mellado Claudia Mellado School of Journalism, Pontificia Universidad Católica de Valparaíso
• https://doi.org/10.1093/acrefore/9780190228613.013.832
• Published online: 25 February 2019

Professional roles are a key topic in journalism research along with the fundamental elements in defining journalism as a profession. For many decades, scholars have devoted their efforts to analyzing normative standards and journalistic ideals, while their analysis through the lens of professional performance has remained in the background. Nevertheless, considerably more attention has been paid over the past decade to the theorization of the different concepts in play when analyzing professional roles in journalism, especially the study of journalistic role performance (i.e., the manifestation of professional roles in both news decisions and the news outcome that reaches the public). Studies on journalistic role performance are able to tell us how or to what extent news professionals have enough autonomy for their role conceptions or perceptions to be manifested in journalistic practices, as well as in the news product made available to the public.

So far, research on journalistic role performance has systematically found patterns of multilayered hybridization in journalistic cultures across and within advanced, transitional, and non-democratic countries. Several studies have also shown significant discrepancies depending on societal, organizational, and individual factors, as well as a wide gap between journalistic ideals and professional practices. Some of these studies have also found significant discrepancies between journalists’ role conceptions and their perceived role enactments.

Future studies need to address the intrinsic capacity of social media platforms to deinstitutionalize communication through parallel channels, which may turn out to be a crucial element when it comes to performing both traditional and new journalistic roles.

• professional roles
• journalistic role performance
• journalistic cultures
• role conception
• role perception
• perceived role enactment
• comparative journalistic research
• professionalism
• journalism studies

Introduction

As an object of study, professional roles are one of the key topics in journalism research (Mellado, Hellmueller, & Donsbach, 2017a , p. 3). They are also the fundamental elements when it comes to defining journalism as a profession. In this respect, journalistic roles become essential components of journalistic cultures.

Journalistic cultures represent the cultural capital that, as “interpretive communities,” journalists share. They may manifest themselves in values and ideals as well as in journalistic practices (Mellado et al., 2017a ; Schudson, 2003 ; Zelizer, 1993 ).

For many decades, the study of professional roles in journalism was addressed mostly from the perspective of normative standards and journalistic ideals, while their analysis through the lens of professional performance remained in the background.

Through surveys of and interviews with journalists, an extensive array of studies around the globe have analyzed both the different roles that they should normatively fulfill in society (Christians, Glasser, McQuail, Nordenstreng, & White, 2009 ) as well as the roles that journalists consider important for their profession (Hanitzsch, 2011 ; Hanitzsch et al., 2011 ; Patterson & Donsbach, 1996 ; Weaver, Beam, Brownlee, Voakes, & Wilhoit, 2007 ; Weaver & Wilhoit, 1986 , 1996 ; Weaver & Willnat, 2012 ; Weinacht & Spiller, 2014 ; Willnat, Weaver, & Choi, 2013 ). Among the classic empirical typologies of roles are the “neutral” and “participant” roles found by Cohen ( 1963 ); the “neutral,” “watchdog,” and “analytical” roles found by Johnstone, Slawski, and Bowman ( 1976 ); the “disseminator,” the “interpreter,” the “watchdog,” and the “populist mobilizer” roles found by Weaver and his team at Indiana University (Weaver & Wilhoit, 1986 , 1996 ; Weaver et al., 2007 ), or the “bloodhound” and “missionary” functions proposed by Köcher ( 1986 ).

Studies on professional role conceptions have been Western-oriented (Josephi, 2005 ; Hallin & Mellado, 2018 ; Mellado, 2015 ). Most early studies on role conceptions were designed and conducted in advanced democracies (especially the United States). 1 Later, researchers from other regions such as Latin America, Asia, and Africa became interested in applying these surveys to their own journalists to establish which professional roles were most important to them (e.g., Deuze, 2002 ; Herscovitz, 2004 ; Muchtar, Hamada, Hanitzsch, Masduki, & Ullah, 2017 ; Pasti, 2005 ; Ramaprasad, 2001 ; Wilke, 1998 ; Zhu, Weaver, Lo, Chen, & Wu, 1997 ). However, most of them relied on measurements or conceptual approaches developed in the West, basically because they were already available.

Over the past six decades, most of these studies have typically shown that journalists worldwide endorse professional roles and values that emphasize neutrality, objectivity, and the scrutiny of official behavior, holding those in power accountable (e.g., Donsbach & Patterson, 2004 ; Hanitzsch et al., 2011 ; Patterson & Donsbach, 1996 ; Weaver, 1998 ; Weaver & Willnat, 2012 ).

One of the basic assumptions of role conception research is that the way in which journalists understand their roles shapes the news content they produce. This assumption is mostly based on psychological studies that have shown a significant relationship between attitudes and behavior in specific working and personal environments (Kim & Hunter, 1993 ).

Indeed, earlier studies on role conceptions argued that there was a significant relationship between role conceptions and journalistic decisions (Kepplinger, Brosius, & Staab, 1991 ; Patterson & Donsbach, 1996 ; Starck & Soloski, 1977 ). However, those studies did not compare the rhetoric on roles with actual journalistic performance but only journalists’ perceptions of different professional roles, as well as what they thought they actually did. However, as we know, surveys will never measure actual behavior.

Moreover, the extent to which journalistic ideals can manifest themselves in practice in the journalistic field depends on myriad factors that frequently yield differences and sometimes contradictions between what professionals would like to do and what they actually do in their work (Mellado et al., 2017a ; Mellado & Van Dalen, 2014 ; Shoemaker & Reese, 2013 ; Van Dalen, de Vreese, & Albæk, 2012 ).

As far back as the 1930s, Rosten ( 1937 ) had already warned of the tension between professional ideals and practical constraints. Since then, gatekeeping theory and the hierarchy-of-influences approach have analyzed a variety of constraints that might influence how journalists perform their roles, suggesting that explanations of journalistic cultures based solely on role conception research should be questioned rather than assumed (Shoemaker & Reese, 2013 ; Shoemaker & Vos, 2009 ).

Due to the problem of taking survey data as valid measurements of journalistic practice (Patterson & Donsbach, 1996 ; Schudson, 2003 ; Vos, 2002 ) and the lack of theorization of professional roles within the journalistic field (Mellado et al., 2017a ), considerably more attention has been paid over the past decade to the study of journalistic role performance, that is, the manifestation of professional roles in both news decisions and the news outcome that reaches the public (Hallin & Mellado, 2018 ; Humanes & Roses, 2018 ; Mellado, 2015 ; Mellado, Humanes, Scherman, & Ovando, 2018 ; Mellado et al., 2017a , 2017b ; Mellado & Van Dalen, 2014 ; Stępińska, Jurga-Wosik, Adamczewska, Selcer, & Narożna, 2016 ; Vos, 2002 ; Tandoc, Hellmueller, & Vos, 2013 ; Wang, Sparks, Lü, & Huang, 2017 ).

The concept of role performance focuses on the connection between journalists’ beliefs about the role of journalism and the actual practice of producing the news (Mellado, 2015 ). It connects studies on the professionalism of roles with those on news production and the sociology of news.

In accordance with gatekeeping and hierarchy-of-influences studies, more recent research measuring both role conceptions and actual role performance has systematically shown a wide gap between professional ideals and professional practices at different levels (Brüggemann & Engesser, 2014 ; Mellado & Van Dalen, 2014 ; Tandoc et al., 2013 ; Vos, 2002 ).

Different studies assert that although journalistic role conception research is an important component for the study of professionalism, where “ideals and values are the tools and skill sets that set journalism apart from other fields and guarantee its autonomy from heteronymous forces, and serve to legitimatize and define journalism” (Mellado, 2015 , p. 596), the constraints to which they are exposed within the profession limit the possibility of living up to their normative standards (Mellado & Van Dalen, 2014 ), even though journalists may have clear ideas about which professional roles are most important to them.

Research on journalistic role performance has also shown that the practice of the profession seems to vary much more widely than the current findings on journalistic ideals would suggest. The most recent empirical studies have found patterns of multilayered hybridization in professional role performance worldwide, with journalistic cultures displaying multiple “types of journalism” that do not fully resemble ideal typologies of roles or media systems (Humanes & Roses, 2018 ; Mellado et al., 2017b ; Mellado, Márquez-Ramírez, Mick, Oller Alonso, & Olivera, 2017e ; Stępińska et al., 2016 ; Wang et al., 2017 ). In other words, these studies have found that when considering the performance of essential roles of journalism in society, ideal or conventional assumptions about journalistic role conceptions in different societal settings fail to work.

Such results reported by the international literature on journalistic role performance in recent years have undeniably re-opened the opportunity to discuss the extent to which professional roles are contingent upon different levels of influence (Mellado et al., 2017a ). In this context, the concept of journalistic role performance becomes quite valuable since it focuses attention on issues relating to structure, agency, culture, and political economy. These issues are key to the understanding of journalism as a social practice (Esser, 2008 ). Journalistic role performance also raises the problem of the connection between journalism’s normative ideals and practice (Schudson, 2003 ), and encourages us to think about how ideals arise and function within a larger social context and how they relate to journalistic performance.

It is relevant to consider that although this entry is squarely positioned in the quantitative research tradition—as it is the tradition of most role performance, role conception, and role perception studies—there is a lot of newsroom and ethnography research that actually addresses these specific issues using a different conceptual framework. Indeed, reading authors from media sociology like Goffman ( 1959 ) or Tuchman ( 1972 ) clearly shows us that the notion of role performance has been quite close to qualitative and ethnographic research that has dealt with how journalists actually “operationalize” their own ideals in their daily practice.

Professional Roles as Object of Study

One of the main challenges of quantitative studies on journalistic professional roles over time has been the lack of theorization and empirical focus on analyzing the link between role conception and journalistic practice. Basically, journalistic roles have mostly been used as an empirical concept to study the functions on which journalists place more importance in society, while the definitions given to the construct “professional role” have varied quite widely (Mellado, Hellmueller, & Weaver, 2017c ).

Overall, role theory defines a role as the expected behavior of people who occupy a particular social status and position in society (Biddle, 1979 ; Montgomery, 1998 ). In other words, a role has been conceptualized as the function undertaken by an individual in accordance with the normative demands placed upon that individual in his or her position (Goffman, 1961 , p. 85). It should be noted, however, that roles might take different forms depending on the level of analysis applied to them.

Mellado et al. ( 2017a ) distinguished four different concepts within the construct of professional roles in journalism: role conception, role perception, (perceived) role enactment, and role performance (see Table 1 ), taking into account the dual empirical and evaluative aspect thereof. The differences between these concepts are not only about how they are defined but also about how they can be empirically measured (Blumer, 1969 ; Lynch, 2007 ; Turner, 2006 ).

Role conception: The purposes of the profession that a journalist conceives as most important at the individual level, and where the journalists’ evaluation of a specific role is not necessarily related to social consensus.

Role perception: Perceived role expectations in society. Role perceptions do not form a mental picture of a role for a particular journalist and do not necessarily have a location in the conceiver. They may instead follow a script that has been internalized and is located in a larger social structure. For example, journalists who perceive a watchdog role as important may have internalized the way of thinking expected by the media outlet where they work.

Role enactment: The implementation of a journalistic role focusing on the individual journalist only. Here, of course, there is a need for consistency between the journalist’s role conception/perception and his or her behavior. Within journalism studies, role enactment has been addressed mostly at the evaluative level of analysis, referring to what journalists think they do (Culbertson, 1983 ; Oi, Fukuda, & Sako, 2012 ; Patterson & Donsbach, 1996 ; Skovsgaard, Albæk, Bro, & de Vreese, 2013 ; Weischenberg, Malik, & Scholl, 2006 ). Some studies have used the term “role enactment” to analyze the manifestation of journalistic roles in news content (Carpenter, Boehmer, & Fico, 2015 ; Tandoc et al., 2013 ; Vos, 2002 ). Nevertheless, given the internal and external constraints that impact journalism, the concept of role enactment may never be fully applied within the journalistic profession. In this respect, role enactment differs from role performance in that it is more difficult to accomplish in journalism, since it implies that journalists have autonomy and freedom over their work, thereby being able to individually put into practice what they think are the most important roles of their profession.

Role performance: The collective outcome of concrete newsroom decisions and the style of journalistic reporting, taking into consideration the different factors that influence journalism as a professional practice (Mellado et al., 2017a , p. 5). Role performance deals with behaviors (Biddle, 1979 ; Burke & Reitzes, 1981 ), and it can be seen as an outcome of dynamic negotiations influenced by different internal and external constraints that potentially inhibit, but may also enable, the practice of journalism (Mellado, 2015 ; Vos, 2005 ). Within this context, role performance can explain the extent to which journalists’ news decisions and reporting styles are influenced by a journalist’s specific role conception, perception, or perceived enactment, or by other variables that are not in the mix of expectations perceived as legitimate.

The performative level of professional roles is the level most likely to be observed by the public and different reference groups through the outcome of newsroom decisions (e.g., the news story). Nevertheless, the study of role performance also enables an analysis that captures the so-called backstage of news production, that is, the negotiations with different reference groups, the search for sources, styles of data gathering, and the verification process.

It should be borne in mind that although its relevant components are inevitably determined by normative criteria, the concept of role performance is not a normative one (Mellado, 2015 ; Mellado et al., 2017a ).

Indeed, roles are not universal, so they are not good or bad per se. They are historical, are situational, and can be mediated constantly depending on the specific contexts. Journalists in advanced democracies, for example, may perform a more detached watchdog role or a more adversarial watchdog role, as happens in the cases of the United States and Spain. Likewise, journalists from transitional democracies may perform a more civic role or stick to performing a disseminator role if, for example, they are still fearful of a dictatorship regime. At the same time, journalists working for a very commercialized media outlet or in specific news beats may use the infotainment role in a much more prominent way, while in non-democratic countries the adversarial role or the loyal-facilitator role (depending on what type of dictatorship it is, of course) would be more appropriate.

Table 1. Role Conception, Perception, Enactment, and Performance in Journalism

Source: Mellado et al. ( 2017a , p. 7).

The differences in the way roles are conceptualized are limited by the applied perspective. The functionalist approach—the one used for most studies on professional role conceptions—sees a role as the set of expectations that a society places on an individual, creating regular rules and patterns of behavior. This definition stresses normative expectations and social consensus, where the performance of a role tends to be taken for granted. 2

This excess of normativism has meant that most research is still limited to “hard” news, giving the impression that journalistic professional roles are more relevant to specific groups of the profession that interact with the power elites, while ignoring those that report on other issues or topics. Indeed, the emphasis that the field has put on the link between politics and journalism has diminished other areas and elements of journalism that do not belong to this relationship (Zelizer, 2011 ). This has certainly affected the configuration of professional roles that are now addressed from both the normative and the empirical levels, and the way in which they have been studied. For example, social media raises important questions about journalists’ roles and their identity in a shared media space (Duggan, 2015 ; Hermida, 2014 ).

Meanwhile, the interactionism perspective—the one used by studies on role performance—does not see roles as being fixed but instead as negotiated and changeable within a particular social context (Blumer, 1969 ; Lynch, 2007 ; Mellado, 2015 ; Vos, 2005 , 2017 ). As it has been already mentioned, role performance studies conceptualize professional roles as flexible, situational, and independent sets of functions that can be combined in various ways across space and time, often subject to adaptation and combination (Hallin & Mellado, 2018 , p. 28). This approach is closer to the one used by earlier research that tackled some of these issues using a different methodology, e.g., newsroom ethnography studies.

The Gap Between Professional Ideals and Journalistic Performance

The gap between ideals of roles or normative values and role performance is measured as the degree of congruence or discrepancy between role conception, role perception, or perceived role enactment and professional performance (Mellado & Van Dalen, 2014 ).

Understanding the gap and the relationship between what journalists want to do (or think they ought to do) and what they and the media actually perform is crucial to the analysis of forces affecting news production.

Journalists may see their main role as that of providing information that citizens need in order to be active in political life, yet they may be unable to perform that role because political stories do not necessarily attract large audiences. They may also see their main role as that of acting as a “watchdog” over those in power, yet they may be unable to perform that role in most countries due to obstacles such as government control, corporate control over the media (e.g., owners have veiled corporate interests that might become compromised by watchdog stories), among others constraints. For journalists, this situation is the rule rather than the exception. Moreover, the threats and harassment that these professionals face are a global challenge, even for advanced democracies (Löfgren & Örnebring, 2016 ).

In this respect, it is important to bear in mind that when the performance of individual journalists differs from the characteristics of their role conception or perception, it is not necessarily an indicator of a malfunctioning news organization. Indeed, such journalists may perform roles that conflict with those they think they do or like or ought to do with those their colleagues perform, those expected by the media outlet where they work, and even those embedded within the macro-organizational structure in which the media is located.

While systematic studies dedicated to measuring the gap are relatively recent and evidence of the forces explaining such a gap is still insufficient, I discuss here several reasons that may explain the distance between professional ideals and journalistic practice.

Professional roles are not mutually exclusive (Mellado, 2015 ); they are historically constructed and sometimes conflicting (Vos, 2017 ), so journalists have to combine them in various ways when they produce a news story. They may praise specific ideals of the profession, but their practices are guided by roles that are context specific and not fixed, which can be applied in different ways, as already exemplified in the section “ Professional Roles as Object of Study ” (Mellado, 2015 ).

Traditional approaches assume that professional roles are fully contracted by behavior and categorized by norms, values, and beliefs (Ashforth, 2001 ). Nevertheless, previous studies have suggested that journalistic roles cannot be considered discrete categories (Mellado, 2015 ; Vos, 2005 ). Indeed, while at the normative level it is possible to talk about ideal types of roles in journalism, in practice this becomes impossible, since professional roles at all stages (conception, perception, perceived enactment, and performance) may overlap in practice (Lynch, 2007 ). In this respect, unlike ideals and values that are easier to transfer, and unlike media system characteristics that have shown more discernible patterns in comparative studies, such as legal frameworks, the size and scope of their media markets, the nature of state intervention, or the extent to which political instrumentalization and parallelism are present, the actual manifestation of journalistic roles is more difficult to fit into existing ideal typologies, since they are constantly changing, being mediated not only by social, organizational, and individual factors, but also by the local context of the news (Mellado et al., 2017b ).

The existence of multiple roles partly reflects the institutional context within which journalists work. Also, since news professionals write for many different audiences simultaneously, journalistic role ideals are often rooted in conceptions of the audience, which often means that several roles overlap when writing a single story.

The gap may also be partly due to differences in the conceptualization of the ideal–practice connection. While some research focuses on the ideal–practice connection as a relationship (Scholl & Weischenberg, 1998 ) (e.g., studying whether journalists who place more importance on the infotainment role of journalism are more likely to perform that role), other studies address the connection between ideals and practice as a gap, by focusing on the distance between the two. Of course, different conclusions may be drawn depending on which of the two approaches are used (Mellado & Van Dalen, 2014 ).

Another possibility is that the gap is, in part, a methodological artifact that may be due to the fact that previous research has not measured role conceptions correctly. For example, in the surveys typically used to measure role conception, the wording of the questions might be too abstract, meaning that journalists do not actually have a common basis on which to reply, thus resulting in non-comparable responses. Another flaw of survey-based research is that journalists might give socially desirable responses that fit normative expectations. Their responses therefore might be about what they would like or are expected to do instead of what they actually do. It is also common for journalists from different contexts to understand concepts differently or give different meanings to concepts such as neutrality or partisanship. For example, the translation of Western ideals such as truth and freedom does not really capture the ways in which those norms are articulated within other contexts. Thus, the fact that they are implemented and understood according to local realities suggests that there are no straightforward universal journalistic practices with unanimous and stable meanings.

It may also be the case that journalists interviewed for studies measuring role conceptions do or do not have great influence over the most important decisions on news content and that the gap should be also addressed at the organizational level (Mellado, Mothes, Hallin, Humanes, 2019 ). Finally, the lack of validation of some of the scales used to measure professional roles at the evaluative level (e.g., Hanitzsch et al., 2011 ) may have contributed to such a methodological artifact too.

There is no doubt that the gap also reflects the heterogeneity of the profession. Not all journalists, media organizations, or news beats embrace the same journalistic role conceptions and/or perceptions, and often there are struggles within the field over which role conceptions should prevail. Advising the consumer, for instance, is a paramount role in soft news and sections such as lifestyle or travel, as much as infotainment is a paramount role for sports or gossip. But even the so-called hard news sections are subject to forces of commercialization, and the inclusion of elements of roles like the infotainment one is becoming more and more common (Mellado, López-Rabadán, & Elórtegui, 2017d ; Uribe & Gunter, 2007 ).

The size of the gap obviously varies by situation, historical period, and social context, resulting in journalistic role conceptions being more or less coupled with role performance in different contexts (Vos, 2017 ). This is undeniably one of the most important advantages of the concept of role performance over the concept of role conception, since it focuses attention on the sources of such variation.

Studying Professional Roles

An important issue when studying not only every stage of professional roles but also the gap between ideals and practice is the way in which these concepts are empirically measured.

The very first studies to measure professional roles by content analysis were important efforts that laid the foundations for the study of journalistic role performance. Nevertheless, most of those studies were not clear about the operationalization of their measurements and were dependent on subjective interpretations rather than an investigation into specific practices that could potentially reveal professional roles in news content. For example, Weaver and Wilhoit ( 1996 ) and Vos ( 2002 ) studied role conceptions and separately evaluated news stories written by the surveyed reporters. However, the operational definition of roles in news content comprises several statements that are the same as the questions put to those journalists (Mellado, 2015 ).

Other studies on reporting styles that indirectly address different functions of the media tended to focus on isolated indicators (Benson & Hallin, 2007 ; Esser, 2008 ; Van Dalen, de Vreese, & Albæk, 2012 ), making it difficult to establish the reliability and validity of scales.

In order to overcome these serious limitations, initiatives like the international Journalistic Role Performance Around the Globe (JRP) Project have generated a common methodology, with valid scales to measure journalistic role performance in news content. Using these scales, scholars are able to analyze different dimensions of journalistic performance in different contextual settings to enable cross-national comparative research.

The performative level of journalistic cultures represents the level most likely to be observed by the public and different reference groups through the analysis of the outcome of newsroom decisions (e.g., the news story). Journalistic role performance as a product, then, is normally measured by content analysis.

Nevertheless, the performance of roles can also take the form of actions performed prior to the output, the so-called backstage of news production (Goffman, 1959 ), where the negotiations with different reference groups, the search for sources, and the verification process are crucial. Journalistic role performance as a process, then, should be measured by techniques such as ethnography and/or in-depth interviews.

The JRP Project focuses on the final output, analyzing how different dimensions of professional roles manifest themselves in news content across different countries and cultures. This project has been the starting point for other related studies, which have also addressed the disconnect between journalists’ norms and professional performance (e.g., Mellado & Van Dalen, 2014 ), as well as comparisons of role performance among media platforms (e.g., Hallin & Mellado, 2018 ).

Based on standardized content-based measures (Mellado, 2015 ), different studies have looked at the practice of six journalistic roles that run along the presence of the journalistic voice in the news, the relationship between journalism and those in power, and the way journalism approaches the audience. Relevant literature has previously suggested similar perspectives as three dimensions to be used for the analysis of journalistic roles (Donsbach, 2008 ; Hanitzsch, 2007 ). Based on this previous work, studies on role performance have indicated and then corroborated (e.g., Mellado et al., 2017b ; Mellado et al., 2013 ; Mellado & Van Dalen, 2017 ; Mellado & Vos, 2017f ; Mellado, Hanusch, Humanes et al., 2013 ) that, instead of dimensions, these are three interrelated domains from which six independent roles emerge: the interventionist, watchdog, loyal-facilitator, service, civic, and infotainment roles.

Following role theory research, studies on role performance also emphasize that these are not the only domains from which role performance can be studied, nor are they the only possible roles that can emerge from these domains (Mellado, 2015 ). Indeed, recent studies on journalistic performance analyze specific sub-dimensions that emerge from those roles (e.g., Márquez-Ramírez et al., 2019 ) as well as new domains of journalistic roles in social media spaces (Hermida & Mellado, 2019 ).

The measures used to analyze the manifestation of different journalistic roles in the news in different political, historical, and cultural contexts, have been inspired by both content analysis on journalistic practices mentioned throughout the text and important qualitative studies of newsroom practices (Bantz, McCorkle, & Baade, 1980 ; Bogaerts, 2011 ; Tuchman, 1972 ; Usher, 2014 ).

The journalistic voice domain deals with the presence of the journalists’ voice in the news. The passive stance of journalists in their reporting has been associated with the neutral and disseminator roles (Cohen, 1963 ; Weaver & Wilhoit, 1986 ), while the active stance has been linked to the participant (Donsbach & Patterson, 2004 ; Johnstone et al., 1976 ), advocate (Janowitz, 1975 ), and missionary roles (Köcher, 1986 ).

The power relations domain deals with the relationship between journalists and those in power. Journalists might defend the idea of monitoring de facto powers and denouncing wrongdoings (Waisbord, 2000 ), but, at the same time, they may see their role as acting as loyal spokespersons for those in power, conveying a positive image of them, supporting official policies, and portraying a positive image of one’s country, thereby encouraging the sense of belonging (Donohue, Tichenor, & Olien, 1995 ; Donsbach, 1995 ; Pasti, 2005 ; Mellado, 2015 ).

The audience approach domain deals with the way in which journalists address the audience. Based on this different understanding of the audience, they can be addressed as citizens, clients, or spectators (Eide & Knight, 1999 ; Rosen, 1996 ; Weaver et al., 2007 ).

Within each of these domains, professional roles can be measured in the news by specific indicators that resemble specific reporting styles. The journalistic voice domain involves the role of the journalist as interventionist vis-à-vis disseminator . The power relations domain involves the watchdog and the loyal-facilitator roles. Finally, the different understanding of the audience can be associated with three independent dimensions of professional roles: the civic, the infotainment, and the service roles.

All these roles—with the exception of the interventionist and the disseminator roles, which are part of a one-dimensional structure (Mellado, 2015 )—have been proved to be independent but relate to each other to some extent (Mellado et al. 2017b ; Mellado et al., 2013 ; Mellado & Van Dalen, 2017 ; Mellado & Vos, 2017 ; Mellado, Hanusch, Humanes et al., 2013 ). A news story may include, for example, elements from the watchdog role and from the loyal role at the same time. They cannot, therefore, be considered poles of a continuum as previous studies have suggested (Hanitzsch, 2007 ).

Each of the six dimensions of role performance is operationalized in terms of its practical manifestations in news content.

Figure 1. Dimensions of journalistic role performance.

As a template for empirical studies, roles are seen as empirical constructs to study role performance in news content within different cultural contexts. However, these are not the only perspectives from which role performance can be analyzed in the news, nor are they the only dimensions of journalistic performance that can be found in news content, particularly in the new media landscape. Moreover, taking into account that several concepts are inevitably culturally bound, it is likely that not all the indicators emerging from the literature will work in the same way in all societies, especially when considering that professional roles can be seen as reflective measurement models, in which the dimensions exist independently from the measures used, and where adding or removing an item does not change the conceptual domain of the dimension (Wirth & Kolb, 2012 ).

Professional Role Conception, Perception, and Perceived Enactment

While it may be true that measuring professional role performance by content analysis or ethnographic work is far more complex than measuring professional roles by surveys of journalists, the challenge of measuring role conceptions/perceptions or perceived role enactment properly is also very big.

Looking at the studies on role conception, role perception, and perceived enactment carried out over the past decades, the operational definition of roles mostly takes the form of abstract statements—such as being a detached observer—that journalists are asked to rate by indicating the extent to which such statements are important to them or to their media outlets, or the extent to which they believe they accomplish those functions. However, they do not take the form of specific practices—such as taking sides in a news story—that journalists may find easier to understand. Of course, what it means “to be a detached observer” for an Indian journalist, for example, is not necessarily the same for an American journalist. In contrast, the terms “opinion” or “interpretation” are globally shared. The same may happen among colleagues from different, or even the same newsroom or news beat. Indeed, the preliminary results of the JRP Project have shown, first, that there is more consistency in journalists’ responses to the questions addressing specific practices than those on abstract aspects of the profession, and second, that abstract and empirical indicators of role conceptions do not belong to the same role dimensions, that is, they do not measure the same thing (Mellado & Helmueller, 2015 ). I mentioned this issue in the section “ The Gap between Professional Ideals and Journalistic Performance ,” when remarking that the gap was a possible methodological artifact.

This critical aspect suggests that (a) possibly, role conceptions have not been measured correctly, and (b) the lack of clarity on the meaning of professional roles at the normative level leads to confusion between role conception, role perception, and perceived role enactment in studies on journalistic professional roles.

Finally, measuring the gap between role ideals and professional practices requires a combination of methods (e.g., surveys or interviews for role ideals and content analysis and/or newsroom observation or ethnography for role performance). The impact of other variables on news decisions and reporting styles can also be assessed through the interviews or data collected at the organizational and/or societal level.

Studying journalistic professional roles through the lens of not only evaluations of journalists’ work but also the way in which they perform their professional roles is far from being a mere academic exercise, since the way in which journalists cover news has a profound impact on shaping public and private spheres, on citizens, on governance, and on the democratic construction of a specific national system.

Studies on journalistic role performance move things a step forward because they are able to tell us how or to what extent news professionals have enough autonomy for their role conceptions to be manifested in journalistic practices as well as in the news product made available to the public. Indeed, one of the advantages of studying the performative level of journalistic roles is that it can help to measure journalistic autonomy in a more indirect yet objective way, by analyzing the extent to which journalists are able to put their professional roles and ideals into practice (Mellado & Van Dalen, 2014 ).

In this respect, unlike studies on professional role conceptions alone, journalistic role performance studies appear to offer different perspectives on the practice of journalism around the world, particularly in countries where evaluative elements are less articulated in practice (Mellado et al., 2017a ).

So far, while international studies on professional role conception/perception support the idea that journalists from different media systems conceptualize their roles as expected (Hanitzsch et al., 2011 ; Weaver & Willnat, 2012 ), research on journalistic role performance has systematically found patterns of multilayered hybridization in journalistic cultures across and within advanced, transitional, and non-democratic countries (Mellado et al., 2017b ). Several studies have also shown a significant gap between journalistic ideals and professional practices, particularly in roles more closely related to the professional ideal of the Fourth Estate; in countries with less press freedom; among journalists who cover “hard” news topics and those who feel more economic and political pressures. Also, some of these studies have found discrepancies between journalists’ role conceptions and their perceived role enactments (Mellado & Van Dalen, 2014 ; Roses & Humanes, 2019 ).

Likewise, studies on role performance have found significant differences in the performance of a variety of professional roles depending on the media platform (Hallin & Mellado, 2018 ), although the latest research suggests that the thematic beat and media audience orientation are more crucial variables in explaining differences in role performance (Humanes & Roses, 2018 ; Márquez- Ramírez et al., 2018 ; Mellado et al., 2018 ). Similar studies have also revealed a significant correlation between the prevalence of different sources and journalistic roles in news (Márquez- Ramírez et al., 2018 ).

One of the main challenges of studying journalistic roles, and, specifically, the performance of such roles today, is that the conceptual boundaries of journalism have become increasingly blurred in the current digital ecosystem. Social media’s intrinsic capacity to deinstitutionalize communication through parallel channels may turn out to be a crucial element when it comes to performing both traditional and new journalistic roles. Professional roles can certainly evolve and be redefined—in both discourse and practice—since their emergence and development not only have historical and contextual components but also a strong grounding in the logic that journalists use to communicate with the audience, and in the expectations of various reference groups. Future studies need to address all these elements.

• Ashforth, B. (2001). Role Transitions in Organizational Life: An Identity-Based Perspective . Mahwah, NJ: Erlbaum.
• Bantz, C. , McCorkle, S. , & Baade, R. (1980). The News Factory. Communication Research , 7 (1), 45–68.
• Benson, R. , & Hallin, D. (2007). How States, Markets and Globalization Shape the News: The French and US National Press, 1965–97. European Journal of Communication , 22 (1), 27–48.
• Biddle, B. (1979). Role Theory: Expectations, Identities and Behaviors . New York: Academic Press.
• Blumer, H. (1969). Symbolic Interactionism: Perspective and Method . Englewood Cliffs, NJ: Prentice Hall.
• Bogaerts, J. (2011). On the Performativity of Journalistic Identity. Journalism Practice , 5 (4), 399–413.
• Brüggemann, M. , & Engesser, S. (2014). Between Consensus and Denial: Climate Journalists as Interpretive Community. Science Communication , 36 (4), 399–427.
• Burke, P. , & Reitzes, D. (1981). The Link between Identity and Role Performance. Social Psychology Quarterly , 44 (2), 83–92.
• Carpenter, S. , Boehmer, J. , & Fico, F. (2015). The Measurement of Journalistic Role Enactments: A Study of Organizational Constraints and Support in For-Profit and Nonprofit Journalism. Journalism & Mass Communication Quarterly , 93 (3), 587–608.
• Christians, C. G. , Glasser, T. L. , McQuail, D. , Nordenstreng, K. , & White, R. A. (2009). Normative Theories of the Media: Journalism in Democratic Societies . Urbana: University of Illinois Press.
• Cohen, B. (1963). The Press and Foreign Policy . Princeton, NJ: Princeton University Press.
• Culbertson, H. M. (1983). Three Perspectives on American Journalism. Journalism Monographs , 83 , 1–33.
• Deuze, M. (2002). National News Cultures: A Comparison of Dutch, German, British, Australian and U.S. Journalists. Journalism & Mass Communication Quarterly , 79 (1), 134–149.
• Donohue, G. , Tichenor, P. , & Olien, C. (1995). A Guard Dog Perspective on the Role of Media. Journal of Communication , 45 (2), 115–132.
• Donsbach, W. (1995). Lapdogs, Watchdogs and Junkyard Dogs. Media Studies Journal , 9 (4), 17–30.

• Donsbach, W. , & Patterson, T. (2004). Political News Journalists: Partisanship, Professionalism, and Political Roles in Five Countries. In F. Esser & B. Pfetsch (Eds.), Comparing Political Communication: Theories, Cases, and Challenges (pp. 251–270). New York: Cambridge University Press.
• Donsbach, W. (2008). Journalists’ Role Perception . In W. Donsbach (Ed.), The International Encyclopedia of Communication . London: Blackwell.
• Duggan, M. (2015). Mobile Messaging and Social Media 2015 . Pew Research Center.
• Eide, M. , & Knight, G. (1999). Public-Private Service: Service Journalism and the Problems of Everyday Life. European Journal of Communication , 14 (4), 525–547.
• Esser, F. (2008). Dimensions of Political News Cultures: Sound Bite and Image Bite News in France, Germany, Great Britain, and the United States. International Journal of Press/Politics , 13 (4), 401–428.
• Goffman, E. (1959). The Presentation of Self in Everyday Life . New York: Anchor.
• Goffman, E. (1961). Encounters: Two Studies in the Sociology of Interaction . Indianapolis, IN: Bobbs-Merrill.
• Hallin, D. C. , & Mancini, P. (2004). Comparing Media Systems: Three Models of Media and Politics . New York: Cambridge University Press.
• Hallin, D. , & Mellado, C. (2018). Serving Consumers, Citizens or Elites: Democratic Roles of Journalism in Chilean Newspapers and Television News. International Journal of Press/Politics , 23 (1), 24–43.
• Hanitzsch, T. (2007). Deconstructing Journalism Culture: Towards a Universal Theory. Communication Theory , 17 (4), 367–385.
• Hanitzsch, T. (2011). Populist Disseminators, Detached Watchdogs, Critical Change Agents and Opportunist Facilitators: Professional Milieus, the Journalistic Field and Autonomy in 18 Countries. International Communication Gazette , 73 (6), 477–494.
• Hanitzsch, T. , Hanusch, F. , Mellado, C. , Anikina, M. , Berganza, R. , Cangoz, I. , . . . Wang Yuen, E. K. (2011). Mapping Journalism Cultures across Nations: A Comparative Study of 18 Countries. Journalism Studies , 12 (3), 273–293.
• Hermida, A. (2014). Tell everyone: Why we share and why it matters . DoubleDay Canada.
• Hermida, A. , Lewis, S. , & Zamith, R. (2014). Sourcing the Arab Spring: A Case Study of Andy Carvin’s Sources on Twitter during the Tunisian and Egyptian Revolutions. Journal of Computer-Mediated Communication , 19 (3), 479–499.
• Hermida, A. & Mellado, C. (2019). The Dimensions of Social Media Logics: Mapping the Impact of Twitter and Instagram on Journalistic Performance . Paper presented at the annual International Association for Media and Communication Research, Madrid, July 7–11.
• Herscovitz, H. G. (2004). Brazilian Journalists’ Perceptions of Media Roles, Ethics, and Foreign Influences on Brazilian Journalism. Journalism Studies , 5 (1), 71–86.
• Humanes, M. L. , & Roses, S. (2018). Journalistic Role Performance in the Spanish National Press. International Journal of Communication , 12 , 1032–1053.
• Janowitz, M. (1975). Professional Models in Journalism: The Gatekeeper and the Advocate . Journalism Quarterly , 52 , 618–626, 662.
• Johnstone, J. , Slawski, E. , & Bowman, W. (1976). The Newspeople: A Portrait of American Journalists and Their Work . Urbana: University of Illinois Press.
• Josephi, B. (2005). Journalism in the Global Age: Between Normative and Empirical. Gazette , 67 (6), 575–590.
• Kepplinger, H. M. , Brosius H. , & Staab, J. F. (1991). Instrumental Actualization: A Theory of Mediated Conflicts. European Journal of Communication , 6 (3), 263–290.
• Kim, M-S. , & Hunter, J. (1993). Attitude-Behavior Relations: A Meta-Analysis of Attitudinal Relevance and Topic. Journal of Communication , 43 (1), 101–142.
• Köcher, R. (1986). Bloodhounds or Missionaries: Role Definitions of German and British Journalists. European Journal of Communication , 1 (1), 43–64.
• Löfgren Nilsson, M. , & Örnebring, H. (2016). Journalism under Threat . Journalism Practice , 10 (7), 880–890.
• Lynch, K. (2007). Modeling Role Enactment: Linking Role Theory and Social Cognition. Journal for the Theory of Social Behaviour , 37 (4), 379–399.
• Márquez-Ramírez, M. , Mellado, C. , Humanes, M. L. , Hellmueller, L. , . . . . Wang, H. (2018). The Detached vs. Adversarial Approaches to Watchdog Reporting: A Comparative Study of Journalistic Role Performance in 18 Countries (under review).
• Márquez-Ramirez, M. , Mellado, C. , Humanes, M. L. , Mothes, C. , et al. (2019). Detached or Interventionists? Comparing the Performance of Watchdog Journalism in Transitional, Advanced and Non-Democratic Countries . Paper presented at the 68th Annual Conference of the International Communication Association, Prague, May 24–28.
• Mellado, C. , Hanusch, F. , Humanes, M. , Roses, S. , Pereira, F. , Yez, L. , . . . Wyss, V. (2013). The Pre-Socialization of Future Journalists. Journalism Studies , 14 (6), 857–874.
• Mellado, C. , & Van Dalen, A. (2014). Between Rhetoric and Practice. Journalism Studies , 15 (6), 859–878.
• Mellado, C. (2015). Professional Roles in News Content: Six Dimensions of Journalistic Role Performance. Journalism Studies , 16 (4), 596–614.
• Mellado, C. , & Helmueller, L. (2015). Assessing Journalistic Role Performance in 31 Countries: Comparative Design and Methodological Challenges. Paper presented at the 65th Annual Conference of the International Communication Association, San Juan, Puerto Rico, May 21–25.
• Mellado, C. , Hellmueller, L. , & Donsbach, W. (Eds.). (2017a). Journalistic Role Performance: Concepts, Contexts and Methods . New York: Routledge.
• Mellado, C. , Hellmueller, L. , Márquez-Ramírez, M. , Humanes, M. L. , Sparks, C. , Stępińska, A. , . . . Wang, H. (2017b). The Hybridization of Journalistic Cultures: A Comparative Study of Journalistic Role Performance. Journal of Communication , 67 (6), 944–967.
• Mellado, C. , Helmueller, L. , & Weaver, D. (2017c). Revisiting Journalists’ Role Conceptions Research. In C. Mellado , L. Hellmueller , & W. Donsbach (Eds.), Journalistic Role Performance: Concepts, Contexts and Methods (pp. 21–40). New York: Routledge.
• Mellado, C. , & Van Dalen, A. (2017). Challenging the Citizen–Consumer Journalistic Dichotomy: A News Content Analysis of Audience Approaches in Chile. Journalism & Mass Communication Quarterly , 94 (1), 213–237.
• Mellado, C. , Humanes, M. L. , Scherman, A. , & Ovando, A. (2018). Do Digital Platforms Really Make a Difference in Content? Mapping Journalistic Role Performance in Chilean Print and Online News . Journalism .
• Mellado, C. , López-Rabadán, P. , & Elortegui, C. (2017d). ¿(Para) ciudadanos o espectadores? Periodismo político chileno y sus modelos profesionales de aproximación a la audiencia [(For) citizens or spectators? Political journalism and its professional models for approaching the audience]. Palabra Clave , 20 (1), 14–46.
• Mellado, C. , Márquez-Ramírez, M. , Mick, J. , Oller Alonso, M. , & Olivera, D. (2017e). Journalistic Performance in Latin America: A Comparative Study of Professional Roles in News Content. Journalism , 18 (9), 1087–1106.
• Mellado, C. , & Vos, T. (2017f). Conceptualizing Journalistic Role Performance Across Platforms. In C. Mellado , L. Hellmueller , & W. Donsbach (Eds.), Journalistic Role Performance: Concepts, Models and Measures (pp. 106–126). New York, NY: Routledge.
• Mellado, C. , Mothes, C. , Hallin, D. , & Humanes, M. L. (2019). In Pursuit of Autonomy: Investigating the Gap Between Journalist’s Role Conception and Media Role Performance in Nine European, Asian, and Latin American Countries . Paper presented at the Annual IAMCR Conference, Madrid, July 7–11.
• Montgomery, J. D. (1998). Toward a Role-Theoretic Conception of Embeddedness. American Journal of Sociology , 104 (1), 92–125.
• Muchtar, N. , Hamada, B. , Hanitzsch, T. , Masduki, A. , & Ullah, M. (2017). Journalism and the Islamic Worldview. Journalism Studies , 18 (5), 555–575.
• Oi, S. , Fukuda, M. , & Sako, S. (2012). The Japanese Journalist in Transition: Continuity and Change. In D. Weaver & L. Willnat (Eds.), The Global Journalist in the 21st Century (pp. 52–65). New York: Routledge.
• Pasti, S. (2005). Two Generations of Contemporary Russian Journalists. European Journal of Communication , 20 (1), 89–115.
• Patterson, T. E. , & Donsbach, W. (1996). News Decisions: Journalists as Partisan Actors. Political Communication , 13 (4), 455–468.
• Ramaprasad, J. (2001). A Profile of Journalists in Post-Independence Tanzania. Gazette , 63 (6), 539–556.

• Rosen, J. (1996). Getting the Connections Right. Public journalism and the Troubles in the Press . New York: Twentieth Century Fund Press.
• Roses, S. , & Humanes, M. L. (2019). Conflicts in the professional roles of journalists in Spain: Ideals and practice . Comunicar (preprint).
• Rosten, L. C. (1937). President Roosevelt and the Washington Correspondents. Public Opinion Quarterly , 1 (1), 36–52.
• Scholl, A. , & Weischenberg, S. (1998). Journalismus in der Gesellschaft. Theorie, Methodologie und Empirie [Journalism in society. Theory, methodology and findings]. Opladen/Wiesbaden, Germany: Westdeutscher Verlag.
• Schudson, M. (2003). The Sociology of News . New York: W.W. Norton.
• Shoemaker, P. , & Reese, S. (2013). Mediating the Message in the 21st Century. A Media Sociology Perspective . New York: Routledge.
• Shoemaker, P. J. , & Vos, T. P. (2009). Gatekeeping Theory . New York: Routledge.
• Skovsgaard, M. , Albæk, E. , Bro, P. , & de Vreese, C. (2013). A Reality Check: How Journalists’ Role Perceptions Impact Their Implementation of the Objectivity Norm. Journalism , 14 (1), 22–42.
• Starck, K. , & Soloski, J. (1977). Effect of Reporter Predisposition in Covering Controversial Story. Journalism Quarterly , 54 (1), 120–125.
• Stępińska, A. , Jurga-Wosik, E. , Adamczewska, K. , Selcer, B. , & Narożna, D. (2016). Journalistic Role Performance in Poland. Central European Political Studies , 2 , 37–52.
• Tandoc, E. , Hellmueller, L. , & Vos, T. (2013). Mind the Gap: Between Role Conception and Role Enactment. Journalism Practice , 7 (5), 539–554.
• Tuchman, G. (1972). Objectivity as Strategic Ritual: An Examination of Newsmen’s Notions of Objectivity. American Journal of Sociology , 77 (4), 660–679.
• Turner, R. (2006). Role Theory. In J. Turner (Ed.), Handbook of Sociological Theory (pp. 233–254). New York: Springer.
• Uribe R. , & Gunter, B. (2007). Are “Sensational” News Stories More Likely to Trigger Viewers’ Emotions Than Non-Sensational News Stories? European Journal of Communication , 22 (2), 207–228.
• Usher, N. (2014). Making News at The New York Times . Ann Arbor: University of Michigan Press.
• Van Dalen, A. , de Vreese, C. , & Albæk, E. (2012). Different Roles, Different Content? A Four-Country Comparison of the Role Conceptions and Reporting Style of Political Journalists. Journalism , 13 (7), 903–922.
• Vos, T. (2002). Role Enactment: The Influence of Journalists’ Role Conceptions on News Content. Paper presented at the 85th AEJMC Conference, Miami, FL.
• Vos, T. P. (2005). Journalistic Role Conception: A Bridge between the Reporter and the Press. Paper presented at International Communication Association conference, New York, NY.
• Vos, T. (2017). Historical Perspectives on Journalistic Roles. In C. Mellado , L. Hellmueller , & W. Donsbach (Eds.), Journalistic Role Performance: Concepts, Contexts and Methods (pp. 41–59). New York: Routledge.
• Waisbord, S. (2000). Watchdog Journalism in South America . New York: Columbia University Press.
• Waisbord, S. (2013). Reinventing Professionalism: Journalism and News in Global Perspective . Cambridge, UK: Polity.
• Wang, H. , Sparks, C. , Lü, N. , & Huang, Y. (2017). Differences within the Mainland Chinese Press: A Quantitative Analysis. Asian Journal of Communication , 27 (2), 154–171.
• Weaver, D. H. (1998). The Global Journalist: News People around the World . Cresskill, NJ: Hampton Press.
• Weaver, D. H. , Beam, R. A. , Brownlee, B. J. , Voakes, P. S. , & Wilhoit, G. C. (2007). The American Journalist in the 21st Century: U.S. News People at the Dawn of a New Millennium . Mahwah, NJ: Lawrence Erlbaum.
• Weaver, D. H. , & Wilhoit, G. C. (1986). The American Journalist: A Portrait of U.S. News People and Their Work . Bloomington: Indiana University Press.
• Weaver, D. H. , & Wilhoit, G. C. (1996). The American Journalist in the 1990s: U.S. News People at the End of an Era . Mahwah, NJ: Lawrence Erlbaum.
• Weaver, D. H. , & Willnat, L. (2012). The Global Journalist in the 21st Century . New York: Routledge.
• Weinacht, S. , & Spiller, R. (2014). Datenjournalismus in Deutschland. Eine explorative Untersuchung zu Rollenbildern von Datenjournalisten. Publizistik , 59 (4), 411–433.
• Weischenberg, S. , Malik, M. , & Scholl, A. (2006). Die Souffleure der Mediengesellschaft: Report über die Journalisten in Deutschland . [The prompters of the media society. Report about journalists in Germany]. Konstanz, Germany: UVK.
• Wilke, J. (1998). Journalists in Chile, Ecuador and Mexico. In D. Weaver (Ed.), The Global Journalist (pp. 433–452). Cresskill, NJ: Hampton Press.
• Willnat, L. , Weaver, D. , & Choi, J (2013). The Global Journalist in the Twenty-First Century. A Cross-National Study of Journalistic Competencies. Journalism Practice , 7 (2), 163–183.
• Wirth, W. , & Kolb, S. (2012). Securing Equivalence: Problems and Solutions. In F. Esser & T. Hanitzsch (Eds.), The Handbook of Comparative Communication Research (pp. 469–485). New York: Routledge.
• Zelizer, B. (1993). Journalists as Interpretive Communities. Critical Studies in Mass Communication , 10 (3), 219–237.
• Zelizer, B. (2011). Journalism in the Service of Communication. Journal of Communication , 61 (1), 1–21.
• Zelizer, B. (2017). What Journalism Could Be . Cambridge, UK: Polity Press.
• Zhu, J-H. , Weaver, D. , Lo, V-H. , Chen, C. , & Wu, W. (1997). Individual, Organizational, and Societal Influences on Media Role Perceptions: A Comparative Study of Journalists in China, Taiwan, and the United States. Journalism & Mass Communication Quarterly , 74 (1), 84–96.

1. As Western journalists developed high levels of professionalism, their understanding of their roles began to take shape within a context in which individual journalists were believed to have a considerable influence over the news stories they produced.

2. The reliance on normative beliefs is not surprising considering that the journalistic field has been predominantly normative in nature (Hallin & Mancini, 2004 ; Mellado et al., 2017a ; Waisbord, 2013 ; Zelizer, 1993 , 2017 ), thus dictating whether journalism can be considered “good” or “bad” journalism.

Related Articles

• Professional Identity and Roles of Journalists
• Public Service Journalism
• Aggregation and Journalism
• Employment Conditions in Journalism
• Journalism as Institution
• Comparative Journalism Research
• Professionalism, Professionalization, and Deprofessionalization in Journalism
• Trends: Women in International Journalism

Printed from Oxford Research Encyclopedias, Communication. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 04 June 2024

• Cookie Policy
• Privacy Policy
• Legal Notice
• Accessibility
• [66.249.64.20|81.177.182.159]
• 81.177.182.159

Character limit 500 /500

Study Finds That 52 Percent of ChatGPT Answers to Programming Questions Are Wrong

Ah yes. and yet..., not so smart.

In recent years, computer programmers have flocked to chatbots like OpenAI's ChatGPT to help them code, dealing a blow to places like Stack Overflow, which had to lay off nearly 30 percent of its staff last year.

The only problem? A team of researchers from Purdue University presented research  this month at the Computer-Human Interaction conference that shows that 52 percent of programming answers generated by ChatGPT are incorrect.

That's a staggeringly large proportion for a program that people are relying on to be accurate and precise, underlining what other end users like writers and teachers are experiencing: AI platforms like ChatGPT often hallucinate totally incorrectly answers out of thin air .

For the study, the researchers looked over 517 questions in Stack Overflow and analyzed ChatGPT's attempt to answer them.

"We found that 52 percent of ChatGPT answers contain misinformation, 77 percent of the answers are more verbose than human answers, and 78 percent of the answers suffer from different degrees of inconsistency to human answers," they wrote.

Robot vs Human

The team also performed a linguistic analysis of 2,000 randomly selected ChatGPT answers and found they were "more formal and analytical" while portraying "less negative sentiment" — the sort of bland and cheery tone AI tends to produce.

What's especially troubling is that many human programmers seem to prefer the ChatGPT answers. The Purdue researchers polled 12 programmers — admittedly a small sample size — and found they preferred ChatGPT at a rate of 35 percent and didn't catch AI-generated mistakes at 39 percent.

Why is this happening? It might just be that ChatGPT is more polite than people online.

"The follow-up semi-structured interviews revealed that the polite language, articulated and text-book style answers, and comprehensiveness are some of the main reasons that made ChatGPT answers look more convincing, so the participants lowered their guard and overlooked some misinformation in ChatGPT answers," the researchers wrote.

The study demonstrates that ChatGPT still has major flaws — but that's cold comfort to people laid off from Stack Overflow or programmers who have to fix AI-generated mistakes in code.

More on OpenAI: Machine Learning Researcher Links OpenAI to Drug-Fueled Sex Parties

Share This Article