How To Design A Good Questionnaire
Updated: Aug 26
Many people tend to have the perception that it’s really easy to write a questionnaire until they actually try to then have difficulty in structuring the questionnaire, when they had only a few questions in mind or don’t know where to start. There is an art in knowing how to design a good questionnaire but it is more straight forward, once you know how. This post will run through the steps with questionnaire examples to help you along the way.
Table of contents:
What is a questionnaire?
4 key parts in designing a good research questionnaire (organising the questionnaire layout), different types of questions to use, 3 types of questionnaire structures to use, top 10 common mistakes to avoid when writing a questionnaire, pre-test questionnaire, remember the main benefits of a questionnaire.
[Disclosure: This post contains affiliate links, meaning we get a commission if you decide to make a purchase through these links at no additional cost to you.]
The table above shows the top 10 tips for how to write a good questionnaire design.
A questionnaire is a data collection technique consisting of a series of questions that are structured to be asked of people taking part using different forms of communication whether it’s verbal or written. The feedback from a questionnaire can be measurable or used to delve deeper into people’s minds.
There are 4 key parts in knowing how to write a good research questionnaire:
1. A short introduction at the beginning stating the purpose of the questionnaire
At the beginning of the questionnaire, there should be an introduction that is short, to the point in stating the purpose of the survey and sound appealing to entice people to take part. Ensure legitimacy is also included and their answers will not be shared to outside parties.
2. Screening section to filter through the target audience of the research
The first questions should be part of the screening section to ensure you are interviewing your target audience by setting criteria specific questions, which maybe a couple of demographic questions if relevant as well as topic specific. Otherwise you don’t want to be interviewing someone for car buying if they don’t drive or are not looking to buy a car in the next 12 months. Keep in mind, the screening section should not be too long.
3. Main body of the questionnaire will be different types of questions of interest to help achieve the market research objectives that have been outlined
The main body of the questionnaire should follow the screening section with all the questions you’re interested in asking your target audience by using many of the question types mentioned previously. It’s best to do a mix but not too much to keep it consistent as well as engaging for the people your interviewing. This section will help to answer the objectives you have set in writing the questionnaire.
Remember if you are using list for multiple choice or scale questions, don’t go crazy and make it too long, otherwise a lot people will drop out of the interview . For a list of behavioural or attitudinal statements it’s best to use a maximum of 10 statements. You have to see it from the participant’s perspective, would you get bored of answering endless lists of statements?
It’s good to get creative with the questionnaire to keep up engagement levels and receive better responses.
4. The final part of the questionnaire tends to be for classification purposes of the target audience such as demographic questions
The final part of the questionnaire tends to be for classification purposes and asking any sensitive questions like income. Normally it includes the rest of the demographic questions you have not asked at the screening section. Don’t forget the closing of the questionnaire survey in thanking the participants for their time and the feedback they have given will be kept confidential.
There are two main question types that you need to be aware of – open ended questions and closed ended questions , where a number of sub type of questions fall under. These questions serve three main purposes behavioural, attitudinal and classification.
Open ended questions
Open ended questions give participants the opportunity to answer the question in their own words, where they can express the reasons why they bought a certain brand or what they don’t like about an advert rather than being prompted to select an answer. The open responses are either recorded by the interviewer or written down/typed by the participant depending on the method used (telephone, face to face, video call or online).
This can add a lot of time to the interview especially if more than two open questions are used in a quantitative survey that will affect whether the questionnaire gets completed or not by many of the participants.
The other thing to keep in mind is it will generate a lot of verbatim comments to analyse, so additional tasks of grouping these comments into the most popular themes let alone taking a further step in coding responses will take a lot time to do. However, verbatim comments can bring results to life through snippets to help back up certain measurable findings.
EXAMPLE – Q. Please tell us your first impressions of this ad. What did you make of it?
Closed ended questions
Closed questions give participants a choice of options to choose from like a single response with a yes or no answer to multiple responses to a question like mobile brands you are aware of from a list.
Below are a number of types of closed questions you can use in your questionnaire.
Single response
Single response as the name suggests is where only one answer is allowed like what gender they are. Even many questions covering attitudes of the participant are single response from a pre-determined list.
EXAMPLE - Q. Have you seen this ad before?
Multiple choice questions
Multiple choice questions allow participants to offer more than one answer. This could be from a list of brands that they may be aware of, consider, bought or used and is easy for people to fill out. If your using a list of answer options, please make sure the list is randomised to avoid bias.
EXAMPLE – Q. Thinking now about all financial products and services, which of these brands would you be prepared to consider for ANY financial products or services? RANDOMISE LIST
Select all that apply
Rating scale questions
Rating scale questions are a scale of numbers that tend to be from 1 to 10 and are used to ask participants to rate a product or service on a number of things. These types of questions are normally used for customer satisfaction .
EXAMPLE – Q. On a scale of 1 to 10 with 1 being not at all to 10 being extremely likely, how likely are you to recommend Brand X to friends or family?
Likert scale questions
Likert scale questions helps to determine the level of agreement from the participant for each statement on a 5 or 7 point scale about a brand or the participants behaviour. Normally this is “strongly agree” to “strongly disagree” but can also be the level of likelihood to buy as well.
EXAMPLE – Q. To what extent do you agree or disagree with the following statements?
I enjoy entertaining people at home
1. Strongly agree
2. Slightly agree
3. Neither agree nor disagree
4. Slightly disagree
5. Strongly disagree
Ranking questions
Ranking questions are needed to find out what aspects participants rate in order of importance for a website, product or service. It’s best practice to ask the participants their top 3 factors in order of first, second and third from a list of factors. The list of factors should be randomised for every interview to avoid any bias towards the factors at the top of the list.
EXAMPLE – Q. You will now be shown a list of factors that could be important to you when selecting a car to buy. Would you look at the list and tell me which is the most important factor in influencing your choice? RANK JUST 3 FACTORS
RANDOMISE LIST
And what would be the second most important factor?
And what would be the third most important factor?
Rank Factors
Competitive price
Model of the car
There are three types of questionnaires that meet the needs of how a questionnaire is administered, whether this is via telephone, face to face or online:
1. Structured questionnaire
This can be used for large surveys (from 30 to 200+) consisting of closed (like yes or no answers) and prompted questions. This is where you add all the possible answer options that a participant can give to your questionnaire. Can be used over the phone, video calls and online for quantitative research (measurable) .
2. Semi-structured questionnaire
This is a mix of open and closed questions, where like a structured questionnaire, you can get measurable results but it also gives the participant the opportunity to express themselves at certain points with open answers such as the reasons why they like brand X.
3. Unstructured questionnaire
This is more of an open way of allowing participants to express themselves through open questions that are not so rigid like the other two types of questionnaires but there is some order to it. The interviewer has the flexibility to probe to explore different lines of questioning around the subject area. This is used for qualitative research in focus groups or depth interviews.
In knowing how to make a questionnaire for research, you have to understand the common mistakes to avoid making them yourselves.
The following is a checklist of the top 10 mistakes to avoid when writing a questionnaire:
Mistake 1: Using two questions in one
Do not include more than one idea or question as this will cause confusion if you have multiple questions in one. Questions need to be short and not overcomplicated.
Mistake 2: Asking leading questions
Questions should not be leading the participants to give a certain answer. For example, excluding “poor” or “very poor” from the scale when asking customers to rate their recent experience of company B.
Mistake 3: Asking vague questions
There should not be any vagueness in the questions you use and need to be very specific as the words “frequently” or “usually” are not specific for a time period. Instead you should use words like “daily”, “weekly” and “monthly”. You can be even more specific by using numbers when asking questions like how often they go to the supermarket in a month.
Mistake 4: Using technical jargon, abbreviations and uncommon words
It’s important not to include jargon and acronyms unless they are used as everyday terms. If you have to include a technical term in the questionnaire, it’s best to include the definition of that term before the question.
Mistake 5: Including words that could be misheard
This is very relevant if the questionnaire is to be used over the phone, face to face or video calls. For example, if you are asking the question ‘what do you think are the differences between the sects?’ This could lead to interesting or amusing responses that are irrelevant to the actual question being asked.
Mistake 6: Asking questions with a negative in them
Understanding of the question will be more difficult if they are asked in a negative way such as “Do you never…?” when it should be “Do you ever…?”.
Mistake 7: Not applying bands for numeric data
This is really good for sensitive type of questions such as income, where bands can be applied for different levels of income like less than $100,000, $100,000-$200,000 and so on. The order of these bands should be sequential and don’t overlap.
Mistake 8: Not keeping in mind the frame of reference for the respondents
There is no point asking them how much milk they consumed in the last 6 months, when it’s better to ask them how many pints of milk they buy in a typical week as it will be hard for the participants to remember the true amount over a lengthy period of time. Besides you can make a simple calculation for a specific period of time from the feedback you get.
Mistake 9: Not including Other (specify) when you are presenting a list of options
Need to include other specify option when using list of options in a questionnaire, as you may not cover everything off such as a list of brands you buy from a supermarket.
Mistake 10: Asking sensitive type of questions at the start of the questionnaire
By having these questions like income or number of people in household at the back of the questionnaire, you have already developed some trust early on to complete the questionnaire rather than the other way round. Remember this is only if they are not an essential part of the screening process at the start of the questionnaire.
Once a questionnaire is drafted it's best to pre-test the questionnaire with colleagues, family or friends. A pre test uses a small number of respondents to help check if a questionnaire reads well, is easy to follow, see if there are any errors to fix, test the routing has been applied correctly and ensuring the survey length is not too long.
If your doing an online survey then it should not take more than 10 to 15 minutes to complete before people start dropping out.
Once all the relevant checks are made and any issues have been fixed then you are ready to go and launch your survey.
The main purpose of a questionnaire is obviously to get information out of participants but there are other benefits of a questionnaire, which are:
Serves as an aid for the interviewer without having to remember the questions.
To provide consistency in how all interviews are conducted.
Avoids having the interviewer using their own discretion throughout in making up questions.
Allows results to be recorded in a consistent way in order for results to be analysed easily later on.
Can be inexpensive when online and mobile survey methods are used.
Survey questionnaire allows you to collect information from a large audience.
An easy way to interview your target audience is by screening participants early on who meet your set criteria.
Allows comparability when questionnaire results are measurable and the questionnaire is asked of participants every quarter or yearly.
If you are considering DLY online survey then survey platforms like 123FormBuilder.com are great to use as they are inexpensive and are really intuitive to use. Plus, you can use them for more than just surveys with all kinds of online forms and templates for apps, tables, stores, approvals and more.
Another alternative if you just want to run a quick poll to compare or rank designs, ideas, logos, book covers or even Amazon sales pages including running short surveys then you should use PickFu, which is straight forward and not expensive to run (starting from $15). You can take advantage of their large online panel of 15 million across countries like the UK, US, Canada, Australia, Germany and Spain to ask questions to.
So give PickFu a try and get 50% off your first poll with the promo code - ANPAR.
Watch the video below for more information.
RELATED HOW TO GUIDE POSTS BELOW FOR SURVEYS AND MARKET RESEARCH
How To Do A Survey In 9 Easy Steps
Market Research Online: Benefits, Methods & Tools
Questionnaire vs Survey: Key Differences You Should Learn
Learn How To Do Market Research Before Starting A Business
How To Create A Poll For Instagram And Twitter
Market Research Online Surveys: Best Practices & Tips
Exploratory Research: Definition & How To Conduct This Research
Quota Sampling: What Is It & How To Do It
NPS Calculation: Learn All You Need To Know
5 Best Survey Maker Platforms To Consider Using
Conversational Forms: Discover What So Good About Them
Top 5 Website Survey Questions About Usability
#goodquestionnairedesign #anparresearch
- HOW TO GUIDES
Related Posts