Types of Survey Question
There are many different types of survey question, each with their own uses, advantages and even disadvantages. Read our guide to question types to pick so you can deploy the styles that best suit your objectives.
There are many different types of survey question, each with their own uses, advantages and potential shortcomings.
When you create a survey or perform other forms of market research, the questions you ask are of the utmost importance. Open and closed questions can impact the quality of your research, so it’s essential you understand when to ask open-ended questions and how to use them to get the best data. See examples and read advice on how to implement them.
Closed questions collect quantitative data and give the respondent a limited amount of options to choose from. They are popular in surveys, as quantitative data is easier to analyse than qualitative data. There are a few versions of close-ended questions, from dichotomous to multiple choice, and here we look at their advantages and when to use them.
A common feature of any survey, multiple choice questions present respondents with either single select options or multi-select options (where more than one answer can be chosen from a list). Age range would be an example of single select whilst picking from a list of foods might be multi-select, should more than one response apply.
Demographic questions are those that look to categorise the identity of the survey participants based on factors such as their age, gender, ethnicity, marital status, household income, employment, education level and location. Capturing this information allows a better understanding of how respondents fit into the general population. Some demographic questions can be sensitive, however, and so need to be asked with care.
Likert scale questions capture respondents’ opinions – specifically the extent to which they agree or disagree with a statement. Often based on a 5, 7 or even 10-point scale, responses might range from “strongly agree” to “strongly disagree”, or “very satisfied” to “very unsatisfied”. A neutral mid-point choice is often included too, such as “neither agree nor disagree”.
Ranking questions ask respondents to indicate their preferences between a list of research subjects (such as product attributes, packaging designs, or holiday destinations), to identify which are the most and least favoured. As few as two choices might be presented, whilst a neutral/don’t know answer may or may not be included. Ranking questions are useful for finding out what customers want, and work well in conjunction with rating scale questions.
Rating questions are used to enable a comparison of different research items (such as product or service features) using a consistent scale. Participants might be asked to rate a series of choices on a rating scale (say 1-5) where 1 = ‘Not at all important’ and 5 = ‘Very important.’ There are a variety of scales these questions can use such as numbers, frequency (daily to annually, never to always), comparative weighting, stars, ‘smiley’ emojis and more.
Hypothetical questions are those based on an imagined situation rather than facts or actual experience. As such they can be used to capture participants’ opinions and beliefs about something that hasn’t happened, but could. The survey must set the parameters or criteria for the scenario being explored. Hypothetical questions assume certain conditions exist (experience, understanding, opinions) and so can be problematic. Generally avoided, only use them with care.
Dichotomous questions are commonly used to elicit a Yes or No, Agree or Disagree, or True or False response. By offering two possible answers they are a useful way to clarify opinions and understanding, and to direct participants to relevant next questions. But they should also be used with care to avoid frustration or compromised results.
Matrix questions are closed-ended (limited set of answers) and present respondents with a number of questions laid out in rows but using the same selection of column choices. Checkboxes are typically used, with users selecting one option per row, for example a rating scale from “very poor” to “excellent”, with each row using that same scale.
Get started quickly
Use one of our pre-built yet customisable template surveys to get up and running quickly.