In user-centred design, surveys are widely used to help us understand more about our users. Survey research involves ‘the collection of information from a sample of individuals through their responses to questions’ (Check & Schutt, 2012, p. 160) and when used effectively, surveys can be extremely powerful components in a user researcher’s toolkit.
Surveys are particularly useful during the discovery phase of a project when learning about your users and their context.
How useful a survey can be in answering our research questions depends on a range of important factors, which I will explore in this series of posts.
In Part 1, I will look at the core reasons why surveys can be useful to researchers.
In Part 2, I will highlight some of the scenarios when surveys can be most useful and will also comment on their limitations and suggest when not to use them.
Part 3 will focus on planning and structuring surveys, while Part 4 will focus on what you should consider when writing a questionnaire.
Finally, Part 5 will discuss the importance of piloting and testing a survey.
Why are surveys useful for researchers?
There are many potential advantages to using surveys in user research.
Surveys enable flexibility, both in terms of the method by which they’re administered, and the timeframe or part of the day in which they’re completed. Whereas in-depth interviews usually require the researcher and the participants to be available at the same time, once an online survey has been launched, participants can complete them at whatever time is most convenient for them.
Although surveys most commonly take the form of online questionnaires, you can also administer paper surveys, or complete them face-to-face or over the phone with the respondent. In some circumstances, especially when the people you’re interested in researching are difficult to reach, you might employ a combination of those methods.
When using a mixture of data collection, however, remember that the way a respondent completes their survey might affect the way they answer. While you might expect that a respondent would act in a relatively similar way if they were completing a paper or online survey, when you introduce an interviewer into the mix, you are unwittingly adding a variable that could change the results. It’s better to streamline the survey channel wherever possible.
Surveys can be relatively inexpensive to conduct – or, rather, the cost per respondent can be very low. In many cases, you can feasibly collect responses from thousands of people for a fraction of what it would cost to speak to each person individually. This is particularly true of online surveys, as administering paper surveys by post or employing telephone or face-to-face interviewers will incur additional costs.
When a survey is designed well, completing the survey doesn’t involve a huge amount of effort on the part of the participant, so it’s usually easier to find willing participants to take part in surveys than for methods which might be more onerous or time-consuming for them.
3. Robust, generalisable data
One of the major advantages of working with survey data is that you can quite quickly and cheaply collect large samples of people within your population (the group of people you’re interested in studying). The larger the sample of respondents you have, the lower your margin of error will be. In practice, what that means is that you can be more confident that the findings from your sample of people can be generalised across the entire group of people you are interested in.
When designed well, surveys are good at giving you measurable data. Using surveys allows you to easily attribute numerical values to responses, which allows you to use statistics to analyse that data. This is one of the most powerful aspects of survey data because you can start looking at whether trends are statistically significant – in other words, the probability that your findings could have occurred through completely random chance or whether you can be confident that what you’re seeing is a real finding.
Through various sampling methods, conducting surveys helps you ensure that certain groups within your population are adequately represented. Sampling is a complex subject area itself, but there are loads of different sampling methods you can use to ensure you’re including everybody you want to include.
One way of doing this is through quotas. If we want to ensure we’re getting a large enough sample of individuals who have a certain job or are within a certain age range, or anything else you might be interested in, we can make sure that we include enough of those people so that we will be able to analyse their results in a meaningful way. We can also make sure that we’re getting a nationally representative sample if that’s what’s important for a particular study.
It is good practice to give your participants the opportunity to remain anonymous when they complete a survey. When it comes to asking about certain topics, especially subjects which are in any way sensitive, enabling participants to retain their anonymity increases the chances that they will respond openly without fear of any repercussions for their honesty.
7. Useful recruitment tools for further research
Surveys can be a very useful tool for recruiting participants who are willing to take part in further research. Although, to some extent, a respondent providing their contact details risks compromising their anonymity, it is your responsibility as a researcher to keep their data safe. You should always make it clear to people that they have a choice whether they provide contact details or not, and be transparent about how you plan to use that information.
Up next: When should and shouldn’t we use surveys?
Having explored some of the main reasons why surveys can be useful in user research in part 1 of this series, it is equally important to think about when surveys are most suitable to use. Part 2 will look at some scenarios in which questionnaires are particularly useful and others where a survey alone is unlikely to provide you with all the information you need.