Editor’s note: Brooke Patton is research and content specialist at market research firm GutCheck, Denver. This is an edited version of a post that originally appeared here under the title, “6 types of survey respondents to watch out for.”
Respondents come in many different shapes and sizes – meaning there are good ones, bad ones and some in between. Since respondents are not one in the same, market researchers have to be aware of the different types and their impact on results. Even if the study is written, executed and analyzed flawlessly, survey research is only as good as its respondents.
1. Professionals
Professionals are one of the few types of survey respondents who are often categorized as good. These survey takers are the model citizens when it comes to being a market research respondent and frequently take studies. And when it comes to survey taking, they treat it as a job. While that’s usually good, sometimes their frequent participation can lead to biased results, particularly if repeating studies on similar subjects. In order to avoid this, screen professionals by asking if they’ve taken a recent study specific to the subject on which you’re conducting research and terminate them if they have.
2. Rule-breakers
Rule-breakers have a difficult time following directions. They may not intentionally be breaking the rules but could be misinterpreting questions. Either way, using quality checks like a screening question that requires a respondent to carefully read the question in order to choose the appropriate answer before moving forward in the study could help. Additionally, remove respondents who prove in open-ends that they are clearly not paying attention.
3. Speeders
Likely the most self-explanatory, speeders move too quickly through a survey to actually provide thoughtful, honest answers. Often, speeders aren’t intrinsically motivated and are only driven to complete a survey in order to receive their incentive. Luckily, speed-related terminates or parameters around the length of time a respondent has to spend on a question or in a study usually removes this type of respondent. Typically all that is required for that is testing and developing an acceptable time range by which to remove speeders.
4. Straightliners
Straightliners, or flatliners, can go one of two ways. For example, overly positive straightliners always or frequently select top box answers like “strongly agree” on Likert scales. On the other hand, negative straightliners always choose bottom box answers, like “strongly disagree.” Whether positive or negative, these types of respondents are just looking to get through a survey or have some sort of unwanted bias that’s causing them to respond in that way, such as acquiescence bias or extreme responding.
5. Cheaters
Cheaters can be both real respondents and fake ones. The fake ones take the form of bots that try to gain access to surveys in order to redeem the rewards without anyone having to do the actual work of taking a study. Real respondents who cheat do so by creating multiple accounts to take the same study more than once or by attempting to take the same study as many times as possible.
Both panels and platforms usually have technology parameters in place to remove bots and check that respondents are not coming from the same IP address more than once. However, reading through open-ends and double-checking contact information can help catch any other cheaters that may have gotten through.
6. Posers
Posers can often be some of the most difficult low-quality respondents to identify. These types of respondents don’t provide honest feedback or choose to follow the group discussions due to social desirability bias. In other words, they don’t provide their true thoughts and feelings on a subject for fear of being different from the crowd or different from what they think the survey provider wants to hear. Posers aren’t necessarily bad, and often only become posers in unique situations or when it comes to specific topics. Making sure studies provide a comfortable, private environment helps to prevent posers from arising.