High data quality is vital for the success of any market research project. We know that everybody has a different approach to measuring bad data quality. While some may purely focus on completing the interview, others concentrate on answer patterns and attention checks. Nevertheless, using a single method is never enough.
Over the years, there has been a shift in how survey data is collected, and online panelists have given rise to panel fraudsters. These fraudsters have evolved, but panel companies have found new ways to arrest unhealthy responses. Panel providers are at constant war with cheaters, speeders, and bots, and both parties find ways to outsmart each other all the time.
Trends that will improve online panel data quality in 2022
Combating bogus respondents and responses is a crucial step to maintaining the integrity of online panel providers. Various methods to spot such respondents are identified and continuously developed, but applying a mix brings out the best results. Let’s look at the different techniques used to eliminate bad respondents and improve panel health and data panel data quality.
Catching fraudsters in the act
One effective method of identifying genuine panelists from the fakes is by asking them a question or a series of questions that are impossible to answer. For example, you can ask the question – Can you name the pet dogs of every NFL player that has ever played? Yes/No. Respondents that answer ‘Yes’ are very likely the fake ones.
Those who answer surveys purely for monetary benefit will not read questions and other texts attentively. Identifying and eliminating them has already become standard practice in many research areas. For example, we asked respondents to read a paragraph and answer questions in a survey. The text contained 152 words, which approximately takes 35 seconds to read. Unfortunately, 28% of respondents clicked through the section in under 10 seconds; some even clicked through under five seconds.
Speeders, too, like skimmers, select options at random and complete the survey within minutes. A 10-minute survey is completed in considerably less time than usual. We can identify how long the respondent has stayed on each page to answer multiple questions or look at the total time to complete the survey. The ones flagged can automatically or manually be taken off.
Straight-liners are the ones that blindly select the same answer option for every question. They can be identified when choosing identical or near similar answer choices. This pattern can be easily recognized and flagged for elimination.
There are multiple questions to use when it comes to checking respondents’ attention. Usually, attention check questions have obvious correct answers and are embedded data at the beginning of the survey. These questions help identify careless panel respondents initially and help distinguish the high-quality responses from the low-quality, unreliable ones. Red herring and manipulation checks, too, are good indicators of inadequate responses.
Panel data quality with QuestionPro Audience
QuestionPro offers the highest quality survey panels. We screen millions of online panelists and ensure that all profile data points are updated. Controlling the quality of panelists ensures that unhealthy respondents are eliminated before they enter the project. We also have the mechanisms above to arrest survey fraud and provide our customers with the highest quality data. Try QuestionPro Audience for a hassle-free data collection experience.