Survey Fatigue: Data killer

After experiencing a certain number of questions a respondent will grow tired of answering questions. The number of questions will vary depending on the complexity of the questions, the repetitiveness of those questions, and the topic of the survey. You have to use your judgment to decide if the survey you are creating could fatigue the respondent.

Issues from Fatigue

Abandonment - A respondent could simply close the survey without completing it never return again. In this case, we might get some partial data, but this is certainly not the goal.

Bad data - A fatigued respondent could start answering your questions randomly with the goal of completing the survey as fast as possible just to get it over. This is a much worst outcome than abandonment because it's nearly impossible to tell when this has happened. This bad data can skew your analysis.

Reducing Fatigue

To help mitigate fatigue consider the number of clicks a question requires to be completed, not just near the end of the survey but throughout the entire survey. For example, if you ask someone their age as a Number input question it takes 1 click to enter the number field then 2 clicks to type in the number (unless they are a child or a centenarian). If your study does not need a precise age it's easier on the respondent to ask age range as a multiple choice. This setup only takes 1 click and they can move on to the next question.

Dropdown menus require at least two clicks to complete, 1 click to open the menu, and at least 1 to make your selection. If a question can be set up with radio buttons instead of dropdowns it is less fatiguing

Long pick lists where the respondent has to scroll through, perhaps when indicating awareness, can be an issue especially for respondents taking the survey on a mobile device. Look for ways to limit the length of pick lists.

Loop iterations, the best practice to limit loop iterations to 3-5 beyond these the respondent will grow tired of the repetition and we risk fatigue.