When designing your survey, plan the order and flow of your survey questions
An important aspect of successfully writing surveys is knowing in what order to ask your questions. Random placement of survey questions may have a negative impact on the quality of survey data you collect from respondents.
If your survey needs to be completed by a particular target audience, ensure that screener questions (also referred to as qualifier questions) are placed at the beginning of the survey. Do not make respondents, who do not qualify to complete your survey, complete any more work than they need to. Don’t upset your respondents, especially if they are customers. It is possible that you may need to contact unqualified respondents in the future to partake in additional research surveys.
If you are planning to create a long survey – one that is greater than 5 minutes long – place the most important or most difficult questions to answer at the beginning of the survey. Respondents tend to lose interest rather quickly, especially when a survey exceeds 5 minutes. Keep in mind, data quality and survey completion diminishes greatly the longer you make your survey.
As mentioned in a previous post, it is important for survey questions to be free from bias response options. A respondent’s answer may be influenced by where the response option appears in the ordering of question response options. The same concept is true for question ordering. A respondent’s answer may be influenced by where in the survey the question is asked. If you are looking for honest, straightforward feedback in open-ended questions (comment questions), it is recommended to place those questions toward the beginning of the survey, before new topics are introduced, which could potentially influence a respondents’ answers.
To keep respondents interested and their thoughts focused, group similar topic questions. It is easier for respondents to focus on one topic and then move on to the next. Grouping questions with similar topics may draw out more information in open-ended questions and may help respondents differentiate between topics in rating scale questions.
