Yesterday, my wife was completing an online survey when she came across a question that asked:
[How far do you agree/disagree with] Name-of-company makes me feel that my custom is greatly valued
(wording might not be exact). Originally, she answered Disagree, clicked Continue and then a follow-up verbatim question was displayed that asked:
We are sorry to hear [...] – why did you answer in that way?
At which point she clicked the ‘Previous’ button, changed her answer to Neither agree/nor disagree hit Continue (no follow up question) and carried merrily on her way mentioned it to me; so I told her change the answer back, and tell them what you mean. Afterall, the response that was eventually entered was only:
You don’t really know who I am apart from what I buy.
When she first saw the question she was thinking it would take more than a few moments to think about and put this into words (although in reality it didn’t) but it demonstrates one reason why verbatim/open-ended questions should be used sparingly.
This particular survey had an incentive (win an iPod/cash) and that perhaps explains this user’s decision to click back and change their earlier answer as they could see this was a longer route to entering the survey’s prize draw.
It made me wonder how normal is this behaviour though? And how often do respondents change their minds about how they have answered something and go back and edit it? I shall try to investigate…
Meanwhile, consider how asking open-ended or verbatim questions affect how the respondent regards your survey.
Dan Wardle
