[BlindResearch] Strategies for avoiding fraudulent responses to online questionnaires

[Arielle Silverman]

Hi all,

I’m curious if any researchers here have experience with this issue. Over
the past few years, online questionnaires have become infested with large
numbers of fraudulent responses. It appears to be a combination of bots and
human beings who pretend to be legitimate participants. In my experience,
the problem occurs only when compensation is offered in exchange for
participating in a study. This problem is very time-consuming to counteract
and it can compromise data integrity.
Question: Do any of you know whether the problem occurs when a
questionnaire is posted within a private group on social media (Facebook,
etc.) or does it only strike when the link to a questionnaire becomes
publicly available?
I believe distributing a questionnaire via email can circumvent the
problem. But we may want to recruit from closed Facebook groups, too. I
also worry about posting to a publicly archived list like on NFBNet.

I'd love any other advice for either preventing the fraud or efficiently
weeding out fraudulent responses, too. Even the automated survey software
features are not reliable (Qualtrics has a fraud detection feature that
suffers both false-negative and false-positive issues when comparing its
work to that of human beings looking intelligently at responses).
Our most effective strategy has been to put a required open-ended question
in that asks about some esoteric blindness issue (we survey only blind
participants) and then deleting anyone whose answers don't make sense in a
blindness context. The fake answers to this question can be wildly
entertaining. But we still occasionally get responses that aren't 100%
distinguishable as real or fake.
Thanks everyone!



*Arielle Silverman, Ph.D.*

Director of Research

American Foundation for the Blind

Direct Phone: 202-469-6832

Email: asilverman at afb.org



Schedule a call with me <https://calendar.app.google/Gbrte5BFxZ2HLpEb7>



*I currently work Monday-Friday from 9:00 a.m. to 5:00 p.m. Eastern time.*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://nfbnet.org/pipermail/blindresearch_nfbnet.org/attachments/20230921/977bd1ac/attachment.html>