When running a sociological experiment where subjects are randomly assigned to treatment and control groups and only the treatment group will be sent some information regularly (for example, weekly), how can the researcher minimize the risk of contamination of the control group (for example, if someone from the treatment group posts their information on social media or the like where anyone from the control group might see it)?

Is there any way (other than informing both groups of the importance of a control group) to reduce this risk of control group contamination?

  • $\begingroup$ In ideal circumstances, study participants should not know their assignments (blindness). This is not always possible, but perhaps in your study you can send information to all participants? Only sending information to one group creates an additional confound, as the groups differ by both information content and time spent reading it. So for example, if you send bogus information to the control group, then there will be no value in sharing it as subjects won't know which version is real, and with both groups spending an equal amount of time reading, the confound is removed as well. $\endgroup$
    – Arnon Weinberg
    Sep 16, 2020 at 19:19


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.