Effect of Consent Form Language on Survey Response Bias

Autor: Robertson, Graeme, Pop-Eleches, Grigore, Tucker, Jacob R., Rosenfeld, Bryn, DeSisto, Isabelle
Rok vydání: 2023
Předmět:
DOI: 10.17605/osf.io/w5c3e
Popis: The required language to be included in consent forms varies widely by institutional review board and often by study. Despite this, there is little systematic evidence about the effect that consent form language has on survey responses. A long history of survey methodological research has shown that the information respondents are primed with before answering questions can affect their responses (e.g. Strack 1992; Singer 1992). It follows that these effects would extend to the language of the consent form that respondents are asked to agree to before beginning the survey. We plan to conduct a survey experiment in Russia to test the degree to which variance in the risk portion of an informed consent document affects response bias, including non-response bias and sensitivity bias. The primary treatment in the experiment will be the consent form that respondents receive. The control group will receive a standard IRB consent form based on minimal risk. The treatment group will receive a “heightened risk” version of the consent form that contains a paragraph warning them that if their phone has been hacked, their survey responses may be observable. We hypothesize that the inclusion of such text will cause fewer potential respondents to decide to take the survey and, further, that those who do take the survey will be more likely to dissemble about their support for regime policies. Our full list of hypotheses is available in the “Study Information” section.
Databáze: OpenAIRE