Improving sexual orientation data coverage by increasing willing disclosure in recruitment
How to cite
Nicks, Leonie; Davidson, Shoshana; Roy-Chowdhury, Vivek; Hardy, Tim. 2019. "Improving sexual orientation data coverage by increasing willing disclosure in recruitment." AEA RCT Registry. AEARCTR-0004537. https://doi.org/10.1257/rct.4537-3.1
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
Disclosure rates for sexual orientation data are often lower than for other demographic data such as ethnicity. It is important to increase willing disclosure of sexual orientation data to ensure that employers can identify potential biases in their recruitment processes and implement appropriate adjustments to make them fairer. We are testing the impact of messaging on increasing willing disclosure rates of sexual orientation data during recruitment. In a 4-armed randomised controlled trial, applicants starting a job application see either the current form or three alternative simple messages that aim to address different key potential reasons for non-disclosure. These are 1) mistrust in the secure handling of their demographic data, 2) the belief that disclosure would harm the application, and 3) not understanding the purpose of providing such data.
Intervention
We are implementing the trial with an online recruitment platform. The platform asks applicants to provide demographic data as soon as they register to apply for a role. The intervention is different versions of the introductory statements at the beginning of the form. Control is the current version of the form, and the three intervention arms address different key reasons for non-disclosure: 1) mistrust in the secure handling of such data, 2) the belief that disclosure would harm the application, and 3) not understanding the purpose of providing such data.
Experimental design
The trial is a four-armed online randomised controlled trial (RCT) randomised on 25,000 unique individual applicants.
Design details
Cited by
No citation data available for this registration yet.