Several years ago a journal article reported on a mailing of hundreds of phony job-application resumés to potential employers. Conspicuously African-American-sounding names were assigned to some of the phony applicants. The researchers found a statistically significant degree of support for the differential response that they had conjectured.
Medical researchers convinced psychiatric hospitals to admit them as patients requiring treatment. Their purpose was to test how hard it was to convince physicians that these patients were sane, after all, and so gain release. In one twist, to see how admission procedures would be affected, one hospital was told, untruthfully, that fake patients would be sent its way (Sam Harris, The Moral Landscape, 141–142).
Research reported in NBER Digest, March 2013, involved sending about 12,000 phony resumés to employers who had posted some 3,000 job vacancies. The resumés showed how long a supposed applicant, if unemployed, had been unemployed. Statistics on “call-backs” from the employers supposedly confirmed discrimination against the long-term unemployed.
Such research raises several questions. Might not some of the employers (or hospitals) subjected to these experiments have vaguely sensed something peculiar and have responded or not responded accordingly? Is it fair to force the unagreed status of experimental guinea pig onto employers, wasting their time and imposing costs, all in addition to their ordinary burdens? Most important, is lying a respectable tool of research? Should academics profit from having their own resumés augmented by such deceptions?
