Fakability of an empirically derived selection instrument

Fakability of an empirically derived selection instrument developed by criterion-keying for a specific occupation and organization was investigated. Ninety-four subjects completed the test under both Honest and Fake instructions. Instructions to simulate a highly motivated job applicant produced significant score increases on 7 of 10 scales and significant changes in narrative statements of predicted job performance based on these scores. Theoretical and practical implications are discussed. The caveat to use tests only when the subject is motivated to answer honestly is repeated.