B. Marcus et al., INTEGRITY TESTS AS A SPECIFIC GROUP OF IN STRUMENTS IN PERSONNEL-SELECTION - A LITERATURE-REVIEW AND METAANALYTIC FINDINGS ON CONSTRUCT-VALIDITY, Zeitschrift fur Arbeits- und Organisationspsychologie, 41(1), 1997, pp. 2-17
During the last two decades, attempts to assess the trustworthiness of
job applicants via standardized questionnaires led to the development
of numerous so-called ''integrity tests'' in the USA. While there is
substantive meta-analytic evidence for their criterion-related validit
y, using narrowly or broadly defined, ''honesty'' as well as overall j
ob performance as criteria, the findings on their construct validity a
re much fewer in number and less conclusive in results. To address thi
s issue, a meta-analysis of correlations between integrity tests and m
ainstream personality questionnaires was conducted, using the five-fac
tor-model of personality as a base for a nomological net. Depending on
the factor in question, between 10 and 14 coefficients, representing
a total N of 1.636 and 2.236, respectively were entered into the analy
sis. Integrity was moderately correlated with three of the ''big five'
'-factors (emotional stability agreeableness, and conscientiousness; w
ith rho about .30, respectively), while for both the others no substan
tial relationship was found. Moderator analysis showed generally highe
r coefficients for the trait-based tests compared to the attitudinal m
easures of integrity.