Computer versus paper-and-pencil administration mode and response distortion in noncognitive selection tests

This study examined the strength of the relationship between computerized and paper-and-pencil noncognitive selection measures. The tendency to engage in response distortion across these 2 modes was also assessed. The raw cross-mode correlations exceeded .90 for all of the substantive selection scales studied. Indeed, latent cross-mode correlations were estimated to be 1.0 for all of these scales. Mean score comparisons of 1 of the response distortion measures (i.e., the Balanced Inventory of Desirable Responding) suggested that response distortion in the computer-administration mode was significantly greater than response distortion observed with a paper-and-pencil format. However, administration mode did not influence scores on another measure of direct response distortion or any of the substantive selection measures administered.