The predictive utility of behavior-based interviewing compared with traditional interviewing in the selection of radiology residents.

RATIONALE AND OBJECTIVES

This study compares the predictive use of measures based on traditional faculty and resident interviews of residency applicants with measures obtained through behavior-based interviewing. A special emphasis was placed on predicting residents' noncognitive abilities.

METHODS

One hundred fifty-one resident applicants, over a 3-year period, were interviewed using standard interviews by faculty and residents. These residents also were interviewed with an experimental behavior-based accomplishment interview. Four years later, during their diagnostic radiology residency, evaluations of performance were gathered on these applicants from their residency director.

RESULTS

Results indicated that scores based on responses given during the accomplishment interviews added considerable predictive utility to the low prediction demonstrated by traditional interviews.

CONCLUSIONS

These findings imply that improving unstructured faculty and resident interviews to obtain, in a more rigorous manner, desired information about noncognitive abilities may be a key to successful resident selection.