nvestigating the Generalizability and Dependability Coefficients of Admission Interview Scores of Ph.D. in Psychology

The scores obtained on the interviews can be the basis for decision making provided that they are embedded with low measurement error. The purpose of this study was to the estimate the generalizability and dependability coefficients of the admission interview scores of Ph.D. at Shahid Beheshti University. To do this end, interview scores of 27 applicants in the field of psychology at Shahid Beheshti University were analyzed using a two-dimensional cross-sectional design. The real variance, which was related to the “purpose of measurement”, accounted for about one-third of variance of observed score.
Besides, unsystematic measurement error and systematic measurement error accounted for about one-fifth and half of the variance of the observed score, respectively. The findings showed that the generalizability and dependability coefficients of the interview scores were acceptable and they were close to the coefficients estimated for the medical fields. However, the interaction between interviewers and applicants had the greatest contribution to the systematic measurement error, that can lower the credibility of the scores.
According to the results, it is suggested that a workshop is organized for psychological training groups so that they can carry out reliability analyses to obtain a better understanding of the quality of the interview scores and to determine which changes in the concerned dimensions can lower the systematic error and boost the reliability of the interview scores.