The convergent validity between self and observer ratings of personality: A meta‐analytic review
JJ Connolly, EJ Kavanagh… - International Journal of …, 2007 - Wiley Online Library
JJ Connolly, EJ Kavanagh, C Viswesvaran
International Journal of Selection and Assessment, 2007•Wiley Online LibraryThe convergent validity between self and observer ratings of the Big Five dimensions of
personality was examined by cumulating research findings across studies. The mean
correlation corrected for coefficient α in self‐ratings and inter‐rater reliability in observer
ratings was. 46 for agreeableness (N= 6359, k= 53),. 56 for conscientiousness (N= 6754, k=
58),. 51 for emotional stability (N= 8000, k= 55),. 62 for extraversion (N= 7725, k= 50), and.
59 for openness to experience (N= 5333, k= 38). Results indicate that, although there is a …
personality was examined by cumulating research findings across studies. The mean
correlation corrected for coefficient α in self‐ratings and inter‐rater reliability in observer
ratings was. 46 for agreeableness (N= 6359, k= 53),. 56 for conscientiousness (N= 6754, k=
58),. 51 for emotional stability (N= 8000, k= 55),. 62 for extraversion (N= 7725, k= 50), and.
59 for openness to experience (N= 5333, k= 38). Results indicate that, although there is a …
The convergent validity between self and observer ratings of the Big Five dimensions of personality was examined by cumulating research findings across studies. The mean correlation corrected for coefficient α in self‐ratings and inter‐rater reliability in observer ratings was .46 for agreeableness (N=6359, k=53), .56 for conscientiousness (N=6754, k=58), .51 for emotional stability (N=8000, k=55), .62 for extraversion (N=7725, k=50), and .59 for openness to experience (N=5333, k=38). Results indicate that, although there is a high degree of construct overlap, both self and observer ratings have substantial unique variance. Moderator effects were analyzed. The duration of acquaintance (strangers vs close relatives) as well as observer type (peers at work vs relatives) were analyzed. Acquaintanceship had a large moderating effect whereas observer type did not moderate the level of convergence.