engleski

Congruence of self- an peer- ratings as a function of the measured construct, type and length of relationship

In this paper we examine the congruence between self- and peer- ratings as a function of the measured construct, and type and length of relationship between the subject and his/her peer. A large sample of students (562), aged from 18 to 28 years used the IPIP50 and LDS to rate themselves (self-ratings) and they were also rated by their parents, relatives, friends or boyfriends/girlfriends (peer-ratings). The International Personality Item Pool (50 items) is a cross-cultural instrument targeted to measure the Big-Five domains, and LDS is an instrument that measures five basic social attitudes factors. We used the classical MTMM approach proposed by Campbell and Fiske (1959) to test the construct validity of subscales of each instrument. The convergent validity of personality scales ranges from 0, 45 to 0, 60, average being 0, 52, while the self-peer correlations between social attitude scales vary between 0, 26 and 0, 76 with the mean value of 0, 46. In both cases the correlations between the same traits measured by different methods (self and peer ratings) substantially exceeds the HTMM and HTHM correlations, providing strong evidence for the construct validity of both instruments. When we divided the whole sample on the basis of the type of the peer relationship, the congruence between self and peer ratings remained high for all groups, but was highest for the boyfriend/girlfriend group. We found no substantial difference in congruence of ratings between longer (more than 3 years) and shorter (less than 3 years) time acquaintances.

MTMM; self-ratings; peer-ratings; IPIP; LDS

nije evidentirano