1994 Volume 8 Issue 2 Pages 95-108
FACES Ⅲ was developed to measure the Circumplex's central dimensions based on curvilinear hypothesis. But a study of FACES Ⅲ original version by Green et al. (1991) failed to confirm the hypothesized curvilinear relationship.
The present study investigates reliability of the FACES Ⅲ Japanese version, and concurrent validity of this instrument and the curvilinear hypothesis of family functionings with a sumpie of college students’ families (N=190).
The lack of support for curvilinear hypothesis is explained by following findings for the two FACES Ⅲ subscales. The factoral analytic structure of the adaptability subscale was unstable, but the cohesion subscale had high internal consistency and similar construction to the original cohesion items. The results compared the family types that comprised each of three major family categories on the family health scale and evaluation of family relations (satisfaction, troublesome, and interest) did not support the curvilinear hypothesis. Better-functioning families did not cluster in the center of the Circumplex. Rather scores on each of the family health scale and evaluation of family relations were more positive for family types on the right side than on the left side of the Circumplex. There was a linear relationship between family health scale and evaluation of family relations and the cohesion subscale. Implications of these findings for the use of the FACESⅢ Japanese version and for the Olson's Circumplex Model were discussed.