Similarities between regression analysis and analysis of variance are noted and it is argued that the former has advantages over the latter. It is also argued that canonical correlation analysis is more suitable than factor analysis in certain cases. The argument is illustrated with four recent pieces of educational research. (DG)

EEG correlates of intellectual functioning in academically handicapped and talented adolescents were investigated. Data from Ertl's Brainwave Analyzer were factor analyzed and used to predict verbal, numerical, reasoning, and spatial abilities and reading and mathematics achievement. Correlational patterns differed by group. Implications for…

The psychometric characteristics of the revised UCLA Loneliness Scale in Iranian groups correlated significantly with measures of other conceptually-related variables such as depression, anxiety, and neuroticism. Factorial structure supported the multidimensionality of the scale. (Author/PN)

Examines the factor structure of the Survey of Work Values for a sample of White South African males. States that number of factors and item composition of factors differed from results of previous research due to cultural differences between South African and American samples. (CMG)

The 10 subtests comprising the Wechsler Intelligence Scale for Children--Revised (WISC-R) were independently factor analyzed for middle (N=107) and low (N=158) socioeconomic status children. After testing for equivalence of covariance matrices between the two groups, coefficients of congruence between factors were calculated. These coefficients…

Investigated sex differences on the Adult Nowicki-Strickland Internal-External Control Scale using factor analysis of student scores (N=194). Three factors emerged for females and five for males. Comparison with other studies shows item consistency on the first two factors which were the same for both sexes. (Author/JAC)

This study (1) provided evidence concerning the construct validity of scales based on the clusters found in a revised, professional-level form of the California Occupational Preference System Interest Inventory and (2) examined the degree of internal consistency of the 16 newly devised subscales. (Author/PN)

The present study compared the factor structures obtained when the same data were factor analyzed: (1) for the total sample (463 fifth grade students) and by sex for collapsed concepts, and (2) by individual concept for the total sample by sex, and for internal and external locus of control groups. (Author/PN)

Used a modified multitrait-multimethod matrix approach to examine the construct validity of the Career Decision Scale for 175 male college students. The results of the investigation provided evidence supporting the construct validity of the Career Decision Scale. (Author/JAC)

A factor analysis of responses from 44 female teachers to the Bem Sex-Role Inventory yielded two factors, Masculinity and Femininity accounting for 20.7 and 16.2 percent, respectively, of the common variance. Twenty-eight of 40 items performed as hypothesized while 12 items were not associated with masculine or feminine roles. (Author/RD)