Beyond Mars and Venus: An Examination of Gender Differences in Academic Performance of High School Students - Innovative Management in Information and Production - page 34

Tabl e 6 Common metric completely standardized solution for

model 4

English Math

Male Female Male Female

USE 0.86 0.83

RHET 0.91 0.95

ALG 0.90 0.90

CGEOM 0.83 0.86

TRG 0.80 0.83

Tabl e 7 Estimates of unique variance for males and females

USE RHET ALG CGEOM TRG GPA

Male 0.23 0.18 0.19 0.27 0.30 0.51

Female 0.19 0.21 0.21 0.32 0.34 0.61

Tabl e 8 Covariance between math, reading, and general ability

English Math General

Male Female Male Female Male Female

Math 1.00 0.99

English 0.82 0.78 1.00 0.99

General 0.89 0.85 0.92 0.91 1.03 1.02

groups have similar covariance matrices. The test of hypothesis B shows that the
invariance of factor patterns does not hold very well because RMSEA, GFI, and
NFI show poor fit. There seems to exist some degree of difference for two groups in
terms of factor structures. Estimates of pattern coefficients, unique variance, and
the covariance of solutions standardized to a common metric were examined to
ascertain the sources of differences. Table 6 shows the result of the standardized
solutions for the factor pattern coefficients of males and females. After comparing
these coefficients with males and females, it seems that the relationship between
latent variables and their indicators was similar for the two groups.
Tab le 7 shows the estimates of unique variances for the two groups. The errors of
measurement factors were larger for females than for males in all cases except for
the English subtest of grammar/usage. Unique variances are high for GPA indicator
for both groups; the possible reason could be that the scale for GPA is different
from the other variables. Finally, the covariances between the latent variables were
examined, and the results are shown in Table 8 . This table shows that the factor
variance and covariance are larger for the males than for females.