By Tsung-Hsun Tsai
In previous studies of gender differences, boys have been found to score significantly higher than girls on mathematics achievement tests. The EXPLORE test, used to test educational development of 8th graders, is used in the Belin Elementar y Student Talent Search (BESTS) as an above-level test for students in grades 3–6. The EXPLORE consists of several subtests, including a 30-item mathematics test. The results (by gender) of BESTS students on this mathematics test, indicate that boys per form statistically higher than girls, which is consistent with other studies on tests of mathematics achievement.
The EXPLORE mathematics test is comprised of four skill areas: pre-algebra, elementary algebra, geometry, and probability/statistics. The purpose of my study was to analyze these four mathematics skill areas by gender.
The study consisted of the test scores of 6,473 BESTS students from 1993 and 1994. The sample consisted of: 815 third graders (427 males, 388 females); 1,859 fourth graders (973 males, 886 females); 1,938 fifth graders (982 males, 956 females); and 1, 861 sixth graders (1,004 males, 857 females).
The responses to the four skill areas were analyzed by gender and grade level using the PRELIS2 and LISREL program.
The results indicated that there were no statistical differences among the four skill areas. The factor pattern and the factor loadings with the four skill areas were similar for boys and girls (Table 1). Additionally, no statistical gender differences were found in any of the skill areas (Table 2). While no statistical differences were found, boys had higher average scores on the elementary algebra and geometry skill areas, and girls had higher average scores on the pre-algebra and probab ility/ statistics skill areas (Table 3).
An important outcome of this study is to note that factor patterns and factor loadings are similar for males and females on the EXPLORE mathematics test. Thus my results support those reported in the study by Colangelo, Assouline, and Lu (1994) where they suggested that the EXPLORE is an appropriate measure for BESTS participants.
Table 1:
Summary table of significance indices for three models
|
| Model |
c2 |
df |
p-value |
RMS |
GFI |
| All |
7063.23 |
405 |
0 |
0.044 |
0.93 |
| PS |
9062.28 |
810 |
0 |
0.048 |
0.92 |
| IN |
9032.17 |
834 |
0 |
0.047 |
0.92 |
Note:
All: Testing of Four Common Factors Model Fit on all BESTS participants;
PS: Testing of Equality of Common-factor Patterns for males and females;
IN: Testing of Equality of Factorial Invariance for males and females;
RMS: Root Mean Square residual;
GFI: Goodness of Fit Index
|
Table 2:
ANOVA summary table (gender effects) on each skill area
|
| Skill Areas |
df |
MS |
F |
p-value |
| Pre-Algebra |
1 |
0.300 |
0.034 |
0.866 |
| Elementary Algebra |
1 |
0.833 |
0.190 |
0.692 |
| Geometry |
1 |
4.800 |
7.200 |
0.075 |
| Probability/Statistics |
1 |
0.133 |
0.033 |
0.870 |
Table 3:
Mean EXPLORE mathematics test scores earned on skill areas by males and females
|
| Skill Areas |
Male
(n=3,386) |
Female
(n=3,087) |
| Pre-Algebra |
7.00
(10 items) |
7.50
(10 items) |
| Elementary Algebra |
3.33
(9 items) |
2.50
(9 items) |
| Geometry |
4.00
(7 items) |
2.00
(7 items) |
| Probability/Statistics |
2.67
(4 items) |
3.00
(4 items) |
Back to Belin-Blank Center Research