Dissertation: Physical Education and Academic Performance – Data Analysis
Chapter 4
DATA ANALYSIS
This study was designed to explore the effects of increased physical education on reading and mathematics test scores in third grade students at Tryon Elementary. Reading and mathematics test scores were used to determine the differences between the control and experimental group. Those differences were measured by using a 2 x 2 ANOVA. In this study the entire third grade at Tryon participated in the study, with only one student not receiving parental permission. There were four 3rd grade classes. Two were randomly chosen to be the experimental group and two the control. The subjects had already been pre-assigned to their classes before selection. Data were collected using a pre and post-test in reading and mathematics. All subjects were given the same test, in the same testing environment. Their classroom teacher gave the tests in the regular classroom. The students were told that the results from this study would not affect their grades. The control group consisted of 41 third grade students. Twenty-two were male; while nineteen were female. This was the same as the experimental group, which consisted of 41 students, twenty-two male and nineteen female. Analysis of the Data
Separate repeated measures 2 x 2 ANOVA were used to determine whether there was a significant difference between the pre and post test scores for both the control and experimental groups for reading and math. A pairwise comparison for groups on the reading and math pre- and post-test scores was also conducted to test for significance of interaction between groups.Results
Subjects in both the experimental group and the control group were given a pre-test and post-test for reading. The data for the sample are shown in Table 1.
Table 1
Summary of Means and Standard Deviations for Both Groups for Pre- and Post Reading Tests
Group N Mean SD Std. Error Mean
Pre-Test Control
Pre-Test Experimental 40
41 41.5632
35.4046 18.23106
15.08687 2.88258
2.35617
Post-Test Control
Post-Test Experimental 40
41 51.2095
51.9495 18.38346
15.65312 2.90668
2.44461
Subjects in both the experimental group and the control group were also given a pre- and post-test for math. The data for the sample are displayed in Table 2.
Table 2
Summary of Means and Standard Deviations for Both Groups for Pre- and Post-Tests for Math
Group N Mean SD Std. Error Mean
Pre-Test Control
Pre-Test Experimental 40
41 44.1663
35.0405 18.50159
16.96613 2.92536
2.64966
Post-Test Control
Post-Test Experimental 40
41 57.1665
61.1380 21.47744
22.65192 3.39588
3.53764
The research question was whether there was a significant difference between the experimental and control groups on pre- and post-test scores for reading and math. It was expected that both groups would improve; however, based on current research, it was expected that the experimental group would show more improvement. Data were analyzed using repeated measures ANOVA on two factors, group and time. The data for the reading scores for both groups relative to time (pre and post) are reported in Table 3.
Table 3
Analysis of Variance Summary Between Groups Relative to Time on Pre- and Post-Test Scores for Reading
Source df Mean Square F Sig.
TIMEPER 1 69 55.664 .000
TIMEPER 1 48.786 3.862 .053
Error (TIMEPER) 79 124.757
Table 3 shows that there was a significant difference between time periods in that both groups improved from pre- to post-test (p = .95).
Data for the math pre-and post-tests scores for both groups were also analyzed using repeated measures ANOVA on two factors, group and time. The data for the math scores for both groups relative to time (pre and post) are reported in Table 4.
Table 4
Analysis of Variance Summary Between Groups Relative to Time on Pre- and Post-Test Scores for Math
Source df Mean Square F Sig.
TIMEPER 1 15475.109 90.734 .000
TIMEPER 1 1736.573 10.182 .002
Error (TIMEPER) 79 170.555
Table 4 shows that there was a significant difference between time periods in that both groups improved from pre- to post-test (p = .95). It should be noted that the experimental group scored much lower on the pre-test than the control group. Thus, a pairwise comparison was made to determine significance of interaction between the two groups. Date are reported in Table 5.
TROY (1)