Master's Theses

Department

Health and Human Performance

Degree Name

Master of Science (MS)

Abstract

The major purposes of this study were to (1) study the reliability of the McDonald Soccer Skill Test, (2) set up norms at Brooks Junior High School for the McDonald Soccer Skill Test, and (3) set up a grading scale that would be used after the completion of the soccer unit for the girls at Brooks Junior High School. RESEARCH METHODS: The subjects used in this study were 265 girls, grades 7-8-9, at Brooks Junior High School who were enrolled in physical education classes during the school year 1961-1962. The McDonald Soccer Skill Test was administered to all subjects after four weeks of soccer participation in class and was then repeated the following week to obtain two sets of raw scores. The mean, standard deviation, chi square, standard error, linear product moment correlation, paired comparison analysis and standard scores were the statistical measures used on the collected data. RESULTS: The means of all the trials did improve after each successive trial. Very little difference was noted when using the average or four trials or the best of three trials. The seventh graders had the lowest mean scores while the ninth graders had the highest mean scores. Each grade improved on test II. The correlation between the two tests was .6248 on the four trials and .5715 on the best of three trials. The values of t obtained through a comparison of the means of test I, best of three trials to test I, four trials; test II, best of three trials to test II, four trials; test I, best of three trials to test II, best of three trials; and test I, four trials to test II, four trials were significant at the .01 level of confidence. This would seem to indicate that the difference obtained was not due to chance factors. Therefore, it is possible that this difference occurred as a result of the skill practice from taking the test the first time. The chi square test for normality exceeded the value necessary for significance at the .05 level, indicating either a normal distribution of soccer ability was not present in this sample or that the test was not able to discriminate between different levels of soccer ability. CONCLUSIONS: The conclusions that have been derived from this study are: 1. All differences obtained were significant when using the average of the tour trials or the best of three trials. 2. The reliability correlation coefficient between test I, four trials and test II, four trials was .6248 and on test I, best of three trials and test II, best of three trials was .5715 and both are low for a reliable test. 3. All of the mean scores improved which could indicate the McDonald Test could be used for the improvement of volleying skill in soccer during the instructional period. 4. The letter grade scales did not show a large range of scores, as in most cases only 2 or 3 kicks would result in a grade change. 5. There are two conclusions possible as a result of the computation of the chi square test. They are: (1) the test was not able to discriminate between different levels of soccer ability in the sample according to a normal distribution or (2) that the range of ability in soccer skill present in the sample of Brooks Junior High School girls was not distributed normally. It would seem logical that the sample obtained was large enough and widely varied enough, according to ability, to obtain a normal distribution of test scores for soccer ability. Therefore, the data indicates that the test did not discriminate between different levels of soccer ability. 6. The reliability of the McDonald Soccer Skill Test was low as shown in this study and would therefore limit the validity of this test for use with the Brooks Junior High School girls. 7. The T-scores that were prepared from this study must be used with some caution and only under certain conditions due to the findings of the significant chi square values which showed a departure from the normal curve.

Keywords

Physical education, Girls, Sports, Soccer, Wichita (Kan.), Sedgewick County (Kan.), Testing

Advisor

Dr. Walter Kroll

Date of Award

Summer 1962

Document Type

Thesis - campus only access

Rights

© The Author(s)

Comments

For questions contact ScholarsRepository@fhsu.edu

Download

Off Campus FHSU Users Click Here

Share

COinS