An evaluation of a battery of standardized tests administered in the seventh grade

In October 1966, three hundred ninety-seven students, who entered the seventh grade of a large metropolitan junior high school in the Southwest, and who met the following requirements, were included in this study: (1) they took all of the tests in the seventh grade test battery, (2) They completed the entire school year in the school where they study was conducted. The scores made by these students on the various tests of the seventh grade test battery and the grade point averages of these students comprise the data upon which this study was made. The purpose of this study was three-fold: (1) to evaluate each test in the battery for use in grouping and general counseling, (2) to determine the most effective, efficient, valid, and reliable combination of tests to be used for the purposes of grouping and counseling, and (3) to reduce the number of tests in an excessively long and laborious test battery without seriously affecting the value and validity of the battery for the purposes intended. The battery of tests used in this study consisted of: The California Short-Form Test of Mental Maturity, and the California Acheivement Test Battery. Sub-scores on these tests included: The language I.Q., the non-language I.Q., and the total I.Q. scores of the CTMM, the total reading score, the total arithmetic score, the total langauage score, and the total battery score of the CAT. The criterion used in this study was the grade point average of each student at the end of the seventh grade. Pearson product-moment coefficients of correlation were computed between each of the tests and the criterion, and intercorrelations between the various tests were computed in teh same manner. The Total Achievement Battery was found to have the highest correlation with the criterion, .644. The highest correlation between components of the test battery was found to exist between non-language I.Q. and total I.Q., .955. The Wherry-Doolittle Test Selection Method was applied, and it was found that the maxinum predictive value of the test battery was obtained by a combination of the Total Achievement Battery score, Total Language Achievement score, and the Total Arithmetic score, and the Total Reading Achievement score. These measures yielded a multiple correlation of .674 with the criterion. This correlation is not what would be considered high, but it be of some limited value for grouping and counseling purposes. It was concluded that, 1) the beat predictor of over, all academic success in the seventh grade at this junior high school was the total aohlevesent score, 2) in the interest of efficiency and accuracy the California Test of Mental Maturity should be dropped from the test battery, 3) the battery night be Improved by the addition of aptitude tests. It was recomended, that to problems of accuracy of the criterion that the study be repeated using the seme test battery in the sme or in a similar school.