This was a study of the relationship between scores made on the American College Test and academic grades made by pre-engineering and industrial arts students to determine the reliability of American College Test scores when used for counseling and predicting academic grades.

The problem of this study was to determine the predictive validities, as measured by multiple correlation, of certain mathematics course grades and the American College Test (ACT) battery when various achievement criteria were used. This study had a two-fold purpose: (1) to determine measures which could be used effectively to predict academic achievement in college mathematics, and (2) to determine whether the multiple correlation between selected predictors and each criterion were significant.

Grading within a single school was studied by comparing the grades with the policy of grading recommended by the high school, and the relationship between the grades and two tests administered by the school. Six hundred fifty-five graduates from the 1964, 1965 and 1966 graduating classes made up the sample. The grades used were those received during their three years in high school. The tests were the Henmon-Nelson Test of Mental Ability administered in the tenth grade and the American College Test which was taken in the twelfth grade. The Pearson r Correlation Coefficient was used to make the correlations. The number of A and B grades given in all subject areas were beyond that recommended by the school policy. A t test showed the differences in grading between required subjects and nonacademic elective subjects in the lower 25 percent of each graduating class to be significant at the .01 percent level. In the upper 25 percent the differences were not significant. The required subjects area grades correlated highest with scores from both tests.

Freshmen at Sul Ross State University are required to take tests which are used for placement purposes. One of the tests given is the Nelson-Denny Reading Test which measures comprehension, vocabulary, and reading rate. The scores are used with American College Test or Standard Achievement Test scores to place students in either remedial or regular freshman English. Remedial students, who score below the tenth-grade competency level, are placed in English 1300. Regular students are placed in English 1301 or 1302. Twelve studies were found which had been done in this area since 1980. One was directly related to this study. The Anglo and Hispanic population of the freshman class of 1987 was tested. Blacks were not included as they comprised less than 9 percent of the freshman class. There were 69 students in the experimental group and 162 in the control group. A pretest-posttest design was used. A three-way analysis of variance set up data for statistical testing. The Alpha level was set at .05. The findings indicate a significant difference for Hypothesis 1, which predicted no significant difference in the posttest performance of students required to take English 1300 and the pretest performance of students who were not. Therefore it was rejected. Because statistical testing yielded no significant difference for Hypothesis 2 — there will be no significant difference in posttest performance of Hispanic and Anglo students who were enrolled in English 1300, and Hypothesis 3—there will be no significant difference in the posttest performance of males and females who enrolled in English 1300, they were retained. Results indicate that while there is a significant difference between the means of the remedial students' posttest scores and means of the regular students' pretest scores, the program raises the performance of remedial students to a level accepted by Sul Ross State University. Therefore, the program is considered successful.