prediction of future percormance is attempted in almost every field of endeavor. The accuracy varies in different lines of study, and perhaps none is as subject to variability as those attempted with human beings as subjects. When an attempt is made to ascertain in advance the performance of college students in their schoolwork, a multiplicity of complicating problems are introduced. Whereas intelligence can be fairly well isolated, it is difficult to control or even enumerated all the other factors that come into the problem of predicting grades from scores received on an intelligence examination. Among the factors that are difficult to objectively measure or control are the transference of past learning, levels of aspiration, efficiency of study habits and time spent in studyingy, attentiveness in class, as well as specific aptitudes of disabilities, varying difficulty of different academic courses, and susceptibility to or freedom fromphysiological or paychological disorders. Evn though correlations between scholastic grades and intelligence test scores will be, due to various factors of limitation, only moderately high at best, their values cannot be doubted. With high correlations, a definite relationship can be established. With lower correlations, trends can be noted. A segment analysis also may prove to be of value in establishing areas of relative stringth and weaknesses in the predictive structure. The thesis problem is one of determining certain predictive values of the American Council on Education Psychological Examination. Inasmuch as specificity is a virtue in educational measurement, the American Council test is a definite step in this direction. The aim of the test is to measure what the authors of hte test consider to be schoolastic aptitude. The purpose of this study is to determine the accuracy of this scholastic aptitude test in predicting grade-point averages of Utah State Agricultural College freshmen students in their first quarter in college.

Research shows academic literacy is a challenge for students classified as Long-Term English Language Learners (LTELLs). In the pseudonymous Windy Desert School District (WDSD), there are 17,365 students classified as LTELLs. Of these students, the majority are falling short of English academic literacy goals on the Assessing Comprehension and Communication in English State-to-State for English Language Learners (ACCESS for ELLs) test and 67% do not graduate from high school. This quantitative study examined the predictive relationship between ACCESS English language proficiency subscale scores in the language domains of speaking, listening, reading, and writing and course semester grades in English 9, English 10, and English 11. This longitudinal study, informed by theorists Cummins and Krashen, followed a cohort of 718 Grade 9 students for 3 years (2012-2015). Of the 718, only 161 participant data sets were valid for the final ordinal logistic regression analysis. ACCESS subscale scores in speaking, listening, reading, and writing comprised the predictor variables and English course semester grades comprised the criterion variables. Results revealed that LTELLs' ACCESS subscale scores in listening, reading, and writing were significant predictors of their English course grades whereas speaking scores were not. For each predictor variable, a 1-unit increase in the predictor decreased the likelihood of receiving a lower grade in the course. Social change can result from the WDSD using ACCESS results to create and implement effective instructional programs that develop LTELLs' proficiency in the language domains found significant in predicting their academic grades, thereby increasing their language proficiency, academic grades, and graduation rates over time.