The construction of a scale for predicting academic success in grade four

The purpose of this study was to construct a scale for predicting academic success in the fourth grade of the elementary school from scores made on selected psychological tests and academic performance in the third grade. The eight variables selected consisted of (1) an intelligence quotient from The Otis Short Form Mental Ability Test, Alpha, Form As, (2) six raw scores from The Stanford Achievement Test, and (3) a grade average of four major subjects in the third grade. The sample group was composed of 161 students from the third grade of three elementary schools in a large metropolitan independent school district. School records provided the data, test scores, and school marks received during the third and fourth grade school years. Frequency distributions of the scores were made for each of the eight variables according to a dichotomy of 'successful' and 'unsuccessful' performance. 'Successful' performance was defined as a yearly grade point average of 2.0 (C) or higher. 'Unsuccessful' performance was defined as a semester grade point average of 1.9 (D) or below. These frequency distributions were then arranged to afford relative percentages of success, by intervals varying directly with the magnitude of the scores achieved and were used to compile the distributions for the 'Score Sheets.' A 'Total Predictive Score' was determined for each individual with the use of the 'Score Sheet' in order to make the 'Predictive Scale.' A frequency distribution of the 161 'Total Predictive Scores' was made and the students again located in a 'successful' or 'unsuccessful' column according to their fourth grade quality point averages. The 'Percent Successful' column of this table became the means by which predictions of success in terms of odds were made. The validation sample consisted of 101 students who met the same criteria as the original sample, except they completed the third and fourth grades one year earlier than the original sample. The scores for these students on the eight variables were converted to 'Total Predictive Scores' with the use of the 'Score Sheets' and located in the appropriate intervals. If 50 percent or more of the students in any interval had been successful originally, it was predicted that all of the students of the validation group within that interval would be successful. If less than 50 percent of the students in any interval had been successful, it was predicted that all of the students of the validation group within that interval would be unsuccessful. The grade point average a student actually received in the fourth grade was then checked against the predictions made. Accuracy of prediction was determined for each class interval and the scale as a whole. Predictions were correct in 75 percent of the cases. Predictive efficiency of each of the single variables when used alone was also determined. Accuracy of these ranged from 64 to 77 percent. The method used here in constructing scales for predicting academic success is relatively simple, and produces a scale which is accurate and useful.