The construction of scales for predicting academic success in the fourth grade of elementary school

The purpose of this study was to construct a scale for predicting academic success in grade four in elementary school from school marks and scores on standardized tests obtained in grade three. The eight predictor variables were as follows: (1) academic average in grade three based on solid subjects only, (2) scores on The Otis Quick-Scoring Mental Ability Test, Alpha Short Form, (3) scores on The Stanford Achievement Battery, Form J; SAT-Paragraph Meaning, (4) SAT-Word Meaning, (5)SAT -Spelling, (6) SAT-Language, (7) SAT-Arithmetic Reasoning, and (8) SAT-Arithmetic Computational. The sample group was composed of 119 pupils of both sexes, who completed grade four in the school year 1965-1966, in the same elementary school, and, for whom all data was available for the eight predictor variables. The school is located in an average to average-plus socio-economic area. Frequency distributions were arranged for each of the eight predictor variables for the 119 pupils. Scores on each variable were recorded according to a successful-unsuccessful dichotomy. Success was defined as a grade point average of 3.0 (B) or above in grade four (basic subjects). Grade point averages below 3.0 were recorded as unsuccessful. The 'Score Sheet' was then used to obtain the 'Total Predictive Score' for each of the 119 pupils. After a 'Total Predictive Score' had been calculated for each of the 119 pupils, a frequency distribution was set up again according to a dichotomy of successful and unsuccessful. This distribution of 'Total Predictive Scores' became the 'Predictive Scale'. The validation group consisted of 39 pupils that met the same criteria as the 119 of the sample group. The 39 were selected at random from an original group of 158 from the same elementary school. A 'Total Predictive Score' for each of them was obtained from the 'Score Sheet'. Success or failure was then predicted according to the interval that the 'Total Predictive Score' was located in on the 'Predictive Scale'. If the appropriate interval was greater than 50, the prediction was made that the pupil would succeed; if however, the percent successful was less than 50, the prediction was made that the pupil would not be successful. The accuracy of prediction on an individual basis for the 39 pupils in the validation group was 82 percent. The accuracy of prediction for the pupils with the larger 'Total Predictive Scores' (413 and Above) was 87 percent; for the pupils with the lower TPS’s (below 413) was 79 percent.Psychology, Department o