The construction of scales for predicting academic success in grade 10



Journal Title

Journal ISSN

Volume Title



The purpose of this study was two-fold: (1) To construct scales for predicting a student's academic performance in Grade 10 using previous academic performance and test scores as the basis for predictions, and (2) to compare the verbal and quantitative variables as predictors. Predictive scales such as these are useful in counseling students regarding future academic endeavors and their probability of success. The method used in this study was developed by Glueck and Glueck for making predictions concerning the recidivism of delinquent behavior among juveniles. The method has since been use to make various academic predictions. In this study, "Success" was defined as an academic year total average of 84 or above in Grade 10. Those who earned an average of less than 84 were considered "Unsuccessful." The sample was composed of 110 white students enrolled in the 9th grade class of West Columbia Jr. High School. Each student was admitted to the sample only if data on all of the 12 predictive variables used in the study were available. Nine of the predictive variables were taken from the SCIENCE RESEARCH ASSOCIATES Achievement Battery. Three of the predictive variables were taken from 9th grade academic marks. The nine scores taken from the SRA battery were as follows: English, Mathematics, Social Studies, Science, Composite, Language, Reasoning, Quantitative, and Total Score. The three predictive variables taken from academic marks were as follows: Verbal, Quantitative, and Total Average. The procedures involved in the construction of these scales were: 1. Frequency distributions were constructed for each of the twelve variables using the Successful-Unsuccessful headings. The percent of successful students was calculated for each interval in all 12 distributions. The twelve distributions with their score intervals and "Per Cent Successful" became the "Score Sheet." 2. From the "Score Sheet" the "Total Predictive Score" of each student was tabulated. 3. On the basis of the "TPS," students were tabulated into a frequency distribution using the dichotomy "Successful-Unsuccessful," for Verbal predictive scores. Quantitative predictive scroes, and combined scores. 4. The per cent successful was calculated for each interval in the three distributions. The "Per Cent Successful" column in the three distributions is most significant to this study. These columns indicate the probable chances of achieving success which a particular student, or group has in future academic work. Predictions made on the basis of combined verbal and quantitative variables were very similar to those made separately by either quantitative or verbal variables. The three predictive scales deviated only slightly from each other in predictions. This similarity may be due to the fact that school success is a function of both verbal and quantitative variables.



Educational tests and measurements, Prediction of scholastic success, Academic achievement