The construction of a scale for predicting academic success in the sixth grade



Journal Title

Journal ISSN

Volume Title



There is a very definite, present need for counselors in the field of education to obtain a better understanding of those factors which can be used to predict academic success, since a large portion of their time is spent scheduling students into different classes. The author of this study took cognizance of this need and constructed a predictive scale which could be used to predict sixth-grade success on the basis of ten variables or factors from the fifth-grade. Six of these variables came from raw scores received on the Iowa Test of Basic Skills. The total scores were used on the following five subtest: Vocabulary, Reading Comprehension, Language Skills, Work-Study Skills, and Arithmetic Skills. The sixth variable was the Composite Score on the Iowa Test of Basic Skills. The seventh variable was the IQ score received on the Otis-Beta, and the eighth, ninth, and tenth variables came from averages in five fifth-grade subjects. The eighth variable was the total average received in English, social studies, and reading. The ninth variable was the total average received in math and science, and the tenth variable was the total average received in all five previously- mentioned subjects. The criterion for success was an "A" or "B" total average in English, social studies, reading, math, and science, in the sixth grade. The fifth-grade scores and grades of the present eighthgrade class in one junior high were used so as to eliminate any possible delay because of incomplete grade averages. Within this class, 266 eighth-grade students were found who had been enrolled in this same school district since the fifth grade and who had complete records. After selecting the group to be studied and the variables to be considered, the scores on all the variables were recorded for each of the 266 students. Next it was determined whether or not each student was "successful" in the sixth-grade. In the following step a frequency distribution was set up to determine the number of "successful" and "unsuccessful" students in each interval of each variable. When this was completed, the percentage of the subjects "successful" or "unsuccessful" in each interval of each of the ten variables was calculated, and these percentage figures constituted the "Score Sheet." Each subject was then rated on the "Score Sheet" to ascertain the interval in which he fit in each of the ten variables. The resulting ten percentage figures for each student were summed, and this total became his "Total Predictive Score." A frequency distribution was set up using the "Total Predictive Scores" for all 266 students, and the number of successful students in each interval was calculated. The resulting figures constituted the "Predictive Scale." The scale was validated on the fifth and sixth-grade scores of 113 randomly-selected seventh-grade students from the same junior high school whp met the same qualifications as the original sample. Each subject was placed in a certain interval in the "Predictive Scale" on the basis of his "Total Predictive Score," and the prediction accuracy was calculated The results showed that a better-than-chance prediction accuracy prevailed on all of the ten intervals. Considering the top four intervals as one group, the accuracy of prediction was 93 percent. Considering the bottom six intervals as one group, the accuracy of prediction was 100 percent. And considering the entire validation sample as one group, the accuracy of prediction was 94 percent. With the full realisation that it is difficult to predict for the individual, the apparent high degree of accuracy associated with this predictive scale should make it worth the sixth-grade counselor's consideration when concerned with the prediction of academic success in large groups whose qualifications meet those of the original and validation groups.



Prediction of scholastic success, Educational tests and measurements