Date of Award

8-1973

Document Type

Dissertation

Degree Name

Doctor of Philosophy (Medical Science)

Abstract

Statement of the Problem

The problem of this study was to determine the relationship of secondary school, personal, and collegiate factors to student achievement in COBOL computer programming.

Purpose of the Study

The primary purpose of this study was to identify those personal and curricular characteristics of students which could be used to predict performance in COBOL programming classes. A prediction equation was developed which could be used to predict performance in COBOL.

A secondary purpose of this study was to determine if there is a relationship b e tween certain student characteristics and the high dropout rate of students from COBOL computer programming classes. A prediction equation was developed which could be used to predict student dropout in COBOL.

Procedures

Data was collected for 25 independent variables during the fall semester of 1972. Five colleges and universities participated in the study with usable information obtained for 232 subjects .

A COBOL Examination was developed to obtain a reliable measure of student achievement or performance in the participating COBOL classes. This objective test consisted of 62 multiple-choice questions.

Setwise multiple regression analysis was used to identify significant independent variables and to determine regression equations. A- number of computer runs were made utilizing different combinations of the variable sets being investigated.

The prediction equation developed for predicting student performance in COBOL is as follows:

The independent variables included in the regression equation were the college variables: accounting, English, mathematics, introduction to data processing, natural science, foreign language, economics, and computer programming. This set of variables accounted for nearly 25 per cent of the variance in the criterion measure.

Conclusions

The following conclusions we are drawn as a result of this study:

  1. The college course variable s were consistently the best set of predictors in the study.

  2. Cumulutive college grade point average was usually the last predictor variable dropped from the regression program. COL-GPA is a significant variable in predicting dropout or performance in COBOL

  3. The age and sex of a student were insignificant in predicting dropout or performance of students in COBOL computer programming classes.

  4. Of all the students participating in the study it was found that none had taken a high school data processing course of any type. This contradicts the current belief that secondary schools are generally providing data processing instruction.

  5. A very high dropout rate, 25 per cent, was experienced in the participating COBOL classes.

  6. Students with prior programming experience performed better in COBOL than those without training in some other computer programming language.

  7. Students who had taken an introduction to data processing course performed better in COBOL than those who had not taken such a course.

  8. The scores on the four sections of the American College Test (ACT) are insignificant as predictors of dropout from COBOL classes. They are, however, significant as predictors of performance in COBOL.

  9. The high school percentile rank of a student had a consistently high correlation with prediction of student dropout and performance in COBOL.

  10. In predicting dropout from COBOL classes, the college variables of natural science, introduction to data processing, and computer programming are the most significant variables in the prediction equation.

  11. Computer programming, economics, and accounting are the three most significant variables of the college predictor variables in the prediction equation developed for predicting performance in COBOL.

  12. High school data, other than percentile rank, contributed very little to prediction of dropout and performance in COBOL

Download

Share

COinS