AN ADMINISTRATIVE MODEL FOR PREDICTING AND MANAGING STUDENT SUCCESS OR NON-SUCCESS ON THE NEW YORK STATE REGENTS COMPETENCY TEST
This study attempted to develop an administrative model for effective decision-making relative to state mandated minimum competency standards as a prerequisite to high school graduation. The model is based on the utilization of data from official records which would be predictive of student success on the New York State Preliminary Regents Competency Test. The students in this study numbered 324 males and 280 females from 26 school districts in the metropolitan area surrounding New York City who completed the Preliminary Regents Competency Tests in Reading and Writing in October 1980. The 40 predictor variables consisted of the subtest scores from the Metropolitan Achievement Test in grade 6; the I.Q. score in grade 5; the 6th grade New York State Pupil Evaluation Program (P.E.P.) Reading Test; the subtest scores from the Differential Aptitude Tests, the final grades in grade 6 and characteristics descriptive of staff, facilities, and instructional organization of the schools in the sample. A stepwise regression program was used to analyze the data and to obtain the multiple correlations. The regression weights of the significant variables were converted to beta weights in order to determine the direct contribution to the variance in the criterion. Females attained significantly higher means on M.A.T. Spelling and Language, and the Competency Test in Writing. Males attained significantly higher means on the final grades in reading, English, social studies and total grade point average. All standardized test score and academic achievement variables correlated significantly but few in-school variables correlated significantly with the competency tests. Only class size, age of building, student-teacher ratio, expenditures per pupil, number of instructional assignments per teacher, student-administrator ratio, student-learning disabilities teacher ratio, and number of years of teacher experience were significantly correlated with the competency tests. Multiple regression analysis indicated that different combinations of predictors emerged for males and females. For males, the P.E.P. Reading test, D.A.T. Sentences, M.A.T. Reading and D.A.T. Space Relations were efficient predictors of reading competency with a multiple correlation of .91. For females, the D.A.T. Sentences, Reading final grade, I.Q. score and M.A.T. Spelling were efficient predictors of reading competency (R = .87).
HOWARD, DONALD ANTHONY, "AN ADMINISTRATIVE MODEL FOR PREDICTING AND MANAGING STUDENT SUCCESS OR NON-SUCCESS ON THE NEW YORK STATE REGENTS COMPETENCY TEST" (1982). ETD Collection for Fordham University. AAI8213607.