A multi-variant analysis of attendance and achievement in the public schools



Journal Title

Journal ISSN

Volume Title



Introduction. Student absenteeism has become a problem of growing concenn confronting administrators in today's public schools. Because of the educational requirements of this society, laws have been written to force school attendance on youngsters, in most states, to the age of seventeen. The laws and the effort required to enforce them have seemingly been profitable in the past. However, for the last ten years, many standardized tests administered by the schools have Indicated a national decline in student achievement and the declining test scores have created a national controversy, regarding the value of conpulsory education. Purpose. The purpose of the study was to determine whether a significant relationship exists between the attendance and academic achievement of students currently enrolled in public schools. Procedure. Eight hundred and thirty-five students were randomly selected fron the sixty public schools In the Corpus Christi, Texas, Independent School District. The sample included elementary, junior high and senior high students. A purusal of accumulative folders provided Information on sex, race. Intelligence Quotient, attendance and achievement test scores for each subject. The Otis-Lennon Mental Ability Test was used as an indicator of IQ. Achievement was measured, for the elementary school subjects, by the Science Research Associates Multi-level Test. At the junior and senior high levels, the Stanford Test of Academic Skills provided scores in English, reading and mathematics. Methods and Analysis. Two statistical methods were used to test the null hypotheses. First, Partial Correlation Coefficient Analysis provided an "F" tested level of significance of attendance to achievement. Findings recorded at the .05 level of significance dictated the rejection or acceptance of the null hypothesis. Secondly, the Step-Wise Multiple Regression Technique rendered a percent value of contribution to the total achievement test scores made by attendance. The regression technique served to verify the partial correlation coefficient and at the same time furnish an observable value applicable to the level of significance. The tests were run horizontally on all three groups over a one and one-half year period. A vertical investigation was accomplished for all of the students on the senior high level who had maintained residence in the Corpus Christi schools, grades one through ten and one-half. Analysis indicated that there was a significant relationship between attendance and achievement of all three horizontal levels tested, with the exception of reading on the elementary level. The percent of contribution by attendance judged by the Step-Wise Multiple Regression was small in all cases, ranging from a high of 1.9 percent for elementary English to a 1.09 percent in junior high reading. In the vertical study of total accumulated attendance of the senior high group to senior high achievement scores, results indicated acceptance of the null hypotheses at all levels except junior and senior high mathematics. The effects of total over-all attendance of persons in the study for the most part were not significant. [...]