Rumberger, Russell W., & Palardy, Gregory J.
Test Scores, Dropout Rates, and Transfer Rates as Alternative Indicators of High School Performance
University of California, Santa Barbara; University of Georgia
Looks at whether schools that are effective in raising test scores are also effective in reducing dropout and transfer rates, and focuses on which school characteristics predict these alternative indicators of performance.
Journal Name or Institutional Affiliation:
American Educational Research Journal
Vol. 42. No. 1, Pp. 3-42
- Schools have relatively small effects on student learning in comparison with student background characteristics: about 26% of variability in learning is attributable to students' high schools.
- Differences in background characteristics of students account for more than half of variability in school drop out rates.
- Differences in background characteristics of students account for one-third of the variability in attrition rates.
- Differences in background characteristics of students account for almost none of the variability in transfer rates.
- Proportion of students from nontraditional families and mean achievement of students in eighth grade had significant effects on all four outcomes (learning, dropout, transfer, and attrition).
Journal Article Empirical Research
Dropouts, Graduation Rates, SES Composition
Secondary Survey Data
Method of Analysis:
High school students in the US
Unit of Analysis:
- Data come from the National Education Longitudinal Study of 1988 (NELS 88).
- Sample includes 14,199 students at 912 high schools.
- - DV: Achievement growth (learning) over four years of high school (mean of scores on standardized achievement tests in math, science, reading, and social science); proportion of tenth grade students who dropped out between grades 10 and 12, proportion of 10th grade students who transferred over the same period, and total proportion of tenth grade students who left school before twelfth grade
- - IV: Student level: minority status; SES (composite of parental education, income, and occupational status); nontraditional family (students who do not live with both parents); family mobility (families who move between 10th and 12th grades); self-reported grades; plans to finish college; misbehavior; retention between the first and eighth grades; and number of school changes between the first and eighth grades. School level: mean SES of students in school; proportion of Black or Hispanic students in the school; proportion of students who come from nontraditional families; proportion of students who changed residences between 1990 and 1992; mean grade point average of students in grade 6-8; mean amount of misbehavior among eighth grade students; proportion of students who were retained between grades 1 and 8; proportion of students who plan to finish college; whether the school was Catholic or another private school; whether the school was urban, rural or suburban; the size of the school; student-teacher ratio; proportions of teachers with full credentials, credentials in the subjects they taught, a bachelors degree in the subject they taught, and advanced degrees; quality of teaching and teacher support reported by students; teachers efficacy and expectations for student achievement; teachers perceptions of school leadership; collaboration with other teachers; teachers control over school policy; parental involvement; academic climate; social climate; and disciplinary climate.