ED201B

 

Homework Assignment 1

Sample Analysis of Survey Research Study

 

 

1)   Study (citation)

 

Rumberger, Russell W.  1995 .  "Dropping out of middle school:  A multilevel analysis of students and schools."  American Educational Research Journal 32:583-625. 

2)   Purpose of Study (descriptive, exploratory, explanatory, evaluation)

 

      Explanatory

 

3)   Questions and/or Hypothesis(es) Addressed

1.   What factors influence a student's decision to drop out of middle school?

2.   How do these factors differ among ethnic groups and high-risk students?

3.   What factors influence middle-school dropout rates?

 

4)   Research Design Employed

 

      Survey:  hierarchical/longitudinal

 

5)   Population, Sampling Procedure, Sample Selection, and Size

 

      --Population: Eighth grade schools and students in 1988

      --Sampling procedure:  Stratified random sample of schools and students with oversampling of some schools and students

      --Sample selection:  all students who were surveyed in 1988 and 1990; all schools surveyed in 1988 with weights and non-missing data

      --Sample size:  981 schools, 17, 424 students

 

6)   Key Variables and/or Domains Studied

 

      Dependent:  student-level: likelihood of dropping out from 8th to 10th grades

                          school-level:  dropout rates from 8th to 10th grades

 

      Independent: student-level:  array of demographic, family background, academic background, attitudes and behaviors in 8th grade, academic performance in 8th grade variables

                           school-level:  array of student composition, structural, organizational variables

7)   Sources of Data/Procedures for Collecting

 

      -- National Education Longitudinal Survey of 1988 (NELS:88)

      --8th grade student, parent, and school questionnaires; 8th grade achievement tests; 10th grade student questionnaire

      --administered through schools and via mail (parents)

 

8)   Methods of Analyzing Data

 

      --Logistic regression, hierarchical linear models (HLM)

 

9)   Controls/Controlled Extraneous Variables  (if applicable)

 

 

10)   Findings

 

      --At the individual level, the results identified a number of family and school experience factors that influence the decision to leave school, with grade retention being the single most powerful predictor.  But disaggregating the analysis also revealed that there are widespread differences in the effects of these factors on white, Black, and Hispanic students.

 

      --At the institutional perspective, the results revealed that mean dropout rates vary between schools and that most of the variation can be explained by differences in the background characteristics of students.  But restricting the analysis to lower SES schools shows widespread differences in both mean dropout rates and social class differentiation among schools, and that much of the variation can be explained by social composition of students and by several structural features of schools and school climate. 

 

11) Author's(s') Conclusions

 

      --School policies and practices have important implications for student dropout behavior

      --Study provides only a piece of the larger puzzle of school dropouts.

 

12) What factors lead you to have confidence or not have confidence in the conclusions?

 

      --Confidence:  large, representative sample of 8th grade students and schools;  large number of variables in statistical models;  sophisticated statistical methods that disentangle effects of students and schools on dropout process