Using data from the evaluation of the Fast Track intervention, this article illustrates three methods for handling attrition. Multiple imputation and ignorable maximum likelihood estimation produce estimates that are similar to those based on listwise-deleted data. A panel selection model that allows for selective dropout reveals that highly aggressive boys accumulate in the treatment group over time and produces a larger estimate of treatment effect. In contrast, this model produces a smaller treatment effect for girls. The article’s conclusion discusses the strengths and weaknesses of the alternative approaches and outlines ways in which researchers might improve their handling of attrition.
Achenbach, T. M.1991. Manual for the Child Behavior Checklist 4-18 and 1991 profile. Burlington: Department of Psychiatry, University of Vermont.
2.
Agresti, A.2002. Categorical data analysis. 2nd ed.Hoboken, NJ: John Wiley.
3.
Allison, P. D.2002. Missing data. Thousand Oaks, CA: Sage.
4.
Asher, S. R., and K. A. Dodge. 1986. Identifying children who are rejected by their peers. Developmental Psychology22: 444-449.
5.
Chamberlain, P., and J. B. Reid. 1987. Parent observation and report of child symptoms. Behavioral Assessment9: 97-109.
6.
Coie, J. D., K. A. Dodge, and H. A. Coppotelli. 1982. Dimensions and types of social status: A cross-age perspective. Developmental Psychology18: 557-569.
7.
Collins, L. M., J. L. Schafer, and C. M. Kam. 2001. A comparison of inclusive and restrictive strategies in modern missing-data procedures. Psychological Methods6: 330-351.
8.
Conduct Problems Prevention Research Group. 1999a. Initial impact of the Fast Track prevention trial for conduct problems, I: The high-risk sample. Journal of Consulting and Clinical Psychology67: 631-647.
9.
Conduct Problems Prevention Research Group. 1999b. Initial impact of the Fast Track prevention trial for conduct problems, II: Classroom effects. Journal of Consulting and Clinical Psychology67: 648-657.
10.
Conduct Problems Prevention Research Group. 2002a. Evaluation of the first three years of the Fast Track prevention trial with children at high risk for adolescent conduct problems. Journal of Abnormal Child Psychology30: 19-35.
11.
Conduct Problems Prevention Research Group. 2002b. The implementation of the Fast Track program: An example of a large-scale prevention science efficacy trial. Journal of Abnormal Child Psychology30: 1-17.
12.
Conduct Problems Prevention Research Group. 2002c. Predictor variables associated with positive Fast Track outcomes at the end of third grade. Journal of Abnormal Child Psychology30: 37-52.
13.
Cook, R. D., and S. Weisberg. 1999. Applied regression including computing and graphics.New York: John Wiley.
14.
Davidson, R., and J. G. MacKinnon.1993. Estimation and inference in econometrics. New York: Oxford University Press.
15.
Diggle, P., and M. G. Kenward. 1994.Informative drop-outin longitudinal data analysis. Applied Statistics43 (1): 49-93.
16.
Elliott, D. S., S. S. Ageton, and D. Huizinga. 1985. Explaining delinquency and drug use. Beverly Hills, CA: Siegel.
17.
Elliott, D. S., D. Huizinga, and S. Menard. 1989. Multiple problem youth: Delinquency, substance use, and mental health. New York: Springer-Verlag.
18.
Enders, C.2001. A primer on maximum likelihood algorithms available for use with missing data. Structural Equation Modeling8 (1): 128-141.
19.
Foster, E. M. 2002.Using services data to adjust for attrition in outcomeanalyses: An application of multiple imputation. Paper presented at “A System of Care for Children’s Mental Health: Expanding the Research Base, 15th Annual Research Conference,”Tampa, FL.
20.
Foster, E. M., and L. Bickman. 1996. An evaluator’s guide to detecting attrition problems. Evaluation Review20 (6): 695-723.
21.
Graham, J. W.2003. Addingmissing-data-relevant variables to FIML-basedstructural equations models. Structural Equation Modeling10 (1): 80-100.
22.
Greene, W. H.1993. Econometric analysis. 2nd ed.New York: Macmillan.
23.
Hall, S. M., K. L. Delucchi, W. F. Velicer, C. W. Kahler, J. Ranger-Moore,D. Hedeker, J. Y. Tsoh, and R. Niaura. 2001. Statistical analysis of randomized trials in tobacco treatment: Longitudinal designs with dichotomous outcome. Nicotine and Tobacco Research3: 193-202.
24.
Heckman, J. J.1974. Shadow wages, market prices and labor supply. Econometrica42: 679-694.
25.
Heckman, J. J.1976. The common structure of statistical models of truncation, sample selection and limited dependentvariables and a simple estimator for such models. Annals of Economic and Social Measurement5: 475-492.
26.
Heckman, J. J.1979. Sample selection bias as a specification error. Econometrica47: 153-161.
27.
Kenward, M. G.1998. Selection models for repeated measurements with nonrandom dropout: An illustration of sensitivity. Statistics in Medicine17: 2723-2732.
28.
Laird, N. M.1994. Discussion of Diggle PJ, Kenward MG: Informative dropout in longitudinal data analysis. Applied Statistics43: 84-84.
29.
Lillard, L. A., and C. W. A. Panis. 1998. Panel attrition from the Panel Study of Income Dynamics. Journal of Human Resources33: 437-457.
30.
Lillard, L. A., and C. W. A. Panis. 2000. aML multi-level multiprocess statistical software. Version 1.0. Los Angeles: Econ Ware.
31.
Littell, R. C., G. A. Milliken, W. W. Stroup, and R. D. Wolfinger. 1999. SAS system for mixed models. Cary, NC: SAS Institute Inc.
32.
Little, R. J. A., and D. B. Rubin. 1987. Statistical analysis with missing data. New York: John Wiley.
33.
Little, R. J. A., and D. B. Rubin. 2002. Statistical analysis with missing data. 2nd ed.Hoboken, NJ: John Wiley.
34.
Lochman, J. E., and Conduct Problems Prevention Research Group. 1995. Screening of child behavior problems for prevention programs at school entry. Journal of Consulting and Clinical Psychology63: 549-559.
35.
Muthen, L. K., and B. O. Muthen. 2001. Mplus user’s guide. Los Angeles: Muthen& Muthen.
36.
Orr, L. L.1999. Social experiments: Evaluating public programs with experimental methods. Thousand Oaks, CA: Sage.
37.
Rosenbaum, P. R.2002. Observational studies. New York: Springer-Verlag.
38.
SAS Institute. 2001a. The MI procedure. Cary, NC: SAS Institute.
39.
SAS Institute. 2001b. The MIAnalyze procedure. Cary, NC: SAS Institute.
40.
Schafer, J. L.1997. Analysis of incomplete multivariate data. London: Chapman and Hall.
41.
Schafer, J. L.1999. Multiple imputation: A primer. Statistical Methods in Medical Research8: 3-15.
42.
Schafer, J. L., and J. W. Graham. 2002. Missing data: Our view of the state of the art. Psychological Methods7 (2): 147-177.
43.
Schafer, J. L., and M. K. Olsen. 1998. Multiple imputation for multivariate missing-data problems. Multivariate Behavioral Research33 (4): 545-571.
44.
Scott, A. J., and D. Holt. 1982. The effect of two-stage sampling on ordinary least squares methods. Journal of the American Statistical Association77: 848-854.
45.
Stata Corp.2001. Statastatistical software: Release 7.0. College Station, TX: Stata Corporation.
46.
Stolzenberg, R. M., and D. A. Relles. 1990. Theory testing in a world of constrained research design: The significance of Heckman’s censored sampling bias correction for nonexperimental research. Sociological Methods and Research18: 395-415.
47.
Stolzenberg, R. M., and D. A. Relles. 1997.Tools for intuition about sample selection bias and its correction. American Sociological Review62: 494-507.
48.
Terry, R., and J. D. Coie. 1991. A comparison of methods for defining sociometric status among children. Developmental Psychology27: 867-880.
49.
Verbeek, M., and T. Nijman. 1992. Testing for selectivity bias in panel data models. International Economic Review33: 681-703.
50.
Verbeke, G., and G. Molenberghs. 1997. Linear mixed models in practice. New York: Springer-Verlag.
51.
Verbeke, G., and G. Molenberghs. 2000. Linear mixed models for longitudinal data. New York: Springer-Verlag.
52.
Wooldridge, J. M.2002. Econometric analysis of cross section and panel data. Cambridge, MA: MIT Press.
