Abstract
In this chapter, I provide a little theory about multilevel data analysis and some basic imputation strategies that match up with the desired analysis. I then describe the automation utility for performing multilevel (mixed model) analysis with SPSS 15/16 and SPSS 17-19 based on Norm-imputed data. Finally, I describe the automation utility for using HLM 6/7 with Norm-imputed data.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
This estimate of k = 40 variables is just a ballpark figure. With your data, you may be able to impute only k = 25 variables. Or you may find that the imputation model is well behaved with as many as k = 60 variables.
- 2.
Note that this limitation does not apply to imputing within clusters using Proc MI in SAS (see next chapter).
- 3.
- 4.
The feature built in to HLM6 for combining results from multiply-imputed data sets is limited in some important ways and will not be used here. First, the feature works with only 10 imputed data sets. Although this does give reasonable preliminary results, it is often desirable to have more than ten imputed data sets (see Graham et al. 2007). Second, the built-in feature in HLM 6 does not calculate the Fraction of Missing Information. This is not a huge omission, but it would be better to have it.
References
Andridge, R. R. (2011). Quantifying the impact of fixed effects modeling of clusters in multiple imputation for cluster randomized trials. Biometrical Journal, 53, 57–74.
Barnard, J., and Rubin, D. B. (1999). Small-sample degrees of freedom with multiple imputation. Biometrika, 86, 948–955.
Graham, J. W., Olchowski, A. E., & Gilreath, T. D. (2007). How Many Imputations are Really Needed? Some Practical Clarifications of Multiple Imputation Theory. Prevention Science, 8, 206–213.
Hansen, W. B., & Graham, J. W. (1991). Preventing alcohol, marijuana, and cigarette use among adolescents: Peer pressure resistance training versus establishing conservative norms. Preventive Medicine, 20, 414–430.
Murray, D. M. (1998). Design and Analysis of Group-Randomized Trials. New York: Oxford Univ. Press.
Little, R. J. A., & Rubin, D. B. (2002). Statistical analysis with missing data: Second Edition. New York: Wiley.
Raudenbush, S. W., and Bryk, A. S. (2002). Hierarchical Linear Models. Thousand Oaks, CA: Sage. 2nd ed.
Schafer, J. L. (2001). Multiple imputation with PAN. In L. M. Collins & A. G. Sayer (Eds.), New methods for the analysis of change (pp. 357–377). Washington, DC: American Psychological Association.
Schafer, J. L., & Yucel, R. M. (2002). Computational strategies for multivariate linear mixed-effects models with missing values. Journal of Computational and Graphical Statistics, 11, 437–457.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer Science+Business Media New York
About this chapter
Cite this chapter
Graham, J.W. (2012). Multiple Imputation and Analysis with Multilevel (Cluster) Data. In: Missing Data. Statistics for Social and Behavioral Sciences. Springer, New York, NY. https://doi.org/10.1007/978-1-4614-4018-5_6
Download citation
DOI: https://doi.org/10.1007/978-1-4614-4018-5_6
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4614-4017-8
Online ISBN: 978-1-4614-4018-5
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)