Abstract
A significant body of education literature has begun using multilevel statistical models to examine data that reside at multiple levels of analysis. In order to provide a primer for medical education researchers, the current work gives a brief overview of some issues associated with multilevel statistical modeling. To provide an example of this technique, we then present a multilevel analysis examining the relationship between two individual-level variables and the “cross-level” interaction between this relationship and a school-level variable. In offering this discussion and example of multilevel modeling, we hope to provide medical educators with a basic introduction to multilevel statistics, including the advantages of utilizing these techniques.
Similar content being viewed by others
Notes
While this notation may be somewhat different than that employed in OLS regression, the importance of using HLM notation is threefold. First, this notation is ubiquitous in much literature dealing with multilevel modeling. Second, because this notation often is employed in statistical packages’ equation-based language, our use of this notation allows one to interpret relevant output and to recognize the correspondence between results appearing in software packages and those in research articles. Third, given that multilevel models may incorporate more than two levels, our using unique symbols for each level of analysis yields less confusion in interpreting results than would using the same symbol applied at multiple levels.
References
Bliese, P. D. (2002). Multilevel random coefficient modeling in organizational research: Examples using SAS, S-PLUS. In F. Drasgow & N. Schmitt (Eds.), Modeling in organizational research: Measuring and analyzing behavior in organizations (pp. 401–445). San Francisco, CA: Jossey-Bass, Inc.
Guiton, G., Hodgson, C. S., Delandshere, G., & Wilkerson, L. (2004). Communication skills in standardized-patient assessment of final-year medical students: A psychometric study. Advances in Health Science Education: Theory and Practice, 9, 179–187.
Hofmann, D. A. (1997). An overview of the logic and rationale of hierarchical linear models. Journal of Management, 23, 723–744.
Hox, J. J. (1995). Applied multilevel analysis. Amsterdam: TT-Publikaties.
Hox, J. J., & de Leeuw, E. D. (2003). Multilevel models for meta-analysis. In S. P. Reise & N. Duan (Eds.), Multilevel modeling: Methodological advances, issues, and applications (pp. 90–111). Mahwah, NJ: Erlbaum.
Kenny, D. A., & Judd, C. M. (1986). Consequences of violating the independence assumption in analysis of variance. Psychological Bulletin, 99, 422–431.
Kozlowski, S. W. J., & Klein, K. J. (Eds.) (2000). Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions. San Francisco: Jossey-Bass.
Nezlek, J. B. (1996). Multilevel random coefficient analysis of even- and interval-contingent data in social and personality psychology research. Personality and Social Psychology Bulletin, 7, 771–785.
Nezlek, J. B., & Zyzniewski, L. E. (1998). Using hierarchical linear modeling to analyze grouped data. Group Dynamics, 2, 313–320.
Pollack, B. N. (1998). Hierarchical linear modeling and the “unit of analysis” problem: A solution for analyzing responses of intact group members. Group Dynamics, 2, 299–312.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Newbury Park, CA: Sage.
Snijders, T. A. B., & Bosker, R. J. (1999). Standard errors and sample sizes for two-level research. Journal of Educational Statistics, 18, 237–259.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Zyphur, M.J., Kaplan, S.A., Islam, G. et al. Conducting multilevel analyses in medical education. Adv in Health Sci Educ 13, 571–582 (2008). https://doi.org/10.1007/s10459-007-9078-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10459-007-9078-y