This study investigates the quality of multilevel model parameter estimates and standard errors as a function of varying magnitudes of correlation among Level 1 predictors and model characteristics. The results of the study showthat with multicollinearity presented at Level 1 of a two-level mixed-effects linear model, the fixed-effect parameter estimates produce relatively unbiased values; however, the variance and covariance component estimates produce downwardly biased values except for Level 1 variance (< 5%). The standard errors associated with the parameter estimates are also biased under varied magnitudes of Level 1 predictor correlation.
Get full access to this article
View all access options for this article.
References
1.
Ahn, B. J. (1994). A ridge-type estimator for generalized linear models. Korean Journal of Applied Statistics, 7, 75-82.
2.
Belsley, D. A. (1991). Conditioning diagnostics: Collinearity and weak data in regression. New York: John Wiley.
3.
Belsley, D. A., & Oldford, R. W. (1986). The general problem of ill-conditioning and its role in statistical analysis. Comparative Statistical Data Analysis, 4, 103-120.
4.
Bonate, P. L. (1999). The effect of collinearity on parameter estimates in nonlinear mixed effect models. Pharmaceutical Research, 16, 709-717.
5.
Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models: Applications and data analysis methods. Newbury Park, CA: Sage.
6.
Chatterjee, S., Hadi, A. S., & Price, B. (2000). Regression analysis by example(3rd ed.). New York: John Wiley.
7.
Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum.
8.
Fabrycy, M. Z. (1975). Multicollinearity caused by specification errors. Applied Statistics, 24, 250-254.
9.
Goldstein, H. (1995). Multilevel statistical models. London: Charles Griffin.
10.
Hamilton, L. C. (1992). Regression with graphics: A second course in applied statistics. Belmont, CA: Wadsworth, Inc.
11.
Harvey, A. C. (1977). Some comments on multicollinearity in regression. Applied Statistics, 26, 188-191.
12.
Hays, W. (1981). Statistics. New York: Holt, Rinehart & Winston.
13.
Huh, M.-H.,& Olkin, I. (1995). Asymptotic aspects of ordinary ridge regression. American Journal of Mathematical and Management Sciences, 15, 239-254.
14.
Johnston, J. (1963). Econometric methods. New York: McGraw-Hill.
15.
Kreft, I. G. G., & De Leeuw, J. (1998). Introducing multilevel modeling. Thousand Oaks, CA: Sage.
16.
Kubitschek, W. N.,& Hallinan, M. T. (1999). Collinearity, bias, and effect size: Modeling the effect of track on achievement. Social Science Research, 28, 380-402.
17.
Leamer, E. E. (1973). Multicollinearity: A Bayesian interpretation. The Review of Economics and Statistics, 55, 373-338.
18.
MacCallum, R. C., & Mar, C. M. (1995). Distinguishing between moderator and quadratic effects in multiple regression. Psychological Bulletin, 118, 405-421.
19.
Malinvaud, E. (1966). Statistical methods for econometrics. Amsterdam: North Holland.
20.
McCullagh, P., & Nelder, J. A. (1989). Generalized linear models(2nd ed.). Boca Raton, FL: Chapman & Hall.
21.
Morrow-Howell, N. (1994). TheMword: Multicollinearity in multiple regression. SocialWork Research, 18, 247-251.
22.
Neter, J., Wasserman, W., & Whitmore, G. (1978). Applied statistics. Boston: Allyn & Bacon.
23.
Paterson, L.,& Goldstein, H. (1991). Newstatistical methods for analyzing social structures: An introduction to multilevel models. British Educational Research Journal, 17, 387-393.
24.
Pedhazur, E. (1982). Multiple regression in behavioral research: Explanation and prediction(2nd ed.). New York: Holt, Rinehart & Winston.
25.
Sarkar, N. (1992). A new estimator combining the ridge regression and the restricted least squares methods of estimation. Communications in Statistics Theory and Methods, 21, 1987-2000.
26.
Sarkar, N. (1996). Mean square error matrix comparison of some estimators in linear regressions with multicollinearity. Statistics & Probability Letters, 30, 133-138.
27.
Segerstedt, B., & Nyquist, H. (1992) On the conditioning problem in generalized linear models. Journal of Applied Statistics, 19, 513-526.
28.
Tong, H. (2002). Generalized ridge estimate of parameters in variance component model. Communications in Statistics Theory and Methods, 31, 119-128
