Linear mixed models involve data in which the observations are dependent. Linear mixed models provide better support analysis for continuous dependent data.
The following are the types of models:
Random Effects model: In random effects model, the set of values of a categorical independent variable are observed as a random set of values. From random effects model in linear mixed models, inferences over a wider population are obtained.
Hierarchical linear effects model: For Hierarchical linear effects in linear mixed models, the measurement of independent variables is done at more than one level.
Repeated measures model: For repeated measures in linear mixed models, the observations are correlated among themselves, rather than being independent.
Linear mixed models include a variety of multi-level modeling, which includes hierarchical linear models, random coefficients models, etc.
Multi-level mixed models in linear mixed models are based on a multi-level theory that specifies the expected direct effects of variables on each other, which are located within one level. Multi level mixed models in linear mixed models also specify cross-level interaction effects between the variables, which are located at different levels.
There are certain terminologies in linear mixed models that will help us to better understand linear mixed models:
Hierarchical data in linear mixed models involve measurement at various levels, like on the basis of individuals or on the basis of groups, etc. Hierarchical data in linear mixed models are normally obtained by conducting multistage sampling.
Centering in linear mixed models is necessary. Centering in linear mixed models is a process of centering the data (i.e., subtracting the mean from the data).
Subject Variables in linear mixed models are those variables which are generally found in the case of psychology research. Examples of such variables in linear mixed models are age, sex, height, etc.
Repeated measure variables in linear mixed models are those variables in which the observations occur repeatedly.
Fixed factors in linear mixed models are those factors which are fixed and do not change. For example, if a person’s religion is Hinduism, then it means that this factor is fixed. These factors in linear mixed models are category variables (i.e. they are in the form of categories).
Random factors in linear mixed models are those factors that vary, like the price values in a time-series that are random in nature. These factors in linear mixed models are category variables (i.e., they are in the form of categories).
There are certain assumptions of linear mixed models. These assumptions are as follows:
- For Multi-level mixed models in linear mixed models, the grouping is done randomly.
- Independent observations are not assumed in linear mixed models.
- Though observations in linear mixed models are not independent, the blocks formed by the subject variables are supposed to be independent.
- An adequate number of sample sizes are assumed in linear mixed models.
- The block structures are specified properly in linear mixed models, especially for random effects and the repeated measures model.
- Random coefficients model, in linear mixed models, assumes a normal distribution.
Linear Mixed Models Resources
Booth, J. G., & Hobert, J. P. (1998). Standard errors of prediction in generalized linear mixed models. Journal of the American Statistical Association, 93(441), 262-272.
Breslow, N. E., & Clayton, D. G. (1993). Approximate inference in generalized linear mixed models. Journal of the American Statistical Association, 88(421), 9-25.
Bryk, A. S., Raudenbush, S. W., & Congdon, R. H. (1996). Hierarchical linear and nonlinear modeling with the HLM/2L and HLM/3L Programs. Chicago: Scientific Software International.
Burton, B. (1993). Some observations on the effect of centering on the results obtained from hierarchical linear modeling. Washington, DC: National Center for Education Statistics.
Goldstein, H. (1991). Nonlinear multilevel models with an application to discrete response data. Biometrika, 78(1), 45-51.
Hoffman, D. A., & Gavin, M. B. (1998). Centering decisions in hierarchical linear models: Implications for research organizations. Journal of Management, 24(5), 623-641.
Jiang, J. (2007). Linear and generalized linear mixed models and their applications. New York: Springer.
Kreft, I. G. G., de Leeuw, J., & Aiken, L. S. (1995). The effect of different forms of centering in hierarchical linear models. Multivariate Behavioral Research, 30(1), 1-21.
McCulloch, C. E. (1997). Maximum likelihood algorithms for generalized linear mixed models. Journal of the American Statistical Association, 92, 162-170.
McCulloch, C. E. (2003). Generalized linear mixed models. Beachwood, OH: Institute of Mathematical Statistics.
McDonald, R. P., & Goldstein, H. (1988). Balanced versus unbalanced designs for linear structural relations in two-level data. British Journal of Mathematical and Statistical Psychology, 42, 215-232.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage Publications.
SPSS. (2005). Linear mixed-effects modeling in SPSS: An introduction to the MIXED procedure. Chicago, IL: SPSS.
Sullivan, L. M., Dukes, K. A., & Losina, E. (2004). Tutorial in biostatistics: An introduction to hierarchical linear modelling. In R. B. D’Agostino (Ed.), Tutorials in biostatistics. Volume 2: Statistical modeling of complex medical data (pp. 35-126). Hoboken, NJ: John Wiley & Sons.
Verbeke, G., & Molenberghs, G. (1997). Linear mixed models in practice: A SAS-oriented approach. New York: Springer.
Verbeke, G., & Molenberghs, G. (2000). Linear mixed models for longitudinal data. New York: Springer.
West, B. T., Welch, K. B., & Galecki, A. T. (2006). Linear mixed models: A practical guide using statistical software. Boca Raton, FL: Chapman & Hall/CRC.
