

GMM Introduction & Sample Size 

Message/Author 


I am interested in investigating the potential heterogeneity of mental health trajectories for a sample of refugees resettled in the United States. I will have access to monthly mental health screening data for a sample of 110 refugees over the course of 1 year (i.e., 12 data points per individual). Each screen results in a score between 1.00 and 4.00. I will also have a number of predictor variables, including sex, age, country of origin, education level, English level, preflight SES, degree of social support…and more. Some of these are categorical and others discrete variables. I would like to get a sense of the number of distinct trajectories of change within this sample and the predictors of class membership. My sense from reading the literature is that Mplus is uniquely designed to handle exactly this type of analysis. How can I determine if this sample size is sufficient? While the workshop videos are helpful, I find myself wishing I had an even more basic introduction to the program and in particular how to get started with preparing data for entry into Mplus and beginning to learn the syntax coding. Thank you for your response. 


Take a look at the following paper: Muthén, L.K. & Muthén, B.O. (2002). How to use a Monte Carlo study to decide on sample size and determine power. Structural Equation Modeling, 4, 599620. Use Example 12.3 as a starting point. It is what you want if you use the same model for data generation and the analysis. 

Back to top 

