Contents: The workshop will focus on the following topics:

1. Review of the Cohen's framework and key concepts – the overall goal of power analysis is to find an appropriate balance among four factors: statistical significance or probability of making type I error, power or probability of rejecting a false hypothesis, the minimum detectable effect of an experiment or effect size, and needed sample size.

2. HLM – the specific frameworks of power analysis for HLM were developed by Raudenbush and his colleagues, including optimal design for cluster randomized trial (Raudenbush, 1997), for multisite randomized trials (Raudenbush & Liu, 2000), and for repeated measures (Raudenbush & Liu, 2001). Statistical precision is a key factor affecting power of HLM. We will demonstrate how to conduct these analyses using free software OD (http://www.ssicentral.com/other/hlmod.htm).

3. EHA – the specific framework of power analysis for EHA was developed by Collett (1994). In EHA, effect size is defined as hazard ratio (i.e., the ratio of hazard rate of one group over that of the other group). We will demonstrate how to conduct the analysis using free software PS (http://biostat.mc.vanderbilt.edu/twiki/bin/view/Main/PowerSampleSize).

4. SEM – among several frameworks of power analysis for SEM, the most popular one was developed by MacCallum and his colleagues (1996), which evaluates power of SEM based on effect size of RMSEA for close fit, for not-close fit, and for comparing nested models. We will demonstrate how to conduct these analyses using the SAS syntax developed by MacCallum.

Pedagogical Techniques: Teaching methods include lecture, PowerPoint presentation, and computer demonstration.

References

Collett, D. (1994). Modeling Survival Data in Medical Research. London: Chapman & Hall. Chapter 9.

Raudenbush, S.W. (1997). “Statistical analysis and optimal design for cluster randomized trials”, Psychological Methods 2(2): 173-185.

Raudenbush, S.W., & Liu, X.F. (2000). “Statistical power and optimal design for multisite randomized trials”, Psychological Methods 5(2): 199-213.

Raudenbush, S.W., & Liu, X.F. (2001). “Effects of study duration, frequency of observation, and sample size on power in studies of group differences in polynomial change”, Psychological Methods 6(4): 387-401.

MacCallum, R.C., Browne, M.W., & Sugawara, H.M. (1996). “Power analysis and determination of sample size for covariance structure modeling”, Psychological Methods 1(2): 130-149.