SAS GCM power - Zhiyong Zhang \'s Psychometric Website

My Research

BAUW

BMEM

Consulting & Collaboration

Dynamic Models

MedCI

SAS GCM power

SAS WinBUGS

WebStatR

Publications

Programs

In Progress

Favorites

Home

Category: SAS GCM power

Below are collections of SAS macros to calculate power for growth curve analysis with/without missing data. These macros are licensed under the GNL General Public License Version 2.0. You can use/modify/distribute those macros. For questions or comments, please contact Johnny Zhang at zhiyongzhang(@)nd.edu. It would be appreciated if you cite the macros in the following way:

Zhang, Z., & Wang, L. (2009). Power analysis for growth curve models using SAS. Behavior Research Methods, 41(4), 1083-1094. Request a copy Download a copy

The complete simulation results for Zhang & Wang (2009) can be found here Power plot final results online.htm.

Contents

Some selected software/programs for power analysis of growth curve models and multilevel models
Here is a list of software/programs we were aware of that can conduct power anlaysis for growth curve models/multilevel models.

SAS power analysis Macros for a latent basis growth curve model
The SAS macros estimate the power for testing nonlinear trajectory using the latent basis growth curve model. The macros ultilize both PROC MIXED and PROC NLMIXED to estimate statistical power.

SAS power analysis Macros for a quadratic growth curve model
The SAS macros estimate the power of testing whether there is a quadratic term in the model by comparing a quadratic growth curve model and a linear growth curve model.

SAS power analysis Macros for a linear growth curve model with group information
The macros estimate the power for a conditional linear growth curve model where the individual rates of change are predicted by the group information (covariates). The macros can be used to estimate the power of detecting the treatment effects.

The SAS power analysis Macros for an exponential growth curve model
The macros estimate the power for evaluting parameters in an exponential growth curve model. R=10000 are recommended. In this example, the mean change between the iniital status and the asymptotic level is tested. It is recommended that a single replication of data is first tested to observe the behavior of convergence. The PROC NLMIXED is used here. The power estimation for the exponential growth curve model can be very slow.

SAS power analysis Macros for a linear growth curve model
The macros estimate the power for evaluating the average rate of change. When there are missing data, R=5000 are recommended. The PROC MIXED is used to estimate the model parameters.

Total 6 ,Display 1 - 6

Maintained by Zhiyong Zhang