STATA Meta-Analysis Macros
Written by David B. Wilson
dwilson@crim.umd.edu

Included are three macros:

	meanes.ado
	metaf.ado
	metareg.ado

To use, place these files either in a directory already part of the "adopath" or in their own directory.  If the latter, add the directory to the adopath with the "adopath +" command.

MEANES

The "meanes" command performs basic central tendency statistics, such as mean effect size, confidence intervals, z-test, and homogeneity test.  It calculates both fixed and random effects models.

Usage: meanes <es_variable_name> [w=<weight_variable_name>]

METAF

metaf.ado performs a test of the differences in the mean effect size across the levels of a categorical variable, similar to a one-way analysis of variance.  It has an optional ", model()" command to specify a fixed effect (default), method of moments estimated, model(mm), maximum likelihood estimated model(ml), or restricted maximum likelihood estimated random effects model.

Usage: metaf <es_variable_name> <group_variable> [w=<weight_variable_name>]

Usage: metaf <es_variable_name> <group_variable> [w=<weight_variable_name>], model(ml)

METAREG

metareg.ado performs meta-analytic regression analysis.  It has an optional ", model()" command to specify a fixed effect (default), method of moments estimated, model(mm), maximum likelihood estimated model(ml), or restricted maximum likelihood estimated random effects model.

Usage: metaf <es_variable_name> <list of IVs> [w=<weight_variable_name>]

PRINT OPTION

meanes and metaf have an optional ", print()" statement for controlling the output and are useful if you are analyzing Zr transformed correlations or logged odds-ratios.  The "ivzr" option converts the result back into correlation coefficients (inverse Zr transformation).  The "exp" statement prints the exponent of the results, that is, it prints odds-ratios if the input effect sizes were logged odds-ratios.

Usage: metaf <es_variable_name> <list of IVs> [w=<weight_variable_name>], print(exp)
