MA_effectsTable {metagear}R Documentation

Generate an ANOVA-like effects table for a meta-analysis.

Description

Generates an ANOVA-like effects table that summarizes the within and between-study homogeneity tests (Q-tests), as well as moderator level Q-tests as originally described by Hedges and Olkin (1985; p. 156).

Usage

MA_effectsTable(model, weights, data, effects_model = "random")

Arguments

model

A two-sided linear formula object describing the model, with the response (effect sizes) on the left of a ~ operator and the moderator variables, separated by +, :, * operators, on the right.

weights

A column label from data.frame of variances to be used as weights.

data

An optional data frame containing the variables named in the model.

effects_model

The default is "random", which specifies a random-effects meta-analysis (DerSimonian and Laird method). Other options include "fixed" which presents fixed-effect analyses.

Value

An lm object of main effects.

References

DerSimonian, R., and N. Laird. 1986. Meta-analysis in clinical trials. Controlled Clinical Trials, 7, 177-188.

Hedges, L.V., and I. Olkin. 1985. Statistical methods for meta-analysis. Academic Press, New York, USA.


[Package metagear version 0.7 Index]