Skip to contents

Fit a multivariate meta-regression model

Source: R/filt_multivar_MA.R
fit_multivar_MA.Rd

Fit a multivariate meta-regression model that models the effect of peer-review ratings on the deviation from the meta-analytic mean (both continuous and categorical ratings), mean Sorensen's index, and/or whether the analysis uses a mixed effects model, or not.

Usage

fit_multivar_MA(data_tbl, N = 5, ..., env = rlang::caller_env())

Arguments

data_tbl

Data for model fitting

N

threshold for the number of analyses that must have been conducted using mixed effects models to include the binary predictor mixed_model in the meta-regression. Defaults to 5.

...

Additional arguments passed to lme4::lmer

env

Environment in which to evaluate the formula, defaults to the calling environment

Value

An object of class lmer.

Details

Depending on whether enough analyses in data_tbl have been conducted with the mixed_model variable, the function will fit a model with or without the predictor mixed_model.

Expects the following columns in data_tbl:

  • RateAnalysis: continuous peer-review ratings

  • PublishableAsIs: categorical peer-review ratings

  • mean_diversity_index: mean Sorensen's index

  • box_cox_abs_deviation_score_estimate: response variable, Box-Cox transformed deviation from the meta-analytic mean effect-size for each analysis

  • mixed_model: binary variable indicating whether the analysis used a mixed effects model or not

  • ReviewerId: reviewer identifier

  • one of study_id or id_col to uniquely identify each analysis for checking that the threshold N is met.

See also

Other Model fitting and meta-analysis: fit_MA_mv(), fit_boxcox_ratings_cat(), fit_boxcox_ratings_cont(), fit_boxcox_ratings_ord(), fit_metafor_mv(), fit_metafor_mv_reduced(), fit_metafor_uni(), fit_sorensen_glm(), fit_uni_mixed_effects(), poss_fit_boxcox_ratings_cat(), poss_fit_boxcox_ratings_cont(), poss_fit_boxcox_ratings_ord(), poss_fit_metafor_mv(), poss_fit_uni_mixed_effects()