Chapter 4 Meta-analysis in R

Learning outcomes

1. Format data for meta-analyses in R.
   
2. Explore the capacity of the R package metafor.
   
3. Pilot statistics for two datasets.

Context

Try out these ideas in a meta-analysis. Securing derived data to replicate an existing synthesis was not common historically, but it is now becoming increasingly viable with open science influences on these contributions. How to do a meta-analysis is well described in the peer-reviewed literature (Field and Gillett 2010) and numerous books to name a few (Koricheva, Gurevitch, and Mengersen 2013). Doing meta-analyses using the R programming language is similarly well articulated - particularly from the documentation associated with the package metafor (Viechtbauer 2017). There are no bad choices in R with well over 100 packages associated with and supporting meta-analyses conservatively listed at 151 packages. The output of Stata (an application specific to many medical meta-analyses), the R package meta, and metafor were virtually identical in several test cases (C. J. Lortie and Filazzola 2020). Ten criteria are proposed in contrasting R packages for this task specifically, but at the current time, metafor is the most commonly used and extensively documented. Hence, consider tackling the challenges here with this package as a robust starting point and entry in meta-statistics.

The 5 primary steps for meta-analyses in R.

1. Secure primary data.
   
2. Build conceptual model for factors, reponses, and moderators.
3. Calculate effect size(s).
   
4. Fit appropriate meta-model.
   
5. Explore significance levels, heterogeneity, and bias.

Challenge

1. Download a dataset and explore applying metafor to the data.
2. Check the published report and examine whether you reached reasonably similar conclusions.
   
3. Repeat with a second example from another discipline and consider testing another metafor function.

discipline	dataset	publication
environmental science	Wind turbine effects on birds 2005 synthesis	Effects of wind turbines on bird abundance
environmental science	Wind turbine effects on birds and bats data 2017 synthesis	Bird and bat species global vulnerability to collision mortality at wind farms revealed through a trait-based assessment
health	Preventing exercise-induced bronchoconstriction data	Nedocromil sodium for preventing exercise-induced bronchoconstriction
community ecology	A synthesis shrub facilitation studies testing for increases in community diversity	Too much of a good thing: Shrub benefactors are less important in higher diversity arid ecosystems
education	Education data meta-analysis 2020	The Power of Feedback Revisited A Meta-Analysis of Educational Feedback Research

Products

1. Two R scripts for meta-analysis.
   
2. A sense of data structures needed for meta-analyses in the R package metafor.

Resources

metafor documentation
Bookdown free books

Reflection questions

1. Did the analytical process differ significantly from primary-data workflows?
   
2. Do meta-models in R sufficiently report outcomes to assess strength of evidence?
   
3. What other steps would be an appropriate addendum to this process?
   
4. What relational or qualitative elements might be worthwhile to consider adding to meta-analyses and their interpretation for stronger evidence framing and reuse?