Learn R Programming

metaSEM (version 1.4.0)

Bornmann07: A Dataset from Bornmann et al. (2007)

Description

A dataset from Bornmann et al. (2007) for three-level meta-analysis.

Usage

data(Bornmann07)

Arguments

Details

The variables are:

ID

ID of the study

Study

Study name

Cluster

Cluster for effect sizes

logOR

Effect size: log odds ratio

v

Sampling variance of logOR

Year

Year of publication

Type

Type of proposal: either Grant or Fellowship

Discipline

Discipline of the proposal: either Physical sciences, Life sciences/biology, Social sciences/humanities or Multidisciplinary)

Country

Country of the proposal: either the United States, Canada, Australia, United Kingdom or Europe

References

Cheung, M. W.-L. (2014). Modeling dependent effect sizes with three-level meta-analyses: A structural equation modeling approach. Psychological Methods, 19, 211-229.

Marsh, H. W., Bornmann, L., Mutz, R., Daniel, H.-D., & O'Mara, A. (2009). Gender Effects in the Peer Reviews of Grant Proposals: A Comprehensive Meta-Analysis Comparing Traditional and Multilevel Approaches. Review of Educational Research, 79(3), 1290-1326. doi:10.3102/0034654309334143

Examples

Run this code
# \donttest{
data(Bornmann07)

#### ML estimation method
## No predictor
summary( meta3L(y=logOR, v=v, cluster=Cluster, data=Bornmann07) )

## Type as a predictor
## Grant: 0
## Fellowship: 1
summary( meta3L(y=logOR, v=v, x=(as.numeric(Type)-1),
               cluster=Cluster, data=Bornmann07) )

## Centered Year as a predictor
summary( meta3L(y=logOR, v=v, x=scale(Year, scale=FALSE),
               cluster=Cluster, data=Bornmann07) )

#### REML estimation method
## No predictor
summary( reml3L(y=logOR, v=v, cluster=Cluster, data=Bornmann07) )

## Type as a predictor
## Grants: 0
## Fellowship: 1
summary( reml3L(y=logOR, v=v, x=(as.numeric(Type)-1),
                cluster=Cluster, data=Bornmann07) )

## Centered Year as a predictor
summary( reml3L(y=logOR, v=v, x=scale(Year, scale=FALSE),
                cluster=Cluster, data=Bornmann07) )

## Handling missing covariates with FIML
## MCAR
## Set seed for replication
set.seed(1000000)

## Copy Bornmann07 to my.df
my.df <- Bornmann07
## "Fellowship": 1; "Grant": 0
my.df$Type_MCAR <- ifelse(Bornmann07$Type=="Fellowship", yes=1, no=0)

## Create 17 out of 66 missingness with MCAR
my.df$Type_MCAR[sample(1:66, 17)] <- NA
summary(meta3LFIML(y=logOR, v=v, cluster=Cluster, x2=Type_MCAR, data=my.df))

## MAR
Type_MAR <- ifelse(Bornmann07$Type=="Fellowship", yes=1, no=0)

## Create 27 out of 66 missingness with MAR for cases Year<1996
index_MAR <- ifelse(Bornmann07$Year<1996, yes=TRUE, no=FALSE)
Type_MAR[index_MAR] <- NA

## Include auxiliary variable
summary(meta3LFIML(y=logOR, v=v, cluster=Cluster, x2=Type_MAR, av2=Year, data=my.df))
# }

Run the code above in your browser using DataLab