What is it?
MetaNSUE is a meta-analytic method...
... that allows an unbiased inclusion...... of studies with Non Statistically-significant Unreported Effects (NSUEs)
Why is this important?
A number of published studies may not report any statistics (e.g. t- or p-values) from those findings that did not reach statistical significance. They just specify that the effects were not statistically significant.
Using simple simulations, it can be shown that exclusion of these studies leads to strong positive biases, whereas considering these effects as null leads to negative biases.
MetaNSUE uses maximum likelihood estimation and multiple imputation techniques to unbiasedly include these studies. Please read (and cite appropriately) the following two articles for details:
Original method: Radua J, Schmidt A, Borgwardt S, Heinz A, Schlagenhauf F, McGuire P, Fusar-Poli P.
Ventral Striatal Activation During Reward Processing in Psychosis: A Neurofunctional Meta-Analysis.
JAMA Psychiatry 2015;72(12):1243-51
Current method: Albajes-Eizagirre A, Solanes A, Radua J.
Meta-analysis of Non-statistically Significant Unreported Effects (MetaNSUE).
Statistical Methods in Medical Research in Press
How do I use it?
To use MetaNSUE, install first the package "metansue" in R (freely available for Linux, Windows or Mac) with the following command:
You can also download it at the CRAN package repository.
Please read the manual for more information.
You can use it for free - but please cite our work ;-)
Please contact Joaquim Radua at radua at clinic dot cat.