Suppose we have two instruments (A and B) and we want to meta-analyze whether instrument B has incremental validity over instrument A in predicting an outcome Y. What effect size would be most appropriate in a meta-analysis to quantify the incremental validity of B? I have found two different approaches:
- Meta-analyze the beta coefficient of instrument B in the model: Y ~ A + B (see for instance: van den Berg et al., 2018)
- Meta-analyze the delta R^2 of the two models Y ~ A and Y ~ A + B (see for instance: Andrei et al., 2016)
Andrei, F., Siegling, A. B., Aloe, A. M., Baldaro, B., & Petrides, K. (2016). The incremental validity of the Trait Emotional Intelligence Questionnaire (TEIQue): A systematic review and meta-analysis. Journal of personality assessment, 98(3), 261-276.
van den Berg, J. W., Smid, W., Schepers, K., Wever, E., van Beek, D., Janssen, E., & Gijs, L. (2018). The predictive properties of dynamic sex offender risk assessment instruments: A meta-analysis. Psychological Assessment, 30(2), 179.
