Meta-analysis - Can I investigate mediation effects for the effectiveness of training

#1
Hello everyone,

I am currently working on my Master thesis on moderators and mediators for the effectiveness of cross-cultural training (CCT). I am actually writing a literature review, but want to end my thesis by judging to what extent the included studies could also be compared statistically via a meta-analysis. I understand how moderating effects can be examined. But I am not sure whether I can prove a mediation effect across various studies.

Many of the researchers in the field believe that CCT can only improve the performance of trainees if it enhances their adjustment in the first place. I therefore hypothesize adjustment as mediator for the CCT-performance relation. This was already investigated by two of the 23 studies in my sample.
My question is: Is it possible to prove the mediation effect of adjustment for all the studies that measure the impact of CCT on adjustment and performance as well as the correlation between adjustment and performance. And if yes, how can I do this?

Thank you so much for your help!
 
#2
Meta –analysis is the statistical procedure for combining data like quantitative , formal, epidemiological from multiple studies . When the effect size is consistent from one study to the next , meta -analysis can be used to identify this common effect meta –analysis is used to evaluate the effect that is described by a no. of publication (“Program Abstracts from the GSA 2019 Annual Scientific Meeting ‘Strength in Age—Harnessing the Power of Networks,’” 2019) .In the step by step method review researches focus on outcome of the study not on the raw data. For example , people having good and peaceful environment at home are have more chances to become a better and successful person in life in comparison to those who have disturbed environment(Cobb, 2013).
Meta -analysis and Mediation effects for the effectiveness of training
It has often reported that the flexibility of the body will decrease with the growing age. Dancing can help improve flexibility, but these improvements could depend on cognitive functions and age groups(Hofstede, 2011).
For example , let 100 people take part in a dancercise training for three months for the age group of above 60 and below 60.
In 60+Training predicted improved muscle movement speed (x=-.32) and cardiorespiratory fitness (y=.25)and the effect of training on muscle movement was fully mediated by cardiorespiratory fitness(xy=-.11) . in <60, training predicted improvement in flexibility (x=-.25)and change in cardiorespiratory fitness(y=.28), but improvement in flexibility was not fully mediated by change in cardiorespiratory fitness (Ur Rehman, Bhatti, & Chaudhry, 2019).
Conclusion
Results are discussed in terms of increase in flexibility following dancercise training in older adults.
 

hlsmith

Not a robit
#3
I have not heard of examining mediated effects in meta-analyses. This does not mean it does not exist or could not be done. However, results are conditional on the model used and the statistical method. It seems feasible to do a meta-analysis on mediating effects given the mediator was examined by all the original researchers in the individual published studies and the models were comparable between the studies, for what they controlled for. Mediated effects are usually deconstructed from the direct effects, and without access to the original data you probably cannot get at them. If the scenario is that the studies conducted and reported the mediated effects, then you could likely conduct a meta-analysis of the mediator effects pooled from the studies - seems easy enough.

However, I am guessing you have already graduated given the age of this message. So congratulation!