1
$\begingroup$

EDIT: I now believe this question was based on misguided information. See my answer.


I'm looking through this meta-analysis (of studies on the effects of mindfulness on healthy individuals) for some personal research.

https://sci-hub.se/10.1016/j.jpsychores.2015.03.009

On page 6 of the document (and in the table on page 7), it gives analysis of heterogeneity 'within-group (pre post analyses)' and 'between group'.

Could you explain what these terms mean, in layman's language?

I understand that in general, heterogeneity between studies included in a meta-analysis means (basically) that the features of the studies included are quite varied (e.g. demographics studied, outcomes measured, measurement criteria used, etc.)

So does heterogeneity between groups mean that... for the subgroups in the metaanalysis, there was a lot of heterogeneity between the studies that contributed data to those subgroups? What about heterogeneity within groups?

Thank you!

$\endgroup$
1
$\begingroup$

I am afraid the article is quite unreadable here as much of my screen is taken up by a girl waving at me so my answer is based on general principles.

If the studies fall into sub-groups then you can look at heterogeneity within sub-groups. This may be less than overall heterogeneity if the sub-groups differ. The amount of heterogeneity may differ between sub-groups as well.

There may also be variability between the sub-group estimates as well. In fact that is presumably why the sub-groups were chosen.

Note that you are describing heterogeneity as a clinical concept rather than just as a purely statistical one. There may be clinical heterogeneity which in fact does not lead to statistical and vice versa.

| cite | improve this answer | |
$\endgroup$
0
$\begingroup$

I have realised that the question was misguided.

I believe this particular meta-analysis uses the term 'within-group' to refer to studies that followed one group (and did pre-post analysis to examine the potential effects of the mindfulness intervention).

And it uses 'between group' to refer to studies that compared the effects between two (or more) different groups - e.g. an intervention group and a control group.

Therefore when it refers to heterogeneity in within group studies, it is referring to heterogeneity among studies that only looked at one group, and when it refers to heterogeneity in between group studies, it is referring to studies that compared the effects on more than one group (probably an intervention and control group).

| cite | improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.