PLEASE NOTE:
When you click 'Reply' to any message it will be sent to all RAMESES List members.
If you only want to reply to the sender please remove [log in to unmask] from the 'To:' section of your email.
Dear Anyone,
This is the first time I have posted on this - so apologies if this is an old topic or a daft question....but i am trying to write a protocol for a grant application for a realist evaluation. I am curious about the use of mediation and / or moderator analyses to add an additional dimension to qualitative exploration of CMOs, and what might be the pros and cons of doing this
Pros
Adds another dimension to analysis that could be used to triangulate qual findings
Allows relative strengths of different CMOs to be explored
Cons
often mechanism is not a directly measurable "variable"
implies successtionist view of CMO rather than generative causation
however to mitigate the cons, you could
a. Only test CMOs where the M is a variable for which a measure existed (e.g. hope), or where a latent variable could be created from group of variables, that map onto this mechanism. So not attempting to test more complex social processes that it wouldn;t make sense to measure in this way
b. accept that the quan approach will always be limited in this sense and should only ever be secondary to qualitative understanding....but could add something?
Really grateful for anyone who can add some thoughts to this and let me know if I am likely to kill my protocol dead in the water if I put any statistical analysis in it? My reading suggests this might be an issue that divides reviewers?
best wishes
Fi
To UNSUBSCRIBE please see: https://www.jiscmail.ac.uk/help/subscribers/faq.html#join
|