+ Reply to Thread
Results 1 to 3 of 3

Thread: Meta-analysis: number of effect sizes needed for meaningful moderator analysis

  1. #1
    Points: 22, Level: 1
    Level completed: 43%, Points required for next Level: 28

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Meta-analysis: number of effect sizes needed for meaningful moderator analysis




    Hi there,
    I would like to conduct a moderator analysis looking at the effects of six different moderators on the relationship between anxiety in children with dyslexia. The number of studies included in this review is small (N=14) and the number of effect sizes that would be included in each "moderator analysis" is 3-4 at most per moderator. I am reluctant to interpret the data using narrative synthesis as I would like to add weight to my conclusions using statistical data. But my question is, is it meaningful to conduct moderator analyses using such few effect sizes? I have a rule of thumb suggesting 10 studies per moderator should be included, but this estimate still seems conservative?
    Any suggestions on this would be greatly appreciated.
    Thanks,
    d

  2. #2
    Omega Contributor
    Points: 38,374, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Meta-analysis: number of effect sizes needed for meaningful moderator analysis

    Yes, it definitely seems like you have sparse data, especially if you are examining multiple moderators, since such an exploratory analysis may also require you to correct your level of significance (alpha) to minimize risk of false discovery.


    I am assuming when you say 'moderator' you are referencing the addition of an effect modifier (interaction term). Also, I didn't follow how this is related to a meta-analysis?
    Stop cowardice, ban guns!

  3. #3
    Points: 22, Level: 1
    Level completed: 43%, Points required for next Level: 28

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Meta-analysis: number of effect sizes needed for meaningful moderator analysis


    Hi,
    Thanks so much for your reply. I planned to do a subgroup analysis to explore potential moderators on the relationship between dyslexia and anxiety, with the subgroups being type of poor reading, type of anxiety, type of attention, age group and sex. This step essentially involves disaggregating the data following the overall meta analysis to see if any of the moderators can explain the overall relationship. And I would definitely have to correct the alpha level as well, perhaps a conservative .01.
    Thanks again,
    d

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats