Meta-Analysis Resources

Tools for Those Who Summarize the Evidence Base

Resources and networking for those who conduct or interpret meta-analyses related to any phenomenon that is gauged in multiple studies.

David B. Wilson has put a new online effect size calculator in at least two spots (here, hosted on Dave's "meta-analysis stuff" site at George Mason; and here, hosted at the Campbell Collaboration). (Thanks to Seth Noar for drawing the former version to my attention and Janet Swim for drawing my attention to the latter.) These are nicely enhanced from the version that Dave posted after publication of their very useful book, Practical Meta-Analysis, in 2001; there are extensive and elaborate procedures available there, although I'd be curious to hear site members' reactions to the solutions. One thing that I offered in my (now ancient!) software was the ability to re-construct ANOVA tables; I'd be curious as well to see if anyone has seen that function included in any software or website.

Another reason to like Dave's work here is that it facilitates finding errors in reports. Any solution for effect size (ES) calculation should routinely examine multiple sources of information available in any study report regarding a particular effect size to make sure that they converge on the right answer. Of course, coarser information (e.g., "p<.001") should be discarded in favor of more precise information (e.g., means and standard deviations). But the precisely reported information should be compared: An F-test (or t-test) comparing 2 means of 2 groups should provide the same ES as the means and standard deviations, for example.

Views: 1240

Replies to This Discussion

Note that David Wilson has updated his effect size calculator. The link is here.


© 2022   Created by Blair T. Johnson.   Powered by

Badges  |  Report an Issue  |  Terms of Service