A Guide to Conducting a Meta-Analysis with Non-Independent Effect Sizes.
Cheung MW. · 2019
Read A Guide to Conducting a Meta-Analysis with Non-Independent Effect Sizes. by Cheung MW. (2019) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc6892772.txt.
Description
Cheung offers a guide to a common statistical snag, in Neuropsychology Review. Meta-analysis pools effect sizes from many studies to reach one verdict. Standard methods assume those effects are independent. Often they are not: one study may report several related outcomes. Pretending otherwise can bend the pooled conclusion, sometimes badly.
Old fixes waste data, such as averaging a study's effects or picking just one. Cheung introduces two fuller methods, multivariate meta-analysis and the three-level model, which use every reported effect. Worked examples in the R language show how to run them. A third approach, robust variance estimation, is also explained. The guide aims at applied researchers rather than statisticians.
Catalog record
| CREATOR | Cheung MW. |
|---|---|
| YEAR | 2019 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |