I am pooling the effect size of a specific intervention. I am calculating the mean difference and 95% confidence interval scores. I came across an outcome measure expressed in Meters in some of these studies and expressed in Feet in one study.
Is it correct if I convert the mean and the standard deviation to Meters and I calculate the mean difference or should I calculate the standard mean difference as an alternative?
Hi Alaa,
the convention is to use the standardizd mean difference (e.g., Cohen's d or Glass' g).
Best,
Holger
Even if it is the same outcome measure? To be more specific the outcome measure I am referring to is the 6 minute walk test. In the other studies it is expressed in meters however in one study it is expressed in feet.
That's why you calculate the *standardized* mean difference. This mean difference is expressed in terms of *difference in SD's*
Best,
Holger
Yes, you can convert meters to feet if you wish to report the (unstandardized) mean difference. Sometimes it is easier to explain the relevance of the results (decision-makers, lay people) by using unstandardized rather than standardized effect sizes.
You could also compute and report both standardized and unstandardized effect sizes. Some researchers have used this approach in the past:
Large, M., Sharma, S., Compton, M. T., Slade, T., & Nielssen, O. (2011). Cannabis use and earlier onset of psychosis: a systematic meta-analysis. Archives of general psychiatry, 68(6), 555-561. doi:10.1001/archgenpsychiatry.2011.5
Wagenmakers, E. J., Beek, T., Dijkhoff, L., Gronau, Q. F., Acosta, A., Adams Jr, R. B., ... & Bulnes, L. C. (2016). Registered Replication Report: Strack, Martin, & Stepper (1988). Perspectives on Psychological Science, 11(6), 917-928. Article Registered Replication Report: Strack, Martin, & Stepper (1988)
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis