It is possible to express the individual study effects in mean/SD units.

However this is not generally recommended as it can make identical results spuriously appear to be different and can even reverse the order of strength of effect. It would be better to try and find a unit that is uniform across studies, not in SD units. tou mayneed to look closely at data provided in the study reports and/or seek information from original authors.

Thank you, James.

In the case of expressing the individual study effects in mean/SD units, how I would obtain the sd for each study to enter this data into the meta-analysis? Would I maintain the standard deviations of the studies? For example:

Study 1 (scale A): mean = 10; SD=2; n= 100

Study 2 (scale B): mean = 25; SD=10; n= 200

Standardisation:

Study 1 (scale A): mean = 10/2; SD=2; n= 100

Study 2 (scale B): mean = 25/10; SD=10; n= 200

Thanks!

• The use of a population-based study minimized selection bias
• The sample would have been decreased to non-response bias
• The sample was prone to volunteer bias
• The results of the study were prone to ascertainment bias

yes, use study sds if available or calculable