I know that SMD is preferred when the variables have different measurement units, but I am not sure. In which situations, standardized mean differences should be preferred instead of mean differences in the meta-analyses?
Hi Ömer Hatipoğlu, SMD is preferred when one wants to meta-analyse an outcome that is measure using the different scales in different studies included in a systematic review. Like knee pain can be measured using various patient-reported outcome scales (PRO) such as WOMAC, NRS, VAS, etc. The outcome result from these scales cant be combined as they have different anchor points like 0-100 mm for VAS with 0= no pain and 100= maximum pain, similarly 0-9 points for WOMAC with 0= no pain and 9= maximum pain.
However if one still wants to combine these studies in one meta-analysis then SMD needs to be used as it standardises the outcome measure and the variation in the anchor point (i.e. 0-100 or 0-9) no longer makes any difference.
There are many examples of this, one is here: https://www.kjim.org/journal/view.php?number=170530
Hope this helps!
KR
Ambrish
http://eprints.whiterose.ac.uk/109207/1/WalwynMeta-analysis%20of%20standardised%20mean%20differences.pdf
Hi Ömer Hatipoğlu,
Pooled mean differences can be computed when every study in the analysis measures the outcome on the same scale or on scales that can be easily converted. For example, total weight can be pooled using mean difference even if different studies reported weights in kilograms and pounds; however it is not possible to pool quality of life measured in both Self Perceived Quality of Life scale (SPQL) and the 36-item Short Form Survey Instrument (SF-36), since these are not readily convertible to one format.
Sometimes different studies will assess the same outcome using different scales or metrics that cannot be readily converted to a common measure. In such instances the most common response is to compute a standardized mean difference (SMD) for each study and then pool these across all studies in the meta-analysis. By dividing the mean difference by a pooled estimate of the standard deviation, we theoretically put all scales in the same unit (standard deviation), and are then able to statistically combine all the studies.
For more- https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/methods-guide-quantitative-synthesis-update.pdf
Hope this was helpful.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis