Hi all, currently doing a systematic review and unfortunately a paper I'm extracting data from has provided median and SD but not the mean. They have not provided the range of values or IQR - is there any way to calculate the mean from just the median and SD?
Hello
Have the authors given results of Normality tests on the data under consideration? , or Normality graphical analysis results?. In case, the authors claim that their data is following Normal/ satisfying normality assumptions in the analysis conducted.... you can take the median as the value of mean under such circumstances.
In addition to what Debopam Ghosh said (which, unfortunately, is unlikely to be the case as normality tests are seldom reported or even performed), try to get in touch with the corresponding author for the values you are looking for or for the raw data so you can compute them yourself.
I agree with Sir Aedrian Abrilla's recommendation. In addition, the median was used instead of the mean may be due to the level of measurement of the data gathered.
Use this paper to compute mean from median and associated measures.
https://bmcmedresmethodol.biomedcentral.com/articles/10.1186/1471-2288-14-135
This calculator can help with conversions
http://vassarstats.net/median_range.html
Most of the clinical researchers don't even perform a normality test and forget about reporting it! You can look into this paper.Article Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and vari...
If you are "currently doing a systematic review" then this raises the possibility of transferring information from other studies to help with questions about distribution shape or skewness correction, depending of course on judgements of how similar the distributions are across studies and on how much information those studies provide,
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts