I am doing a meta-analysis on GDM, I have values for Parity and Gravida in the median with IQR where the N is known. I want to convert them in mean with std dev.
Mathematically, it is possible. Assuming a normal distribution of the data, the mean equals the median and the standard deviation equals the interquartile range divided by 1.35. However, the median and interquartile range are usually not reported if the data are normally distributed. This question has been discussed before at length.
It can be done in several ways. On this webpage there two methods: The quantile estimation (QE) method or the Box-Cox (BC) purpose by McGrath et al. The ref.: Preprint Estimating the sample mean and standard deviation from commo...