Estimation of mean and standard deviation in meta-analysis in case of median, interval, and / or quartiles reporting

Meta-analysis enables researchers to combine and interpret the data of more than one study on a specific subject with a statistical method. One of the most important advantages of meta-analysis is that it increases the precision of the predictions by increasing the sample size. Two commonly used statistics in data analysis are the sample mean and standard deviation. However, in recent years, instead of directly specifying the sample mean and standard deviation, the studies include the median, first and third quartiles, minimum and maximum statistics. In order for researchers to benefit from the information of the studies reported in this way, they need to be converted into mean and standard deviation. In this thesis, transformation methods are discussed to convert the median, first and third quartiles, minimum and maximum statistics reported in the studies considered for meta-analysis to mean and standard deviation. These methods were analyzed under three main scenarios according to the reported statistics. In the application section, studies dealing with the psychiatric and neuropsychiatric consequences of the COVID-19 pandemic were examined for the age variable. If the age variable was reported aside from mean and standard deviation, it was converted to mean and standard deviation with appropriate transformation methods, and metaanalysis was applied to determine the mean age. In addition, transformation methods were compared. The analyzes were made by using the "metamean" function of the "meta" package of the R program.