There are four types of central tendency measures - the arithmetic mean, the weighted mean, and the median - which are always used in scientific research
Hi. There is no best method of measuring the central tendency. It much depends on your data and the distribution of your data. If your data follows a Gaussian distribution (normal distribution), mean, median and weighted mean all can be used, as in a perfectly normally distributed data, all these are same. In a scientific experimental scenario, however, such a data is rarely seen. So if your data is not exactly gaussian, but very close to it, it is still recommended to use mean. In such cases, you use parametric analysis of statistics.
On the other hand, if your data is skewed or does not follow a normal distribution, median will be a much better estimate of central tendency and in such cases, non parametric statistics may be used.
Please refer to the book by Jerrold Zar on Biostatistical analysis
"The median is one of the best of these measures, because the median is the average value between the values after ranking them in ascending or descending order, therefore, the median always exists and is single and is not affected by extremism values and then from each other."