An investigator gets a positively skewed data on account of having only a small number of simple numerical observations at extremely high values. Which measure of central tendency should he select if data is numerical?

| |