Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I studied in East Europe and post Soviet mathematical education program have no Median, Mode, and Range terms.

Mean (or average) on other hand was studied (with root mean square and sometimes with geometric mean).

Looking to education English sites I see a lot of lessons about Median, Mode, and Range.

These statistical parameters are strange for me. Why west schools use them?

Have Median and Mode any sense in mathematical statistics (in science)?

share|improve this question
5  
Mode, like all too many topics, is taught mainly so that one can ask test questions about it. Median is an important notion, which happens to be mathematically less pleasant than mean, but ought to be taught. –  André Nicolas Aug 1 '13 at 14:24
    
As long as you accept them with their statistical definition, these are good statistics; real starting values might be worse, and your derived statistics worse as well. Sometimes you need to keep track of the error and not only the result. But you'll learn that with experience. –  Eric Tressler Aug 1 '13 at 14:34

2 Answers 2

For certain data sets, such as income, mean can be strongly affected by outliers, while median is much more robust and better represents the data. More generally robust statistics investigates data sets that are not normally distributed.

Perhaps a regime that has political interest in having income appearing normally distributed, would prefer to not consider this example.

share|improve this answer
    
Ah,sorry, had in mind the same example as yours –  Evgeny Aug 1 '13 at 14:20
    
That's the usual example since most things are normal... –  vadim123 Aug 1 '13 at 14:20

Of course they have their use. For example, roughly speaking mean and median serve the one purpose: to find the most typical value from dataset. But when median is more robust to so called outliers, mean is very sensitive to them. Suppose you have a dataset with hundreds of value around, say, 1, and one value around million. Mean will show value that is much higher than real typical value, but median will catch such situation. That's why typical salaries are usually computed with average, not median :)

Of course, it's all roughly speaking and specialists will gave deeper answer.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.