squirtyflower
Well-Known Member
Only on bluemoon would posters be arguing on which average should be used rather than discussing the findings
Median isn't the mid-range sir baconfaceMode is the most commonly occurring value. Median is mid-range. Mean is the sum total divided by the number of cases.
Mean and median are both valid measures in this context. As shown above, are very similar in City's case. What the median does is reduce the effect of rogue outlying values. If City had a centenarian, for example, the mean average would rocket, whereas the median wouldn't be affected.
Fortunately we've moved our geriatrics on, so this is a hypothetical example.
Median isn't the mid-range sir baconface
Yeah, but that's still not the mid-rangeIf you array a load of numbers in ascending value, it's the middle number. Half are below it and half above. That's what I meant by mid-range, without boring everyone about the exact calculation.
Only on bluemoon would posters be arguing on which average should be used rather than discussing the findings
As my wife was a Maths teacher I'm actually impressed how many are able to discuss more than one type of averageOnly would a mod think footballers' ages are more interesting than debating alternative measures of distribution. ;)
Yeah, but that's still not the mid-range
As my wife was a Maths teacher I'm actually impressed how many are able to discuss more than one type of average
Well the median isn't a measure of distribution either. It's just a big arrow pointing to the middle of a data set.It is in the sense of weight of numbers.
Presumably you are saying add the smallest value to the largest and divide by two? If so, technically correct I grant you, but that's not a measure of distribution.
The arithmetic mean works well hereIn normally distributed data the mean and median are usually the same or approximately so.
(we should of course decide whether to use the arithmetic mean the geometric mean or the harmonic mean ;-)