[SUZY] So, a doctor just told me...

[Tugela]

hey tug, i'm not sure what you are saying here. i always thought ages were data, and that the calculation of central tendencies was very straightforward, but i'm always open to being re-educated

http://en.wikipedia.org/wiki/Central_tendency

No, the age isn't the data, the n numbers are the data. Ages are just a property used to group the data.

For example, you might use individual years to group people, but there is nothing intrinsically special about a year. You could equally well group them as 0-5, 6-13, 14-25, 26-45, 46-77, 78-100. What is important is the n number associated with each group, the property of the group is not what is being averaged.

 

[Tugela]

actually, using your example, the average age would be either 1 or 2 depending on which method you use the mean average being 2 yrs old ,and both median and mode being 1 yr old

Ok, do the same thing with 9 1 year olds and 1 100 year old, all expressing infantile behaviour. Can you conclude then that the average 11 year old is infantile? No. And the reason would be a fundamental misunderstanding of what statistics actually looks at. The error you are making is ignoring the n number and focussing on the property.

As I said before, replace the property "age" with something else, such as arbitrary colors. If you have a collection of 10 red boxes and 9 blue boxes and 1 green box you are not going to say that the average box is green, because it obviously is not. You can blend the n numbers, but you cannot blend the properties of the grouped objects. So why would you do that when the property is age? You are looking at the wrong thing.

 

[Horse99]

The median and average are two very differing things in statistics.....the median, is the midpoint of the sample numbers.....for example, 13,13,40.....the median is 13, as that is the middle number in the series....the average number is the average.....the sum of the sample numbers divided by the number in the sample......there has been some confusion by using the median number in correctly earlier in the thread.

Does anyone want to talk standard deviations? It was a long time ago when I suffered through this stuff in college.


My previous posting is confusing because I copied and pasted a someone else's comment without putting it in italics.....and therefore the quote is unattributed to the original author. Posting #36 by tugela is what I was referencing.....and trying to correct....I read since then numerous posts correcting the original error.