Talk:Average absolute deviation

Average Absolute Deviation vs. Mean Absolute Deviation
This article is VERY poorly worded. In particular the "average absolute deviation" section claims that the 'average absolute deviation' is the same as the 'mean absolute deviation' which makes sense. However the following section is titled "mean absolute deviation" and gives an alternative definition which implies that the 'average absolute deviation' and the 'mean absolute deviation' are in fact different. The alternative definition also utilizes a variable 'Fi' which is not defined anywhere in the article. — Preceding unsigned comment added by 59.156.80.115 (talk) 05:27, 14 December 2011 (UTC)

I came to say the same thing. The two sections should be merged into one, and whoever created the secondmost of the two (rather than editing whatever was there first) must be shot at dawn. Glenbarnett (talk) 05:04, 11 July 2012 (UTC)

also Need to clarify what M means in "MAD" it's used for both mean absolute and median absolute. Please change the mean to AAD or only let it reference the one it actually represents... --184.158.56.62 (talk) 21:12, 10 November 2012 (UTC)


 * I've gone ahead and tried to clean up the section on the Average Absolute Deviations. It is still somewhat messy, but I no longer think there are duplicitous definitions. The article still seems unfocused, and could be better served by framing the uses of absolute deviations with contexts rather than just using formulas. Douglas Whitaker (talk) 18:53, 20 February 2013 (UTC)

The terminology usage was still poorly woven. I hope my updates are an improvement. LarryLACa (talk) 02:40, 24 August 2014 (UTC)

expected value?
As a summary statistic, this isn't an expected value because it's not in regard to a random variable, but the actual data. Should the "expected value" be removed, or should it be described aas the expected value of something else? Cretog8 (talk) 20:12, 2 July 2008 (UTC)

Contrast Section Needed
A contrast with standard deviation would help make this more clear for those less adept at math. I would add this, except I am less adept at math. 64.122.190.43 (talk) 16:36, 9 July 2010 (UTC)

A Greek Symbol for the Mean Absolute Deviation?
For standard deviation, the Greek letter, σ, is used in statistics. Is there a Greek letter conventionally used for the Mean Absolute Deviation? Roland 01:19, 29 May 2013 (UTC)

Assessment comment
Substituted at 21:35, 26 June 2016 (UTC)

Lead is confusing
This article suffers from WP:TECHNICAL and is not understandable to a general audience. For starters, the lead has two sentences that are confusing.


 * Furthermore, as described in the article about averages, the deviation averaging operation may refer to the mean or the median.

The article about averages mentions the word "deviation" once which is in the See also section link back to this article. I believe this sentence is trying to say the the word "average" in the phrase "Average absolute deviation" can refer to taking either the mean or median. However, if that's the case the sentence is redundant as the previous sentence is "In this general form, the central point can be the mean, median, mode, or the result of another measure of central tendency." Thus, I'm not sure why we have the "Furthermore, as described ..." sentence.


 * Thus the total number of combinations amounts to at least four types of average absolute deviation.

This sentence makes a little more sense if we remove the "Furthermore" sentence as four methods are listed earlier (mean, median, mode, or "another measure of central tendency"). However, why does this sentence use the word "combinations?" Is it "combinations" in a mathematical sense? Can't we have "Thus, there are at least four types of average absolute deviation." There is still confusion as there three types of average absolute deviation listed plus average absolute deviation allows for using any other number as the measure of central tendency.

I believe the lead would be clearer if it just listed the three methods of "average", said that there are three of them, and than added "An average absolute deviation can also be computed using other measures of central tendency."

I wished this article had explained the reasoning and/or benefits of computing and using an average absolute deviation versus standard deviation which I suspect is the most common and well defined method of computing amount of variation or dispersion of a set of data values.

I'm not enough of a subject matter expert to be making changes though. Based on the previous comments from people like me I've also hatted this article with expert needed. --Marc Kupper&#124;talk 16:00, 11 October 2017 (UTC)

Max absolute deviation?
The following paragraph is nonsense and should be deleted. Plugging the sample maximum into the formula simply gives the mean absolute deviation around the max.
 * The maximum absolute deviation around an arbitrary point is the maximum of the absolute deviations of a sample from that point. While not strictly a measure of central tendency, the maximum absolute deviation can be found using the formula for the average absolute deviation as above with $$m(X)=\max(X)$$, where $$\max(X)$$ is the sample maximum. — Preceding unsigned comment added by 128.196.225.90 (talk) 02:17, 4 April 2019 (UTC)