If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Lesson 8: Mean absolute deviation (MAD)

Mean absolute deviation (MAD) of a data set is the average distance between each data value and the mean. Mean absolute deviation is a way to describe variation in a data set. Mean absolute deviation helps us get a sense of how "spread out" the values in a data set are.

## Want to join the conversation?

• I still don't get how to find the MAD, can anyone pls help me •   1. finding the mean(average) of the set of numbers
2. find the distance of all the numbers from the mean.
3. Find the mean of those numbers.
• Wait, so we have to find the mean and then the absolute value right? • Is there an easier way to calculate MAD? So much writing! • but how do you do these things and not get them wrong:{
(1 vote) • Is this different from standard deviation? I find that I get different answers from both, but they seem like the same concept. Can you please explain the difference and purpose of each? • what how to do it • Mean Absolute Deviation (MAD) is a way to measure how spread out a set of data is.
The first step is to calculate the mean (average) of the set of data. If we have the set of data [-1,2,3,7,9,12,17], the mean would be [-1+2+3+7+9+12+17] / 7, so 7 is the mean.
The second step is to measure how far each point of data is from the mean, so [7-(-1)] + [7-2] + [7-3] + [7-7] + [7-9] + [7-12] + [7-17]. If we have a number that is bigger than the mean, like 9,12, and 17 in this case, we take the absolute value, so usually 7-9 = -2 (but with absolute value) 2.
The third step is to add all the values above, and divide them by
the number of data points, so 7. Eventually, we get : 34/7, or [4 6/7]
Hope this helps.
• There was a distinction made between a sample variance/standard deviation and a population variance/standard deviation. The population variance is calculated by taking the sum of the squared deviations from each data point to the population mean, and then dividing by the number of data points in the population. On the other hand, the sample variance goes through the same process as above, except it's with respect to the sample mean, and you should also divide by one less than the number of data points in your sample, to correct the bias (Bessel's Correction). I'm wondering if a similar notion exists for the Mean Absolute Deviation (MAD)? In other words, whether it's a sample or population we're dealing with, is there any significant difference in the way that the MAD is calculated for either of them? • I still don't under stand how you come up with the two different data sets do I split my data in half? • Sal uses two completely different data sets to show how MAD describes the variability of a single data set.
2,2,4,4 - number of donuts I ate each of the last four days
1,1,6,4 - number of times I scored in my last four soccer games
Both data sets have a mean of 3. On average, I eat 3 donuts a day, and score 3 goals per game [I wish].
The MAD of the donut data is 1, showing that I am pretty consistent on eating donuts. The average day is 2 to 4 donuts (1 donut more or less than 3).
However, The MAD of the soccer data is 2, showing that there is more variability in my goal scoring. An average game is 1 to 5 goals (2 goals more or less than 3).  