Question

Synthesis
How does the mean absolute deviation (MAD) help you compare data sets?
Use the examples to the left if they help with your thinking.
Mean: 5, MAD: 2 Polina - 3,3,4,4,7,9
Mean: 5, MAD: 1 Rishi - 2,5,5,5,6,7
Mean: 6 MAD: 1.5 Deven - 1,5,6,6,7,7,7,9

Answer 1

Mean absolute deviation (MAD) measures the average distance of the data values from their mean. It tells you how spread out or how consistent a data set is around its center. A smaller MAD means the values are closer to the mean (more consistent); a larger MAD means they are more spread out.

Quick check with your examples (absolute deviations from the mean, then average):

- Polina (mean 5): data 3,3,4,4,7,9 → |deviations| = 2,2,1,1,2,4 → sum = 12 → MAD = 12/6 = 2
Interpretation: on average her values are 2 units from the mean — relatively spread out.

- Rishi (mean 5): data 2,5,5,5,6,7 → |deviations| = 3,0,0,0,1,2 → sum = 6 → MAD = 6/6 = 1
Interpretation: his values are closer to the mean (more tightly clustered) than Polina’s.

- Deven (mean 6): data 1,5,6,6,7,7,7,9 → |deviations| = 5,1,0,0,1,1,1,3 → sum = 12 → MAD = 12/8 = 1.5
Interpretation: Deven’s values are more spread than Rishi’s (1.5 > 1) but less spread than Polina’s (1.5 < 2). Note his center (mean = 6) is higher than the other two, but MAD compares variability, not location.

Uses and cautions:
- Use MAD to compare consistency/variability between data sets (lower MAD = more consistent).
- MAD is easier to interpret than variance/SD because it’s in the same units as the data.
- Be careful: MAD doesn’t tell you about the mean itself (so two sets can have different centers but similar MADs), and different sample sizes or strongly skewed distributions can affect interpretation. For robustness to outliers you might consider the median absolute deviation (a different measure).

Bottom line: MAD is a simple, interpretable way to compare how tightly data are clustered around their mean.