Comparing Data Sets
Comparing data sets is a Grade 11 Algebra 1 statistics skill from enVision Chapter 11 that uses both center measures and variability to determine if differences between groups are meaningful. The key insight: express the gap between means as a multiple of the MAD (mean absolute deviation). Basketball players averaging 80 inches vs. soccer players at 70 inches, with MAD of 2 inches: the 10-inch difference is 10/2 = 5 times the MAD, indicating a significant difference. A 3-minute travel time gap with MAD of 5 minutes — only 0.6 times the MAD — suggests the groups are similar.
Key Concepts
Use measures of center and measures of variability to compare two data sets. To determine if there is a meaningful difference between groups, express the difference between their centers as a multiple of their measure of variability. When the difference between means is large compared to the variability, the data sets show a significant difference.
Common Questions
How do you determine if the difference between two data sets is significant?
Divide the difference between means by the MAD. If the result is large (typically >= 2), the difference is meaningful. If small, the sets are similar despite different means.
Basketball players average 80 in, soccer players 70 in, both with MAD=2 in. Is the difference significant?
Difference/MAD = 10/2 = 5. The gap is 5 times the variability, indicating a highly significant difference between the groups.
Two schools have average travel times of 15 and 18 minutes with MAD of 5 minutes. Is this significant?
Difference/MAD = 3/5 = 0.6. The gap is less than the variability, suggesting no meaningful difference between the schools.
Why compare the mean difference to the MAD instead of just the difference alone?
A 10-inch difference is huge if typical variation is 2 inches but negligible if typical variation is 20 inches. Relative comparison provides context.
What is MAD?
Mean Absolute Deviation (MAD) measures the average distance of all data points from the mean. It represents the typical spread within a data set.
Can you use IQR instead of MAD to compare data sets?
Yes. IQR (interquartile range) can also be used as the spread measure, especially with box plots. The comparison logic is the same: difference/IQR.