Comparing Distributions - Pre Algebra
- Which measures of center and spread are best for comparing distributions that are skewed or have outliers?
- The median and the IQR, because neither is strongly affected by extreme values.
- How can the difference between two means be expressed as a multiple of variability?
- Divide the difference in means by the MAD (when the two MADs are similar). A larger result means less overlap.
- Which measures of center and spread are typically used to compare roughly symmetric distributions without outliers?
- The mean and the mean absolute deviation (MAD).
- Class A has median score 78 with IQR 10. Class B has median 85 with IQR 10. What is the best comparison?
- Class B typically scored 7 points higher, and the two classes have the same spread.
- Two groups have means 20 and 26, and both have MAD 3. How many MADs apart are the means?
- 2 MADs. The difference is , and .
- Two teams both have a mean height of 60 inches. Team A has MAD 2 inches and Team B has MAD 5 inches. Which team's heights are more consistent?
- Team A. Its smaller MAD means heights are closer to the mean.
- Which two features should be described when comparing two numerical distributions?
- Their centers (such as mean or median) and their spreads (such as MAD or IQR).
- Why compare medians instead of means when one group contains an extreme outlier?
- The median is resistant to outliers, while the mean is pulled toward the extreme value.
- What does a larger IQR in one of two groups indicate?
- The middle half of that group's data is more spread out, so its values are less consistent.
- Group A's values range from 40 to 60, and Group B's values range from 70 to 90. How do the distributions overlap?
- They do not overlap. Every value in Group B is greater than every value in Group A.
- Store A wait times have median 5 minutes and IQR 6 minutes. Store B has median 5 minutes and IQR 2 minutes. Which store's waits are more predictable?
- Store B. The typical waits are equal, but B's smaller IQR means its middle half of waits varies less.