Lecture Notes: Measures of Center in NBA Salaries
Overview
- Today's activity focuses on understanding measures of center using NBA salaries as an example.
- Previously covered measures of spread such as standard deviation.
Key Concepts
Measures of Center
- Distribution: A representation or picture of a variable (e.g., NBA salaries).
- Center of Distribution: The middle of the data, which can be represented by different measures of center.
Decision-Making Scenario
- Consideration of a college basketball player dropping out of college based on financial prospects in the NBA.
- Opinions may vary: stay in college for long-term benefit vs. grab the opportunity for immediate financial gain.
- Statistics help make informed decisions based on data.
Measures of Center: Mean vs Median
Definitions
- Median: The middle value in a list of numbers.
- Stack numbers from smallest to largest, the median is the middle point.
- Mean: The average of numbers.
- Sum all numbers and divide by the count.
Characteristics
- Median: Not affected by outliers; gives a true central tendency for skewed data.
- Mean: Easier to calculate but is affected by outliers and skewed data.
- Can be misleading if data has outliers or is not symmetric.
Analysis of NBA Salaries
- Dot Plot Analysis: Visual representation of NBA salaries.
- Example: James Harden earns $28 million vs. Chris Johnson $25,000.
- Distribution Shape: Identified as skewed right due to a long tail to the right.
- Outliers like James Harden’s salary can skew the mean.
Estimating Salaries
- Typical Salary Range: Between $0 to $2.5 million.
- Median Salary: $1.6 million.
- Mean Salary: Over $5 million.
Comparing Mean and Median
- Percentage Calculation: Estimating percentage of players earning above the mean.
- Only about 28% earn more than the mean.
- Above the Median: Median divides the dataset into two equal halves, about 50% above and below.
Misleading Claims
- Use of mean to represent "typical" salaries can be misleading due to skewness and outliers.
- Median is a more appropriate measure for skewed data as it is resistant to outliers.
Application Examples
- Income in Santa Barbara: Right-skewed distribution due to a few high-income earners.
- GPAs at UCSB: Left-skewed distribution due to few low scores.
- Body Temperatures: Normally distributed, symmetrical.
Summary of Relationship
- If data is skewed right, mean > median.
- If data is skewed left, mean < median.
- If data is symmetrical, mean ≈ median.
Important Takeaways
- Median: Resistant to outliers and skewed data.
- Mean: Not resistant to outliers; should be used with caution in skewed distributions.
Final Thoughts
- Be cautious of misleading statistical claims.
- Use median for better accuracy in skewed distributions.
Note: Practice further to internalize these concepts.