The Constructs of Comparison: Defining Ourselves in a Bell Curve World

  • The Nature of Comparison:

    • Western societies tend to compare individuals to a reference group (e.g., height relative to North Americans vs. Latin Americans, university selectivity relative to other institutions, income relative to socioeconomic class).
    • These comparisons are often implicit and deeply ingrained in how we perceive ourselves and others.

  • Understanding the Bell Curve (Normal Distribution):

    • Many human characteristics and societal measures (e.g., height, weight, intelligence, income, test scores) tend to cluster around an average, forming a bell-shaped curve when plotted.
    • The bell curve illustrates that most people fall near the middle, with fewer individuals at the extremes.
    • However, comparisons are complex and can be swayed by external factors such as geographical location, societal norms, and economic disparities (e.g., education or income levels can differ drastically based on where one lives).

  • Measures of Central Tendency (The Middle):

    • Mean: The average score (e.g., a typical test score in this class might be between 7272 and 7676, or specifically 7575 by adding all scores and dividing by the number of students, e.g., 3939).
    • Median: The middle score when all data points are arranged in order. The median is crucial because it is less affected by extreme outliers than the mean.
    • Example: If the mean income is high due to a few trillionaires (like Larry Ellison, whose net worth might distort the average), the median income might provide a more accurate representation of the typical person's earnings.
    • While simple means and medians can be useful for broad comparisons (e.g., comparing educational attainment in the U.S. vs. Zimbabwe, where the world average is 11 year and the U.S. average is closer to a high school education), they offer limited insight into the spread of data.

  • Measures of Variance (Dispersion or Width):

    • Range: The difference between the highest and lowest scores. This provides a quick insight into the spread of data.
    • Example 1: A test with scores ranging from 1414 to 9999 indicates a wide dispersion, suggesting varying performance levels or possibly a mixed difficulty test.
    • Example 2: A test with scores from 8080 to 9595 suggests a narrower spread, implying that most students performed well or the test was relatively easy.
    • It is statistically impossible for