Course contentsShow
← Data Science
Lesson 45 of 2,1452. Descriptive Statistics FoundationsFree lesson

Central Tendency for Grouped Data

Estimating mean and median from frequency tables and histograms when raw data isn't available.

Central Tendency for Grouped Data

What You'll Learn

You'll learn how to estimate the mean and median when you don't have individual data points, only frequency tables or histograms showing grouped ranges.

Why This Matters

Sometimes you receive data already summarized—like "15 people scored between 60-69, 22 people scored between 70-79"—without seeing each person's exact score. You can't calculate the precise mean or median, but you can make excellent estimates using statistical techniques.

Estimating the Mean from Grouped Data

To estimate the mean from a frequency table:

  1. Find the midpoint of each group (class interval)
  2. Multiply each midpoint by its frequency (how many values fall in that range)
  3. Sum all those products
  4. Divide by the total number of observations

Think of it like a weighted mean: you're assuming all values in a group cluster at the middle of that range.

Example: If 10 people scored 60-69 and 20 scored 70-79:

  • Midpoints: 64.5 and 74.5
  • Products: (64.5 × 10) + (74.5 × 20) = 645 + 1490 = 2135
  • Mean estimate: 2135 ÷ 30 = 71.2

Estimating the Median from Grouped Data

For the median, you locate which group contains the middle value(s), then use linear interpolation within that group to estimate where exactly the median falls. This assumes values are evenly spread within each interval.

When You'll Use This

Grouped data appears in published reports, surveys with age brackets, income ranges, and historical datasets where raw numbers were never recorded—only summaries.

Key Takeaway: When raw data isn't available, use group midpoints to estimate the mean and interpolation techniques to estimate the median from frequency tables—your answers will be close approximations of the true central tendency.