Measures of Central Tendency (Grouped Data)

When data is collected in large quantities, it is often organized into a frequency distribution table — grouping individual values into class intervals such as 0–9, 10–19, 20–29, and so on. These grouped frequency distributions are far more compact than listing every single data point, but they come at a cost: we no longer know the exact value of each observation. To find measures of central tendency (mean, median, and mode) for such grouped data, statisticians use special formulas that rely on the midpoint of each class interval and the cumulative frequencies. Understanding these techniques is essential in real-world contexts. National census data, exam score reports, income surveys, and medical studies almost always present data in grouped form. A government economist estimating the average household income, a school principal analysing test performance, or a doctor studying patient ages all need to extract meaningful central values from grouped tables. The methods covered in this article allow you to do exactly that with mathematical precision. In this article you will learn how to calculate the estimated mean using the midpoint method, find the median class and apply the interpolation formula, and identify the modal class along with its precise modal estimate. Each technique gives a slightly different perspective on where the 'centre' of a grouped distribution lies, and knowing when to use each one — and understanding their limitations — is a core skill in statistical data exploration.

Key takeaways

Set Up the Frequency Table with Midpoints
Calculate the Estimated Mean
Find the Median Using Interpolation
Find the Mode
Interpret and Compare the Three Measures

Set Up the Frequency Table with Midpoints

Write out each class interval and its frequency f. Then calculate the midpoint x of each class: x = (lower boundary + upper boundary) / 2. For example, if the class is 20–29, the midpoint is (20 + 29) / 2 = 24.5. These midpoints represent all values in that class, which is why the mean calculated this way is an estimate.

Calculate the Estimated Mean

Multiply each midpoint x by its frequency f to get fx. Sum all fx values to get Σfx, and sum all frequencies to get Σf (which equals n, the total number of data points). The estimated mean is x̄ = Σfx / Σf. To simplify calculations with large midpoints, you can use the coded mean method: choose an assumed mean A, compute d = x − A for each class, then x̄ = A + (Σfd / Σf).

Find the Median Using Interpolation

The median is the value at position n/2 in the ordered data. Build a cumulative frequency (CF) column by adding frequencies progressively. The median class is the first class where the cumulative frequency reaches or exceeds n/2. Once you identify the median class, apply the interpolation formula: Median = L + [(n/2 − CF_prev) / f_m] × h, where L is the lower boundary of the median class, CF_prev is the cumulative frequency before the median class, f_m is the frequency of the median class, and h is the class width.

Find the Mode

For grouped data, the modal class is the class with the highest frequency. To get a more precise modal estimate, use the formula: Mode = L + [Δ₁ / (Δ₁ + Δ₂)] × h, where L is the lower boundary of the modal class, Δ₁ = f_modal − f_before (frequency of modal class minus the frequency of the class before it), Δ₂ = f_modal − f_after (frequency of modal class minus the frequency of the class after it), and h is the class width. This formula (sometimes called the Czuber or grouping formula) assumes the distribution within the modal class is approximately triangular.

Interpret and Compare the Three Measures

For a perfectly symmetrical distribution, the mean, median, and mode will all be equal. For a positively skewed distribution (long tail to the right), the mean > median > mode. For a negatively skewed distribution (long tail to the left), the mean < median < mode. The mean is most sensitive to extreme values; the median is robust to skewness; the mode tells you the most common class. Always state that values from grouped data are estimates, since exact data within each class is unknown.

The Key Equation

\bar{x} = \frac{\sum f_i x_i}{\sum f_i}, \quad \text{Median} = L + \frac{\frac{n}{2} - CF_{\text{prev}}}{f_m} \times h, \quad \text{Mode} = L + \frac{\Delta_1}{\Delta_1 + \Delta_2} \times h

\bar{x} — Estimated mean (same as data): The arithmetic average estimated from grouped data using midpoints
f_i — Class frequency (count): Number of data values falling in the i-th class interval
x_i — Class midpoint (same as data): Middle value of the i-th class interval, representing all values in that class
n — Total frequency (count): Total number of observations; equals Σf
L — Lower class boundary (same as data): The exact lower boundary of the median or modal class
CF_{\text{prev}} — Cumulative frequency before (count): Sum of all frequencies in classes before the median class
f_m — Median class frequency (count): Frequency of the class interval containing the median
h — Class width (same as data): The width of the class interval (upper boundary minus lower boundary)
\Delta_1 — First modal difference (count): Frequency of modal class minus frequency of the preceding class
\Delta_2 — Second modal difference (count): Frequency of modal class minus frequency of the following class

Worked Example

Problem. The ages (in years) of 50 members at a sports club are recorded in the grouped frequency table below. Find the estimated mean, median, and mode. Class: 10–19, Frequency: 4 Class: 20–29, Frequency: 11 Class: 30–39, Frequency: 18 Class: 40–49, Frequency: 12 Class: 50–59, Frequency: 5

\bar{x} \approx 35.1, \quad \text{Median} \approx 35.1, \quad \text{Mode} \approx 34.9 \text{ (all in years)}

Answer. Estimated Mean ≈ 35.1 years, Median ≈ 35.1 years, Mode ≈ 34.9 years. The three values are very close, indicating a roughly symmetrical distribution centred in the 30–39 age group.

Figure

Diagram

Origins · 1 minKarl Pearson and how this was discovered

Karl Pearson

1857–1936 · British

Karl Pearson was the founder of modern mathematical statistics. He developed the method of moments — a systematic way of summarising frequency distributions using their mean, variance, skewness, and kurtosis — and formalised much of the theory behind frequency curves. His work in the 1890s–1900s, including the Pearson chi-squared test and the product-moment correlation coefficient, gave statisticians rigorous tools for analysing grouped data distributions and relating their central tendency to their shape. Pearson was so devoted to statistics that he founded the world's first university statistics department (at University College London in 1911) and single-handedly produced a journal called Biometrika for decades, often writing most of it himself.

Measures of Central Tendency (Ungrouped Data) — Grouped data methods are extensions of the simpler ungrouped formulas; mastering exact mean, median, and mode first makes the interpolation approach much easier to understand.
Frequency Distributions and Histograms — Grouped data is presented in frequency distribution tables and histograms; understanding how to construct and read these is a prerequisite for computing central tendency.
Measures of Spread for Grouped Data — After finding the centre of a distribution, the natural next step is to quantify its spread using variance and standard deviation formulas adapted for grouped data.
Cumulative Frequency and the Ogive — The median interpolation formula relies directly on cumulative frequencies; the ogive (cumulative frequency graph) provides a visual way to read off the median and quartiles.
Skewness and Symmetry of Distributions — The relationship between mean, median, and mode — and how they separate in skewed distributions — is a key application of the three central tendency measures for grouped data.