How to Find the Median on a Frequency Table
Finding the median from a frequency table is a common task in statistics, especially when data are grouped into intervals or when raw observations are summarized by their frequencies. In practice, the median represents the middle value of a data set, splitting the observations into two equal halves. And when the data are presented in a frequency table, you cannot simply pick the middle row; instead, you must work with cumulative frequencies to locate the interval or value that contains the median. Below is a step‑by‑step guide, complete with a worked example, explanations of the underlying logic, and practical tips to avoid common mistakes The details matter here..
Steps to Find the Median on a Frequency Table
-
Organize the table
Ensure the frequency table is sorted in ascending order of the data values (or class intervals). Each row should show a value (or class) and its corresponding frequency Worth keeping that in mind.. -
Calculate the total number of observations (N)
Add up all the frequencies.
[ N = \sum f_i ] -
Determine the median position
- If N is odd, the median is the value at position (\frac{N+1}{2}).
- If N is even, the median lies between the positions (\frac{N}{2}) and (\frac{N}{2}+1); for grouped data we usually take the (\frac{N}{2})th position as the cutoff and interpolate within the class that contains it.
-
Compute cumulative frequencies
Starting from the lowest value, add each frequency to the running total. The cumulative frequency (CF) tells you how many observations fall at or below each value Small thing, real impact.. -
Locate the median class
Find the first row where the cumulative frequency is greater than or equal to the median position identified in step 3. The value (or class interval) in that row is the median class The details matter here.. -
Apply the median formula for grouped data (if needed)
When dealing with continuous class intervals, use:
[ \text{Median} = L + \left(\frac{\frac{N}{2} - CF_{prev}}{f_{med}}\right) \times w ]
where:- L = lower boundary of the median class,
- CF₍ₚᵣₑᵥ₎ = cumulative frequency of the class preceding the median class,
- f₍ₘₑd₎ = frequency of the median class,
- w = class width.
-
Interpret the result
The number you obtain is the median: the point that divides the data set into two equal halves Not complicated — just consistent..
Worked Example
Suppose a teacher records the number of books read by 30 students in a month and summarizes the data as follows:
| Books read (x) | Frequency (f) |
|---|---|
| 0 | 4 |
| 1 | 6 |
| 2 | 8 |
| 3 | 5 |
| 4 | 4 |
| 5 | 3 |
Step 1–2: The table is already ordered. Total observations:
(N = 4+6+8+5+4+3 = 30).
Step 3: Since N is even, the median position is (\frac{N}{2}=15). We look for the 15th observation Simple, but easy to overlook. That alone is useful..
Step 4: Cumulative frequencies:
| x | f | Cumulative (CF) |
|---|---|---|
| 0 | 4 | 4 |
| 1 | 6 | 10 |
| 2 | 8 | 18 |
| 3 | 5 | 23 |
| 4 | 4 | 27 |
| 5 | 3 | 30 |
Step 5: The first CF ≥ 15 occurs at x = 2 (CF = 18). Hence the median class is the value 2 Small thing, real impact..
Step 6: Because the data are discrete values (not intervals), the median is simply that value: 2 books.
If the same data were grouped into intervals, say 0‑1, 2‑3, 4‑5, we would apply the formula in step 6 to interpolate within the median class.
Why the Method Works
The median is defined by the rank order of observations, not by their magnitude. By converting raw frequencies into a cumulative count, we translate the problem into “how many observations are less than or equal to a given value?” The median position tells us the rank that splits the data in half. The first value whose cumulative count reaches or exceeds that rank must be the median because all lower values account for fewer than half the observations, and adding the frequency of that value pushes us past the halfway point.
When data are continuous and presented in class intervals, we assume observations are evenly spread within each interval. The interpolation formula therefore estimates where the exact halfway point falls inside the median class, based on the proportion of the class needed to reach the median rank.
Tips and Common Pitfalls
- Double‑check ordering – If the table is not sorted, cumulative frequencies will be meaningless.
- Watch for open‑ended classes – If the first or last class lacks a boundary, you may need to estimate a reasonable width or use alternative methods (e.g., assuming a distribution).
- Even vs. odd N – For discrete data, some textbooks define the median as the average of the two middle values when N is even. In a frequency table, you can locate both middle positions and see whether they fall in the same class or different ones; if different, compute the average of the two corresponding values (or interpolate).
- Class width consistency – Ensure all intervals have the same width w before applying the formula; otherwise, adjust the calculation per class.
- Rounding – Keep enough decimal places during intermediate steps to avoid rounding error, especially when the median falls near a class boundary.
- Software verification – After manual calculation, you can cross‑check with a spreadsheet or statistical package to confirm the result.
Frequently Asked Questions
Q: Can I find the median directly from the frequency column without cumulative frequencies?
A: No. The frequency column alone tells you how many observations fall at each value, but you need a running total to know where the middle observation lies.
Q: What if the median position lands exactly on a cumulative frequency?
A: If the cumulative frequency equals the median position, the median is the upper boundary of that class (for continuous data) or
