What Is Cumulative Frequency?
Cumulative frequency is a running total of frequencies. For an ordered set of classes, the cumulative frequency at any class is the sum of that class's frequency plus all the frequencies before it. It answers questions like "how many observations fall at or below this point?" and is the foundation for cumulative frequency graphs (ogives), medians, quartiles, and percentiles.
How to Use This Calculator
Type your class frequencies into the box in order, separated by commas or spaces (for example 5, 8, 12, 4, 6). The calculator adds each frequency to the running total and displays a full table showing every class, its frequency, and its cumulative frequency. The final row gives the grand total, which equals the last cumulative value.
The Formula Explained
The cumulative frequency of the k-th class is written $$CF_k = \sum_{i=1}^{k} f_i \quad \text{where } f_i \in \text{Frequencies}$$ In plain terms, start with the first frequency, then keep adding each subsequent frequency to the previous total. The last cumulative frequency, \(CF_n\), always equals the sum of every frequency in the data set.
Worked Example
Suppose the frequencies are 5, 8, 12, 4, 6. The cumulative values are: 5, then \(5+8=13\), then \(13+12=25\), then \(25+4=29\), then \(29+6=35\). So the cumulative frequency column reads 5, 13, 25, 29, 35, and the total number of observations is 35.
Interpreting Your Cumulative Frequency Results
The cumulative frequency (CF) column is a running total of the class frequencies. The last CF value always equals the total number of observations, N. If it does not match your data count, a frequency was entered incorrectly.
Take the frequencies 5, 8, 12, 4, 6. The running totals are 5, 13, 25, 29 and 35, so N = 35.
| Class | Frequency | CF | Relative CF |
|---|---|---|---|
| 1 | 5 | 5 | 14.29% |
| 2 | 8 | 13 | 37.14% |
| 3 | 12 | 25 | 71.43% |
| 4 | 4 | 29 | 82.86% |
| 5 | 6 | 35 | 100% |
Locating the median, quartiles and percentiles
- Median class: the first class where CF reaches or exceeds \(N/2\). Here \(N/2 = 35/2 = 17.5\). CF is 13 for class 2 (too low) and 25 for class 3 (at least 17.5), so the median lies in the 3rd class.
- Quartiles: use \(N/4\) for Q1 and \(3N/4\) for Q3. Here \(N/4 = 8.75\) falls in class 2 (CF 13), and \(3N/4 = 26.25\) falls in class 4 (CF 29).
- Percentiles: use \(kN/100\) for the k-th percentile. For the 90th percentile, \(90 \times 35 / 100 = 31.5\), which falls in class 5 (CF 35).
For grouped data, you can estimate the exact position inside the class by linear interpolation:
$$\text{Median} \approx L + \left(\frac{N/2 - CF_{prev}}{f}\right) \times w$$
Here \(L\) is the lower boundary of the median class, \(CF_{prev}\) is the CF of the class before it, \(f\) is the median class frequency, and \(w\) is the class width. In the example, \(CF_{prev} = 13\) and \(f = 12\), so the fraction is \((17.5 - 13)/12 = 0.375\) of the way through the class.
Key properties
CF never decreases, because each step adds a frequency that is zero or more. A class with frequency 0 simply repeats the previous CF. A decrease means a negative frequency or a typing error.
Reading an ogive
On an ogive (cumulative frequency graph), each point is plotted at the upper class boundary of a class, with the CF on the vertical axis, because the running total is only complete once the whole class has been counted. The curve starts at CF = 0 at the lower boundary of the first class and rises to N at the upper boundary of the last class. To find the median, go up to N/2 on the vertical axis, across to the curve, then down to read the value.
Relative cumulative frequency
Divide each CF by N and multiply by 100 to get a percentage: \(\text{Relative CF} = \dfrac{CF}{N} \times 100\%\). For example, class 3 gives \(25/35 \approx 71.43\%\), meaning about 71% of observations fall at or below that class's upper boundary. The final value is always 100%. This is useful for comparing datasets of different sizes.
Key Terms: Frequency, Class, and Cumulative Frequency
- Frequency (f)
- The number of observations that fall in a given class or take a given value.
- Class / class interval
- A range of values used to group continuous or large datasets, such as 10–19 or 20–29. Each observation belongs to exactly one class.
- Class boundaries
- The exact dividing points between adjacent classes. For integer classes 10–19 and 20–29, the boundary is 19.5, so the class 20–29 runs from 19.5 to 29.5. Ogive points are plotted at upper boundaries.
- Cumulative frequency (CF)
- The running total of frequencies up to and including a given class: \(CF_k = f_1 + f_2 + \dots + f_k\).
- Relative frequency
- A class frequency divided by the total, \(f/N\), often shown as a percentage. The relative frequencies sum to 1 (100%).
- Relative cumulative frequency
- The cumulative frequency divided by the total, \(CF/N\). It gives the proportion of observations at or below a class's upper boundary.
- Ogive
- A line graph of cumulative frequency (vertical axis) against upper class boundaries (horizontal axis). It is used to estimate the median, quartiles and percentiles.
- Less-than vs more-than cumulative frequency
- A less-than CF counts observations below each upper boundary, accumulating from the lowest class upward, and is the type this calculator builds. A more-than CF counts observations at or above each lower boundary, accumulating from the highest class downward. It starts at N and decreases.
- Total (N)
- The sum of all frequencies, which equals the final cumulative frequency. It is the number of observations in the dataset.
FAQ
Can I enter decimal frequencies? Yes — weighted or relative frequencies with decimals are supported, though raw counts are typically whole numbers.
Does order matter? Yes. Cumulative frequency depends on the order of the classes, so enter them from the lowest class to the highest.
What is the last cumulative frequency? It is always equal to the total of all frequencies — the size of your data set.