Statistics and data

Relative frequency and cumulative percentages

Turn raw counts into a percentage distribution: relative frequency is each count ÷ the total, and cumulative percentage is the running total of those. A worked frequency table.

PercentSwiftPublished 2 min read

Short answer

Relative frequency = count in a category ÷ total count × 100. Cumulative percentage = the running total of relative frequencies. If 17 of 50 people commute 15–30 minutes, that’s 34%, and with the 16% who commute under 15 minutes, 50% commute 30 minutes or less.

On this page

A frequency table counts how many observations fall in each category. Converting the counts to percentages makes the table easier to read and to compare with tables of different sizes.

The table

Commute times reported by 50 employees:

Commute Count Relative frequency Cumulative
Under 15 min 8 16% 16%
15–30 min 17 34% 50%
30–45 min 14 28% 78%
45–60 min 7 14% 92%
Over 60 min 4 8% 100%
Total 50 100%

Relative frequency

Divide each count by the total. 15–30 minutes: 17 ÷ 50 = 34%.

Try it: 17 of 50 employees

Open in calculator
Column chart of commute times for 50 employees: under 15 minutes 16%, 15 to 30 minutes 34%, 30 to 45 minutes 28%, 45 to 60 minutes 14%, over 60 minutes 8%.
Each column is the band’s count divided by 50. The cumulative column in the table adds them up from the left. Tap the image to open it full size.

The relative frequencies always add to 100% (before rounding), since every observation is in exactly one category.

Cumulative percentage

Add the relative frequencies from the top down. The cumulative column answers “what share is at or below this level?”

  • 50% commute under 30 minutes (16 + 34).
  • 78% commute under 45 minutes (16 + 34 + 28).

You can also get these directly from counts: 8 + 17 + 14 = 39 people under 45 minutes, and 39 ÷ 50 = 78%.

Try it: 39 of 50 commute under 45 minutes

Open in calculator

Reading the median from the cumulative column

The median is where the cumulative percentage crosses 50%. Here it reaches exactly 50% at the end of the 15–30 band, so the median commute is about 30 minutes. If the 50% mark falls inside a band, the median is somewhere in that band.

Comparing two groups

Relative frequencies let you compare groups of different sizes. If another office of 120 employees has 30 people commuting over an hour, that’s 25%, versus 8% here. The raw counts (30 vs 4) exaggerate the difference because the offices differ in size, and the percentages correct for that.

Rounding

With awkward totals, rounded relative frequencies might add to 99% or 101%. For example, three categories of 1, 1 and 1 out of 3 each round to 33%, totaling 99%. See why percentages don’t add to 100.

Percentile ranks use cumulative percentages too: see percentage vs percentile. For two-way tables with row and column totals, see percentage of total in a table.

Questions

What's the difference between relative frequency and probability?

Relative frequency describes data you’ve collected. If the sample is large and representative, it’s often used as an estimate of probability.

Should a band boundary like 30 minutes go in the lower or upper band?

Pick one rule and state it, for example “15–30 means at least 15 and less than 30.” Otherwise values on the boundary can be counted twice or not at all.

The calculator links in this guide are checked against the PercentSwift calculator every time the site is built. How we calculate explains the rounding rules. If you spot a mistake, email hello@percentswift.com.