How to make a frequency table
- Paste 1–200 finite numbers in the list field. Separate items with commas, spaces, semicolons, or line breaks.
- Choose Individual values to count each distinct number, or Equal-width intervals to group the range.
- For grouped data, choose 2–20 intervals. The calculator starts at the minimum and ends at the maximum.
- Select Calculate, then read the ordinary, relative, and cumulative columns.
- Copy the text version when you need to move the result into notes or a document.
What each column means
Frequency is the number of observations in a row. Relative frequency is that count divided by the total sample size n. Cumulative frequency adds the current count to every earlier count, while cumulative proportion divides that running count by n. The last cumulative frequency must equal n and the last cumulative proportion must equal 1.
The table computes every proportion from integer counts before formatting. A displayed column can appear to total 0.99 or 1.01 when its individually rounded values are added; this is a display effect, not lost data.
Individual values and equal-width intervals
Individual mode is best when exact repeated values matter. Values are sorted numerically, so negative values come before zero and decimals retain their position. Duplicate observations are counted, not removed.
Grouped mode divides max − min into equal widths. Every class except the last uses a closed lower boundary and open upper boundary: [a, b). The last class is [a, b], so an observation equal to the overall maximum is included exactly once. Empty classes remain visible. Constant data produces one closed row because a positive-width range does not exist.
Choose the mode according to the question you need to answer. Individual rows show precisely which values repeat and suit a short discrete list. Intervals make a longer continuous-looking list easier to scan, but each row then describes a range rather than one observed value. Changing the interval count changes the grouping and can change the apparent shape, even though the sample size and original observations stay the same.
Worked examples
For 2, 2, 3, 5, 5, 5, individual mode gives frequencies 2, 1, and 3. Their relative frequencies are 2/6, 1/6, and 3/6; the cumulative counts are 2, 3, and 6.
For 0, 1, 2, 3, 4 with two intervals, the classes are [0, 2) and [2, 4]. Their frequencies are 2 and 3. The boundary value 2 belongs to the second class and the maximum 4 belongs to the final class.
Limits and common input problems
- The list must contain 1–200 finite numbers. Words, empty items mixed with stray punctuation, Infinity, and NaN are rejected.
- Grouped output is limited to 20 intervals. If very large or extremely close values cannot produce distinct displayed boundaries, the calculator reports an error instead of showing a misleading table.
- Use a period for the decimal separator in the input, such as 1.25. Localized punctuation is used only when results are displayed.
- Changing any input, switching the table type, or clearing the form immediately invalidates the previous result and disables Copy.
Frequently asked questions
Why are there empty interval rows? Equal-width grouping describes the whole range, so an interval remains part of the distribution even when its frequency is zero. Can I use negative numbers and decimals? Yes. They are sorted and classified using their full numeric values. Does grouping change my observations? No; it changes only which row counts each observation. Why might rounded relative frequencies not total exactly 1? Each displayed row is rounded independently while all internal totals use exact counts. Is this a histogram maker? No. The requested output is a table, without a chart editor or statistical dashboard.