Frequencies
Frequencies counts how many cases fall into each value of a variable and gives percentages, so it is the first thing to run on categorical data (gender, group, Likert items) and a quick check for typing errors.
Menu: Analyze → Descriptive Statistics → Frequencies…
When should I use Frequencies?
- Describing categorical or ordinal variables (counts and percentages).
- Checking data entry: unexpected codes (e.g. a 7 on a 1–5 scale) show up immediately.
- Getting the mode and percentiles of a numeric variable.
What are the assumptions?
- None — it is a descriptive procedure.
How do I run it in StatLab?
- Choose one or more variables.
- Keep “Statistics table” ticked for numeric variables (mean, median, mode, SD, percentiles …).
- Optionally add a bar chart (categories) or a histogram with a normal curve (scale variables).
How do I read the output?
| Output | Meaning |
|---|---|
| Frequency | Number of cases with that value. |
| Percent | Share of all cases, including missing ones. |
| Valid Percent | Share of cases with a valid (non-missing) value — usually the one to report. |
| Cumulative Percent | Running total of Valid Percent; useful for ordinal scales (“62% agreed or less”). |
| Missing rows | User-missing codes (e.g. 9 = no answer) are listed separately, then “System” for blank cells. |
| Percentiles | Weighted-average definition (HAVERAGE): position p(n+1), the same default as most desktop statistics packages. |
Worked example
Run Frequencies on gender in the bundled sample dataset (File → Sample dataset: 150 survey respondents with gender, programme group, age, pre/post test scores, five satisfaction items, income, smoking and education). Of 150 respondents, 69 (46.0%) are coded 2 = Female.
How would I report it in APA style?
Of the 150 respondents, 69 (46.0%) were female and 81 (54.0%) were male.
Numbers above were computed by StatLab’s validated engine from the sample dataset when this page was built.
What are common mistakes?
- Reporting “Percent” when there are missing values — use “Valid Percent”.
- Running Frequencies on an ID or a continuous variable with hundreds of distinct values: use Descriptives instead.