Frequency analysis counts how often categories or value ranges occur and converts those counts into proportions. Its apparent simplicity makes denominator choices, missing values, weights, and multiple-response rules especially important.

The central question is: “How often does each category occur, and is its percentage calculated from the full sample, valid responses, or a weighted population?” Defining it before opening a menu prevents a page of statistics from replacing an explanation.

The purpose is category counts and percentages tied to an explicit denominator. Interpret counts, overall percent, valid percent, and missing responses in one view, because the same count can produce different percentages under different denominators.

Frequency is a count; relative frequency divides by a stated denominator. Valid percent excludes missing responses, so it can differ from percent of the full sample. Cumulative percent is meaningful only when categories have an order.

No single statistic is sufficient. Reconcile totals with sample size, ensure missing codes are not treated as categories, and explain why multiple-response totals may exceed 100%. Small cells can be unstable and disclose identities.

The purpose is category counts and percentages tied to an explicit denominator. Interpret counts, overall percent, valid percent, and missing responses in one view, because the same count can produce different percentages under different denominators.

If 40 of 100 people select “satisfied” but only 80 answer the item, the overall percentage is 40% and the valid percentage is 50%. Both are correct answers to different denominator questions.

Report sample size and missingness alongside the primary estimate, comparison standard, uncertainty, and visible pattern. A descriptive result becomes inferential or causal only when the design supports that claim.

Treat the worked number as the beginning of interpretation. Read counts, overall percent, valid percent, and missing responses in one view; then state what the sample supports without turning a descriptive pattern into a population or causal claim.

Frequency is a count; relative frequency divides by a stated denominator. Valid percent excludes missing responses, so it can differ from percent of the full sample. Cumulative percent is meaningful only when categories have an order.

No single statistic is sufficient. Reconcile totals with sample size, ensure missing codes are not treated as categories, and explain why multiple-response totals may exceed 100%. Small cells can be unstable and disclose identities.

The purpose is category counts and percentages tied to an explicit denominator. Interpret counts, overall percent, valid percent, and missing responses in one view, because the same count can produce different percentages under different denominators.

A defensible workflow is Audit labels → Define missingness → Choose denominator → Calculate frequencies → Review sparse cells. Record the variable definitions, exclusions, transformations, denominators, and decision rules. Reproducibility begins with these choices, not with the final number.

Begin by auditing type, unit, coding, missingness, and total N, valid N, weights, and multiple-response rules. Defaults are only starting points; document every denominator, exclusion, transformation, and decision rule needed to reproduce the result.

Common uses include survey option distributions, customer and product mix, defect and event types, demographic and response-rate reporting. Before operational use, define how the same quantity will be recomputed for new data, how subgroup differences will be monitored, and what action the result is meant to support.

The purpose is category counts and percentages tied to an explicit denominator. Interpret counts, overall percent, valid percent, and missing responses in one view, because the same count can produce different percentages under different denominators.

Reconcile totals with sample size, ensure missing codes are not treated as categories, and explain why multiple-response totals may exceed 100%. Small cells can be unstable and disclose identities.

The denominator is the core of a frequency table. Report total N, missing N, valid N, weights, and multiple-response rules. Choose between methods by the question and output, not by sophistication. The target here is category counts and percentages tied to an explicit denominator; an alternative may be useful precisely because it preserves a different feature of the data.

Reconcile totals with sample size, ensure missing codes are not treated as categories, and explain why multiple-response totals may exceed 100%. Small cells can be unstable and disclose identities.

Strengths include Direct summary for categorical data; Quickly exposes coding errors and rare categories; Easy to communicate with tables and bars. Important limitations are Percentages change with denominator, weights, and missing rules; Counts alone do not test differences or explain causes; Over-collapsing categories can erase structure.

Diagnostics are part of the result because the same count can produce different percentages under different denominators. Compare reasonable alternatives and inspect the observations that drive the summary before treating one output as stable.

The denominator is the core of a frequency table. Report total N, missing N, valid N, weights, and multiple-response rules. Recommended sequence: Audit labels → Define missingness → Choose denominator → Calculate frequencies → Review sparse cells.

A complete report links the question, analysis population, total N, valid N, weights, and multiple-response rules, the primary result, and the diagnostic evidence. A reader should be able to reconstruct the same quantity from those decisions.