# Quantitative analysis

Canonical page: https://www.questionpunk.com/support/quantitative-analysis

Run descriptive statistics, group comparisons, and cross-tabs on your survey data, and save reproducible analyses.

Time: 10-20 minutes

Level: Intermediate

Audience: Researchers, Analysts, Product teams

## Steps

### 1. Open quantitative analysis

Click **Quantitative analysis** in the sidebar's Analysis section to see every study you have available, or pick **Quantitative** from the switcher on a survey's Results tab to open the same workspace scoped to that survey. You can also import an external CSV dataset directly into its own quantitative workspace.

### 2. Review your data and variables

In the **Data & variables** tab, check how each column was typed (numeric, categorical, multiselect, or text) and exclude any rows you don't want counted in the analysis.

### 3. Choose an analysis method

In the **Analysis** tab, pick a variable to analyze and a method: **Descriptive** statistics (mean, median, min, max, standard deviation, and frequencies), **Group comparison** (break the same statistics down by a second variable), or **Crosstab** (cross the variable against a group variable).

### 4. Filter and weight

Narrow the analysis with filters on any variable, and optionally apply response weights so results reflect a target population rather than raw counts.

### 5. Save and reproduce analyses

Save a named analysis to return to its settings later. Saved analyses are flagged as stale automatically if the underlying data changes, so you know when to rerun them before exporting.

## Details

Quantitative analysis is a standalone, cross-survey workspace reached from the sidebar's **Analysis** section, and it is also embedded in a native survey's Results tab alongside Report and Qualitative analysis.

Three analysis methods are available: **Descriptive** statistics for a single variable, **Group comparison** to break those statistics down by a second variable, and **Crosstab** to cross two variables against each other.

Response weighting lets you correct for over- or under-sampled groups so summary statistics better reflect your target population.

Saved analyses are reproducible: each one stores its definition (variable, group, weight, filters, and excluded rows) and is marked stale if the dataset changes, prompting a rerun before you export or present it.

You can analyze responses from an existing survey or import an external CSV dataset that has no parent survey.

## Related articles

- [Qualitative analysis and coding](https://www.questionpunk.com/support/qualitative-analysis.md)
- [Viewing the report](https://www.questionpunk.com/support/view-report.md)
- [Exporting results](https://www.questionpunk.com/support/data-exports-exporting-results.md)