Ratio statistics
Ratio statistics summarise the ratio of two variables per case (for example assessed value / sale price) with the median, weighted mean, coefficient of dispersion (COD), price-related differential (PRD) and coefficient of variation.
Menu: Analyze → Descriptive Statistics → Ratio…
When should I use Ratio statistics?
- Appraisal and assessment studies, cost ratios, or any comparison of a numerator and a denominator per case.
What are the assumptions?
- Cases with a zero or missing denominator are left out.
How do I run it in StatLab?
- Choose the numerator and the denominator; optionally a grouping variable.
How do I read the output?
| Output | Meaning |
|---|---|
| Median and Mean | Typical ratio. |
| Weighted Mean | Sum of numerators / sum of denominators. |
| COD | Average absolute deviation from the median, as a percentage of the median. |
| PRD | Mean / weighted mean; above 1 suggests high-value cases have lower ratios. |
Worked example
Ratio post / pre 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): median 1.06, COD 8.88%, PRD 1.00.
How would I report it in APA style?
The median post/pre ratio was 1.06 (COD = 8.88%, PRD = 1.00).
Numbers above were computed by StatLab’s validated engine from the sample dataset when this page was built.
What are common mistakes?
- Using ratio statistics when a regression of numerator on denominator answers the question better.