Correlation Calculator: Pearson and Spearman with Significance
Paste two columns to get Pearson and Spearman correlations, p-values, confidence intervals and a scatter plot, or find the sample size to detect a correlation.
inquiry@globalvoxpopuli.com · globalvoxpopuli.com
How to use the correlation calculator
- Paste two columns of numbers, such as satisfaction and spend, one respondent per row. A header row is used to label the chart.
- Read Pearson's r (linear relationship), Spearman's ρ (any consistent rising or falling pattern) and their significance.
- Check the scatter plot for outliers or curves that a single number can hide.
Switch to Sample size to find how many respondents you need to detect a correlation of a given size.
Pearson or Spearman?
Pearson's r measures how closely points follow a straight line and is sensitive to outliers. Spearman's ρ correlates the ranks, so it captures any steady upward or downward pattern and resists outliers. For rating scales and skewed data such as spend, report Spearman alongside Pearson. If the two differ noticeably, look at the chart.
Interpreting the size of a correlation
| |r| | Usual description |
|---|---|
| Under 0.1 | Negligible |
| 0.1 – 0.3 | Weak |
| 0.3 – 0.5 | Moderate |
| 0.5 – 0.7 | Strong |
| 0.7 and above | Very strong |
With large samples, even weak correlations are significant. R² (r squared) is a useful reality check: r = 0.3 means the two measures share only 9% of their variation.
Correlation in key driver analysis
Correlating each attribute rating with overall satisfaction or likelihood to recommend is the simplest form of key driver analysis. It is quick and transparent, but attributes are often correlated with each other, so the strongest correlate is not always the true driver. Use relative-importance methods such as Shapley value or Johnson's relative weights for final recommendations.
Frequently asked questions
Does correlation mean causation?
No. Two measures can move together because one causes the other, because a third factor drives both, or by chance. Experiments and careful modelling are needed to show cause.
How many respondents do I need to detect a correlation?
At 95% confidence and 80% power: about 783 for r = 0.1, 194 for r = 0.2 and 85 for r = 0.3.
Can I correlate a yes/no variable?
Yes. Coding yes = 1 and no = 0 gives the point-biserial correlation, which is Pearson's r on that coding.
Need the respondents, not just the numbers?
Global Vox Populi runs quantitative and qualitative fieldwork in 170+ countries through its own proprietary panels of consumers, B2B and IT decision-makers, physicians, nurses, patients, caregivers and payers. ISO 9001, ISO 20252 and ISO/IEC 27001 certified and HIPAA compliant, following ESOMAR and Insights Association guidelines.
Check feasibilityLink to this tool
Free to use in proposals, courses and articles. If it helped, a link back is appreciated:
<a href="https://globalvoxpopuli.com/tools/correlation-calculator/">Correlation Calculator</a> by Global Vox Populi