Syntax
RSQ(known_y, known_x)
| Parameter | Description |
|---|---|
| known_y | Parameter of the RSQ function. |
| known_x | Parameter of the RSQ function. |
Examples
Model fit quality
Formula
=RSQ(B2:B50, C2:C50)
Returns 0.92 — 92% of the variance in B is explained by C. This is a strong linear relationship.
Advertising effectiveness
Formula
=RSQ(D2:D13, E2:E13)
If D is monthly sales and E is ad spend, an RSQ of 0.75 means 75% of sales variation is explained by ad spend variation.
Weak relationship
Formula
=RSQ(A2:A100, B2:B100)
An RSQ of 0.08 means x explains only 8% of the variation in y — the linear model is a poor fit.
Common Errors
#N/A
The arrays have different lengths.
#DIV/0!
Zero variance in one or both arrays (all values identical).
Tips
RSQ = CORREL squared
RSQ is simply the square of the Pearson correlation coefficient. CORREL of 0.9 gives RSQ of 0.81.
Interpreting the value
RSQ > 0.7 is generally considered a good fit, 0.4-0.7 moderate, below 0.4 weak. But thresholds vary by field — social sciences accept lower values than engineering.
Only measures linear fit
RSQ measures linear fit. A perfect quadratic relationship (y = x^2) may show low RSQ because it is not a straight line. Always visualize with a scatter plot.
Try RSQ in Viztab
Import your data and use RSQ with 370+ other formulas. No signup required.
Open Viztab