Calculate Syy, the sum of squared deviations of y, from y values or from Σy, Σy², and n with step-by-step work.

Syy Calculator

Syy = Σ(yᵢ − ȳ)² — the sum of squared deviations of y.

From y values
From Σy & Σy²

Related Calculators

Syy Formula

Syy is the sum of squared deviations of the y values from their mean. Two equivalent formulas:

Syy = Σ(yᵢ − ȳ)²
Syy = Σyᵢ² − (Σyᵢ)² / n
  • yᵢ — the i-th y value
  • ȳ — mean of the y values, Σyᵢ / n
  • n — number of y values
  • Σyᵢ — sum of the y values
  • Σyᵢ² — sum of the squared y values

Both formulas give the same result. The second (computational) form avoids computing the mean first and is what the calculator uses in the Σy / Σy² mode. Syy is always ≥ 0, and Syy = 0 only when every y value is identical. Divide Syy by n − 1 to get the sample variance of y, or by n for the population variance.

Reference values

Syy on its own is just a sum, so its size depends on n and the spread of y. The table below shows how Syy connects to other common statistics.

Quantity Formula using Syy
Sample variance s²Syy / (n − 1)
Population variance σ²Syy / n
Sample standard deviation s√(Syy / (n − 1))
Total sum of squares (SST)SST = Syy
Coefficient of determination R²1 − SSE / Syy
Pearson correlation rSxy / √(Sxx · Syy)
Regression slope b₁Sxy / Sxx (Syy used for R²)

Quick example: for y = {4, 7, 9, 12, 15}, ȳ = 9.4, and Syy = 30.96 + 5.76 + 0.16 + 6.76 + 31.36 = 75.0.

Dataset (y values) n ȳ Syy
5, 5, 5, 5450
2, 4, 6, 84520
10, 20, 30, 40, 505301000
4, 7, 9, 12, 1559.475.2

Worked example and FAQ

Example using the computational formula. Suppose Σy = 47, Σy² = 515, n = 5.

  • (Σy)² = 47² = 2209
  • (Σy)² / n = 2209 / 5 = 441.8
  • Syy = 515 − 441.8 = 73.2

Is Syy the same as SST? Yes. In simple linear regression, the total sum of squares about ȳ is exactly Syy.

Can Syy be negative? No. It is a sum of squares. If you get a negative value from the computational formula, you have a rounding error in Σy² or Σy.

Difference between Syy and variance? Syy is the numerator. Divide by n − 1 for sample variance, or n for population variance.

When do I use Syy vs Sxx? Sxx uses x values, Syy uses y values. Both appear in the regression slope, correlation, and R² formulas.