PMCC calculator
PMCC, r, from 8 pairs
0.9994
Significant: the critical value is 0.6215.
- r
- 0.9994
- Critical value
- 0.6215
- Decision
- Reject H₀
Working
- n = 8 pairs. Σx = 36, Σy = 72.2.
- Sxx = Σx² − (Σx)² ÷ n = 42; Syy = Σy² − (Σy)² ÷ n = 167.795; Sxy = Σxy − ΣxΣy ÷ n = 83.9.
- r = Sxy ÷ √(Sxx × Syy) = 83.9 ÷ √(42 × 167.795) = 0.9994.
- H₀: ρ = 0. H₁: ρ > 0. One-tailed at the 5% level, n = 8.
- Critical value: 0.6215 (from t on 6 degrees of freedom: r = t ÷ √(n − 2 + t²)).
- 0.9994 is beyond the critical value, so reject H₀: there is evidence of positive correlation in the population.
Your result
0.9994. Significant: the critical value is 0.6215.
Practise correlation questions
A-level Maths Pro
Get an A* or your money back
£129 once
Price rises to £199 on 1 January. Save £70 by joining before then.
At least 12 months' access
No VAT is charged. Conditions
A PMCC calculator turns your pairs into r = Sxy ÷ √(Sxx × Syy) and tests it: with 10 pairs, a one-tailed test at 5% needs r above the critical value 0.5494.
Questions people ask
What does the PMCC measure?
How close the points lie to a straight line: r = 1 is a perfect positive line, r = −1 a perfect negative one and r near 0 no linear correlation.
How do you test the PMCC?
H₀: ρ = 0 against H₁: ρ > 0, ρ < 0 or ρ ≠ 0. Reject H₀ if r is beyond the critical value for your sample size and significance level.
Where do PMCC critical values come from?
From the t distribution on n − 2 degrees of freedom. The tables in exam papers are made this way, so a computed value matches them.
Does a significant PMCC prove one variable causes the other?
No. Correlation shows association only; a third factor can drive both.
More free tools
Keep going
How we work this out
What it does
Paste your x and y values to get r, then test it at the level you choose with a critical value computed for your sample size.
Product-moment correlation for 4 to 200 pairs, and the test of H₀: ρ = 0.
Method
- Sxx, Syy and Sxy come from the sums of the data; r = Sxy ÷ √(Sxx × Syy).
- The critical value is computed from the t distribution on n − 2 degrees of freedom: r = t ÷ √(n − 2 + t²). This gives the values printed in exam-board tables, to 4 decimal places, for any n.