Scenario A
Same r, small class
r
0.45
n
8
t
1.234
p
0.26
You found r = 0.45 in an 8-person class. The p-value stays high because the sample is too small to separate signal from noise.
Conclusion: Not significant.
Correlation significance
P-Value for CorrelationWhat It Means & How to Use It
You calculated r. This page answers the next question: is that relationship statistically meaningful, or could it be random noise from your sample?
Featured answer
The p-value for a correlation tells you the probability that the observed relationship occurred by chance. A p-value < 0.05 means the correlation is statistically significant.
Quick test
Enter r and sample size
Two-tailed p-value
0.26
t statistic
1.234
df
6
test
Two-tailed
The relationship is not statistically significant and may be explained by chance or low power.
Formula used: t = r * sqrt((n - 2) / (1 - r^2)), df = n - 2, two-tailed p from the t distribution.
The intuition
Same r, Different Confidence
The p-value is not only about the size of r. It also depends heavily on n. Larger samples make the same pattern harder to dismiss as coincidence.
Scenario B
Same r, large study
r
0.45
n
200
t
7.091
p
< 0.0001
You found the same r = 0.45 in 200 people. Now the p-value collapses because the same pattern is repeated across many more observations.
Conclusion: Highly significant.
Core insight: the same r value can be non-significant in a tiny sample and highly significant in a large sample. The correlation coefficient tells you relationship strength. The p-value tells you how much statistical evidence you have that the relationship is not random.
How to calculate it
Correlation Hypothesis Test
t = r * sqrt((n - 2) / (1 - r^2))
df = n - 2
two-tailed p = P(|T| >= |t|)Step 1: Calculate the t statistic
Plug your r and n into the formula. Stronger r and larger n both increase the absolute t value.
Step 2: Determine degrees of freedom
For a Pearson correlation test, df = n - 2.
Step 3: Compare against the t distribution
A calculator or t distribution table converts the t statistic into a two-tailed p-value.
Common mistakes
What p-Value Does Not Say
Significance is useful, but it is easy to overread. These are the three mistakes that cause the most bad interpretations.
Mistake
"p < 0.05 means the correlation is strong."
p-value only tests statistical significance. Correlation strength is measured by r.
Review the Pearson correlation calculatorMistake
"p > 0.05 means there is no relationship."
A non-significant p-value may mean the sample is underpowered, especially when n is small.
Mistake
"A significant correlation proves causation."
Statistical significance does not remove confounding variables, reverse causality, or study design limits.
Read correlation vs causationFAQ
Correlation P-Value FAQ
Short answers to the questions that usually come up after a calculator outputs r and p together.
What does p-value mean in correlation?+
It tells you the probability that the observed relationship could occur by chance if the null hypothesis were true. In practice, p < 0.05 usually means the correlation is statistically significant.
What p-value is acceptable for correlation?+
In most fields, p < 0.05 is the standard threshold. In medical research or higher-risk work, p < 0.01 is often preferred.
Can I have a high r but high p-value?+
Yes - if your sample size is very small, even a high r may not reach significance. For example, n = 5 can leave r = 0.7 non-significant.
My calculator shows r = 0.3, p = 0.001. Is this useful?+
It is statistically significant, but the effect size is still weak. Significance does not equal practical importance, so interpret r and p together.
Does a significant correlation prove causation?+
No - a significant correlation does not prove causation. It only says the observed relationship is unlikely to be random under the null hypothesis.
Next step
Let the calculator compute r and p
If you have paired data instead of a finished r value, use the Pearson calculator so the coefficient and significance test stay connected.
Related links