What is the Student's t-Distribution Calculator?
This tool evaluates the Student's t-distribution for a given value x (the t statistic) and degrees of freedom ν. It returns three quantities: the probability density f(x), the lower cumulative probability P(T ≤ x), and the upper cumulative probability Q = P(T > x) = 1 − P. This is pure mathematics and applies universally; no region-specific rules are involved.
How to use it
Enter any real number for x (it may be negative) and a positive number for the degrees of freedom \(\nu\) (commonly an integer such as 5, 10 or 30, but any \(\nu > 0\) is accepted). Press calculate to obtain the density and both tail probabilities. As \(\nu\) grows large the distribution converges to the standard normal \(N(0, 1)\).
The formula explained
The density is $$f(\text{x}) = \frac{\Gamma\!\left(\frac{\nu+1}{2}\right)}{\sqrt{\nu\pi}\;\Gamma\!\left(\frac{\nu}{2}\right)}\left(1 + \frac{\text{x}^{2}}{\nu}\right)^{-\frac{\nu+1}{2}}$$ where \(\Gamma\) is the gamma function. The cumulative probability uses the regularized incomplete beta function: with \(z = \nu/(\nu + \text{x}^{2})\), \(P(T \le \text{x}) = 1 - \tfrac{1}{2}\cdot I_{z}(\nu/2, 1/2)\) for \(\text{x} \ge 0\) and \(\tfrac{1}{2}\cdot I_{z}(\nu/2, 1/2)\) for \(\text{x} < 0\). We compute the density in log space using a Lanczos log-gamma approximation and the incomplete beta via a Lentz continued fraction for numerical stability.
Worked example
For \(\text{x} = 1.0\) and \(\nu = 10\): the density \(f(1) \approx 0.2304\). The lower cumulative probability \(P(T \le 1.0) \approx 0.8303\), so the upper cumulative probability \(Q \approx 0.1697\), matching standard t-tables.
Critical Values of the t-Distribution
The table gives the value t such that the upper-tail probability P(T > t) equals the column heading. The two-tailed level is twice the one-tailed level. These are standard textbook values rounded to three decimals.
| df (ν) | 0.10 (two-tailed 0.20) | 0.05 (two-tailed 0.10) | 0.025 (two-tailed 0.05) | 0.01 (two-tailed 0.02) | 0.005 (two-tailed 0.01) |
|---|---|---|---|---|---|
| 1 | 3.078 | 6.314 | 12.706 | 31.821 | 63.657 |
| 2 | 1.886 | 2.920 | 4.303 | 6.965 | 9.925 |
| 3 | 1.638 | 2.353 | 3.182 | 4.541 | 5.841 |
| 5 | 1.476 | 2.015 | 2.571 | 3.365 | 4.032 |
| 10 | 1.372 | 1.812 | 2.228 | 2.764 | 3.169 |
| 15 | 1.341 | 1.753 | 2.131 | 2.602 | 2.947 |
| 20 | 1.325 | 1.725 | 2.086 | 2.528 | 2.845 |
| 30 | 1.310 | 1.697 | 2.042 | 2.457 | 2.750 |
| 60 | 1.296 | 1.671 | 2.000 | 2.390 | 2.660 |
| 120 | 1.289 | 1.658 | 1.980 | 2.358 | 2.617 |
| ∞ (normal) | 1.282 | 1.645 | 1.960 | 2.326 | 2.576 |
As the degrees of freedom grow, the critical values decrease toward the standard normal values in the last row.
Interpreting the Output
- Density f(x) is the height of the curve at x. It is not a probability; for a continuous distribution, probabilities come from areas under the curve between values.
- Lower cumulative P(T ≤ x) is the area to the left of x. Upper tail Q = 1 − P is the area to the right of x.
- One-tailed p-value: for an observed t statistic, Q (for a positive statistic in a right-tailed test) is the p-value. For a left-tailed test with negative t, the p-value is P(T ≤ t).
- Two-tailed p-value: use 2Q computed at |x|, because the distribution is symmetric and both tails count.
- Symmetry: the t-distribution is symmetric about 0, so P(T ≤ −x) = Q(x) = 1 − P(T ≤ x), and f(−x) = f(x).
- Comparing with critical values at α = 0.05: a two-tailed test rejects when |t| exceeds the 0.025 one-tailed critical value (for example 2.228 at df = 10, or 2.042 at df = 30). A one-tailed test at α = 0.05 uses the 0.05 column (for example 1.812 at df = 10). Equivalently, the p-value is below 0.05.
- Heavy tails: for small ν, extreme values are more probable than under the normal curve. At x = 2, Q is about 0.0510 at df = 5 versus 0.0228 for the normal. As ν increases, the t-distribution approaches the standard normal.
Example of the two-tailed rule: with df = 5 and |t| = 2, Q ≈ 0.0510, so the two-tailed p-value is about 2 × 0.0510 = 0.102, which is above 0.05.
FAQ
Can \(\nu\) be a non-integer? Yes. The formula is valid for any real \(\nu > 0\); the calculator accepts decimal degrees of freedom.
What does the upper cumulative probability mean? It is the area in the right tail, \(P(T > \text{x})\). For a two-sided p-value with positive x you would use \(2\cdot Q\).
Why does it look like a normal curve for large \(\nu\)? As degrees of freedom increase, the t-distribution's heavy tails shrink and it approaches the standard normal distribution.