Is it possible to calculate the percentile from a T-score?
If yes, what formula can be used for it, and what other variables (e.g. standard deviation, degrees of freedom, etc.) are needed for it?
Is it possible to calculate the percentile from a T-score?
If yes, what formula can be used for it, and what other variables (e.g. standard deviation, degrees of freedom, etc.) are needed for it?
It is possible to calculate the percentile given a t-statistic and the degrees of freedom. This is the CDF at the given t-value.
The formula for this is as follows, where $\nu$ is the degrees of freedom, $t^{2} < \nu $, and $_{2}F_{1}$ is a hypergeometric function.
$\int^{t}_{-\inf} f(u) du = \frac{1}{2} + t [ \frac{\Gamma(\frac{\nu + 1}{2})}{\sqrt{\pi\nu}\Gamma(\frac{\nu}{2})}] [ _{2}F_{1}(\frac{1}{2}, \frac{\nu + 1}{2}; \frac{3}{2}; -\frac{t^2}{\nu})]$
This can be easily done in R using pt()
.
Yes - to do so would require a complicated calculus calculation, however.
Textbooks usually include a table of t-values and their corresponding approximate percentiles based on the degrees of freedom.
There are also many free web-based technologies that can do this quickly, such as