Fundamentals of Statistics contains material of various lectures and courses of H. Lohninger on statistics, data analysis and here for more.


T-scores are sometimes used to characterize results in the medical literature (for example, in ivestigations on osteoporosis). T-scores are comparable to z-scores and can be calculated from z-scores by the following simple formula:

T = z*10 + 50

Thus the T-scores are scaled in a way that the mean becomes 50 and the standard deviation becomes 10. By means of this scaling most of the T-scores fall into the range between zero and hundred, which is sometimes said to be interpreted easier than z-scores (I sometimes ask myself, whether medical personell is actually thought to be incapable of interpreting z-scores).

In some papers though the term "T-scores" are used synonymically to z-scores. Further there is the possibilitiy to mix up the T-scores and the t-values obtained in the one sample t-test - which contributes to even more confusion. In summary, T-scores should be avoided whenever possible.