Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

A table from a psychological test is giving percent ranks (PR).

How can I transform those into t-values by calculation? (Assuming normal distribution.)

I found a lot of references on how to transform z --> t, but none on PR --> t.

Way back in university, we had tables with the bell-curve, to look up PR, z-values, t-values and stanines - but how to calculate?

share|improve this question
    
check this one en.wikipedia.org/wiki/Student's_t-test –  Seyhmus Güngören Feb 22 '13 at 22:43
1  
When you say $t$-values, it suggests that you're using one Gosset's $t$-distributions (not the normal distribution, which always uses $z$-values). You need the $\textit{inverse}$ of the cumulative distribution function (CDF) for the particular distribution. –  Sammy Black Feb 22 '13 at 22:44
    
@SeyhmusGüngören: I think this is different from what I need. –  michi Feb 22 '13 at 22:50
    
@SammyBlack: I am not very experienced in statistics, but I am quite sure that those psychological tests at least assume normal distribution. You're right, z-values are common, but t-values, too. I found a lot of references on how to transform z --> t, but none on PR --> t. Way back in university, we had tables with the bell-curve, PR, z-values, t-values and stanines - but how to calculate? –  michi Feb 22 '13 at 22:54
    
try stats.stackexchange.com –  john mangual Feb 22 '13 at 23:26
show 3 more comments

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.