GraphPad Prism report exact P values with most statistical calculations using these algorithms, adapted from sections 6.2 and 6.4 of Numerical Recipes.
PFromF(F_Ratio, DF_Numerator, DF_Denominator) =
BetaI(DF_Denominato /2, DF_Numerator/2, DF_Denominator / (DF_Denominator + DF_Numerator * F_Ratio))
PFromT(T_Ratio, DF) = BetaI(DF /2, 1/2, DF / (DF + T_Ratio^2))
PFromZ(Z_Ratio) = PFromT(Z_Ratio, Infinity)
PFromR(R_Value) = PFromT(|R_Value| / SQRT((1 - R_Value^2)/DF) , DF)
PFromChi2(Chi2_Value, DF) = GammaQ(DF / 2, Chi2Value /2)
Note that BetaI is the incomplete beta function, and GammaQ is the incomplete gamma function. The variable names should all be self-explanatory.
If you want to compute P values using newer (2010 and later) Excel, use these functions:
P value from F |
=F.DIST.RT (F, DFn, DFd) |
P value from t (two tailed) |
=T.DIST.2T(t, df)
|
P value from Chi Square |
=CHISQ.DIST.RT(ChiSquare, DF) |
P value from z (two tailed) |
=2*(1.0-NORM.S.DIST(z,TRUE)) |
If you want to compute P values using older (pre 2010) Excel, use these functions:
P value from F |
=FDIST (F, DFn, DFd) |
P value from t (two tailed) |
=TDIST (t, df, 2) |
P value from Chi Square |
=CHIDIST (ChiSquare, DF) |
P value from z (two tailed) |
=2*(1.0-NORMSDIST(z)) |
Numerical Recipes 3rd Edition: The Art of Scientific Computing, by William H. Press, Saul A. Teukolsky, William T. Vetterling, IBSN:0521880688.