Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: MCKelvey and Zavoina

From   Richard Williams <>
To, <>
Subject   Re: st: MCKelvey and Zavoina
Date   Sat, 02 Jun 2012 11:05:46 -0500

At 07:19 AM 6/2/2012, enrico wrote:
Dear Statalist,

my name is Enrico and I am an italian student. I'm working with Stata but I
have a problem. I know that the formula to calculate the Pseudo R-square
MCkelvey and Zavoina is R^2=Var(y*)/(Var(y*)+Var(error)), where Var(error)
in the Logit is equal to 3.14^2/3 while in the Probit is equal to 1.

Why if I use this formula in this case I don't obtein the exact value? In
fact  6.84/(6.84+3.29)=0.675 and not 0.519. Which formula do you use

Wrong formula. V(y*) already includes V(e). You need to SUBTRACT V(e) from the numerator rather than add it to the denominator. Then it becomes the logical analog of Explained variance/ Total variance. So,

(6.84 - 3.29)/6.84 = .519

logit honcomp female read science
fitstat, sav(r2_1)

Measures of Fit for logit of honcomp

Log-Lik Intercept Only:       -115.644   Log-Lik Full Model:
D(196):                        160.236   LR(3):
                                         Prob > LR:
McFadden's R2:                   0.307   McFadden's Adj R2:
ML (Cox-Snell) R2:               0.299   Cragg-Uhler(Nagelkerke) R2:
McKelvey & Zavoina's R2:         0.519   Efron's R2:
Variance of y*:                  6.840   Variance of error:
Count R2:                        0.810   Adj Count R2:
AIC:                             0.841   AIC*n:
BIC:                          -878.234   BIC':
BIC used by Stata:             181.430   AIC used by Stata:

Thank you,

*   For searches and help try:

Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index