Postestimation output from logistic regression

[Joceynne McRae]

Hello,

I am new to logistic regression and I am working through some exercises. I have undertaken a postestimation test using the command 'estat class' which produces the following output:

I write because, in a text book it refers to the specificity as the percentage of the cases that were incorrectly classified for those having no good health (Y = 0). In this example, the specificity = 6.42%.

However, if I understand statistic correctly, this should mean the percentage of cases that were correctly classified as not have the outcome of interest. ie classified '-'.

A penny for anyone's thoughts?!

With thanks
Jocelynne

[Clyde Schechter]

I write because, in a text book it refers to the specificity as the percentage of the cases that were incorrectly classified for those having no good health (Y = 0).

Either you are misunderstanding that book, or that book needs to be burned!

However, if I understand statistic correctly, this should mean the percentage of cases that were correctly classified as not have the outcome of interest. ie classified '-'.

That is correct.

[Joceynne McRae]

Thanks Clyde.
Since I'm not into book burning I'll assume it's a typo!!!