Does the delta output from a k-fold cross validation indicate the estimated classification error?

Question

I'm currently working on a logistic regression analysis and want to determine if my model validates well. I used the following R code using the "boot" package:

results <- glm(R0A1 ~ SeasonNew * (MP_Scaled + MPHW_Scaled + HW_Scaled + YP_Scaled + AG_Scaled + Shrub_Scaled), family = binomial, data = turkey2nd)
cost <- function(r, pi = 0) mean(abs(r - pi ) > 0.5)
(output <- cv.glm(turkey2nd, results, K=10)$delta)

My estimated delta was 0.23; therefore, is the correct classification 1 - 0.23 = 0.77 or 77% accuracy? Thanks for the assistance!

Answer 1

Yes.

The output of the cv.glm function in the case of binary classification is the estimation of the classification error rate on the test set (it is computed by averaging the classification error rate of the corresponding K validation sets).