Hi all,
I am using the MICE package to handle missing data in my logistic regression (I have missing data for my binary predictor, pls see the complete data below). However, I got the completely same results between with vs without MICE (which is strange). Am I doing anything wrong in terms of my coding below?
data_1$SmokingNA <- ifelse(data_1$Smoking.Status == "", NA, data_1$Smoking.Status) ## replacing empty cells with NA
m <- glm(disease ~ SmokingNA, data=data_1, family=binomial) ## non-imputed logistic regression
library(mice)
imp1 <- mice(data_1, m = 5) ## MICE
fitm <- with(imp1, glm(disease ~ SmokingNA, data=data_1, family=binomial)) ## Imputed logistic regression
summary(m)
summary(fitm)
## both show the same results (coefficient, p-value, standard error, etc)
Below is the data (Predictor: smoking status; Outcome: disease):
structure(list(Smoking.Status = c("smoking", "smoking", "smoking",
"smoking", "smoking", "non-smoking", "smoking", "non-smoking",
"non-smoking", "non-smoking", "smoking", "non-smoking", "non-smoking",
"smoking", "non-smoking", "smoking", "smoking", "non-smoking",
"non-smoking", "", "", "", "", "", "", "", "non-smoking", "",
"", "non-smoking", "smoking", "non-smoking", "non-smoking", "smoking",
"non-smoking", "non-smoking", "non-smoking", "non-smoking", "non-smoking",
"", "non-smoking", "smoking", "non-smoking", "non-smoking", "smoking",
"non-smoking", "smoking"), disease = structure(c(1L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 2L, 1L, 2L, 2L, 2L, 1L,
1L, 2L, 1L, 2L, 1L, 2L, 2L, 1L, 2L, 2L, 2L, 2L, 1L, 2L, 2L, 1L,
2L, 1L, 1L, 2L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L), .Label = c("none",
"disease"), class = "factor")), row.names = c(NA, 47L), class = "data.frame")