It would be most useful if you can provide a reproducible example (reprex) and not a screenshot. Share the results of dput(head(dataframename, 20))
is one way to do this. Here's some more info about making a reprex: FAQ: What's a reproducible example (`reprex`) and how do I create one?
As to your question, I think Ever_Married has values of "" and not NA. To remove these, there's a few options. Here's one:
library(tidyverse)
newdata <- data %>%
filter(Ever_Married!="")