Why is STATA giving me (.) correlation? How do I justify this?

[Liv Fernandez]

I do not know how to go about this. Please help

[Jared Greathouse]

Please read the FAQ to see how to ask a question.

[Nick Cox]

It seems that you have (this may not be a complete list) one of.


1. at most one observation in which both variables are defined

2. one variable being constant as the other varies

3. both variables being constant for observations in which both are non-missing.


Your first statistics course should have encouraged the use of a scatter plot whenever you calculate a correlation, especially if a correlation seems puzzling. Here

scatter wage inlf

will, I predict, reveal the problem as an empty graph, a single data point (possibly repeated), or a stripe of constant values for one variable.

You can also go

summarize wage inlf if !missing(wage, inlf) 

Once you have worked out what is going on, whether there is a fix and if so what it is should be clearer.