Deleting the missing values (not entire observations) of a variable

Suppose I have the following table:

ID	Preferences
1	2
1	.
1	.
2	.
2	3
3	.
3	.
3	4
4	1

I have other columns before 'ID' and after 'Preferences'. What I want to get is two columns of the following type on the right of the above table:

ID	Preferences
1	2
2	3
3	4
4	1

In other words, I want to delete the missing values of the variable 'Preferences'. I don't want to delete observations. I only want to record 'Preferences' with non-missing values. So

Code:

drop Preferences if Preferences != .

would not work in this case. This would delete observations with missing values for Preferences, which is not what I want.

How can I do this?

Share on FacebookPlus on Google+