st: finding incorrectly entered data

[Michael Eisenberg <meisenberg713@gmail.com>]

Colleagues,

I have what I hope will be a simple question.

I have a list of several thousand patients with a lab test and days
from the test recorded.  I'd like to be able to analyze the days from
test data as a continuous variable.  Unfortunately, some of the data
in the days from lab test are not numeric and have odd characters
(e.g. ?, +-5, na, etc).  See below.  Thus, I cannot run any summary
statistics on it.

I cannot manually look through all the data due to the number of men
to find the nonnumeric data.  Is there a way I can have stata somehow
flag the rows with nonnumeric data?

Thank you in advance.

Mike



id    test   daysfromtest
1     41          2
2     43          4
3     45          6
4     44          10
5     48          ?
6     44          +-5
7     24          na
8     74          2
9     94          3
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/