st: Problem identifying a specific value of a string variable

[Brad Humphreys <brhumphreys@gmail.com>]

Dear Statalisters,

I have a string variable that I cannot seem to get Stata to perform
operations on.  This is part of a multiple step string to numeric
conversion. The string variable is col1_d

. desc col1_d

              storage  display     value
variable name   type   format      label      variable label
-------------------------------------------------------------------------------------------
col1_d          str5   %9s

. tab col1_d

     col1_d |      Freq.     Percent        Cum.
------------+-----------------------------------
         2e |          1        0.00       29.09

The variable takes many values.  I have not reported the rest, as I
want to keep them. I can't get that observation, or about 10 more,
identified, using -list-, -drop-, or -replace-.  I tried:

. l if col1_d=="2e"

Returns nothing.  I tried:

. drop if col1_d=="2e"
(0 observations deleted)

I tried -trim-, -ltrim-, strltrim- on the variable, thinking there
could be leading blanks in there.  No luck.  I tried copying the
offending value for col1_d from the data editor into the do file
command.  No luck.  The variable col1_d is created using this command

gen col1_d=col1 if frac_col1==0

where col1 is a string variable.

. desc col1

              storage  display     value
variable name   type   format      label      variable label
-------------------------------------------------------------------------------------------
col1            str10  %10s


Any help would be appreciated.   I am at wits end.

Brad Humphreys

*
*   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/