Binary, Ordinal, and Categorical Data Analysis - Agricultural Statistical Data Analysis Using Stata - page 219

he expperc( 50 50 ) is equivalent to a 1:1 ratio and the observed
frequencies do not differ from this ratio, p = 0.3447.
Finally, when dealing with ratios of just two categories, as with
these simple inheritances where the degrees of freedom will be 1,
using Yates' correction will give more accurate results. Yates' correc-
tion is nothing more than subtracting 0.5 as in the following equation:

( )

2

k

∑ OB

−− Ex

Ex

05 .

χ 2 =

i = 1

To incorporate this modification to the csgof command enter

which csgof

This will give you the pathname to this command, which you should
copy. Then select the View… menu item under the File menu and
paste the pathname in the dialog box. This will open a Viewer win-
dow with a list of the csgof ado' command. Select this entire file
and copy it into a Do-File Editor window. At this point you can make
modifications to this file. You can run these modifications and use the
modified command immediately or you can save the file for later use.
I would suggest that you not overwrite the original csgof command,
but store the modified file elsewhere. It is standard and safe advice to
rename the modified file/command. This also will require changing
the program lines in the file. To change the program lines find the
following and change them as indicated:
Original lines:

capture program drop csgof
program define csgof

New lines:

capture program drop csgofy
program define csgofy

Find the following line in the csgof file:

quietly gen `chi1' = ( obsfreq - expfreq )^2/ expfreq