I am running a logistic regression with multiply-imputed data and using the svy: command for complex survey data. I then use outreg2 to send the output into a table. For statistical significance, I am setting 3 levels .05, .01., and .001. Everything works fine, but a few of the results have asterisks that do not match the P-value that is listed in the raw stata output. In the example below, the P value is 3.07 in the raw output...but only 1 asterisk shows up on the outreg2 table that is produced. I can easily check and change these by hand...but I am wondering why this mismatch may be occurring and if I am missing something conceptually.
Here is the syntax:
mim,storebv: svy:logit money_action par_exp4 par_aspr4 par_bach grandpar_bach age_parent income_ten_thou numkids ///
female asian hispanic black par_nativeng parttime nowork catholic private ///
par_involv heldback ztest_score daycare nursery headstart kinder, or
outreg2 using moneyaction-0915, se replace ef alpha(0.001, 0.01, 0.05) bdec(3) tdec(3) rdec(3) word
Here is the raw output of the "nursery" variable:
money_action | Odds Rat. Std. Err. t P>|t| [95% Conf. Int.] MI.df
nursery | 1.17315 .061089 3.07 0.002 1.0587 1.29997 211.7
here is the row in the outreg2 table:
nursery 1.173*
*
* 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/
