Why two analogous codes generating different results?

[Phuc Nguyen]

Dear all, I have two codes for xtdidregress :

     xtdidregress (lpay_turn_w1  $logshortcov_np con_of_corr rul_of_law reg_qua gov_eff poli_stab voice_acc) (post1treatvn), group(TYPE2) time(year1)
     estat ptrends

and

          xtdidregress (lpay_turn_w1  $logshortcov) (post1treatvn), group(TYPE2) time(year1)
     estat ptrends

The results from code 1 is

Parallel-trends test (pretreatment time period)
H0: Linear trends are parallel

F(1, 10233) =  10.00
   Prob > F = 0.0016

And for code 2 is:

Parallel-trends test (pretreatment time period)
H0: Linear trends are parallel

F(1, 8230) =   0.41
  Prob > F = 0.5205

when $logshortcov = $logshortcov_np con_of_corr rul_of_law reg_qua gov_eff poli_stab voice_acc

Could you please tell me why there is such a difference?

[Andrew Musau]

Show the exact Stata output from the part where you define the globals to the results of the two regressions.