Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • interpreting lincom output

    Dear members of the community,

    I am having difficulty interpreting the lincom output:

    Code:
    Logistic regression                                     Number of obs = 19,118
                                                            Wald chi2(37) =      .
                                                            Prob > chi2   =      .
    Log pseudolikelihood = -3725.0714                       Pseudo R2     = 0.2564
    
                                                            (Std. err. adjusted for 39 clusters in PSUID)
    -----------------------------------------------------------------------------------------------------
                                        |               Robust
                                   PRS | Odds ratio   std. err.      z    P>|z|     [95% conf. interval]
    ------------------------------------+----------------------------------------------------------------
                 TORT_FB1_FBN_PW_NW1230 |
                                     1  |   1.534704   .2820355     2.33   0.020     1.070527    2.200146
                                     2  |    1.54518   .1708339     3.94   0.000     1.244145    1.919054
                                     3  |   1.438381   .1204091     4.34   0.000     1.220727    1.694842


    Code:
    . lincom  2.TORT_FB1_FBN_PW_NW1230- 1.TORT_FB1_FBN_PW_NW1230, or
    
     ( 1)  - [PRS]1.TORT_FB1_FBN_PW_NW1230 + [PRS]2.TORT_FB1_FBN_PW_NW1230 = 0
    
    ------------------------------------------------------------------------------
            PRS | Odds ratio   Std. err.      z    P>|z|     [95% conf. interval]
    -------------+----------------------------------------------------------------
             (1) |   1.006826    .220022     0.03   0.975     .6560565     1.54514
    ------------------------------------------------------------------------------

    How to interpret the odds ratio (1.006826 ) in the second code?

    regards,
    ajay
    Last edited by ajay pasi; 26 Dec 2022, 23:39.

  • #2
    Ajay:
    I'd say that the ratio between the two =Rs does not reach statistical signifcance as the indifference value (that is, 1) is included within the 95% CI.
    What does this mean substantively, I cannot say.
    Kind regards,
    Carlo
    (Stata 19.0)

    Comment


    • #3
      Thanks, Carlo. Appreciate the help.

      Comment

      Working...
      X