Announcement

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

  • getting different "population sizes", coefficients and p values from survey weighted logistic on same dataset

    I am obtaining different results from what I believe is the same survey weighted logistic regression on the same data set.

    The main difference is that one says "Population size = 25,327" and the other "Population size = 43,676

    Can anyone explain why this might be the case? I thought my weights were the same in each sample but I puzzled as to why it is generating a different population size.


    Here is my code followed by a sample output:

    mi estimate, or: svy: logistic visa foreigner fujian pregnant otherpaidtrip othersapply age male migrant agent disadveth geotravex1 geotravex2 pargeotravex1 pargeotravex2 ocpr0 lsal satsal educyear schpc9foreig parocpr pareducyear ifoplive lwealthwosqm humanities socialscience professions stem business married numsib jobyear chimin parret numfaminus famtiegravsponsor



    Multiple-imputation estimates Imputations = 40
    Survey: Logistic regression Number of obs = 2,395

    Number of strata = 1 Population size = 43,676
    Number of PSUs = 2,395
    Average RVI = 0.0792
    Largest FMI = 0.2823
    Complete DF = 2394
    DF adjustment: Small sample DF: min = 387.79
    avg = 1,697.47
    max = 2,361.48
    Model F test: Equal FMI F( 35, 2368.0) = 5.58
    Within VCE type: Linearized Prob > F = 0.0000


    example coefficient:
    ifoplive | 1.420166 .2310829 2.16 0.031 1.031706 1.954889

    mi estimate, or: svy: logistic visa foreigner fujian pregnant otherpaidtrip othersapply age male migrant agent disadveth geotravex1 geotravex2 pargeotravex1 pargeotravex2 ocpr0 lsal satsal educyear schpc9foreig parocpr pareducyear ifoplive lwealthwosqm humanities socialscience professions stem business married numsib jobyear chimin parret numfaminus famtiegravsponsor


    Multiple-imputation estimates Imputations = 40
    Survey: Logistic regression Number of obs = 2,395

    Number of strata = 1 Population size = 25,327
    Number of PSUs = 2,395
    Average RVI = 0.0939
    Largest FMI = 0.3811
    Complete DF = 2394
    DF adjustment: Small sample DF: min = 232.51
    avg = 1,616.43
    max = 2,363.46
    Model F test: Equal FMI F( 35, 2359.5) = 5.67
    Within VCE type: Linearized Prob > F = 0.0000

    ifoplive | 1.365883 .2233156 1.91 0.057 .9907228 1.883105

    Thanks!
Working...
X