+ Reply to Thread
Results 1 to 2 of 2

Thread: interpreting interaction results in ols regression

  1. #1
    Points: 573, Level: 11
    Level completed: 46%, Points required for next Level: 27

    Posts
    2
    Thanks
    1
    Thanked 0 Times in 0 Posts

    interpreting interaction results in ols regression




    Dear all,

    I am trying to examine the association between household wealth and expenditure on education of children aged 11-17 and 5-10 in a developing country using the following equation. In particular, I am trying to find out, using interactions, whether there is a gender bias in education expenditure when households become wealthier. HHwealth is a continuous variable while femalechild is a binary variable that equals 1 if the child is female, and 0 otherwise.

    educexp = β0 + β1hhwealth+ β2 femalechild+ β3 hhwealth*femalechild + εij

    The following results (I’m only pasting relevant sections of the results) are obtained using Stata’s regress command.

    Would it be correct to say that the results in Model 1 suggest that household wealth is significantly associated with in an increase in educational expenditure for both boys and girls in the older age group? And Model 2 shows that increasing household wealth is significantly associated with an increase in educational expenditure for only boys aged 5-10, thus suggesting a possible gender bias? Thanks in advance!


    Code:



    Code: 
    
    regress educexp hhwealth fem hhwealth_fem `xlist' 
     
    //edu=amount spent on education, hhwealth=wealth score for household (ranges from 0 to 1), fem=1 if child is female, 0 if otherwise, hhwealth_fem =interaction of hh_wealth and fem, ‘xlist’=other controls   
     
     
    /*tests for the total coef of education expenditure for girls */    
     
    test hhwealth + hhwealth_fem = 0
    
    //Model 1, children aged 11-17
    
    Linear regression                                      Number of obs =    1389
                                                           F( 25,  7363) =   69.86
                                                           Prob > F      =  0.0000
                                                           R-squared     =  0.2122
                                                           Root MSE      =  1.0857
    
    ---------------------------------------------------------------------------------
                    |               Robust
         dietscore9 |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
    ----------------+----------------------------------------------------------------
           hhwealth |  0.2369617   .1049415    -2.26   0.024     -.442677   -.0312463
       hhwealth_fem |  -.0409221   .1408595    -0.29   0.771    -.3170471    .2352029
    
    ( 1)  hhwealth + hhwealth_fem = 0
    
           F(  1,  7363) =    7.83
           Prob > F =    0.0052
    
    
    
    //Model 2, children aged 5-10
    
    Linear regression                                      Number of obs =    1983
                                                           F( 28,  1954) =   16.53
                                                           Prob > F      =  0.0000
                                                           R-squared     =  0.1798
                                                           Root MSE      =  .95551
    
    ---------------------------------------------------------------------------------
                    |               Robust
         dietscore7 |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
    ----------------+----------------------------------------------------------------
           hhwealth |  -.4282509    .168332    -2.54   0.011      -.75838   -.0981218
       hhwealth_fem |   .1345255   .2562779     0.52   0.600    -.3680813    .6371323
    
    
     ( 1)  hhwealth + hhwealth_fem = 0
    
           F(  1,  1954) =    2.16
           Prob > F =    0.1421

  2. #2
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: interpreting interaction results in ols regression


    not a stata person, but I don't see a main effect term for female in the models?
    Stop cowardice, ban guns!

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats