Close

Complete Captcha

• 9 years - 46 posts - Latest - RSS
• Thread: 868 Goods vs 48 No Goods

1. Economist
   4c2e
   

   So I am a sociologist trying to estimate the effects of gender. Not too hard using dummies.

   For gender identity. I have two dummies. One if the person identifies male and one if they identify as nongendered.

   Two more for gender expression, one for maleness, one for agendered.

   Two more for biological sex. One for maleness, one for intersex

   three more for orientation. One for attracted to men, one for attracted to nobody, one for attracted to both males and females.

   Then i added a bunch of interaction terms. Because the effect of being biologically intersex may be different for a male identified woman expression versus a female identified woman expression.

   I have a smaple size of 100. I run Stata and it just refuses to estimate some of these dummies, it just throws them out. Something about "colinearity," which I presume is code for heteronoratism. Disgusting. It should be ashamed.

   9 years ago # QUOTE 502 Volod 6 Vlad !

2. Economist
   f28a
   

   Stata is an expression of microfascism, and the robust option just microraped me.

   9 years ago # QUOTE 78 Volod 2 Vlad !

3. Economist
   7b10
   

   Alas it's true. I've been micro-raped many times by Stata when an effect turned out not to be robust after clustering the standard errors. Raped I tell you. Oh it hurted, and it wasn't worth it. We should sue the makers of Stata!

   9 years ago # QUOTE 21 Volod 4 Vlad !

4. Economist
   3021
   

   I want to let my residuals to be able to express freely their heteroskedasticity.

   Also I hate minimizing the sum of the squares . It makes me feel oppressive.

   9 years ago # QUOTE 64 Volod 3 Vlad !

5. Economist
   8cd7
   

   You fools, you don't know the secret command of Stata! Only after you run 1kkk regressions\cite(Xavier) will you be allowed a glimpse into the secret commands that let you keep the whole set of controls. Keep regging!

   9 years ago # QUOTE 4 Volod 6 Vlad !

6. Economist
   1c5a
   

   I guess in the next edition of Baby Woolwridge he can change the name of this from "dummy variable trap" to "dummy sociologist trap".

   9 years ago # QUOTE 20 Volod 1 Vlad !

7. Economist
   83ba
   

   "dummy sociologist trap"

   Kind of redundant, don't you think?

   9 years ago # QUOTE 17 Volod 2 Vlad !

8. Economist
   8a86
   

   Your specification is flawed. Add a dummy for transitioning.

   9 years ago # QUOTE 29 Volod 1 Vlad !

9. Economist
   837d
   

   Deserves more love, I lol'd

   9 years ago # QUOTE 4 Volod 1 Vlad !

10. Economist
    6fb1
    

    Simple solution: recode all of this into one single variable and consecutively number the cases, i.e.

    male:male:male:homo = 0
    male:male:male:bi = 1
    .
    .
    .

    and so on. BAM collinearity problem solved!

    9 years ago # QUOTE 3 Volod 2 Vlad !

11. Economist
    9f03
    

    9/10

    9 years ago # QUOTE 1 Volod 0 Vlad !

12. Economist
    d5c1
    

    10/10 OP

    9 years ago # QUOTE 9 Volod 1 Vlad !

13. Economist
    f2c7
    

    WTF? As a furry, I demand a dummy for people like me.

    9 years ago # QUOTE 22 Volod 1 Vlad !

14. Economist
    bad2
    

    lol op

    9 years ago # QUOTE 0 Volod 1 Vlad !

15. Economist
    d962
    

    Check this out:

    http://itspronouncedmetrosexual.com/2012/03/the-genderbread-person-v2-0/

    9 years ago # QUOTE 13 Volod 1 Vlad !

16. Economist
    1608
    

    ^

    Its funny because the guy who came up with it probably included it in his tenure dossier.

    9 years ago # QUOTE 0 Volod 1 Vlad !

17. Economist
    5246
    

    The comments are awesome:

    I for one...think you nailed it! As a bigender pansexual who expresses hirself fluidly...I'm in love with this!

    *Sigh* I do thank you for doing this but I really just want to be called what I am. I am Agender. Not Nongendered. Because that implies that I believe in gender which I don't.

    Check this out:
    http://itspronouncedmetrosexual.com/2012/03/the-genderbread-person-v2-0/

    9 years ago # QUOTE 15 Volod 1 Vlad !

18. Economist
    ec72
    

    My cis-privlege to add this to the 100 good club. best post in a long time.

    9 years ago # QUOTE 9 Volod 1 Vlad !

19. Economist
    09d8
    

    The problem is that you don't, for example, have a dummy for female identity. I identify as both female and male, it is not either or. Stata recognizes this and is expressing its outrage at your privilege.

    9 years ago # QUOTE 24 Volod 1 Vlad !

20. Economist
    4b2f
    

    unlike most troll posts on ejmr, this one is genuinely well written

    9 years ago # QUOTE 4 Volod 4 Vlad !

Reply »