+ Reply to Thread
Results 1 to 2 of 2

Thread: T Test

  1. #1
    Points: 705, Level: 14
    Level completed: 5%, Points required for next Level: 95

    Thanked 0 Times in 0 Posts

    T Test

    I am trying to do a weighted mean ttest.

    The statistics I'm using are the vote shares for republican and democrat women candidates in primary elections Idaho 2008-2014 (state legislature).

    I am currently sorting out the percentage of votes obtained by rep and dem women in ratio to the men in each district.

    I would like to control for voter turnout, but am not sure which step of the test I need to be multiplying by voter turnout %.

    Do i do it with the raw #s of dem women/dem men, or do I do it as a total of women/men. Also, I assume that I have to multiply both numerator and denominator by the %voter turnout. Is that correct?



  2. #2
    Points: 3,457, Level: 36
    Level completed: 72%, Points required for next Level: 43

    Thanked 43 Times in 40 Posts

    Re: T Test

    I am not exactly sure what you are trying to do, but I don't think you want a t-test. When you want to add control variables it's time to go from a t-test to ANOVA/Regression (or possibly logistic regression). However, since you have (apparently) longitudinal data it may be that you need a more complex model.

+ 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