+ Reply to Thread
Results 1 to 2 of 2

Thread: Weighing male/female responses (30/70 distribution)

  1. #1
    Points: 5, Level: 1
    Level completed: 9%, Points required for next Level: 45

    Thanked 0 Times in 0 Posts

    Weighing male/female responses (30/70 distribution)

    Hi all!

    I've got a question. I received 617 responses to a questionnaire however, 445 of them are female (72%). There are two ways to fix this:

    1. I could remove 230 female respondents from my data set, bringing my sample back to 385 people. This would leave me with 44% male and 56% female, which would still not be ideal and I would lose massive amounts of beautiful data (I only have 172 male respondents )

    2. I could assign a certain weight to each respondent. However, because of the large difference between male/female respondents, I'm afraid my data won't be representative anymore.

    I could also just separate males and females for the questions I want to answer, thus avoiding a group average But this won't look nice in the report, so I'd rather not.

    So, the question for you guys: What would you do to make this sample more representative?


  2. #2
    Points: 12,676, Level: 73
    Level completed: 57%, Points required for next Level: 174
    Master Tagger
    GretaGarbo's Avatar
    Thanked 462 Times in 402 Posts

    Re: Weighing male/female responses (30/70 distribution)

    Down weight the the women so that their sum of weight is 0.5. That will also mean that each individual man will have a higher weight.

    Don't throw away any data. Of course it is an advantage to have a larger sample, than a smaller!

    (The result is an effect of that women are more friendly and more responsible (no kidding))

    ...and I would lose massive amounts of beautiful data
    (You could keep only the females. Then you would ony have "beautiful data" [joke! ] )

+ Reply to Thread


Tags for this 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