+ Reply to Thread
Results 1 to 2 of 2

Thread: Generalized Linear Model: help interpeting interaction

  1. #1
    Points: 488, Level: 9
    Level completed: 76%, Points required for next Level: 12

    Posts
    5
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Generalized Linear Model: help interpeting interaction




    Hello!

    I apologise in advance if I missed it, but I couldn't find an answer in previous threads.

    I run a generalized linear model with 1 factor (gender) and 5 covariates. No single variable appears to have a predictive value on my VD. However, when I test for interactions, many seem to have a significant effect.

    First thing I thought is that probably a single variable has not enough predictive power over the VD, but it does when is combined with other(s).

    Does this make sense? Theoretically, I can find an explanation to this, but how should I report it? «Individually, variables had no predictive effect on the VD but when considering the combined effect, they did»?

    Thank you so much in advance =)

  2. #2
    Omega Contributor
    Points: 38,303, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,993
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: Generalized Linear Model: help interpeting interaction


    Can you really predict Gender?


    Well there could be a synergistic interaction on the multiplicative scale. So one group in the presence of another results in a greater likelihood of the outcome. This is completely feasible, though we don't know your context so who knows if it makes sense. What you should do is calculate the probabilities for the groups along with the odds ratios with 95% CI. This way you can see the effect.


    P.S., When doing this, did you leave the main terms in the model, so y = x1 + x2 + x1*x2 ? Which is highly recommended.
    Stop cowardice, ban guns!

  3. The Following User Says Thank You to hlsmith For This Useful Post:

    ernesthor (06-29-2016)

+ 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