+ Reply to Thread
Page 2 of 2 FirstFirst 1 2
Results 16 to 23 of 23

Thread: Doubt on statistical test

  1. #16
    Points: 80, Level: 1
    Level completed: 60%, Points required for next Level: 20

    Posts
    13
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: Doubt on statistical test




    Rightio! Let's go for it then!
    Thanx a bunch

  2. #17
    Points: 80, Level: 1
    Level completed: 60%, Points required for next Level: 20

    Posts
    13
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: Doubt on statistical test

    Hi!

    Does a significant p.value mean that either one or both factors (context and rate) have a significant effect on the dependent variable? If so, how can I pin it down? Is there a way to find out how much each factor explains the variability observed in the final results compared against the expected values?

    Thanks!

    sid2016

  3. #18
    TS Contributor
    Points: 12,227, Level: 72
    Level completed: 45%, Points required for next Level: 223
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,470
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: Doubt on statistical test

    Hi,
    yes, the significant p tells you that somemfactor combinations do have an effect. In the output of the test you generally have the chi-squared value for each cell. Those with a high value are probably the contributors to the effect.

    regards

  4. #19
    Points: 80, Level: 1
    Level completed: 60%, Points required for next Level: 20

    Posts
    13
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: Doubt on statistical test

    Thanks!! Let's see how it goes...

  5. #20
    Points: 228, Level: 4
    Level completed: 56%, Points required for next Level: 22

    Location
    Tampa, FL
    Posts
    24
    Thanks
    0
    Thanked 11 Times in 10 Posts

    Re: Doubt on statistical test

    Just to clarify, are these repeated measurements on each subject? As in, each person receives a sequence of all five factor combinations?

    Also, is the outcome the rate at which they choose a particular response out of the three (the "right" response vs. the "wrong" response)? Or do the rates of all three responses matter?

  6. #21
    Points: 80, Level: 1
    Level completed: 60%, Points required for next Level: 20

    Posts
    13
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: Doubt on statistical test

    Hi ab-stats,

    1) each listener will hear the different items five times in a random order.
    2) response rate is not collected here

    Best
    sid2016

  7. #22
    Points: 228, Level: 4
    Level completed: 56%, Points required for next Level: 22

    Location
    Tampa, FL
    Posts
    24
    Thanks
    0
    Thanked 11 Times in 10 Posts

    Re: Doubt on statistical test

    OK, since you're dealing with repeated measurements here, be aware that responses within each subject might be correlated. This would violate the independence assumption of Pearson's chi-square test. One option is to fit a mixed-effects logistic model that includes a random intercept for each subject. Categorical Data Analysis by Alan Agresti is a good resource on this topic.

  8. #23
    Points: 80, Level: 1
    Level completed: 60%, Points required for next Level: 20

    Posts
    13
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: Doubt on statistical test


    Right on!! This is getting even better : ) I'll check out your suggestion!
    Thanks!

+ Reply to Thread
Page 2 of 2 FirstFirst 1 2

           




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