+ Reply to Thread
Results 1 to 5 of 5

Thread: Repeated Measures ANOVA and baseline control

  1. #1
    Points: 1,887, Level: 25
    Level completed: 87%, Points required for next Level: 13

    Posts
    24
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Repeated Measures ANOVA and baseline control




    Iíve been asked to work on a cross-over design study for an MSc project. The study is looking at the effects of sleep deprivation on word recall. Each participant was asked to recall a list of words before and after a night of total sleep deprivation or following a normal nightís sleep. I took baseline measures before each night and then there were two tests following each condition Ė totalling four. Would it be best to just do a straight forward repeated measures ANOVA? Also, seeing as I have them would it be useful to pop my two baseline measures into the model to control for any effects stemming from them? Iím unsure whether I should include both or take an average of the two baseline measures?

    Thanks for any advice offered!

  2. #2
    Points: 3,631, Level: 37
    Level completed: 88%, Points required for next Level: 19
    staassis's Avatar
    Location
    New York
    Posts
    226
    Thanks
    2
    Thanked 41 Times in 39 Posts

    Re: Repeated Measures ANOVA and baseline control

    The answer depends on your data set. How many participants, between-subject factors and covariates? What is the distribution of the recalled number of words in each test? Gaussian or non-Gaussian? All of that matters.

  3. #3
    TS Contributor
    Points: 17,749, Level: 84
    Level completed: 80%, Points required for next Level: 101
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,540
    Thanks
    56
    Thanked 640 Times in 602 Posts

    Re: Repeated Measures ANOVA and baseline control

    Each participant was asked to recall a list of words before and after a night of total sleep deprivation
    or following a normal night’s sleep. I took baseline measures before each night and then there were two tests following each condition – totalling four.
    So you did the same test 6 times, one before each
    of the two conditions, and two afterwards?

    Looks like 2 repeated measures factors,
    condition (2 levels) and time of meaureent
    (before, afterwards 1, afterwards 2).
    But see the previous answer for all the aspects
    whch have to be taken into account first.


    With kind regards

    K.

  4. #4
    Points: 1,887, Level: 25
    Level completed: 87%, Points required for next Level: 13

    Posts
    24
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: Repeated Measures ANOVA and baseline control

    Thanks for your responses.

    We have 18 participants who completed both conditions. The data is normally distributed.

    For my initial model I wasn't going to include co-variates but I was considering whether my baseline measures should be entered in as such. I also have measure of attention and anxiety that I might include. I suppose I thought it depended on my sample size. Any advice offered would be much appreciated.

    Also, apologies my original post was quite unclear - I have for measure. Two baselines and two (one after each condition)

    Thanks again

  5. #5
    TS Contributor
    Points: 17,749, Level: 84
    Level completed: 80%, Points required for next Level: 101
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,540
    Thanks
    56
    Thanked 640 Times in 602 Posts

    Re: Repeated Measures ANOVA and baseline control


    We have 18 participants who completed both conditions. The data is normally distributed.
    It isn't. But whether this matters depends on the model.
    For ANOVAs, residuals should be from normally distributed
    populations, at least if sample size is small.
    Also, apologies my original post was quite unclear - I have for measure. Two baselines and two (one after each condition)
    As I wrote before, looks like (an ANOVA with) 2 within-subject
    factors, "condition" and "time of measurement". If your data
    don't fulfill assumptions of ANOVA, maybe baseline/after-difference
    scores could be compared using Wilcoxon signed rank test.

    With kind regards

    K.

+ 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