+ Reply to Thread
Results 1 to 5 of 5

Thread: t-test or ANOVA in a validationstudy?

  1. #1
    Points: 1,284, Level: 19
    Level completed: 84%, Points required for next Level: 16

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    t-test or ANOVA in a validationstudy?




    Hello,

    i am doing reasearch for my bachelor thesis in physiotherapy and i need help with the statistics.
    my reasearch is about the validation of devices that count steps.
    so i got one group equiped with 2 devices counting their steps while walking. while walking someone counted the steps by hand.
    This gives me 3 stepcounts per person taken the same moment.

    i chose to do a dependent t-test to accept or discard the hypothesis that the devices count as many steps as were counted by hand.

    but is this the right test for the job?
    someone said, that i should use an ANOVA. but for an ANOVA i would need different people in the groups to be aplicable? am i right? so i thought maybe a repeated ANOVA? but then again this is normally used for pre/during/posttreatment measurements which is not the case for my study.

    so now i don't know what test is should use.
    i hope you can help me!

    Thanks

  2. #2
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: t-test or ANOVA in a validationstudy?

    How many people did you have doing this?
    Stop cowardice, ban guns!

  3. #3
    Points: 1,284, Level: 19
    Level completed: 84%, Points required for next Level: 16

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: t-test or ANOVA in a validationstudy?

    I had 20 people. so a pretty small sample size.

  4. #4
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: t-test or ANOVA in a validationstudy?

    I may think you could conduct pairwise comparisons on the differences between devices using Wilcoxon signed rank test. Correcting the level of significance if you are comparing one group to the other than to the actual value (counted).
    Stop cowardice, ban guns!

  5. #5
    Points: 1,284, Level: 19
    Level completed: 84%, Points required for next Level: 16

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: t-test or ANOVA in a validationstudy?


    Is that not the same as the paired sample t-test? except for the assumtions that it doesnt have to be normally distributed and can have outliers?
    my data is about normally distributed and has no outliers.

    so there is no way i could use a ANOVA or somthing like that?

+ 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