+ Reply to Thread
Page 2 of 2 FirstFirst 1 2
Results 16 to 18 of 18

Thread: Normality Assumption for Independent t-test

  1. #16
    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: Normality Assumption for Independent t-test




    Interesting. I attended a lecture on dementia (focus on Alzheimer's disease) earlier this week and the presenter mentioned MMSE and MoCa. Funny enough, now that I think about it they made mention of the effects of pharmaceutical efficacy on metrics related to mind and said such and such drug showed a difference of say 2 points on the MMSE, but does that really translate into any real world application related to say activities of daily living - was their follow-up. This was done in contrasting the drug benefits vs. negative effects and costs. So there is that too. What are the metrics gonna be used for. I remember using the MMSE to screen elderly patients for their participation in a research study.


    Yeah, too busy, but papers discussing the foundations and constructs of the tool would be interesting. But I am so crammed on time, I would prefer to listen to a podcast version.


    Have a good weekend.
    Stop cowardice, ban guns!

  2. #17
    Points: 4,824, Level: 44
    Level completed: 37%, Points required for next Level: 126

    Posts
    41
    Thanks
    19
    Thanked 1 Time in 1 Post

    Re: Normality Assumption for Independent t-test

    Hi All,

    Thank you so much for interest and input, it's great having a forum like this to discuss these sometimes tricky questions!

    To give you some more clarity on the research question I am trying to answer, I am aiming to find out whether or not there is a difference in level of cognitive function (as assessed using the Mini-Mental State Exam (MMSE) between those who did/did not receive an anxiety disorder diagnosis.

    I thought the most obvious way to do this was via a t-test, but you have brought up some good points that the MMSE might be an ordinal level of data. I did some quick searching on Google and found this: http://pb.rcpsych.org/content/30/9/354.3 which suggests that MMSE scores should be treated as ordinal.

    If this is the case would I therefore use a Wilcoxon test instead? And if so, does that mean that the normality assumption is not as important?

    Also, if someone could expand on why the MMSE scores would be seen as ordinal rather than interval data, that would be much appreciated. I have seen a number of papers treat it as interval data so it would be great to get some clarity on this.

    Thanks!

  3. #18
    Points: 1,741, Level: 24
    Level completed: 41%, Points required for next Level: 59

    Posts
    230
    Thanks
    37
    Thanked 68 Times in 59 Posts

    Re: Normality Assumption for Independent t-test


    Quote Originally Posted by bozatron View Post
    Hi All,

    Thank you so much for interest and input, it's great having a forum like this to discuss these sometimes tricky questions!

    To give you some more clarity on the research question I am trying to answer, I am aiming to find out whether or not there is a difference in level of cognitive function (as assessed using the Mini-Mental State Exam (MMSE) between those who did/did not receive an anxiety disorder diagnosis.
    Do you have any other variables that might be important in determining cognitive function such as age, comorbidities, history of stroke, etc. ?

    Quote Originally Posted by bozatron View Post
    I thought the most obvious way to do this was via a t-test, but you have brought up some good points that the MMSE might be an ordinal level of data. I did some quick searching on Google and found this: http://pb.rcpsych.org/content/30/9/354.3 which suggests that MMSE scores should be treated as ordinal.

    If this is the case would I therefore use a Wilcoxon test instead? And if so, does that mean that the normality assumption is not as important?

    Also, if someone could expand on why the MMSE scores would be seen as ordinal rather than interval data, that would be much appreciated. I have seen a number of papers treat it as interval data so it would be great to get some clarity on this.

    Thanks!
    If the Wilcoxon rank sum is your tool of choice, you need not worry about the normality assumption. For the ordinal vs. interval question: we can agree that a 28 represents better cognitive function compared with a 26, and 26 is better than 24, therefore 28 is better than 24, but we can't necessarily make the argument that this 2 point difference is equal elsewhere. Take the change from a 25 to a 23 (going from unimpaired to impaired cognitive function). Sure, 2 points is two points, but is the decline in cognition from 25 to 23 really the same as 28 to 26 (or any other jump of 2)? Long story short, the changes aren't necessarily the same throughout. (By the way, I do know the points can move in increments of 1, but I chose 2 for my example.)

    Hope this helps!
    Last edited by ondansetron; 01-14-2017 at 01:14 AM.

+ 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