+ Reply to Thread
Results 1 to 4 of 4

Thread: Which statistical test?

  1. #1
    Points: 4, Level: 1
    Level completed: 7%, Points required for next Level: 46

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Which statistical test?




    Hi everyone,

    It's been a couple of years since I've taken a statistics class and I'm struggling to remember which test I should use. Essentially, what I'm trying to analyze is if new technology has decreased a trend of accidents and severity.

    Thanks!
    Attached Images  

  2. #2
    TS Contributor
    Points: 14,811, Level: 78
    Level completed: 91%, Points required for next Level: 39
    Miner's Avatar
    Location
    Greater Milwaukee area
    Posts
    1,171
    Thanks
    34
    Thanked 405 Times in 363 Posts

    Re: Which statistical test?

    A chi-square test would be appropriate.

  3. #3
    Points: 371, Level: 7
    Level completed: 43%, Points required for next Level: 29

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Which statistical test?

    Quote Originally Posted by cruiselife View Post
    Hi everyone,

    It's been a couple of years since I've taken a statistics class and I'm struggling to remember which test I should use. Essentially, what I'm trying to analyze is if new technology has decreased a trend of accidents and severity.

    Thanks!
    Chi-square, but since the count in one of the cells is less than 5, Fisher Exact would be an appropriate one.

    Thanks!

  4. #4
    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: Which statistical test?


    Quote Originally Posted by bburugap View Post
    Chi-square, but since the count in one of the cells is less than 5, Fisher Exact would be an appropriate one.

    Thanks!
    It's a small point, but typically, the observed cell count is not the focus. The assumption is with respect to the expected cell count being at least 5. The smallest expected count in this table is around 4.80, so strictly following the rule, a FET probably wouldn't hurt too much. OP should be careful, though, as the FET is frequently less powerful than the chi-square (if I recall, it's ameliorated by larger a larger sample size), and these differences don't appear to be too large. Looking at both a FET and chi-square would probably be useful. If they agree, you don't have to think as hard about it as in the case of the tests disagreeing (personally, I'd probably be more conservative and stick with FET if they disagree). If this is for a research project or something in the real world, the OP should definitely decide beforehand how disagreement between the tests will be handled.

+ 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