+ Reply to Thread
Results 1 to 2 of 2

Thread: Am I choosing the right 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

    Am I choosing the right test?




    My daughter is doing a science fair project testing the efficacy of insect repellants. She has 3 trials each for 3 repellants and 1 control. In each case she measured the number of insects crossing a line in a 5 minute period. The number counted in each trial ranges from 1 to about 50, with most trials being in the low teens. She would like to determine if the repellants work at all, and which ones are superior to the control.

    I am thinking of having her do a 1 way ANOVA on all 4 groups together. If a difference is seen, she will then do a Mann-Whitney test between each repellant and the control.

    Does this seem appropriate?

    Thanks in advance

  2. #2
    Human
    Points: 12,676, Level: 73
    Level completed: 57%, Points required for next Level: 174
    Awards:
    Master Tagger
    GretaGarbo's Avatar
    Posts
    1,362
    Thanks
    455
    Thanked 462 Times in 402 Posts

    Re: Am I choosing the right test?


    Yes, I would do something that is close to anova. Since she is counting the number of crossings it seems more natural to assume a Poisson distribution. Thus a generalized linear model with Poisson distributed dependent variable.

    Then I would test the different pairs, not with the Mann-Whitney test, but with the same model, comparing treatment 1 versus treatment 2, treat1 vs trest3 etc.

+ 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