Exam 1

  1. Metainformation

    Tag Value
    fileInferential_Statistics_uu-Effect-size-815-en_uu-Effect-size-815-en
    nameuu-Effect-size-815-en
    sectionInferential Statistics/Effect size
    typeschoice
    solutionFALSE, FALSE, TRUE, FALSE
    TypeCase
    LanguageEnglish
    LevelStatistical Literacy

    Question

    A study of the relationship between location of deployment (Iraq or Afghanistan) and depression of soldiers primarily examines only whether soldiers became depressed, yes or no. A χ2\chi^2-test is performed and the variables were found, in a sample of 60 soldiers, not to be independent: χ2(1,n=60)=5.88\chi^2(1, n = 60) = 5.88, p=.015p = .015. What is the value of the effect size that should be reported here?


    1. FALSE: 0.0150.015
    2. FALSE: 0.0980.098
    3. TRUE: 0.3130.313
    4. FALSE: This cannot be determined because the cross table is not given.

    Solution

    Location is a dichotomous variable (Iraq/Afghanistan), as is depression (yes/no). To calculate the effect size in this case, the phi coefficient can be calculated:

    ϕ=χ2n=5.8860=0.313\phi= \sqrt{\frac{\chi^2}{n}} = \sqrt{\frac {5.88}{60}} =0.313


    1. .015.015 is pp-value and not the effect size.
    2. Thist answer (.098.098) forgot to root.
    3. ϕ=χ2n=5.8860=0.313\phi= \sqrt{\frac{\chi^2}{n}} = \sqrt{\frac {5.88}{60}} =0.313
    4. This can indeed be determined. Location is a dichotomous variable (Iraq/Afghanistan), as is depression (yes/no). To calculate the effect size in this case, the phi coefficient can be calculated.