Skip to main content

How can we help?

Frequently searched: Android devices, cloning, ADC Gallery, Glossary,...

Significancy tests (Significativity, formulas, conditional formatting, best test to use, maxdiff, ...)

You just want to know which is the best sig test to use, how we calculate, this section is for you!

In this article

    What is the best Sig Test to use?

    It's not always obvious which significance test to use when analysing your project results. But here is a decision tree chart to help you pick the most appropriate test . . . . 

    signifincance_fig.PNG

     Table of contents:

      Type Samples Sample type Test Askia

    A

    		 Mean 2 Independent T-Test Significance (numeric)
    B Mean 2 Paired Paired T-Test Col significativity on numeric
    C Mean 2 Paired Wilcoxon rank test Col significativity on numeric
    D Mean 3 Independent ANOVA with One Way Col significativity on numeric
    E Mean 2 Paired ANOVA with Random subject Col significativity on numeric
    F Proportion 2 Independent Z-Test for proportions Col significativity on closed
    G Proportion 2 Independent Khi² Significance (Closed)
    I Proportion 3 Independent Khi² + Marscuilo post-hoc Significance (Closed)
    J Proportion 3 Independent Khi² + Cochran adjustment Col on Col Sig (Closed)

     

    A - Significance (numeric) (T-test)

    A

    		 Mean 2 Independent T-test Significance (Numeric)

     

     

    A-T-test.PNG

    Go back to the Top

    B - Col significativity on numeric  (Paired T-test) 

    B Mean 2 Paired Paired T-Test Col significativity on Numeric

    B-PairedT-Test.PNG

    Go back to the Top

    C - Col significativity on numeric (Wilcoxon rank test)

    C Mean 2 Paired Wilcoxon rank test Col significativity on Numeric

    C-Wilcoxon.PNG

    Go back to the Top

    D - Col significativity on numeric (ANOVA One Way + Tuckey HSD test)

    D Mean 3 Independent Anova With One Way Col significativity on Numeric

    D2-AnovaOneway.PNG

    mceclip0.png

    D2_step4.PNG

     

    D2_step5.PNG

     

    Go back to the Top

    E - Col significativity on numeric (ANOVA with random subject)

     

    E Mean 2 Paired ANOVA with random subject Col significativity on Numeric

     E1.PNG

     E2.PNG

    E3.PNG
     

     E4.PNG

     

     

     

    Go back to the Top

    F - Col significativity on closed  (Z-Test for proportions)

    F Proportion 2 Independent Z-test for proportions Col significativity on closed

    f1.PNG

     

    Go back to the Top

    G - Significance (closed) (Khi²)

    G Proportion 2 & above Independent Khi² Significance (Closed)

    g1.PNG

     

    Go back to the Top

    I - Significance (closed) (Khi² + Marscuilo post-hoc)

    I Proportion 3 Independent Khi² + Marscuilo post-hoc Significance (Closed)

    g1.PNG

     

    Go back to the Top

    J - Col on Col Sig (closed) (Khi²+ Cochran adjustment)

    J Proportion 3 Independent Khi² + Cochran adjustment Col on Col Sig (Closed)

    j1.PNG

    J2.PNG

    With the Cochran–Mantel–Haenszel (CMH) test adjustment

    Which allows the comparison of two groups on a dichotomous/categorical response. It is used when the effect of the explanatory variable on the response variable is influenced by covariates that can be controlled. It is often used in observational studies where random assignment of subjects to different treatments cannot be controlled, but influencing covariates can.

    Related articles