Please enable JavaScript.

Coggle requires JavaScript to display documents.

QM #7 ANOVA (One-way ANOVA (One-way between ANOVA (:old_key: values (eta…

- - - - Avoiding doing a lot of t-tests
      - Testing the equality of means
    - - :old_key: values
        
        Mean
        
        SD
        
        eta-squared/
        partial eta-squared
        
        effect size
        
        based purely on sums of squares from the sample and no adjustment is made for the fact that we’re trying to estimate the effect size in the population
        
        η2=.2 :arrow_forward: small effect
        η2=.4 :arrow_forward: medium effect
        η2=.6 :arrow_forward: large effect
        
        f statistic/ratio
        
        mean square (MS) or variance between treatments
        :heavy_division_sign:
        mean square (MS) or variance within treatments
        
        :hammer_and_wrench: Confirming that overall differences exist, but not which group means differ from which
        
        :key: 1/more is significant
        
        Post-hoc comparison
        (“pairwise” comparison)
        
        :bulb: Consist of pairwise comparisons that are designed to compare all different combinations
        of the treatment groups
        
        :hammer_and_wrench: To control the family-wise error by correcting the level of significance for each test such that the overall Type I error rate across all comparisons remains at .05.
        
        :key: Don't do this if ANOVA ~~significant~~
        
        p value
        
        p < .05 :arrow_forward: significant
      - Does the data meet
        the assumptions?
        
        DV should be measured at the continuous level (i.e., they are interval or ratio variables).
        
        :memo: revision time (measured in hours),
        intelligence (measured using IQ score),
        exam performance (measured from 0 to 100),
        weight (measured in kg), etc.
        
        IV or factor should consist of at least 2 categorical, "related groups" or "matched pairs"
        
        ~~significant outliers~~
        in the related groups
        
        The distribution of the DV in the 2/more related groups should be approximately normally distributed
        
        Known as sphericity, the variances of the differences between all combinations of related groups must be equal.
  - - - Mean
      - SD
      - eta-squared
        
        The overall effect size of ANOVA
        
        Biased & ~~specific~~
        
        partial eta-squared
        
        :bulb: unbiased correction of eta-squared
        
        How to interpret by
        RStatsInstitute
      - f statistic/ratio
        
        mean square (MS) or variance between treatments
        :heavy_division_sign:
        mean square (MS) or variance within treatments
        
        :hammer_and_wrench: Confirming that overall differences exist, but not which group means differ from which
        
        :key: 1/more is significant
      - Post-hoc comparison
        (“pairwise” comparison)
        
        :bulb: Consist of pairwise comparisons that are designed to compare all different combinations
        of the treatment groups
        
        :hammer_and_wrench: To control the family-wise error by correcting the level of significance for each test such that the overall Type I error rate across all comparisons remains at .05.
        
        :key: Don't do this if ANOVA ~~significant~~
      - p value
        
        p < .05 :arrow_forward: significant
      - Interpreting
        the result
        
        Source #1 YouTube
        
        Source #2 Laerd
    - - DV is measured at the interval or ratio level (i.e., they are continuous) (technically, it is the residuals that need to be normally distributed, but the results will be the same)
      - There is homogeneity of variances
        
        Interpreting Levene's test for homogeneity of variances spss
      - Independence of observations
      - IV should consist of 2/more categorical, independent groups
      - There should be no significant outliers
      - DV should be approximately normally distributed for each category of the IV