Please enable JavaScript.

Coggle requires JavaScript to display documents.

statistics - Coggle Diagram

- - - - SE (diff = )
    - - Assumptions for the One-way ANOVA include the following:
        
        The observations are independent of each other
        
        The data are normally distributed in each group
        
        The variances in the groups are all equal
        
        Lavenes test
        
        if not equal use welch or transform the outcome of Y (eg. log). if the welch test is postive you can use a post hoc rocedure for equal variance not assumed
      - which group is different use : Post Hoc procedures: LSD and Bonferroni.
        
        bonferronia is one step procedure where the significance is devided by the amount of pairs, usually this is correct by the program
  - - - chi square test
        
        assumptions
        
        All observations are mutually independent (in our example: the children are all independent of each other)
        
        The expected numbers in all cells are larger than 1
        
        The expected numbers in at least 80% of the cells are larger than 5
        
        if this is not fulfilled the fisher's exact test
        
        for big calculations to get the degrees of feedom, the number of degrees of freedom, in a table with r rows and c columns, equals (r – 1)·(c – 1), see Table 8.13.
        
        Yates correction for more conservitve result
    - - Wilcoxon Ranksum test
      - Mann-Whitney test , is for compare continuous or ordinal data
      - more than two groups use the Kruskal-Wallis test it is an extention of matt-whitney test
        
        Post-hoc Mann-Whitney tests between the groups
      - try to get the root or log of the data maybe its normal
  - - - singed test + and - for all differences , not senstive
      - wilcoxon's signed rank test, reasonably symmetric distribution is a requirement
    - - McNemar’s test
        
        this leads to a chi square test
  - - - fishers Z for this to become a normal dis
- - - - to get a measurement of the area use the z table