Inferential Statistics:- Non-parametric Tests

Inferential Statistics:- Non-parametric Tests

1. P Values

   • The purpose of a statistical test is to get a P value
     • P = is the probability that the results are due to chance
       • p< 0.05 (1 in 20, 5% chance)
   • This is the likelihood that the null hypothesis is true (the likelihood that the differences are not significant).

2. Tests

   • Chi-squared (x squared) - nominal data, repeated measures/matched pairs
     • Calculated value must be greater or equal to critical value to be significant
   • Binomial sign test (s) - nominal data, repeated measures/matched pairs
     • Calculated value must be less than or equal to the critical value to be significant
   • Spearman's Rho (r) - correlation or relationship
     • Calculated value must be greater or equal to critical value to be significant
   • Mann-Whitney U (U) - experiment, independent measures
     • Calculated value must be less than or equal to critical value to be significant
   • Wilcoxon Signed Ranks (T) - experiment, repeated measures/matched pairs
     • Calculated value must be less than or equal to critical value to be significant

3. Symbols

   • = equal to
   • < less than
   • << much less than
   • ~ approximately
   • proportional to
   • > greater than
   • >> much greater than

Imagens e arquivos do mapa