KINE 2050 Chapter Notes - Chapter 13: Chi-Squared Distribution, Type I And Type Ii Errors, Level Of Measurement
Document Summary
13. 2 the chi-square distributions: chi-square distributions = a family of sampling distributions which are useful for testing hypothesis of nominal data, a normally distributed population whose scores are converted to z-scores, would have a mean of. If significant, the assumptions causing for expected frequencies is rejected. B/c it is the expected frequencies that are hypothetical. If not significant, assumptions causing fro expected frequencies are accepted same as retaining the null hypothesis. 13. 5 relationship between nominal variables: the test of independence. In above example, null hypothesis is that gender and age are independent, alternate hypothesis is that they are not independent: find the expected frequency for each box in the contingency table. Expected frequency = (row sum x column sum)/total sum. E. g. , expected frequency for old men = (20 x 25)/50 = 10. The expected frequencies in each box must sum up to the row sum, the same applies for the column sum.