SOC 282 Lecture Notes - Chi-Squared Distribution
Document Summary
Lab notes 3. 27. 14: chi squared goodness of fit test, a one way x squared test can determine whether or not the observed frequencies differ significantly from an expected distribution. H0: there is no significant difference across categories. H1: there is significant difference across categories. 127. 4: chi squared test for statistical association between two variables, tests if there"s a significant relationship (at least statistically) between two variables, also used to see if there is statistical independence. H0: there is no significant association between education and health in the population. H1: there is significant association between education and health in the population: fe = (column marginal)(row marginal)/n, n= total number in the population/sample. **(picture of the rest of the table on my phone) If after adding a control variable, the original significant bivariate association becomes insignificant, that suggests the relationship between the two variables is spurious. They are both determined by the third variable.