SOCI 328 Lecture Notes - Lecture 11: Null Hypothesis, Test Statistic, Chi-Squared Distribution
Document Summary
We therefore create the null hypothesis ho: the variables are statistically independent in the population and the alternative hypothesis ha: the variables are statistically dependent in the population. It is based upon what are called expected counts and observations i. e. they are related. It represents what you would expect to find in each cell if there was no relationship at all between the two variables. Expected count = (row total) x (column total)/table total. The chi- square test statistic essentially determines the magnitude of the differences between the observed and expected values. More precisely, the chi- squared statistic equals the sum of the squared difference between each observed and expected value divided by the expected count. We look to the chi- square distribution that applies in a given distribution (as defined by table size) - - the chi square statistic and the degrees of freedom are both pertinent in this regard.
