CRIM 320 Lecture Notes - Lecture 4: Chi-Squared Distribution, Null Hypothesis, Security Level
12 Jun 2017
School
Department
Course
Professor
Document Summary
Jan 31st, 2017 lecture 4: analysis of categorical data i. How do we conduct a chi-square? (observed expected) ^2 / expected: calculate marginal (total, calculate expected counts. 3: then, sum across all cells (this will give you the chi-square statistics, significant or not, calculate degrees of freedom, compare observed with expected, to determine if we can reject the null hypothesis. Less than 0. 05, reject null hypothesis -> significant. Data taken from table 1 in camp and gaes articles. Whether or not intensities of incarceration would make inmates engage in more misconduct: min level 1 security vs. max level 3 security; n = 561. Step 1, column and row marginal are same as totals represent each category variable. Row total x column total / grand total. These are the numbers that we expect if the two variables are independent of each other. Step 3, calculating the (observed expected counts) ^2 / expected.
