Logistic Regression: The Odds Ratio and Contingency Tables

Introduction to Logistic Regression: The Odds Ratio and Contin" rel="nofollow">ingency Tables Order Description Introduction to Logistic Regression: The Odds Ratio and Contin" rel="nofollow">ingency Tables For the most part, you have only been exposed to statistical methods that require a contin" rel="nofollow">inual dependent variable. You have likely been thin" rel="nofollow">inkin" rel="nofollow">ing that there must be some way to predict the myriad of categorical variables that exist in" rel="nofollow">in social data. What if you want to predict the odds of whether a student passes or fails, a disease occurs or does not, or whether a recently released in" rel="nofollow">inmate offends or not? You will notice all of these examples are categorical and have a dichotomous outcome (yes/no). Bin" rel="nofollow">inary logistic regression will allow you to answer questions where you can predict the odds of an event occurrin" rel="nofollow">ing from a combin" rel="nofollow">ination of categorical and contin" rel="nofollow">inual variables. This week you revisit a basic approach to examin" rel="nofollow">inin" rel="nofollow">ing categorical relationships, the crosstab table, and consider how to expand on it. More specifically, you construct conditional probabilities of an event, create odds ratios to analyze differences between groups, and estimate slopes from a basic crosstab analysis. • Calculate values derived from a contin" rel="nofollow">ingency table • Create a tool to calculate statistical values • This week’s readin" rel="nofollow">ings discuss conditional probabilities, conditional odds, logits, odds ratios, relative risk, and slopes. These can all be confusin" rel="nofollow">ing terms but the good news is that all these values have some relationship to each other. Researchers have their own opin" rel="nofollow">inions on which values makes the most sense to report. • In a 2- to 3-paragraph post, construct a persuasive argument for the value (conditional probability, odds, odds ratio, etc.) that, in" rel="nofollow">intuitively, makes the most sense for you to report as a result to your audience. Be sure to provide a specific rationale for your choice. Osborne, J. W. (2015). Best practices in" rel="nofollow">in logistic regression. Thousand Oaks, CA: SAGE Publications.