Logistic Regression
Manage episode 437612813 series 3589860
Logistic regression is a statistical method used to predict a binary outcome (yes or no, true or false). It calculates the odds ratio, which is the probability of the outcome occurring. This method does not require the same assumptions as linear regression. It can be used in research studies where the outcome is categorical rather than numerical, and can consider multiple variables and adjust for confounding factors. A study on prescribing habits among podiatrists is used as an example of the application of logistic regression. Cohort studies are a type of longitudinal, observational study that follow participants over time to assess the relationship between an exposure and an outcome. They are similar to randomized control trials in that they both track outcomes over time, but they differ in that cohort studies are observational and do not randomly assign participants to an exposure group. Retrospective cohort studies, also known as chart review studies, are used to examine associations between exposures and outcomes in cases where randomized control trials are not possible or ethical.
5 bölüm