POPM 3240 Midterm: POPM 3240; Midterm II Review Notes

• School: University of Guelph
• Department: Population Medicine
• Course: POPM 3240
• Professor: Andria Jones

Document Summary

Regression line uses notation of y = 0 + 1x1: if 1 is positive positive relationship with outcome, if 1 is negative negative relationship with outcome, if 1 = 0 no relationship with outcome. Linear regression: continuous outcomes, y = 0 + 1x1 + 2x2 + + nxn, estimation of linear effect of an explanatory variable (x"s) on the outcome (y) after controlling for potential cfvs, need to have data on suspected cfvs, 1 represents average change in y for a unit change in x1, when the other explanatory variables (x2, x3, etc. Logistic modeling: dichotomous outcomes, logit (y) = 0 + 1x1 + 2x2 + + nxn, y is the natural logarithm of the odds of outcome, so e coefficient (i. e. e s ) = odds ratio of the outcome for a unit increase in the explanatory variable, keeping the values of the other variables in the model constant.