Statistics: Linear and Logistic Regression and Repeated Measures Analysis
Full course description
In this course the statistical techniques linear regression, logistic regression, and analysis of repeated measurements are introduced. With these techniques a broad range of statistical analyses of biomedical data can be conducted.
Course objectives
Goals:
Linear regression analysis, logistic regression analysis, analysis of repeated measurements. The student learns the most important concepts associated with these techniques. The student is able to apply these techniques with the statistical package SPSS on real biomedical data and can interpret the obtained results.
Concepts:
dependent variable, independent variable, intercept, slope, standard error, t-test for coefficient, t-value, p-value, confidence interval for coefficient, continuous and categorical independent variables, dummy variables, F-test for set of independent variables, residuals, residual plot, histogram of residuals, normal probability plot, interaction, R-square, sum of squares, multiple comparisons, Bonferroni adjustment, relation between regression analysis and independent-samples t-test or Pearson correlation coefficient, dichotomous dependent variable, risk, relative risk, odds, odds ratio, confidence interval for odds ratio, relation between relative risk and odds ratio, different sources for correlated data, linear mixed model analysis, fixed effect variable, random effect variable, random intercept model, interpretation of fixed effects parameters, variance of random intercepts, variance of residuals, intra-class correlation (ICC).