Biostatistics II
Objectives
After a hands-on practical revision of the statistical methodologies introduced in Biostatistics I, focus will be given to the multivariable analysis using, mainly, a practical approach. The aim of this curricular unit involves the learning of data modelling when the outcome variable is a continuous or a binary variable. More in-depth concepts and knowledge will be given to linear and logistic regression models. Detail will be given to the checking of the assumptions required for each model and to the use of residual analysis for this purpose. This curricular unit should promote the following skills: to identify the distribution of the outcome variable, to identify the independent variables and potential confounders, to select and implement the most adequate regression model, to verify models assumptions, and to interpret the results. Factors to be considered in the design, conduct and reporting of intervention studies to evaluate the health benefits of foods, will also be addressed.
General characterization
Code
41040
Credits
Available soon
Responsible teacher
Ana Luisa Trigoso Papoila da Silva,SOFIA ALEXANDRA FRIAS MENDES DA GRACA POETA,ANA FILIPA DA SILVA ALVES,ANA CATARINA DOMINGUES PEREIRA SANTOS,SUSANA ISABEL MATEUS SANTOS
Hours
Weekly - Available soon
Total - 0
Teaching language
PT
Prerequisites
Not applicable
Bibliography
Not applicable
Teaching method
Teaching is mainly practical based in the resolution of exercises using SPSS. Interaction between students and teachers is both at the classroom and by e-mail. Classes will take, at most, 180 minutes and will take place at a classroom with computers (1 for each student).
To assess students performance, a dataset will be made available that students, after setting up work groups, should analyze using the statistical methodologies taught.
Evaluation method
Teaching is mainly practical based in the resolution of exercises using SPSS. It will also be part of the evaluation, the elaboration and discussion of a report with the results obtained in the previous analysis.
Subject matter
Hypotheses tests revision: tests for two independent samples, tests for paired samples, Chi-square test, Fishers exact test and McNemars test, non-parametric tests for more than two independent (Kruskal-Wallis) and related samples (Friedmans test). Logistic regression model: model fitting, residual analysis and interpretation. Linear regression model: model fitting, residual analysis and interpretation. Guidelines for the Design, Conduct and Reporting of Human Intervention Studies to Evaluate the Health Benefits of Foods.