Scientific training

Academic year: 2010-2011 Course code module 3BGEN-12 Semester: 1st semester Credits: 3 Study load (hours) 84 Theory (hours): 20,00 Practice/Exercises(hours): 8,00 Other (hours): Part-time program: Instructor(s) Joost Weyler Language of instruction: Dutch Semester exam information: exam in the 1st semester Contract restriction information: faculty decision 1. Prerequisites *Algemene competenties The student must have general basic knowledge how to use a personal computer as enduser. Also basic knowledge of English is required, since the student has to look up information via international scientific publications which are usually written in English. *Sequentiality None all courses of Ba1 and Ba2 2. Objectives (expected learning outcomes) To know Study aim 1. Make the difference between random error and the systematic error and indicate with which methods theses errors are contested in the medical scientific research. Study aim 2. To define confounding and this on the basis of notions component causes, sufficient causes.  Indicate that confounding occurs in the study population and enumerate the conditions for confounding. Study aim 3. To define effect modification and to indicate that effect modification occurs at the level of the theoretical population. Study aim 4. To indicate what the difference is between confounding and effect modification. Study aim 5. To formulate the diagnostic problem in terms of a multiple determined probability in the presence. Study aim 6. To formulate the prognostic problem in terms of a multiple determined probability with a direction in the future. Study aim 7. To formulate the etiognostic problem as a multiple determined probability in the presence with a direction in the past (causal antecedents). Study aim 8. To make the difference between descriptive and causal functions. Study aim 9. To indicate what correlation is. At the same time the student knows the difference between the Pearson and Spearman correlation coefficient. Study aim 10. To indicate what explained variance is and how this can be derived from the correlation coefficient. Study aim 11. To clarify the concept of regression and to make the difference between the correlation coefficient and the regression coefficient. Study aim 12. To write the basic form of the multiple regression, (linear, logistic and Cox) and to indicate how effect modification can be presented by an interaction term. Study aim 13. To indicate that repeated measurements of a same person lead to correlated data and that these data have to be analysed with appropriate statistical technics.   To understand Study aim 1. To indicate what the hierarchy is concerning the importance of the different kind of errors (random error and systematic error) on the design and analyse of medical scientific research. Study aim 2. To give a critical discussion concerning the manner in which confounding is treated in medical scientific research. Study aim 3. To explain conflicting research results in the light of effect modification. Study aim 4. To translate the different objects of the medical scientific research to adequate mathematic models on the basis of examples. Study aim 5.  To consider the concepts of temporality and directionality when creating a research design. Study aim 6. To explain the relation between correlation and regression. Study aim 7. To explain the relation between linear and logistic regression. Study aim 8. To explain the relation between  Cox regression analysis and Kaplan-Meier survival analysis. Study aim 9. To indicate what the dangers are of the stepwise building up or breaking down of a regression model. Study aim 10. To interpret the output of a multiple regression analysis (linear, logistic, Cox) and to translate the results into the different measures of effect. Study aim 11. To indicate what the advantages are of applying longitudinal data analysis on correlated data.   To apply Study aim 1. To perform the different multiple regression techniques for diagnostic, prognostic and etiognostic problems on basis of a simple dataset. Study aim 2. To perform the basic techniques for longitudinal data analysis (dependent t-test and R.M.- ANOVA). 3. Course content - Multicausality - Sources of errors in medical scientific research - Confounding and effect modification - Diagnosis: prevalence functions, multiple diagnostic indicators… - Prognosis: non experimental prognostic research - Etiognosis: causal functions - Correlation and regression - Simple linear regression - Multiple linear regression - Multiple logistic regression - Multiple Cox regression - Introduction to longitudinal data analysis 4. Teaching method Direct contact: Lectures Practical sessions Personal work: Supervised self-study 5. Assessment method Exam: Practical exam Continuous assessment: Assignments Written assignment: With oral presentation 6. Compulsory reading – study material Syllabus (to find on the website) Manual statistics 7. Recommended reading - study material 8. Tutoring laatste aanpassing: last update: 14/01/2010 10:27 david.kums