STATISTICS FOR EXPERIMENTAL RESEARCH
The course aims to make students aware of the usefulness of the statistical method in order to take decisions on a rational basis and to critically evaluate information from scientific literature, concerning analyses of data collected in Biomedical and health sector, and to rationally justify some diagnostic procedures and some indicators of health care and the evaluation of service activities.
The course aims to provide the basic knowledge of research in the biomedical field
• Evaluation of diagnostic tests. Sensitivity definitions (SE), specificity (SP), positive predictive values (VPP), and negative (VPN).
• Techniques of determination of the curves ROC
• Elements of probability calculation. Definition of probability. Calculation of the probabilities of mutually exclusive, independent and conditioned events.
• Critical reading of articles related to topics related to the CdS with significant content of biostatistics.
• Copy of the slides used during the lessons and bibliographic material available on-line on the site www.biostatistica.unich.it
The teaching is structured in 10 hours of frontal didactics, each lesson will be divided into two parts. A first part in which will be presented the topic of the lesson and in a second part in which will be proposed by the teacher of the exercises that allow to verify the practical application of the topics treated in a theoretical way.
The verification of the preparation of the students will be done by written examination composed of multiple answer questions and/or exercises that allow to verify the practical application of the subjects covered. The total points (30) will be divided on the basis of the questions present on the task and reported on the text in relation to each individual test question. There is no oral test.
The duration of the test is 30 minutes.
The teaching methodology and the methods of verification will be presented by the teacher at the beginning of the course.
Students will be received by the teacher by appointment to be agreed by e-mail to the address marta.dinicola@unich.it