Duration
15h Th, 15h Pr
Number of credits
Lecturer
Coordinator
Language(s) of instruction
French language
Organisation and examination
Teaching in the first semester, review in January
Schedule
Units courses prerequisite and corequisite
Prerequisite or corequisite units are presented within each program
Learning unit contents
The course focuses on the prospective determination of sample size using statistical power anlaysis, considering effect size, alpha level, statistical power, and sample size jointly. Particular attention is paid to the reasoned selection and scientific justificationof the hypothetical effect size, and to the critical use of previous literature while taking into account distortions related to low power, the winner's curse, and Type M error. The course also addresses the control of multiplicity across tests and dependent variables, other uses of power analysis, and the interpretation of results in terms of magnitude, précision, and uncertainty. G*Power is used in a reasoned way as a tool for statistical understanding rather than as a simple calculator, using examples based on tests of differences between means.
Learning outcomes of the learning unit
This course aims to provide students with the conceptual framework needed to understand the logic of power analysis, recognize the main issues involved in statistical planing, and avoid naïve or mechanical uses of this statistical approach. Students will be made aware of the need to select and scientifically justify hypothetical effects sizes, to interpret estimations together with their uncertainty, to consider biases that may effect the literature, and to distinguish the different uses of power analysis. They will also be inroduced to a reasoned use of G*Power for the situations covered in the course.
Prerequisite knowledge and skills
Students are expected to have acquired, through previous courses in psychostatistics, applied statistics, research methodology, and quantitative methods, the main statistical concepts required for the course: distributions, sampling, means, variance and standard deviation, confidence intervals, statistical hypotheses, Type I and II errors, p-values, commonly used statistical tests, and effect-size measures. Some prior familiarity with power analysis and power-analysis softwares may be assumed, but no advanced knowledge of statistical planning or of how to justify its parameters is required.
Planned learning activities and teaching methods
The course is based on detailed slides, generally made available to students before the corresponding sessions, and on an interactive teaching format in which students are encouraged to ask questions at any time. Course materials may be supplemented, or even corrected, as the course progresses in response to questions or comments raised by students. The pace of teaching will prioritize understanding of the concepts rather than exhaustive coverage at the cost of an overly rapid presentation. A small number of G*Power demonstrations will be used to connect statistical reasoning, methodological choices, and practical implementation, without organizing the course around sets of conventional statistical exercices.
Mode of delivery (face to face, distance learning, hybrid learning)
Face-to-face course
Course materials and recommended or required readings
Platform(s) used for course materials:
- MyULiège
Further information:
The main course materials consist of relatively detailed slide sets, which frequently include references to the statistqical or methodological sources used. A supplementary list of recommended references is also provided in these materials. Students are encouraged to explore some topics in greater depth by consulting these sources. There are no compulsory readings.
Exam(s) in session
Any session
- In-person
written exam ( open-ended questions )
Written work / report
Work placement(s)
An internship is possible. It will involve the use of statistical méthods, either covered in the course and further developed at the beginning of the internship, or introduced specifically for the internship, in order to assess the methodological quality of a set of scientific articles within the same field.
Organisational remarks and main changes to the course
No remarks.
Contacts
Prof. Ezio TIRELLI; ezio.tirelli@uliege.be
Association of one or more MOOCs
There is no MOOC associated with this course.