Bayesian Models in Psychology 2500-EN-F-204
Bayesian data analysis is an alternative to the classical (frequentist)
approach to statistics, that deals directly with problems of uncertainty
and probability in research problems. As pointed by many authors, power,
flexibility, and easiness of interpretation of Bayesian data analysis makes
it a natural candidate to approach problems in modeling of psychological
processes.
During this lab we will learn the basics of Bayesian approach to statistics.
We will learn strengths of Bayesian alternatives to t-test, ANOVA,
correlation and regression analyses, and how to perform them using
open-source software (R and JAGS).
Type of course
Learning outcomes
Upon completion of this course:
students know basics of Bayesian data analysis and it’s theoretical
underpinnings
students know how to perform basic statistical computations with R
students know how to perform basic Bayesian analyses with R and
JAGS
students know potential applications of Bayesian models in social and
behavioral sciences and are able use some of them in their own
research
Assessment criteria
Students are allowed to miss 2 classes without excuse, 2 more classes in
case of excuse, but will not pass the course in case of more than 4
absences.
Additional work is assigned in case more than 2 classes are missed (even
in cases of valid excuse).
The Final grade will be determined by three components: midterm and
final exam scores and amount of points gathered from the home
assignments throughout the semester.
The final grade will be the weighted average computed according to the
following formula: 30% * (midterm score) + 30% * (home assignments) +
40% * (final exam score) = total score
Total score and both exam scores should be at least at the 50% level to
pass the course.
Grading scale:
95%+ = 5!
90-94% = 5
80-89% = 4.5
70-79% = 4
60-69% = 3.5
50-59% 59% = 3
below 50% = 2 (fail)
Additional information
Additional information (registration calendar, class conductors, localization and schedules of classes), might be available in the USOSweb system: