19 July 2013
on course website
My First Bayes: Why and How to Run Your First Bayesian Model using MPlus
Did you ever wonder why:
- A p-value can be used to falsify a null-hypothesis but not to validate it?
- You are testing the null hypothesis even when it is never among your hypotheses of interest?
- It is sometimes difficult to interpret the results of classical hypothesis testing?
Did you ever encounter one of the following issues:
- A data set too small for your complex model?
- Non-normally distributed variables?
- Negative variances or correlations larger than one?
Or did you ever want to compute:
- The probability that your hypothesis is correct after observing the data? Note that this cannot be done with the classical p-value.
- A 95% probability that your estimate (e.g. mean, regression coefficient) is in between two values. Note that, again, this is not the interpretation of the classical confidence interval.
- A degree of support for each of the models in your model selection competition.
If you answer ‘yes’ to one of these questions, this workshop might be of interest for you!
During this five-day workshop you will be gently introduced into Bayesian statistics. Bayesian statistics are becoming more and more popular among applied researchers to answer the research question at hand. This is especially due to the availability of Bayesian estimation methods in popular software like Mplus where it is very easy to switch to Bayesian statistics. The purpose of this 5-day workshop is to explain why and how to use Bayesian statistics without going into the mathematical details. No prior knowledge of Mplus is assumed, but if you are not familiar with this software we will ask you to prepare a homework exercise.
Rens van de Schoot
If you want to learn when or how to switch to Bayesian analyses using the software Mplus (because it is very easy to do so in Mplus), this course is for you. You should have some (basic) experience with regression analysis and preferable other SEM software, for example AMOS, LISREL, openMX, or SAS. No previous knowledge of Mplus is assumed. You do not need to know matrix algebra, calculus, or likelihood theory, or any knowledge on Bayesian statistics. Participants from a variety of fields, including sociology, psychology, education, human development, marketing, business, biology, medicine, political science, and communication, will benefit from the course.
Aim of course: After attending this course students are able to:
- argue when switching to Bayesian statistics is beneficial
- understand the basic ingredients of the Bayesian toolbox
- formulate models and hypotheses in the Bayesian way
- execute Bayesian analyses using the Mplus software
- interpret the output and note the differences with classical output
Certificate of Attendance
EUR 795: Course + course materials + housing
EUR 595: Course + course materials
Utrecht Summer School doesn't offer scholarships for this course.Register for this course
on course website