Designing Better Studies
The international doctoral school REASON ENB is happy to announce the upcoming visit and workshop:
Friday 27.09.2019, 9:00-16:00
Room 3232, Leopoldstr. 13, 80802 Munich
Recent discussions in psychology have led researchers to acknowledge that there is room to improve our research practices. In this workshop, we will discuss state of the art statistical and methodological insights that will teach you how to draw better inferences from the scientific literature, and design more reliable and efficient experiments. I will explain why the scientific literature is biased, how you can detect this bias, and how you can design informative studies that should allow you to publish both true effects as null effects, and thereby contribute to a more reliable scientific knowledge base.
- Type 1 and Type 2 error control
- Why error inflation quickly destroys the evidence in your data, and how to prevent you from fooling yourself
- Approaches to sample size justifications
- Planning for accuracy
- Planning for power
- Sequential analyses: How can you design studies by repeatedly collecting data without inflating error rates? How can we specify falsifiable predictions? How can you examine null effects?
- Equivalence tests, Bayesian Rope, Bayes factors
- Equivalence testing: How do you perform an equivalence test? How do you calculate the power for a test aimed at showing the absence of a meaningful effect?
- Why the scientific literature is biased, how we know this, and what an unbiased literature should look like
- Evaluating the evidence in mixed results
This course aims to help you to draw better statistical inferences from empirical research, improve the statistical questions you ask when you collect data, and design better and more efficient studies. In practical, hands on assignments, you will learn techniques and tools that can be immediately implemented in your own research, such as sequential analysis, thinking about the smallest effect size you are interested in, and equivalence testing.
Assumed quantitative knowledge
You should have some basic knowledge about calculating descriptive statistics, and how to perform t-tests, correlations, and ANOVA's. A lot of the examples will be focused on experimental research, so having performed experimental research is useful as well.
We're looking forward to many interested participants!
Participation is only possible after prior registration at Arianne.Herrera-Bennett@psy.lmu.de