Daniel Lakens

When it’s ok to use p-values to make dichotomous decisions | Note: seminar at 4pm instead of 12pm

Personalise
Headshot image of Professor Daniel Lakens
Time

01/06/2021 - 16:00 - 17:00

Address

Zoom meeting (online only)

Description

  • June 01, 2021
  • Speaker: Daniel Lakens
  • Topic: When It’s OK to Use P-Values to Make Dichotomous Decisions

Abstract

The never-ending criticism of p-values might make researchers doubt if the use of p-values is in any way defensible. Attempts to ban p-values, or to move beyond p < 0.05, suggest p-values are mainly harmful, and will lead to undesirable dichotomous statistical inferences. In this presentation I will discuss when p-values are a useful tool to answer a research question, and how dichotomous decisions were intended to be used in a methodological falsificationist framework. I hope to help researchers for who p-values answer a question of interest to continue their use with confidence, while also pointing out that p-values are a very specific answer to a very specific question, which means they should probably be used less often than they currently are. 

About the speaker

Daniel Lakens is an experimental psychologist working at the Human-Technology Interaction group at Eindhoven University of Technology. In addition to his empirical work in cognitive and social psychology, he works actively on improving research methods and statistical inferences, and has published on the importance of replication research, sequential analyses and equivalence testing, and frequentist statistics. He was involved in establishing dedicated grants for replication studies by the Dutch science funder NWO, and co-edited the first special issue on Registered Reports in psychology in 2014. His lab is funded until 2022 by a VIDI grant on a project that aims to improve the reliability and efficiency of psychological science. He teaches about better research practices on Coursera, and received the Leamer-Rosenthal Prize for Open Social Science in 2017 for his course ‘Improving Your Statistical Inferences’ in which more than 50.000 learners have enrolled.