Conférence d'Éloi Gagnon

Le 26 octobre prochain, à 13h00, un ancien étudiant du GRIP fera une présentation intitulée : When correlation IS causation: Gaining causal insights using regression analyses. 

La présentation sera faite en ligne. Pour y assister, merci d'envoyer un courriel au La présentation sera faite en anglais.

Au plaisir de vous y voir!

Résumé de la présentation:

When correlation IS causation: Gaining causal insights using regression analyses 

One of the earliest lessons we encounter as aspiring researchers is the familiar adage: “Correlation does not imply causation.” I vividly recall the initial draft of my first research paper I had submitted to my master’s supervisor. Every instance of “X increases Y” was systematically edited to read, “X is associated with an increase of Y.”  

As researchers, we often ask causal questions: “What factors contribute to the development of anxiety?” or “Would the administration of drug X alleviate symptoms of disease Y?” Our primary focus seldom revolves around mere prediction. Thus, a disconnect arises between the questions we ask and the methods and language we use. 

In this short interactive presentation, my primary objective is to dispel some common misconceptions regarding causality. In the process, I hope you learn how to harness the powers of regression analyses to address the causal questions you yearn to answer.

En ligne