Guest
What is p-hacking? In 2011, Joe Simmons, Leif Nelson, and Uri Simonsohn published an academic paper titled False-Positive Psychology: Undisclosed Flexibility in Data Collection and Analysis Allows Presenting Anything as Significant. In this paper, they first demonstrated just how dangerous what is now well known as p-hacking can be. In short, they showed that what were once conventional research practices could result in incredibly high levels of what we call Type 1 error, or false positive results. These are results that allow researchers to publish papers with results that actually have little basis in reality. Welcome to Data Demystified. I’m Jeff Galak, and this is my series of videos on p-hacking. In part 1 of this series we’re going to dig into just how p-hacking works and why, when abused, it can undermine our confidence in science. In the next four parts we’ll dig into specific techniques that unethical researchers might use to actually p-hack their data with the goal of developing an intuition for just why those techniques are so bad. And in the final video I’ll explain a brilliant tool that these same researchers devised called a p-curve that helps detect these types of bad practices. P-Hacking Series Playlist: Video 1: P-Hacking Explained (Released on 6/24/2021): Video 2: Dropping Conditions that “Work“ (Released on 7/1/2021): Video 3: Multiple Measures Misuse (Released on 7/8/2021): Video 4: Covariate Misuse (Released on 7/15/2021): Video 5: Selective Stopping Rules (Released on 7/22/2021): Video 6: P-Curve (Released on 7/29/2021): Link to video about statistical significance testing: Link to video about False Positives: Link to academic paper by Simmons, Nelson and Simonhson: Link to academic paper by John, Lowenstein, and Prelec: Follow me at: LinkedIn: Patreon: Equipment Used for Filming: Nikon D7100: Softlight: Yeti Microphone: iPad for Teleprompter: Camtasia for Video Editing:
