CHDH Seminar Series 2020: What can we do about generalisation? MRP in psychology research — Lauren Kennedy
Date: Monday 31 August
Time: 12 - 1 PM
Location: Via Zoom. Please use this link at the time of the event to join: https://unimelb.zoom.us/j/95160457106?pwd=ZnNNeGw3SG41TzRDQm5tejBSY01Rdz09
Psychology research has long relied on convenience samples. While it has been argued that the type of sample (from undergraduate student samples to Amazon MTurk samples) increases generality of results, this is hard to argue without clearly stating a population that we wish to generalize to. In this talk we consider multilevel regression and poststratification (MRP), a method used in political polling and survey research to adjust from sample to population. We consider whether this method could be applied to psychological research, what would need to be considered to do so, and what we need to know about the method to be secure in our findings.
Read Lauren's preprint on this topic here: https://arxiv.org/abs/1906.11323
Lauren Kennedy is a lecturer in business analytics at Monash University. She completed her PhD with A/Prof Amy Perfors and A/Prof Danielle Navarro in 2018, before undertaking a postdoc with Prof Andrew Gelman at Columbia University. Her work aims to understand and model heterogeneity using Bayesian statistics and survey statistical approaches.