Advice-Taking Metaanalysis
Analysis Report
Overview
This site is my log of work on converting the Himmelstein (2022) advice-taking dual-hurdle model to better model advice-taking in a metaanalytic context.
Metaanalysis Model
First, I converted the Himmelstein (2022) model from a two-level cross-classified model, with responses cross-classified within judges and items, into a three-level cross-classified model, with responses cross-classified within judges and items and then nested within studies.
One key initial finding from this metaanalysis is that some studies have a high proportion of Weight of Advice (WOA) values of exactly 0.5. To address this, we explored adding a third hurdle to the model.
Tri-Hurdle Model
With just one study that was high on 0.5s, we modified the original model to include an additional Midpoint hurdle.
Tri-Hurdle Metaanalysis Model
With that model successfully estimated, we then merged the tri-hurdle and metaanalytic model frameworks.