In this presentation, Professor Edward (Ted) Miguel explores recent efforts to improve research transparency, replicability, and credibility across economics and related fields. The presentation reviews the gap between core scientific ideals and real-world practices, focusing on publication bias and selective reporting. The talk highlights major changes over the past two decades, including the rise of study pre-registration, data and code sharing, and other open science practices across the social sciences.
New preliminary evidence from an ongoing study shows that while most studies are eventually published, many pre-specified hypotheses remain unreported, indicating persistent publication bias. The presentation concludes by discussing strategies to strengthen reproducibility and transparency, emphasizing the need for coordinated institutional and cultural change across the research community.