How broken is social science? (with Matt Grossman)

Sep 2, 2021

Join Matt Grossmann, Director of the Institute for Public Policy and Social Research, as he delves into the state of social science. He discusses the difficulties in studying human behavior versus other sciences and the ongoing replication crisis affecting credibility. Grossmann highlights advancements in methodologies and the challenges in political predictions, particularly post-2016 election. The conversation also touches on the importance of diversity in perspectives and the evolving self-reflection within research communities, especially amidst the recent pandemic.

Ask episode

AI Snips

Chapters

Books

Transcript

Episode notes

INSIGHT

Social Science vs. Natural Science

Studying humans is inherently harder than other sciences due to human diversity and researcher subjectivity.
Unlike geology, where rocks share similarities, human experiences vary greatly, and our interaction with evidence influences our interpretations.

INSIGHT

Increased Rigor in Social Science

Social science boasts more data, methods, and diverse interpretations than ever before.
These combined with improved tools, offer better research design and more definitive conclusions.

INSIGHT

Predicting Political Events

Predicting future political events depends on the timeframe and closeness of the race.
While easily predicting most district outcomes is easy, forecasting close presidential races years in advance remains challenging.

Get the Snipd Podcast app to discover more snips from this episode

Get the app

How Social Science Got Better

Matt Grossmann

In 'How Social Science Got Better', Matt Grossmann argues that social science is undergoing a renaissance, becoming more relevant, rigorous, and self-reflective. He emphasizes how scholars now better analyze biases and incorporate diverse perspectives, leading to significant progress in the field. The book provides a comprehensive overview of recent trends and improvements in social science research.

The better angels of our nature

Why Violence Has Declined

Steven Pinker

In this book, Steven Pinker presents a detailed argument that violence has significantly decreased over the course of human history. He uses extensive data and statistical analysis to demonstrate this decline in various domains, including military conflict, homicide, genocide, torture, and the treatment of children, homosexuals, animals, and racial and ethnic minorities. Pinker identifies four key human motivations – empathy, self-control, the moral sense, and reason – as the 'better angels' that have oriented humans away from violence and towards cooperation and altruism. He also discusses historical forces such as the rise of the state (which he terms 'Leviathan'), the spread of commerce, the growth of feminist values, and the expansion of cosmopolitanism, which have contributed to this decline in violence[1][4][5].

Expert Political Judgment

How Good Is It? How Can We Know?

Philip E. Tetlock

In this book, Philip E. Tetlock evaluates the predictive abilities of experts by analyzing 27,451 verifiable predictions made by 284 world-renowned experts over more than 20 years. He finds that expert predictions are often no better than random guesses and identifies two types of thinkers: 'hedgehogs,' who are confident in their single, overarching idea, and 'foxes,' who are more eclectic and adaptable. Tetlock argues that foxes are generally more accurate in their predictions due to their ability to consider multiple perspectives and adapt to new information. The book also explores how the media often favors hedgehogs for their confident and decisive statements, despite their lower accuracy.