El Podcast

E142: How to Lie With Research (Even If You’re Not Trying) - Alex Edmans

Episode Summary

Alex Edmans, a finance professor at London Business School, breaks down how studies and statistics are often twisted to fit a narrative—even by smart people and big institutions. His book May Contain Lies shows how to spot bad information and think more clearly in a world full of spin.

Episode Notes

Finance professor Alex Edmans joins to expose how research, statistics, and stories are often weaponized to mislead us—and what we can do to resist confirmation bias in a post-truth world.

👤 Guest Bio

Alex Edmans is a Professor of Finance at London Business School, a former investment banker, TED speaker, and the author of May Contain Lies: How Stories, Statistics, and Studies Exploit Our Biases—and What We Can Do About It.

🧠 Topics Discussed

• CEO pay, testimony before UK Parliament, and research misrepresentation
• The problem with cherry-picked or manipulated studies
• Diversity, ESG, and performance: what research really shows
• Why “smart” people still fall for bad ideas (e.g., Theranos)
• How confirmation bias affects policymakers, businesses, and parents
• The myth of breastfeeding and IQ
• Using ChatGPT and AI tools more critically
• Why post-truth thinking is dangerous—and how to fight back

📌 Main Points

• Misquoting research is rampant—even in government policymaking.
• Studies can be cherry-picked or selectively framed to “prove” anything.
• Diversity and ESG don’t always lead to better performance, especially when oversimplified.
• Smart investors, policymakers, and academics are just as vulnerable to bias.
• AI tools like ChatGPT can reinforce misinformation unless prompted with skepticism.
• It's not about learning statistics—it’s about applying common sense and open-mindedness.
• Be as skeptical of studies you like as those you dislike.

💬 Top 3 Quotes

• “Even if a statistic is 100% correct, the interpretation of it can still be completely misleading.”
• “We know how to poke holes in studies we disagree with. The challenge is doing the same when we agree with the findings.”
• “AI can’t fix misinformation—it reflects it. You need to know how to interrogate it.”