The Science of Second Chances: A Revolution in Criminal Justice

by Jennifer Doleac

Freakonomics for criminal justice , The Science of Second Chances presents a groundbreaking approach to criminal justice reform, revealing how small-scale interventions can reduce people’s chances of reoffending and break the incarceration cycle.

When criminal justice expert Jennifer Doleac thinks about reform, she’s not just hopeful, she’s optimistic that second chances are possible—for the justice-involved population and the system as a whole. In The Science of Second Chances, she reveals her powerful approach to reducing crime and incarceration. Drawing on cutting-edge economic research and real-world experiments, the book presents a blueprint for reform that runs all the way through the system. Doleac shows how economists like herself approach big, complicated problems as if they were scientists in a lab, carefully testing different approaches and following the data to maximize impact. She explains how shifting the incentives people face can produce dramatic changes in the decisions they make, significantly reducing the number of people cycling through the prison system. From DNA databases that increase the likelihood of catching reoffenders to leniency programs for first-time defendants, she reveals a series of surprising interventions that actually work, along with cautionary tales about great ideas that never panned out.

Doleac doesn’t have a “burn it all down” mentality but seeks to empower readers with practical, achievable solutions. She demonstrates that we can have both public safety and a smaller, less intrusive justice system—without waiting for big, structural reforms that might never come. By shifting focus from the enormity of the problem to the power of small, evidence-based changes, she offers a transformative approach to prosecution and punishment. The Science of Second Chances is essential for anyone seeking data-driven strategies to revolutionize the criminal justice system and create the society we want and need.

Genres: Nonfiction, Science, Crime, Psychology, Sociology

304 pages, Hardcover

Expected publication February 24, 2026

Reviews

[Brendan (History Nerds United) · Rating 5 out of 5]

I love me some data! Luckily for me, there is plenty of that in Jennifer Doleac's The Science of Second Chances. No emotions, no arguments, just cold, hard data. My favorite!

Doleac looks at various experiments which have been conducted around the world to improve the criminal justice system. The first question anyone may have is how partisan is it. It's not at all. Doleac even points out repeatedly that she won't ask questions which have value based responses. How long should someone be in jail for it to be considered justice by society? Totally immaterial. Instead, the author may explain an experiment where we look at how more usage of ankle monitoring devices affect recidivism.

What I found most interesting is that while Doleac focuses mostly on experiments with positive effects, there are a slew of examples where you would expect one thing and get another. This is not an anomaly either. As with most data driven science, Doleac explains that the vast majority of theses will fail. It's a feature, not a bug.

This book is best for folks who want to see unambiguous (when the data supports it) and unbiased looks at various criminal justice reforms. There isn't any anecdotes where the author pulls on your heartstrings. This is all about the science. I loved it.

(This book was provided as an advance reader copy by Henry Holt.)

[Savannah Quick · Rating 5 out of 5]

I was so excited when I saw that Jennifer Doleac had written this book. I think she's the perfect person for this type of exploration of the data we have on criminal justice interventions. She touches on aspects of criminal justice ranging from supervised release to the impact of pollution exposure on impulsivity and subsequent incarceration.

I think this is critical reading for policymakers at all levels. I'll certainly be calling Erik Bottcher's office and recommending he read this when it comes out. I think policymakers and practitioners should especially pay attention to Doleac's conclusion. Failure isn't unexpected, and it's better to fail fast and know that a program isn't working than to drag out programs that have adverse impacts.

My only critique of the book is that there are parts that get repetitive. She explains the quasi-experimental setup and how it approximates randomness each time she introduces a non-RCT study. This isn't necessarily bad! I believe her audience is one she's trying to train into thinking about evidence. However, I wonder if her intended audience really understands why RCTs are the gold standard. I think this book would benefit from a chapter offering a crash course on casual inference. It could be framed as optional reading and it could even be placed at the back. However, I think a section folks could reference to understand why randomization is the gold standard (and perhaps something that does a quick overview of the intuition of commonly used quasi-experimental methods) could be really useful to practioners and policymakers without much experience thinking about what makes research reliable.

Thank you to NetGalley and the publishers for the ARC.

[Natalie · Rating 4 out of 5]

The biggest takeaway I had from listening to this audiobook was how important it is to base societal change on evidence-based data supported by appropriate statistical analysis. This message feels especially relevant, as people often support ideas that reinforce what they want to believe or think should work, rather than what the evidence actually shows. A lot of the criminal justice reform research that was discussed in this book I had heard about before, but I did learn about some new viable ideas for change. This book does not discuss the psychological understandings or racial disparities of criminal justice besides some very surface level statements. I will have to read something else to fill my desire to read something touching on those (if anyone has recommendations).

[M. · Rating 5 out of 5]

It was a good read.