Sam Norris is a candidate in the economics department at Northwestern University. His research focus is applied microeconomics, specifically crime, education, and development. With the goal of answering policy-relevant questions with understandable, transparent research, his projects often cross into areas of sociology and public policy.
Norris is from Victoria, British Columbia, a short ferry ride from Vancouver on Canada’s western coast. He earned his bachelor’s degree in Economics from Simon Fraser University in 2010 and a master’s degree from the University of Toronto in 2012. Between his master’s and PhD studies, he spent a transformative year as a research fellow at Harvard University studying under Monica Singhal and Rohini Pande.
Welcome to Chicago, and congratulations on your fellowship. This semester at UChicago is the latest chapter in your academic career. You have focused on economics since your undergraduate degree at Simon Fraser University. How did you become interested in this research area?
I have always been interested in economics because it’s so central to everything we do. Economists are constantly in the news because they are tackling important issues in sociology, policy, finance, and of course, the economy. Right around the time I started university, I read Steve Levitt’s book [Freakonomics: A Rogue Economist Explores the Hidden Side of Everything], which really solidified my interest in economics and got me thinking about the statistical tools that economists have to answer research questions in a way that is both causal and policy-relevant.
My interests have evolved into trying to understand different aspects of policy within education, crime, and development. My goal is always to write research that is comprehensible to non-economists and can be readily incorporated into public policy.
You just submitted a paper to the Journal of Human Resources on the effect of school start time on student success. Can you describe your project and how it adds to the large body of research on this topic?
Well, the impetus behind this research is based on my and my co-author’s experiences in our younger days – it is widely known by students and parents everywhere that teens do not like waking up early for school. There is a large body of physiological evidence showing a relationship between start time and alertness, which probably has ramifications for academic achievement. We’ve been talking about this for 20 years or more, yet very few jurisdictions in the US have made changes. One of the main reasons is parent work schedules; moving school start time an hour later for adolescents is not logistically feasible for many families. Additionally, older kids arrive home earlier than their younger siblings, so there is a babysitting aspect to the current school schedule as well.
That said, there may still be good reasons to consider adjusting or swapping start times for younger and older students. Instead of having high school kids start school before their younger siblings — which is the case in most schools — we might do better if we had high school kids start last, letting them sleep later. What we wanted to do in this project was quantify how much that sort of arrangement would improve academic achievement.
We started with the insight that wake schedules are largely driven by sunrise and sunset. Most of the physiological changes that occur for children during the act of waking are a result of sunlight hitting their eyes, meaning that, from a biological standpoint, the policy-relevant measure of start time for a given school is relative to sunrise. We tracked test scores of children of various ages who moved from one side of a time zone boundary across into the adjacent zone. When students moved from east to west, they experienced earlier sunrises (and later relative start times), and their test scores significantly improved. The improvement was mostly driven by older children – girls at age 11, and boys at 13, which lines up with the gender-specific transition into puberty.
The take away message from this piece of research is that we need to look at this issue from a biological as well as from a developmental standpoint. Our contribution is that we’ve shown a strong relationship with puberty and with age. Our findings are consistent with those in physiology and indicate that we can do a lot better by our students if we reverse school start times, with the youngest students starting first, followed by middle school and high school.
You have another work in progress on a completely different topic. Can you discuss your current project on judicial discretion?
In the US, there are over 100 million court cases each year, ranging from criminal and civil cases to decisions on social security benefits, visas, and health inspections. In the Chicago Metro area alone, there are over a million court cases annually. Those statistics got me wondering about the decision-making process and potential ways to quantify how often judges are making mistakes. This question is fundamentally difficult to answer because in all cases, we never actually hear what the objective truth is, but I wondered if there was a way to learn something about judicial process and discretion without having to know the truth.
The environment I’m using to answer these questions is unusual because two judges separately hear each case and decide whether a person qualifies for the program to which he or she is applying or not. In this unique situation, we can look at cases where the judges disagree; while we won’t always know which judge is right or wrong, we can begin to make some inferences. We started by categorizing judges as either lenient or strict. For instance, in criminal cases, a lenient judge might incarcerate 20% of defendants, and a strict judge may be closer to a 45% incarceration rate. Our question here is whether the judges are sending the same people to jail – or in this case, admitting the same people to the program. In general terms, we are defining a “mistake” as a situation in which someone is approved by a strict judge, who typically admits only a small number of people, but denied by a lenient judge who admits almost everyone.
We are still analyzing data, but the central finding is that, although all judges make mistakes, some make more mistakes than others. And there is enough difference in accuracy between judges that replacing the inaccurate judges with accurate ones would make a substantial difference in the overall accuracy of the system. What I’m doing in this study is something that can be implemented by judicial systems to help policy-makers learn about higher and lower quality judges. We can also compare this estimate of judge quality to Bar Association surveys answered by other players in the system (e.g., prosecutors, defense lawyers, other judges, etc.). To the extent that the two rankings match, the Bar survey may be a useful proxy for judge quality, or we may find it is too noisy to make inferences on individual judges. This approach is a way to learn more about that process and what it means.
What are your expectations for your time in Chicago, and how do you think it might benefit your research?
In basic terms, Becker’s contribution was to take economic tools and apply them to a broader set of questions. That’s what I do in my own research, so BFI is a great place to be. In my time here, I am most eager to meet the folks at the BFI, talk about my research, and get feedback from multiple new perspectives. I have heard so much about the Price Theory class and the Chicago style of seminars, and I’m already familiar with much of the work done here at the University. Many of the big names come up over and over again in the literature – Mogstad, Heckman, Levitt, Bertrand, and Murphy to name a few – and I see a lot of overlap with my interests.
Sam Norris is one of three 2016 Price Theory Fellows attending UChicago this semester. The award, supported by the Searle Freedom Trust, is intended to introduce PhD students from economics programs around the country to the Chicago price theory tradition.