Jordan Graham, MPP Class of 2021

Jordan Graham, MPP Class of 2021, previously wrote a post about his favorite class fall quarter, so we thought we’d check back in with him to see what his favorite winter quarter 2020 class was.

Here’s a sentence I never thought I’d write: I love learning about statistics.

During my second quarter of Harris’s quantitative-focused core, Statistics for Data Analysis II with Professor Austin Wright has been my favorite course. I enjoy the broad applicability of the regression models we’re learning, the course’s methodical pacing that emphasizes a deep understanding of the statistical concepts behind these models, and Wright’s ability to fluidly incorporate students’ wide range of policy interests into his lectures.

My background is in newspaper journalism. For the six years before I enrolled at Harris, I worked as journalist for the Orange County Register, covering the government, policy, and politics of America’s sixth most populous county. While I had enough mathematical knowledge to perform pivot table analyses of public data in Microsoft Excel, I lacked a meaningful understanding of how to interpret and evaluate large data sets in ways that could uncover hidden connections between variables. Those shortcomings are one of the many reasons that I chose to attend Harris.

Harris touts its focus on evidence-based policy. The school’s tagline, “social impact, down to a science,” felt most fully realized in Stats II.

The course introduces students to regression models and emphasizes the value, obstacles, and commonly made mistakes in using regressions to determine correlation and causality between variables and outcomes. In Stats II, that sometimes entailed analyzing the effect of education on income, assessing whether construction of a border wall influenced rates of car theft, and even examining how acidity affects ratings of wine quality. We then used this information to determine the statistical significance of our models, to see how much explanatory power they had, and to evaluate how we could create regressions that more accurately described real-world relationships.

The class teaches students to think critically and remain skeptical during each step of analysis. That includes scrutinizing how randomized controlled trials are conducted, considering how variables might interact in ways that bias our models, learning how to deal with outliers in data, and accounting for trends in order to find connections in cross-sectional or panel data. Each solution presents a new set of challenges. But as the class progresses, the process begins to feel familiar. The value of the scrutiny is evident.

Midway through the quarter, Chicago’s former Chief Data Officer and current director of Analytics at KPMG, Tom Schenk Jr., spoke at an after-hours, on-campus event about how data is used in municipal government to create policy. During the event, I asked him, “For students who don’t want to focus on data analysis but want to partake in data-informed policymaking, what is the most valuable skill we can learn?”

His answer was simple: “A fundamental and deep understanding of statistical concepts.” Data fluency and the ability to recognize mistakes in research, he said, can be the difference between implementing the right evidence-based policy—one that helps people—and the wrong one. Knowing how to speak and understand the language of statistics is essential to that process, he said.

In Statistics for Data Analysis II class, that is exactly what we learned. And with each additional lesson, it has become clear to me how we’re building a skill set to help empower us when we leave Harris.

Did you enjoy reading this post? Read more recent posts from students:

Home