Joe is a digital transformation expert and scientist, focusing on the integration of data science and artificial intelligence into executable corporate strategies. He is the Head of Data Science at Search Discovery, where he leads groundbreaking data science initiatives for a diverse clientele of Fortune 500 and startup clients, including some of the most admired companies in the world. His career has spanned numerous technical and operational roles at venues including The White House, Columbia and Princeton. He was a founder at the artificial intelligence startup Peachtree AI, and an initial team member at the fintech startup Prattle, both of which successfully exited in 2019. From 2011-2013, Joe worked in the Obama White House as a Presidential Advance Staffer. Prior to that, he served as a freelance software developer, running a single-shingle web development shop called CLEVER Analytics from 2006. He is presently affiliated as a Research Fellow at Johns Hopkins University.
- PhD, MPhil, Master’s, Columbia University
- Bachelor’s, Washington University in St. Louis
My principles are twofold. First, we will not be saved by big data—we will be saved by the analytical techniques that allow us to draw actionable, clear insights from them. Second, a method is only as useful as the substantive conclusions you are able to draw because of it. At the core of my research is an obsession with understanding how the digital transformation of our society – first niche television, then the internet, now cell phones and predictive personalization at scale – has changed both the actual distribution of power in our society, and the way we study this distribution of power. The agenda I pursue to satisfy this obsession is held up by an applied leg and a methodological leg.
My applied studies focus on political behavior, which includes topics of polarization, nationalization, and self-censorship. My greatest interest is in the present phenomenon of affective polarization, in which voters express extreme dislike for each other on the basis of political in and out-groups, without reasoned concern for issue disagreement. One of my favorite findings, which Jon Rogowski and I published in the journal Political Behavior, is that we may be able to overcome affective polarization by presenting individuals with “humanizing” information about the other side. The RCT suggested that when voters were treated with small vignettes about the out-group’s day-to-day life, even the most extreme haters came around and moderated their evaluations of the out-group. Due to partisan sorting, issue consistency, and economic trends in technology and inequality, perhaps we just don’t see enough of the other side these days.
My methodological research demonstrates how big data methods, particularly analyses involving machine analysis of text, can introduce non-ignorable bias and inefficiency into results and undermine conclusory exercises. I then develop theoretically supported frameworks to overcome this bias and inefficiency, and derive statistical methods to operationalize them. The subject matter is part of the emerging field of computational social science, which combines data science, politics, and economics to conduct research at scale.
One of the biggest threats to our sustained progress is, in my opinion, the difficulty of integrating instruction in the social and computer sciences. It’s tempting to blame an over-indexed promotion of “STEM” majors for it…“those arts and sciences students just weren’t made for engineering.” But the difficulty is in my opinion due to a lack of access to cross-pollinated courses of study. What are we doing to demonstrate to computer science students that they can thank Bernard Berelson, a sociologist, for that NLP algorithm they just ran? What are we doing to show arts and sciences students that qualitative inference is an exercise in probability?
The implications of these integration difficulty extend well beyond the ivory tower. We have hundreds of AI-powered companies predicting when you need your laundry picked up and washed, or your food delivered, or your personal finances managed. Yet we atrociously don’t have a widely-used and accessible business intelligence solution in the legal profession. Crossover specialists – people who understand the challenges of their chosen field, but who also understand how a system might be engineered to overcome them – are what we need to mitigate this problem, but they seem to exist primarily in fields where the challenges are commonplace! When we fail to grow “crossover” specialists in the academy, we fail to provide the talent necessary to overcome the important, niche-specific issues that are endemic to every industry, country, and profession.
Particularly, graduates of political science and economics – many of whom go onto law school, business school, and other professional degrees – may never get another opportunity to fully understand the power of AI technology for the prediction and analysis of the world around the. This places these graduates at a significant disadvantage, and hobbles the potentially incredible contributions they might have offered.
I aim to overcome this difficulty by introducing new courses into the ordinary social science curriculum that introduce these students to such “engineering” concepts. The courses are anchored and driven by substantive research applications that require deeper consideration of the methods used to develop them. The method is never as important as the actionable insights generated with it. The benefit is that engineering student may cross-register for the courses, and be exposed to an entirely different way of thinking, while entering into collaborative relationships with their peers.
Initiatives of Interest
Digital transformation can (and will, with the right leadership!) create a more thoughtful and productive world. Better measurement strategies mean we’re all on the same page about how we approach a problem; better predictions about the future mean we can free people up to work on more important judgment calls based on those predictions. The more people we can get on the same page, investing brainpower in the things that matter, the more people will be better off tomorrow than they were today.
A great example of an organization walking the walk is Mentor Collective, an ed-tech startup in Boston that runs impactful mentorship programs for students who traditionally would not have the resources to navigate the jungle of higher education. We’re leading digital transformation and insights generation in the education sector by deploying ongoing “voice-of-student” surveys in academia, which enables tracking outcomes and melt probability at a much higher frequency. This transformation has allowed us to deploy more than 15 randomized controlled trials studying the effect of educational interventions on student success—and we’ve shown a causally identified, positive effect for thousands of students (insights, not just data!).