Women in Data Science @ Penn Conference
Agenda

8:30–9:00 a.m.

Check In + Grab-and-Go Breakfast

9:00–9:15 a.m.

Welcoming Remarks

9:15–10:05 a.m.

Keynote Address

Predictability in the Unpredictable: An Exercise in Whole-Brain Marketing

After leading customer experiences across traditional to digital brands such as Zillow, Starbucks, and Campbell Soup, Aimee Johnson is now synthesizing her experiences to mentor startups and C-suites on whole-brained marketing. Learn how this CMO excelled in creating a storied career by intersecting human behavior, creativity, data, analytics, and technology to drive impactful digital transformation. Aimee will share her career journey from programmer to CMO and important career tips she has found valuable along the way.

10:05–10:35 a.m.

What Big Data Reveals About Media Biases

An understudied facet of the information ecosystem is the scale of misleading and biased information in traditional media and its impact on society. High-profile anecdotes of misinformation and echo chambers on the web may raise (legitimate) concerns about social media platforms such as Twitter and Facebook, but it is important to keep in mind that Americans still get most of their news from TV. Homa Hosseinmardi will present a recent work that assesses the TV news consumption habits of tens of thousands of American adults in a longitudinal observational study to uncover their news viewership patterns.

10:35–10:50 a.m.

Break

10:50–11:20 a.m.

Responsible Data Management

Incorporating ethics and legal compliance into data-driven algorithmic systems has been attracting significant attention from the computing research community, most notably under the umbrella of fair and interpretable machine learning. While important, much of this work has been limited in scope to the “last mile” of data analysis and has disregarded both the system’s design, development, and use life cycle (What are we automating and why? Is the system working as intended? Are there any unforeseen consequences post-deployment?) and the data life cycle (Where did the data come from? How long is it valid and appropriate?). In her presentation, Julia Stoyanovich will argue two points. First, the decisions we make during data collection and preparation profoundly impact the robustness, fairness, and interpretability of the systems we build. Second, our responsibility for the operation of these systems does not stop when they are deployed.

This talk is based in part on an article by the same title that appeared in Communications of the ACM in June 2022. Read more here.

11:20–11:35 a.m.

Hey, what’s up? Gas prices: Analyzing Influences of Gas Price Trends

The United States retail gasoline prices have dramatically fluctuated over the past two decades. Recent events, such as COVID-19 and the Russia-Ukraine conflict caused extreme fluctuations in gasoline prices. Furthermore, American citizens’ growing concerns over concerns gasoline prices drive a divide between political parties, states, and countries. This study aims to identify the major factors that influence retail gasoline prices in the United States using multiple linear regression, LASSO, text mining, and random forest. The results will display factors influencing the price of gasoline and predict future gasoline prices.

11:35–11:50 a.m.

COVID-19 Contact Tracing App Reviews Reveal Concerns And Motivations Around Adoption

Google and Apple’s Exposure Notifications System (ENS) was developed early in the COVID-19 pandemic to complement existing contact tracing efforts while protecting user privacy. An analysis by the Associated Press released in December 2020 estimated approximately 1 in 14 people had downloaded apps in states one was available. In this study, Sukanya Joshi and team assessed the motivation and experience of individuals who downloaded ENS apps from the Google Play and Apple App Stores.

11:50 a.m.–12:05 p.m.

Lunch

12:05–12:30 p.m.

Academic Breakout Discussion & Lunch

Join Drs. Susan B. Davidson and Julia Stoyanovich to continue the conversation about Responsible Data Management.

12:30–12:55 p.m.

Career Journeys in Data Science Discussion & Lunch

Join Mary Purk and special guest to discuss career advice in the industry sector.

1:05–1:45 p.m.

Data Science in Finance, Retail, and Tech

1:45–2:00 p.m.

McDonald’s Complaint Analysis

Using sentiment analysis, keyword extraction, and topic modeling, the Wharton Analytics Accelerator student team derived business insights from McDonald’s complaint data. After using the language analyses above, we provided McDonald’s with clusters of complaint customers that would guide complaint management and handling.

2:00–2:15 p.m.

Grading Defenders in Coverage Using NFL Player Tracking Data

Analyzes wide receiver – cornerback matchups to evaluate defenders and create game plan strategies.

2:15–2:50 p.m.

Effects of CCT Programs on Academic Achievement

This work develops and estimates a dynamic model, which integrates value-added and school choice models, to evaluate grade-by-grade and the cumulative impacts of the Mexican Prospera conditional cash transfer (CCT) program on educational achievement. The empirical application advances the previous literature by estimating policy impacts on learning, accounting for dynamic selective school attendance and incorporating both observed and unobserved heterogeneities. A dynamic framework is critical for estimating cumulative learning effects because lagged achievement is an important determinant of current achievement. The model is estimated using rich nationwide Mexican administrative data on schooling progression and math and Spanish test scores in grades 4-9 along with student and family survey data. The estimates show significant CCT learning impacts, inter alia, particularly for students from poorer households. School-choice decisions, especially whether to attend tele-secondary schools, play crucial roles.

2:50–3:00 p.m.

Closing Remarks