Data Deluge: Villanova Students Earn "Best Insight" at DataFest Philly 2018
Indeed is the No. 1 job site in the world. With more than 200 million unique monthly visitors, 500 million salaries, 100 million resumes, 72 million user ratings and 9.8 jobs added per second globally, the site generates an abundance of data—all the requirements for a weekend of data dives and analyses. Four Villanova students took on the challenge and won "Best Insight" at the 2018 ASA DataFest Philadelphia competition in April.
DataFest Philly is part of the national American Statistical Association DataFest, a 48-hour case competition where teams of two to five students receive a rich, complex dataset from a large company and analyze it for insights. This program takes data analysis beyond the constraints of a typical classroom setting by enabling students to work with real-world data as a team—uniting students in statistics, computer science, engineering, mathematics, social science and other fields for an intense weekend of data wrangling and analysis.
The event was organized by the Center for Statistics Education in the College of Liberal Arts and Sciences under the direction of Michael Posner, PhD. This year, Villanova University hosted DataFest Philly for the second consecutive year and included students from Temple University, Ursinus College and Villanova. The weekend kicked off in the Innovation, Creativity and Entrepreneurship Institute’s Idea Accelerator. University professors and local professionals joined the students to offer expert advice throughout the competition.
Once they received the data, Data Team Six, a Villanova team comprised of Elizabeth Ryan ’19 CLAS, Demetrios Papakostas ’18 CLAS, Thomas Longo ’19 CLAS and Vincent Mutolo ’19 CLAS, immediately developed a three-pronged strategy. First, they laid out a plan of attack for each day. Second, they developed an efficient way to share analyses with each other and work as a team.
“The third strategy that we employed was focusing on a specific issue, and we decided that this was going to be how to make Indeed more money,” Thomas said. “We approached the assignment as if we were consultants for the company.”
The team found that states with the most job postings also had the most postings with no clicks, and found a similar trend in densely populated counties. These areas were completely saturated with job postings, and essentially there were not enough people to click on all of them. The team suggested Indeed could set a limit of postings in certain areas in order to maximize clicks and profit.
“When you’re given such a large dataset like the one we were given for this competition, it’s difficult to narrow down what insights are important to share,” Elizabeth said. “Making the decision to keep or cut data was a great professional exercise for me because in the workplace, it’s crucial that you offer the most relevant and impactful information."
Along with the "Best Insight" award for their analysis and presentation, the Villanova students received a free membership to the ASA—offering professional development and networking opportunities with world’s largest community of statisticians. DataFest offered students real-world exposure, insight from industry experts and best practices for group work. Data Team Six agreed that getting creative with the data was exciting but working as a team was the most rewarding aspect of the competition.
“Sometimes working with big data among multiple people can be challenging, but our team did a great job of sharing the workload, the insights and the data manipulation codes along the way,” Elizabeth said.