Duke Wraps Fourth Annual DataFest Stats Competition
Duke University held
ASA DataFest 2015 March
20-22, 2015. It was the university's fourth-annual 48-hour statistics and
data analysis competition.
More than 220 students from multiple universities and a variety of academic
disciplines competed in teams of two to five students. When the competition
began at 6 p.m. on Friday night, each team was presented with a single data set,
which consisted of five tables and millions of rows of visitor information from
a comparison shopping site. Consultants from businesses and area schools were on
hand to answer questions. The teams had until 5 p.m. on Sunday to identify and
communicate insights into the data and present their ideas to the judges. Prizes
were awarded for best visualization, best insight and best use of outside data.
The team that won for best visualization developed a flow diagram and network
graph to help the company target its advertising based on trends in customer
preferences. The team that won for best insight found a way to help the company
identify customers likely to spend more than they planned based on their
demographics and Web browsing behavior. The team that won for best use of
outside data combined their assigned data set with outside data on political
preferences to find out if people from left- or right-leaning cities were more
likely to buy environmentally friendly products.
The competition fostered interdisciplinary collaboration between students
from diverse fields such as economics, sociology, political science, math and
engineering, in addition to statistics, according to Justin Yu, a participant in
the competition. For many of the students, it was their first time working with
real-world data sets.
"In most courses, the problems are guided and it is very clear what you need
to accomplish and how," said Tori Hall, a junior at Duke, in a prepared
statement. "DataFest is much more like the real world, where you’re given data
and have to find your own way to produce something meaningful."
ASA DataFest 2015 was organized by the by the Department of Statistical
Science at Duke University and co-hosted by the Department of Statistics and
Operations Research at the University of North
Carolina and Department of Statistics at
North Carolina State University.
About the Author
Leila Meyer is a technology writer based in British Columbia. She can be reached at [email protected].