Tuesday, March 28

Edmunds.com sponsors data set for fifth ASA DataFest at UCLA



The original version of this article incorrectly stated that Edmunds.com is a website with information on used car sales. In fact, it is a car shopping website.

Hundreds of tired college students poured over a vast data set, arguing about correlation and causation and trying to learn new programs at UCLA early Saturday morning.

ASA DataFest, a 48-hour data analysis marathon, lets students analyze a trove of data any way they want and present conclusions to judges, said Robert Gould, undergraduate vice chair at the statistics department and founder of the event. The event was held this year for the fifth time.

The data this year came from Edmunds.com, a car shopping website, and included information on users’ advertisement success rates, purchases, frequency of visiting the site and other data.

The event was first held in 2011 with less than 100 students, but has now grown to several sites across the country. UCLA holds the largest event with students from different schools participating, including USC, UC Riverside and California Polytechnic State University.

As interest in the event grew, attendance rose to the hundreds and organizers moved ASA DataFest from the Institute of Pure and Applied Mathematics on campus to De Neve Plaza. ASA DataFest partnered with UCLA Residential Life this year, allowing the event’s venue to expand.

“We had to put some students on a waitlist this year and had to turn some away because there just wasn’t enough room,” said Linda Zanontian, program director for ASA DataFest.

In past years, the Los Angeles Police Department and eHarmony.com have provided data sets for the event. Companies provide huge data sets and in return see conclusions from hundreds of college students.

“Class is not the same as the real world. During class we have rules for predictions, now we get to ask questions by ourselves,” said Ze Jia, a third-year statistics and economics student who attended the event.

Representatives from Edmunds.com and professional statisticians also attended the event to help students and to scout for possible internships and job openings as statisticians.

“Students get to interact with professionals in a natural way in this setting,” said Annie Flippo, a data scientist from Edmunds.com. “I’m on a secret mission to find a new intern.”

Some teams said they were friends from classes, clubs or groups from class projects who decided they wanted to analyze data for the weekend.

Most students came from math or other science-based majors, and experience varied greatly between participants, Zanontian said.

One group came with a desktop computer and matching T-shirts, but others treated the event as more of a bootcamp to learn the programs very rapidly.

“I know in the next 10 years we’re entering a new age of big data,” Jia said. “And now I can have the know-how and skills to gain insight from that data.”

Share on FacebookTweet about this on TwitterEmail this to someoneShare on Google+Share on Reddit

Comments are supposed to create a forum for thoughtful, respectful community discussion. Please be nice. View our full comments policy here.