Millions of Americans, including President Obama, fill out their "brackets" when the NCAA Tournament field is announced each March, but does that really affect their work? It certainly appears to, at least among a segment of the population who use research libraries, a Duke University professor has found.
Charles Clotfelter, Z. Smith Reynolds Professor of Public Policy at Duke's Sanford School of Public Policy, used data from 78 research libraries in the U.S. to determine the number of articles viewed from February through April in 2006, 2007 and 2008. The number of articles viewed on Monday through Wednesday of those weeks averaged more than 1,000 a day per library.
Clotfelter found that the number of articles viewed through the JSTOR digital repository of academic journals increased an average rate of 5 percent a week in the weeks leading up to "Selection Sunday," but fell 6 percent in the week right after the NCAA field was announced. The following week, library usage resumed its increase, at a rate of 3 percent a week.
"I observed similar patterns in each of the three years, 2006 to 2008, and the post-selection dip occurred both in libraries not connected to universities with Division I teams as well as those with them," said Clotfelter, who is currently writing a book about the role of big-time athletics in American universities. "This drop in research activity in these libraries is quantitative evidence of the NCAA tournament's power to influence patterns of work."
Clotfelter also measured the impact of library usage at universities whose teams won "toss-up" games, those with no clear favorite, and compared that to schools that lost such games. He found that at winning schools, daily article viewing fell by 10 percent in the seven days after the game and even more, 14 percent, through the end of the tournament, compared to schools whose teams lost these games.
"By all appearances, fans of losing teams shook off the disappointment and returned to work in greater numbers or with greater diligence, while the fans of winning teams continued to follow their team into the next round," he said. "Such an effect would imply an ironic sort of ‘winner's curse,' where students and researchers at universities whose teams win unexpectedly do less work than those whose favorite teams are also-rans."
In 2008 an estimated 37 million workers participated in "March Madness" office pools, and CBS, which owned the rights to broadcast the tournament on television and other media, reported that 4.8 million people used video and audio streaming on their computers to follow games, Clotfelter noted. "Examining the amount of work done in research libraries before, during and after the tournament can provide a window for assessing whether such a media event effect is real or imagined," he said.