WHO GETS INVITED TO THE NCAA TOURNAMENT? MERIT ISN'T THE WHOLE STORY, A NEW STUDY FINDS
April 1, 2005
For more information, contact: Benjamin Haimowitz, HHaimowitz@aol.com
"History does not play any role in our consideration. We take each year on its own merits. You play yourself in or play yourself out based on this year alone." ---Chairman, 2004 NCAA Selection Committee, quoted in the Chicago Tribune
Well, not quite. With the NCAA tournament having just concluded, a new study takes a close look at the tourney selection process and finds somewhat more to be involved than merit in determining who is invited.
As one would expect, the study, in the current issue of the Academy of Management Journal, finds that teams' performance during the current season predominantly determines whether they receive bids to the nation's leading post-season invitational.
But something else is involved too -- what the report's authors call "status." They find that a team's status or lack of it has a significant effect on whether or not the team receives an NCAA tournament bid, even though status-influencing factors have little or nothing to do with current-season performance.
According to the study's authors, Marvin Washington of Texas Tech University and Edward J. Zajac of Northwestern, the inclusion of status factors in their analysis of the selection process enhanced their ability to predict invitations by more than 15 percent.
The issue is important to management scholars because of the light it casts on the role status plays in competitive situations every day in business or commerce, where it may confer competitive advantages on a product or company that have little to do with its actual quality. "For example," write Profs.
Washington and Zajac, "the Jaguar automobile has long suffered from a reputation of poor perceived quality, while at the same time benefiting from the privilege of being a car with high social status."
What makes the results of the NCAA study "particularly compelling,"
according to the authors, "is our ability to control for dominant and unambiguous performance measures, such as a team's won/loss record and the won/loss record of all its opponents." Controlling for such measures, they continue, enabled them to highlight "the complementary, significant, quantifiably measurable, and yet subtle importance of status differences in the competitive selective process."
What status elements come into play in the NCAA selection process? One factor is whether a team received an at-large invitation during the previous season
-- that is, an invitation extended to a team that did not win its conference but is considered worthy of competing for the national championship. Having received a bid in the prior year, the research finds, more than doubles a team's chance of receiving a bid, regardless of how it fared in the previous year's tournament.
A second status factor is the number of games played against other invited teams, quite apart from a team's record in those games.
A third factor is the number of games played against teams outside Division I, the top competitive division in college athletics. Since Division I teams are prohibited from playing more than four contests outside the division, the effect of such games on a team's record is necessarily small. Yet, such is the loss of status in those games, that every one reduces a team's chance of receiving an NCAA bid by close to one third.
In other words, if, in the course of a season (which typically consists of about 25 games), a team plays just one game outside of Division I, it reduces its chance for an at-large invitation by nearly one third. If it plays a second game, its chance for a bid fades to about half what it would have been had it played none at all.
Such is the power of status.
The study's findings are based on an analysis of the bid process over seven years, in which about 270 Division I colleges were considered each year for some 32 at-large tournament berths. (About half the colleges in the 64-team competition receive automatic invitations by winning conference championships, but only the optional at-large berths were analyzed in this
study.) A committee of athletic directors and educators makes the at-large selections in early March at a closed-door meeting that lasts about three days and is governed by two
principles: 1) select the best available teams to fill the at-large berths regardless of conference affiliation; 2) no limit on the number of teams the committee may select from one conference.
In their analysis, Washington and Zajac started out with data that have specifically to do with performance -- a team's win percentage, the win percentage of its opponents, its win percentage over the prior four years, and its membership in a conference where the champion receives an automatic bid. Then, with these performance factors as controls, the authors analyzed the effect of four factors that have to do with status -- 1) the average number of times a college had received an at-large invitation in the prior four years; 2) whether or not it received such an invitation in the previous year; 3) the number of games it had during the current season with other teams playing in the tournament; and 4) the number of games played against teams outside Division I.
All four of the status factors turned out to be statistically significant, with the second and fourth having the strongest effects.
Given the clear influence of these status factors, what accounts for the selection-committee chairman's insistence that its choices are based solely on merit? The authors surmise that the same factors are at work here that come into play every day when people pay for status irrespective of performance.
"We are not suggesting that this effect is consciously motivated by the selection committee," Washington and Zajac write. "Indeed, our view is that status can exert its influence in competitive situations in ways that are largely non-conscious."
In this way, status, they conclude, should be distinguished from reputation, with which it is often lumped. Reputation, they write, "is fundamentally an economic phenomenon that deals with differences in perceived or actual quality or merit that generates earned, performance-based rewards...Status differences, on the other hand, reflect a fundamentally social phenomenon that can be related to -- and exists independently of -- the product/service quality differences...of primary interest to economists."
Related to status, they add, is its poor cousin discrimination. "Both come from social beliefs that have little to do with assessments of quality, and much more to do with social or cultural categorization schemas or stereotypes."
The Academy of Management Journal, a peer-reviewed publication now in its 48th year, is published every other month by the academy, which, with about 15,000 members in 90 countries, is the largest organization in the world devoted to management research and teaching. The academy's other publications are the Academy of Management Review, the Academy of Management Executive, and Academy of Management Learning and Education.