Last modified: Friday, May 23, 2003
IUB researchers examine the relationship between football and domestic violence
EDITORS: This release is embargoed until 2 p.m. EST on Sunday (May 25). The presentation will be given at 3:45 p.m. PST at the International Communication Association conference at the San Diego Marriott Hotel, 333 West Harbor Drive. Gantz will be available for interviews in his office on Tuesday (May 27) and can be reached at 812-855-1621 or firstname.lastname@example.org.
BLOOMINGTON, Ind. -- Research conducted at Indiana University Bloomington being presented Sunday (May 25) at an international conference finds that the relationship between televised professional football and domestic violence may be less prevalent than previous reports would suggest.
Not all games matter in the final standings, and neither do all football games elicit arousal of fans' emotions leading to domestic violence. However, the IUB researchers found that expectation about a game, rather than its outcome, was an even greater contributing factor in increased reports of domestic violence in 14 National Football League cities highlighted in the study.
Previous studies have raised concerns about a pattern of violent behavior against women around the time of the Super Bowl. But the preliminary study, presented at the International Communication Association conference in San Diego, found that the number of cases that day was relatively small compared to those reported on holidays such as Christmas or Memorial Day.
The study, "Televised NFL Games and Domestic Violence: A 14 City Study," was prepared by Walter Gantz, chairman and professor of telecommunications at IUB, and two doctoral students, Sam Bradley and Zheng Wang.
"Not all football games are alike. The nature of the game, the time of year, the consequences of the outcome, the expectations going in; these factors and others make the relationship between NFL football games and domestic violence quite complex," said Gantz, who has looked at the relationship between television and family life for more than a decade.
The researchers contacted police departments in every city with an NFL franchise between 1996 and 2001. They were able to obtain data from 14 cities. The study is based on 24,102 days of domestic violence data from these cities, including days when 1,108 games were played.
First, they sought to examine the effect of a game itself after controlling for month, weekday, year and holiday status. "There was only a marginally significant effect of having an NFL game on that day's domestic violence dispatches," the authors wrote. "The results suggest that the presence of an NFL game does slightly increase the number of domestic violence reports."
They next looked at other factors, such as expectations about whether teams would win or lose. They found that the number of domestic violence cases was inversely proportional to the point-spread. "That is, the more the team was expected to lose, the greater the number of domestic violence dispatches on game day," the authors wrote.
"As a team was expected to win, domestic violence decreased," Bradley said. "The interesting thing that we found is that actual outcome of the game did not, in terms of domestic violence reports, show any significant relationship with what actually happened."
Games that were expected to be close were thought to lead to greater fan anxiety and arousal of emotions leading to domestic violence. Gantz, Bradley and Wang found no such effect on game day. However, they noticed a pattern of domestic violence on the third day after the game. This finding is similar to a 1983 study, which discovered an increase in homicides on the third day after a heavyweight prize fight.
While they can only speculate, the researchers observed that unlike baseball, basketball and hockey fans, football fans must live with the results of a bad game for a week. "Combined with other mid-week frustration, maximal distance from the weekend, simmering resentments and for some, lost bets, this frustration may need several days after the game to finally boil over," they wrote.
Lastly, Gantz, Bradley and Wang tested theories about whether the Super Bowl -- the most widely viewed single television program of the year and an unofficial holiday for many -- could be a predictor for domestic violence. While they did find there to be an increase in domestic violence cases, it was small when compared to the number reported on holidays.
"The third analysis provides a link between the Super Bowl itself and domestic violence, resulting in an average of 244 additional cases of domestic violence across the 14 cities studied," they wrote. "Since every holiday tested in the study except for Valentine's Day was a significant predictor of domestic violence, the Super Bowl reported here may be less of a football effect than that normally encountered with holidays."
Gantz added, "If we think of Super Bowl Sunday as a holiday -- much as these other holidays that we've talked about -- it may be the game itself or it may be a holiday function."
While the researchers believe their results are preliminary and welcome additional data from other NFL cities, they think their findings provide a sobering warning to fans of all sports. "Sports can be exciting and be physiologically arousing, which fans need to realize when they are wrapped up in the action and after the game is over," Gantz said.
The authors contacted the police department in every city with an NFL franchise between 1996 and 2001. They were able to obtain data from 14 cities. The study is based on 24,102 days of domestic violence data from these cities, including days when 1,108 games were played. There were 1,320,235 reports of domestic violence over the years studied in those cities.
Cities providing information on domestic violence dispatches were Baltimore, Chicago, Cincinnati, Cleveland, Dallas, Denver, Green Bay, Kansas City, Miami, Oakland, Phoenix, San Diego, Seattle and Tampa Bay. Researchers plan to continue to work with authorities in 13 other NFL cities to get their data. Since the study was unfunded, the cost of records from New Orleans and Atlanta was prohibitive. The city of Indianapolis responded that it did not keep such records.
Because the Tennessee Titans, formerly known as the Houston Oilers, moved twice during the study period, no data was sought from Houston, Memphis or Nashville.
The number of domestic violence incidents for each city was converted in standardized z-scores because of variations in city populations. Variables used included the times of week and season when games were played, whether opponents were in the same division and the published point-spread.