Sport/Media effects

Beth Adubato asks, “Does a highly identified sports fan feel a strong bond while watching his favorite football players and then exhibit violent, copycat behavior?” Using the media, copycat framework, this research looked at five categories of domestic violence arrests in the city of Philadelphia on Eagles’ “gamedays,” for an 8-hr period, beginning with kick-off time. These relationships were tested using comparison of means tests. The mean average of domestic violence arrests on football was statistically significantly different from both comparison Sundays and other sports “gamedays.” As predicted, there was no statistically significant difference between home and away games, removing the possible bias that fans were at the game and then became violent.

She does this in an article in the Journal of Sport and Social Issues February 2016 vol. 40 no. 1 22-37. This came too late for inclusion in my Sports Criminology book to be published in summer by Policy. However, I do address the issue and this is my take on the issue.

Super Bowl effect

What of the Super Bowl effect in the States or the ‘Old Firm’ derby effect in Scotland? It is alleged and frequently repeated by the media that domestic violence rises on the occasion of these matches. The Super Bowl is played annually between the winners of the American and National Football Conferences. The ‘Old Firm’ derby is played between Glasgow Rangers and Celtic; the two most successful Scottish football teams who have played each other about 400 times. When Rangers were liquidated (for tax offences) at the end of the 2011/12 season and demoted to the fourth tier of the league another 3 years was to pass before they met again in a Cup Match.

Crowley et al (2014) were commissioned by the Scottish Government to explore the correlation between ‘certain matches football and domestic abuse’. They found that research based upon incidents reported to the police found that relative to various comparators, there was an increase in recorded domestic abuse incidents on the day the fixtures were played between 13-138.8%, depending on a number of variables: the day of the week the match took place; the comparator day/event; and the salience/outcome of a match. Such nuance and recognition of confounding variables is rare. They are critical of Williams et al (2013) for failing to take into account the comparators they use; for instance Scotland International Matches, take place during the week not the weekend and there are issues about the time period covered around the match. Though Williams et al (2013) only claim their analysis to be preliminary they confirm the association between Old Firm matches and reports of domestic violence. The natural experiment of 3 years gap in such derbies may spark further investigation.

Both Crowley et al (2014) and Williams et al (2013) cite Gantz et al (2009) who examined the Super Bowl using police recorded domestic violence incidents taken from 15 cities with NFL teams over a six year period. The analysis suggested that domestic violence in a city increased both when the local team played during the season, and during the Superbowl (whether or not there was local allegiance to the competing teams). The Superbowl was seen to result in an average increase of 244 domestic violence incidents per city which represented 6.5% of incidents that day. Factors to be considered are that the Superbowl is the last game of the season and a public holiday which introduces factors, such as people being cooped up together, it being winter and lots of alcohol being consumed.

As with claims of trafficking their is a danger of media hype and use by campaigning groups of media opportunities around high profile events. Thus Snopes investigative website lists the claimed rise of domestic violence on Super Bowl Sunday as ‘false’. Following Gantz et al (2009) and others (uncited) they found Christmas to the busiest time for domestic violence shelters. This raises the issue of whether we should ban christmas because of this.

Snopes also mention the work of Kirby et al (2014) which does not appear to assist them as they claim, ‘Every time England loses the World Cup, domestic violence against women raises 38%.’ But Crowley et al (2014) argue much of this effect might be attributed to other factors, particularly alcohol. Kirby et al (2014) note the deficiencies of their own study too and conclude where and when the violence occurred, the precipitating factors prior to it occurring; the level of violence; consumption of drugs or alcohol; and a comparison between the expected and final result. This latter issue is taken up by Card and Dahl (2011) who, as economists, discuss ‘family violence’ in terms of ‘intra-family incentives’! They tested this in respect of reports to police on Sundays associated with American Football. Controlling for variables like, weather, time and ‘pre-game point spread’ etc found that if the home team was expected to win by more than 3 points but lost there was an 8% increase in male-on-female violence. This is taken to show the significance of non-instrumental factors.

In the preface to my I acknowledge that I am a sports fan but that some of my opinions on drugs and gender would not find favour with all. She is also a fan and a media professional and her husband commentates ice hockey.

Her argument is not about sport so much as media coverage of it. She picks up on some of the same issues I do and notes the media criticisms and misuse of earlier work. So I am grateful now to know of Janet E. Katz and Garland F. White’s ‘Engaging the Media: A Case Study of the Politics of Crime and the Media Social’ on the reception of their work of them with Kathryn E. Scarborough on ‘The Impact of Professional Football Games Upon Violent Assaults on Women’. I think we agree that it is not sport’s ‘fault’ - though sports authorities not blameless - and that masculinities may have something. Think we might part company on the ’media effects’.  I incline to Gauntlett on these matters.