ABSTRACT
I engage debates about racial media bias by analysing newspaper coverage of professional tennis players in France and the United States. Tennis is an elite sport that typically does not have many non-white players and may be especially sensitive to racial boundaries. Tennis also offers a new solution to the methodological challenge of establishing that any difference in newspaper coverage across racial groups is due to bias and not actual differences across the groups. I use the professional tennis ranking system, which assigns an objective marker of how good a player is (and therefore the media coverage that s/he should receive) at any point in time. I explore two types of bias (the amount and tone of media coverage) and uncover no systematic racial differences in the relationship between ranking and media coverage. My findings have several implications for our understanding of racial boundaries and the media.
Acknowledgement
A previous version was presented at the Council for European Studies Conference of Europeanists. The author would like to thank Mindy Foster for valuable research assistance and Christopher Bail, Frank Baumgartner, Erik Bleich, Irene Bloemraad, John Brigham, Els de Graauw, Paul Statham, Matthew Wright and anonymous reviewers for helpful comments on earlier versions.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. One prominent issue is media bias in the coverage of black American tennis players Serena and Venus Williams, although they are beyond the scope of this article because there are no highly ranked female French tennis players to serve as a comparison.
2. There are differences in the composition of racial categories across the two countries. Racial diversity in the US has historically been organized around a white–black divide whereas in France the non-white ‘other’ has been Jewish people, African-origin blacks and North African-origin Arabs at various historical times. For my purposes, these differences are less important than the fact that in both countries there is a hierarchical distinction between whites and non-whites.
3. Another potentially relevant difference between France and the US is national media systems. For more see the supplementary appendix.
4. See the supplementary appendix for more on other newspaper options.
5. Different types of tennis tournaments offer different numbers of ranking points. The most prestigious ‘Grand Slam’ events (Australian Open, French Open, Wimbledon and U.S. Open) offer the most (2000 points to the winner, 1200 points to the runner up and a sliding scale that declines in points-awarded for each round). Regardless of how many points are available, players who win more matches have more points and are more highly ranked. For more details see: http://www.atpworldtour.com/Rankings/Rankings-FAQ.aspx
6. As seen below, there are relatively few non-white players in each country, so it was important to get the full dataset for their careers.
7. Each search of newspaper articles was for ‘First Name + Last Name + Tennis’ to avoid articles about other people with the same name as the tennis players. Comparing these results with those from searches using only ‘First Name + Last Name’ did not lose any articles about tennis players. Once obtaining the set of articles in which a player is mentioned, the computer makes three calculations. One is the number of mentions of ‘First Name’, a second is for ‘Last Name’ and the third is for ‘First Name + Last Name’. Each calculation was provided with a 95% confidence interval, as was the overall calculation of how many times the player’s name was mentioned across each of the three measures for each of the weeks.
8. Historical ranking data for each player were obtained from: http://www.atpworldtour.com/Rankings/Rankings-Home.aspx.
9. Appendix , and 3 provide detailed data on weekly name mentions for each player in each newspaper.
10. The results are also consistent when estimated with negative binomial and poisson regression models.
11. Another exception comes from articles about James Blake and Donald Young where there are extensive references to family members and at times the players are referred to as ‘James’ or ‘Donald’ to distinguish them from family members. There are also occasional references to Jo-Wilfried Tsonga as ‘Jo’. This likely reflects his superstar status as other top (white) players are also referred to by nicknames (e.g. Novak Djokovic is ‘Djoker’, Roger Federer is ‘Fed’ and Rafael Nadal is ‘Rafa’) and nicknames are generally reserved for prominent athletes.
12. Words are considered in association with a player if they occur within ten words of any part of a player’s name. This excludes stopwords.
13. Although Blake is known for his powerful forehand shot, he is mainly described in French and American media with words like grace, elegance and intelligence, which celebrates his mental attributes and counters typical racialized tropes for non-white athletes. Ram and Young have not been successful enough to receive detailed coverage of their personalities or playing styles.