Partisanship, Chauvinism, and Reverse Racial Dynamics in the 2003 Louisiana Gubernatorial Election

Abstract

Skinner and Klinkner (2004) argue that attitudes reflecting racial prejudice were present in the 2003 Louisiana governor’s race, and were even stronger in north Louisiana. Utilizing as did they aggregate data with an approach that better fits theory and available data, this study shows that they overstate the significance and importance of presumed racial prejudice in the election, especially statewide. Across the state, attitudes reflecting racial prejudice had no impact on the vote decision, and where they did in north Louisiana, the effects were small enough that they did not change the outcome of the contest. However, using individual-level data showed that partisan effects strongly controlled voting in this contest; in fact, non-Republicans displayed a chauvinistic tendency in their voting while a Republican partisanship negated this effect. Conflating various meanings to variables and inferior indicators explains the less-valid results and interpretations achieved by the use of aggregate data compared to the individual-level data.

Beyond the Politics of Race? Explaining the Vote in Recent Louisiana Gubernatorial Elections

“…Skinner and Klinkner (2004) use the vote received by David Duke in the 1991 gubernatorial election as a proxy for such racial attitudes at the parish level. However, Sadow (1996, 2005) correctly notes that the parish‐level vote for Duke likely measures phenomena other than just racial attitudes. Of greatest consequence is that Duke's 1991 vote also measures partisanship, something that seems particularly plausible given the loathing by many Republicans for Duke's opponent Edwin Edwards.…”

“…Indeed, there is a fairly strong correlation between the vote for Duke in 1991 and the Republican presidential vote in 1988 for George Bush (r = .70). Sadow (2005) also argues that Duke's vote in the 1996 U.S. Senate primary may be a more valid indicator of racial backlash than his support in 1991. However, in their response to Sadow, Skinner and Klinkner (2004) point out that by 1996 Duke had been reduced to the status of a fringe candidate, and hence his 1991 vote is the more valid indicator of racial‐backlash voting.…”

“…These findings were challenged by Sadow (2005), who questioned using Duke's 1991 vote as a surrogate for racial backlash, and more generally claimed that Skinner and Klinkner (2004) were guilty of relying exclusively on aggregate‐level data. Using individual‐level data, Sadow claims to find no evidence of racial backlash voting.…”

What Goes Around, Comes Around

“…Louisiana's 2003 gubernatorial race was previously examined by both Skinner and Klinkner (2004) and Sadow (2005). Skinner and Klinkner compared that race to the pattern of votes received by David Duke in the 1991 gubernatorial election and found a significant relationship between the number of votes received by Duke and Blanco, concluding that Louisiana's racial divisions led many white Republican voters to give their support to a white Democrat, rather than a nonwhite Republican.…”