Original Source:
http://www.ssc.wisc.edu/~bhansen/vote/florida.pdfI recreated a lot of the work on Stata on my own using the 2000 Census and Atlas vote total.
Analyzed Variables- The natural logarithm of the Buchanan Vote (P). In the form of Log(P).
- A linear function of the percentage of people aged 65 or older (A65).
- The percentage of Blacks (B)
- The percentage of Hispanics (H)
- The percentage of College-educated (C)
- Median-family income in dollars (I)
Regression Functionlog(P) = 2.146 -0.0415*A
65-0.0132*B-0.0349*H-0.0193*C-0.0658*I
Summary Statisticsn=66 (for all counties in Florida)
R
2=0.73
σ = .334
Interpretation- Increasing the population over 65 years of age by 1% decreases Buchanan's vote by 4.1%
- Increasing the Black population by 1% decreases Buchanan's vote by 1.3%
- Increasing the Hispanic population by 1% decreases Buchanan's vote by 3.5%
- Increasing the college-educated population by 1% decreases Buchanan's vote by 1.9%
- A $1000 increase in median household income decreases Buchanan's vote by 6.6%
ConclusionThe results certainly suggest that Buchanan did far better in Palm Beach than what he would have normally expected to do. Or it could be the result of great staff placement in Palm Beach itself.
The butterfly ballot itself definitely had the potential to be a bigger spoiler than Nader. Considering Palm Beach County has a high proportion of elderly, college-educated and high income people.
In general it also is likely true that Buchanan hurt Bush amongst downscale but conservative White voters.