A Reassessment of `The Methods behind the Madness: Presidential Electoral College Strategies, 1988–1996’
This article reassesses Shaw (1999) and finds that the original study's substantive conclusions about presidential campaign strategy are unsupported when the correct statistical methods are applied.
Why It Matters
The publication begins with a motivating question: Do the findings of Shaw (1999) regarding the systematic effects of campaign strategy on presidential electoral college strategies hold when the correct statistical methods are applied?
Its central contribution is to show that this article reassesses Shaw (1999) and finds that the original study's substantive conclusions about presidential campaign strategy are unsupported when the correct statistical methods are applied.
It matters because the findings connect institutional choices to the way authority, public responsibility, and political behavior are experienced in practice.
Key Findings
- The analyses in Shaw (1999) were actually least-squares regressions, not ordered probit or 2SLS as claimed.
- When the correct ordered probit and 2SLS methods are applied, the substantive findings in Shaw (1999) disappear.
- Electoral college strategy is not shown to be a strong function of competitiveness, electoral votes, or TV Ad Cost using appropriate methods.
- Errors in new data used in Shaw (2003) account for differences in findings.
Research Design
- Design
- Article
- Data
- Data from Shaw (1999); Data from CQ’s Guide to U.S. Elections 4th Edition, Volume 1 (Moore, Preimesberger, and Tarr 2001)
- Geography
- United States
- Time Period
- 1988–1996
- Unit of Analysis
- state-year
- Methods
- Replication of published results using least-squares regression.; Implementation of ordered probit analysis.; Attempted implementation of two-stage least squares (2SLS) analysis.
Citation
Journal of Politics 66 (2): 616-620.
- Venue
- The Journal of Politics
- Volume
- 66
- Issue
- 2
- Pages
- 616–620