A growing set of measures aims to place incumbent and nonincumbent congressional candidates on a single ideological scale, offering the promise of new tests of accountability and representation. This paper evaluates how well six recent candidate-orientation measures actually capture the same underlying concept.

🔎 What Was Compared

• Six recently developed measures that estimate candidates' political orientation for both incumbents and challengers on a common scale.
• Each measure is built from different choices, incentives, and contexts, and is often treated as a proxy for candidate ideology.

📊 How the Measures Were Evaluated

• Examined the properties and relationships among the six measures across party lines.
• Assessed the measures' ability to distinguish moderate versus extreme roll-call voting records within each party.

📌 Key Findings

• The measures show only a weak relationship with one another within parties, indicating low concordance across approaches.
• Evidence suggests the measures largely capture domain-specific factors rather than a single, shared ideological signal.
• The measures perform poorly at separating moderate from extreme roll-call voting records within parties, limiting their usefulness for classifying legislators by extremity.
• Because of these weaknesses, the measures are limited as tools for empirical tests of accountability and representation in Congress.

💡 Why This Matters

• Treating these candidate-orientation measures as synonymous with ideology risks misleading inferences about accountability and representation.
• Future research should exploit the conceptual and empirical variation across measures rather than assuming they measure the same thing, and should be cautious when using these scores to draw conclusions about legislative behavior or electoral accountability.