It depends: Partisan evaluation of conditional probability importance.

Area of Science:

• Cognitive Psychology
• Political Science
• Behavioral Economics

Background:

• Policies restricting rare events often impact broader categories, like immigration.
• Evaluating such policies necessitates understanding conditional probabilities, which can be complex.
• Misinterpreting probabilities can lead to flawed policy judgments and societal polarization.

Purpose of the Study:

• To investigate how partisan evaluation of conditional probabilities influences judgments about restrictive policies.
• To examine whether political affiliation affects the perceived importance of different statistical probabilities.
• To explore the role of numeracy in motivated reasoning regarding policy-relevant statistics.

Main Methods:

• Two studies were conducted involving participants evaluating probabilities related to terrorism and immigration.
• Participants' political leanings and support for specific restrictive policies were assessed.
• The perceived importance of 'hit rate' probabilities versus inverse conditional probabilities was measured.
• Interventions to adopt an expert's perspective and numeracy assessments were included.

Main Results:

• Partisans prioritized probabilities supporting their policy stances, showing motivated reasoning.
• Supporters of restrictive policies valued 'hit rates' more, while opponents valued inverse probabilities.
• These partisan differences occurred across policies favored by both Republicans and Democrats.
• Adopting an expert perspective reduced but did not eliminate these differences.
• Higher numeracy correlated with larger partisan differences in Study 2, suggesting motivated reasoning.

Conclusions:

• Political partisanship significantly biases the evaluation of statistical evidence, impacting policy judgment.
• Understanding conditional probabilities is crucial for rational policy assessment, but is subject to motivated reasoning.
• These findings highlight challenges in evidence-based policymaking and contribute to understanding political polarization.