Abstract

Numerous health behavior models have emphasized absolute risk perceptions as prominent predictors of future behavior and intentions, but more recent research has shown that people also attend to comparative risk information. While most research highlights external (social) comparisons as the primary way people contextualize risk, it is also possible that people use internal comparisons, such as comparing their current risk to their past risk (temporal comparisons) or comparing their risk for one health threat to their risk for another health threat (dimensional comparisons).

Objective: The current research sought to examine differences in absolute, external, and internal comparative risk perceptions across a variety of health threats.

Design: This study utilized a cross-sectional design wherein participants completed all study materials online. Main outcome measures: MTurk workers (N = 654) responded to questions about absolute and comparative risk perceptions, concern, and precautionary intentions.

Results: Hierarchical regression analyses revealed that although absolute risk perceptions were positively associated with concern and precautionary intentions for all health threats, comparative risk perceptions also accounted for unique variance in all models. Internal comparisons were uniquely associated with concern and intentions even when accounting for absolute and external comparative risk perceptions in nearly all models.

Conclusion: These findings provide the first systematic evidence for the utility of incorporating internal comparisons into models of risk perception.

摘要

许多健康行为模型都强调绝对风险感知是未来行为和意图的重要预测因素，但最近的研究表明，人们也会关注比较风险信息。虽然大多数研究强调外部（社会）比较是人们将风险背景化的主要方式，但人们也可能使用内部比较，例如将他们当前的风险与过去的风险进行比较（时间比较）或将他们对一种健康威胁的风险与他们面临另一种健康威胁的风险（维度比较）。

目标：当前的研究旨在检查对各种健康威胁的绝对、外部和内部比较风险认知的差异。

设计：本研究采用横断面设计，参与者在线完成所有学习材料。主要成果措施： MTurk 工人 ( N  = 654) 回答了关于绝对和比较风险认知、关注和预防意图的问题。

结果：分层回归分析表明，虽然绝对风险认知与对所有健康威胁的关注和预防意图呈正相关，但比较风险认知也解释了所有模型中的独特差异。即使在几乎所有模型中考虑绝对和外部比较风险感知时，内部比较也与关注和意图有着独特的联系。

结论：这些发现为将内部比较纳入风险感知模型的效用提供了第一个系统证据。