Testing temporal frames of colorectal cancer risks: evidence from randomized online experiments

Abstract

Low perceived risk of is associated with a low probability of engaging in risk preventive behaviours. Temporal framing has been suggested to make the risk seem higher, as narrower time frames appear to increase perceived risk and intentions to engage in risk-preventive behaviours. In two online experiments, we tested how manipulating risk information about colorectal cancer (CRC) influenced risk perception and preventive behaviours. 902 men and women aged 45–54 were recruited from an English online panel. Study 1 (N = 132) investigated how communicating CRC mortality instead of incidence rates influences risk perception, using both yearly and daily frames. In study 2 (N = 770), CRC incidence was described as occurring yearly, daily or hourly. The primary outcome measures were risk perception, intention to engage in preventive behaviours and whether participants chose to read information about ways to reduce CRC risk or skip it to finish the survey. Study 1 did not find a difference in intentions based on information about the risk of developing vs the risk of dying of CRC. Study 2 found that, while communicating CRC incidence occurring as every day increased information-seeking behaviour compared to every year and every hour (57.4% vs 45.9% vs 45.6%, p = 0.012), individuals in the daily condition were less likely to get all comprehension questions right (69.9% vs 78.2% vs 85.0%, respectively, p = 0.015). Temporal framing had no impact on perceived risk or intentions to engage in risk-reducing behaviours. While manipulating risk information in terms of temporal framing did not increase perceived risk, it may influence people’s decision to engage with the information in the first place.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.