The 1-in-X effect in perceptions of risk likelihood differences

Abstract

In risk communication, 1-in-X and N-in-10(p) ratios are used frequently to present likelihood information. This research examines the effect of numeric formats (1-in-X vs. N-in-10(p)) on perceived difference in multiple risk likelihoods through six experiments (N = 2,254). We find that the perceived difference in risk likelihoods is greater when the information is presented as the 1-in-X ratio than the N-in-10(p) ratio in the context of risk in-crease and reduction, which in turn also affects risk-related downstream behavioral consequences such as behavioral intentions, willingness to pay, and donation target selection. Furthermore, the numeric format effect is greater as risk levels become lower. The reason is because the two formats differ in terms of which fraction element is fixed (i.e., a fixed numerator in the 1-in-X format and a fixed denominator in the N-in-10(p) format) and how the varying fraction element changes with probability levels. Study results imply that the numeric format effect is because of the overestimation of the difference of the risk likelihoods presented in the 1-in-X format. The findings have important theoretical and practical implications for risk communication.