Varying risk comparison elements: effects on public reactions

Our plain-English summary of Varying risk comparison elements: effects on public reactions, by B.B. Johnson (Risk Analysis, 2004).

The study

• Author: B.B. Johnson
• Journal: Risk Analysis
• Published: 2004;24(1):103-14
• Cited by: 9 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Expanding a limited empirical base on effects of risk comparisons, a pilot experiment explored how varying elements of such comparisons might influence public response. The scenario used was a hypothetical trial of an asbestos-installing firm for putting students and staff at a junior high school at "unreasonable risk," first used by Slovic et al. (1990). Study participants played the role of jurors in the trial, asked to rate the risk, the firm's guilt, whether it should be made liable for future health effects, whether the school should be occupied in its current condition, and whether the asbestos should be removed at a cost of 3 million dollars. Opportunity samples of New Jersey residents (n= 309) received information intended to vary four comparison attributes: number of dimensions of comparison; single versus multiple ways of expressing mortality; a narrative to explain and justify the risk comparison; and the hypothetical role of the participant (juror vs. parent). The dependent variable was a hazard scale constructed from four of the five postcomparison judgments. ANOVA found format variations swamped in their effect by concern about asbestos, with the absence of a narrative and the parental role being the only attributes that increased negative risk reactions. Multiple regression analyses found that multidimensionality, narrative, role, and the presence of any risk comparison at all had significant effects when these (plus death formats) were the only independent variables. However, only multidimensionality (which increased negative reactions) retained significance when asbestos concern, risk beliefs (no safe level of exposure to a carcinogen; any exposure leads to cancer), and demographic variables were added to the analysis. Concern and risk beliefs alone explained 33% of variance in hazard scores; adding demographics and risk comparison variables only raised explained variance to 36% (having children at home and being exposed to multidimensional risk comparisons raised scores; age and income lowered them). The results underline the potentially small effect of risk comparison information on risk views (at least between subjects), but offer some insight into aspects of message design left unexplored in empirical literature to date.

In plain English

This pilot experiment studied reactions to different ways of comparing risks, not people exposed to asbestos. The author gave 309 adults, recruited through civic groups mostly in central New Jersey, a hypothetical case of an asbestos-installing firm accused of putting junior high school students and staff at "unreasonable risk." In most versions a school board sued the firm and participants acted as jurors; in the parental version, parents made the claim. Students attending for three years were said to have an increased lifetime risk of death of 0.23 per million. The seven versions also varied the risk comparison table and whether a defense expert's explanation was included. Leaving out the explanation, and judging as parents, raised hazard scores in simpler analyses. Neither effect stayed statistically significant once concern, risk beliefs and demographics were included; only a comparison covering several dimensions did. Concern about asbestos and beliefs about cancer-causing agents explained 33% of the variation in scores; adding other variables raised this only to 36%. The author concluded that the effects of risk comparisons are "at least contingent, if not weak."

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Pilot experiment (questionnaire study with seven randomly sorted versions); participants role-played jurors or parents in a hypothetical asbestos scenario
• Who was studied: 309 adults recruited as opportunity samples through 23 civic groups, mostly in central New Jersey; 62% female, aged 17 to 93 (mean 53); 43% had children living at home
• Exposure: No real exposure was studied. Participants read a hypothetical scenario about properly installed asbestos in a junior high school. It stated an increased lifetime risk of death of 0.23 per million for three years' attendance, based on an average air concentration of 0.001 fibers per cubic centimeter. In most versions a school board sued the installer and participants acted as jurors; in the parental version some parents made the claim instead. The versions varied the risk comparison: none; the original comparison with unrelated hazards; one or several measures of school-hazard deaths; a multi-attribute table; or no defense expert narrative. No fiber type was named.
• Main results:
  • Without the defense expert's narrative, mean hazard scores were higher (4.01, n=44) than with it (3.03, n=42) in a t-test (p less than 0.01). This effect did not remain significant in the full regression.
  • Participants asked to judge as parents had higher mean hazard scores (3.88, n=42) than those judging as jurors (3.03, n=41), p less than 0.05. This effect did not remain significant in the full regression.
  • In analyses of variance, concern about asbestos had main effects in five of six contrasts. The risk comparison manipulations were significant at p less than 0.05 in only three.
  • When concern was taken into account, receiving any risk comparison versus none had no main effect on hazard scores
  • Presenting deaths in one versus several ways produced no significant differences in the analysis of variance or regression analyses
  • Concern about asbestos and two risk beliefs alone explained 33% of the variance in hazard scores. The two beliefs were that there is no safe level of exposure to a cancer-causing agent, and that a person exposed will probably get cancer.
  • Adding demographics and risk comparison variables raised the explained variance only to 36%. In this full model, having children at home and the multi-attribute comparison raised scores. Income lowered scores; for age, the table shows only p less than 0.10.
  • When risk beliefs and demographics were controlled as well as concern, only multidimensionality remained statistically significant among the risk comparison variables, and it increased negative reactions
• The authors' conclusion: The author concluded that the impacts of risk comparison information on risk responses are at least contingent, if not weak, and that much remains to be learned about the contingencies.
• Limitations the authors state:
  • This was a pilot study with a limited sample size
  • The no-comparison group was small, so the results on having any comparison at all might overgeneralize
  • Questions about asbestos concern were asked after the scenario, because the budget did not allow collecting them in advance
  • Having children at home was used in place of parental status, which was not assessed directly
  • All hazards in the multi-attribute table were identical on control, which may have removed information salient to readers

Topics

Medical terms in this study, explained in our Lexicon: No Safe Level of Exposure.

Related studies

• Apples to apples: the origin and magnitude of differences in asbestos cancer risk estimates derived using varying protocols — D.W. Berman (2011)
• An integrative histopathologic clustering model based on immuno-matrix elements to predict the risk of death in malignant mesothelioma — M.L. Balancin, W.R. Teodoro, et al. (2020)
• Stability and inoculation of risk comparisons' effects under conflict: replicating and extending the "asbestos jury" study by Slovic et al — B.B. Johnson (2002)
• Mineral particles of varying composition induce differential chemokine release from epithelial lung cells: importance of physico-chemical characteristics — J. Ovrevik, T. Myran, et al. (2005)
• Estimating the potential disease rate reduction under varying conditions of combined effects — H. Checkoway, J.L. Hickey (1988)
• Influence of Varying Concentrations of Epoxy, Rice Husk, Al2O3, and Fe2O3 on the Properties of Brake Friction Materials Prepared Using Hand Layup Method — A.P. Irawan, D.F. Fitriyana, et al. (2023)

Browse more research in Risk Analysis, or return to medical research.