There is a growing demand by the civil society for relevant information on the environment quality and related health risks. The state should be able to satisfy this demand and this makes the present research truly vital. It concentrates on correlating expert and non-expert opinions expressed when perceiving risk quantification. Our goal was to answer two following questions: 1) How does an average unprofessional person quantify a probability and severity when he or she hears certain verbal expressions that denominate them? 2) How can we possibly identify the assessment of health risks associated with environmental pollution factors given by the population in general or specific social groups? To find answers to these questions, we applied quantitative methods for data collection and analysis. The first stage involved collecting data on subjective correlation of a verbal probability scale with its numeric expression among people living in industrial cities. The second stage focused on testing the methodology for studying assessments of health risks associated with ambient air pollution given by the population/social groups. This methodology relied on the results obtained at the previous stage. We established that only 70 % of people actually correlated words with figures. We determined that experts tended to rate probabilities approximately by 10 % higher than “average people” did when it came down to such words as “Virtually certain” and “Very likely”. Such words as “Likely”, “Similarly likely” and “Unlikely” were also rated differently but with a smaller gap between the opinions. The study also provides a method for determining the public assessment of health associated with ambient air pollution. The research results give an opportunity to solve a practical task related to informing the population about health risks and to overcome a so-called language barrier between experts and ordinary people. For example, messages aimed for decision-makers can be adapted considering all the identified perception peculiarities.

There is a growing demand by the civil society for relevant information on the environment quality and related health risks. The state should be able to satisfy this demand and this makes the present research truly vital. It concentrates on correlating expert and non-expert opinions expressed when perceiving risk quantification. Our goal was to answer two following questions: 1) How does an average unprofessional person quantify a probability and severity when he or she hears certain verbal expressions that denominate them? 2) How can we possibly identify the assessment of health risks associated with environmental pollution factors given by the population in general or specific social groups? To find answers to these questions, we applied quantitative methods for data collection and analysis. The first stage involved collecting data on subjective correlation of a verbal probability scale with its numeric expression among people living in industrial cities. The second stage focused on testing the methodology for studying assessments of health risks associated with ambient air pollution given by the population/social groups. This methodology relied on the results obtained at the previous stage. We established that only 70 % of people actually correlated words with figures. We determined that experts tended to rate probabilities approximately by 10 % higher than “average people” did when it came down to such words as “Virtually certain” and “Very likely”. Such words as “Likely”, “Similarly likely” and “Unlikely” were also rated differently but with a smaller gap between the opinions. The study also provides a method for determining the public assessment of health associated with ambient air pollution. The research results give an opportunity to solve a practical task related to informing the population about health risks and to overcome a so-called language barrier between experts and ordinary people. For example, messages aimed for decision-makers can be adapted considering all the identified perception peculiarities.