The starting point was a survey that was included in the annual report from the SOM Institute at the University of Gothenburg and was presented in June 2016. Tobacco facts participated in the section on tobacco. A distinctive result was the strong support for smoke-free outdoor environments. 66-92 percent supported smoke-free outdoor cafes, entrances, train platforms, bus stops and balconies in apartment buildings. The proportion of unsure, "no opinion" or "neither good nor bad" was a low 5-17 percent.

However, the unsure category was significantly larger in the answers to the four other questions that dealt with tobacco taxes, invisible tobacco in stores, advertising-free cigarette packages and a national political decision on a smoke-free society, 21-40 percent.

Göran Boëthius hypothesized that the low, uncertain figures regarding support for smoke-free outdoors are due to the fact that most people have experience and knowledge of tobacco smoke in outdoor environments.

–Therefore, it is easy to understand the meaning of that question, while, for example, the question of introducing requirements to store cigarette packages out of sight in stores provides too little information for the participants in the survey to feel they know enough to take a position, says Göran Boëthius.

– More information is needed in the questions that improve the knowledge of those who respond. This makes it easier to take a position, but the opinion polls fear that the questions risk becoming leading if they contain more information. Therefore, they are kept brief.

Boëthius had the idea to compare the results from two groups in a survey where one is questioned with the briefly formulated questions from the SOM survey and the other with more comprehensive informative question formulations that provide increased knowledge on the subject. After consultation with the SOM Institute and with a grant from the Jacob Wallenberg Foundation, the opinion institute Novus was commissioned to investigate to what extent more comprehensive information given together with the questions affects the outcome of the survey. It was the four SOM questions that did not concern outdoor environments that were tested.

The experiment showed a significant effect of the questions with more informative wording. The proposal to increase taxes on cigarettes was met with insignificant differences by the three groups, original SOM, Novus with and Novus without explanatory texts.

However, the proportion of positive responses increases and the proportion of undecided decreases significantly in the answers to the questions about exposure bans, advertising-free tobacco packages and a long-term strategy for phasing out tobacco in Swedish society. They are supported by seven out of ten, while one out of ten is negative and two out of ten do not take a clear position.

–This outcome shows how crucial information and knowledge are when deciding important public health issues, says Göran Boëthius.

– The conclusion that the politicians we are trying to support should draw is to better inform the population about why measures need to be introduced and what they entail. Our survey shows that with greater knowledge, understanding increases for tightening tobacco policy.