Department of Human Sciences, Ohio State University, DIW-Berlin, NBER, 1787 Neil Avenue, Columbus, OH, 43210, USA.
Faculty of Business Administration, Bilkent University, Bilkent, 06800, Ankara, Turkey.
Eur J Health Econ. 2019 Feb;20(1):149-162. doi: 10.1007/s10198-018-0988-9. Epub 2018 Jun 22.
We investigate whether individuals are less likely to start and more likely to quit smoking in years when newspapers publish more articles about the health risks of smoking. With data from 9030 respondents to the 2008 Global Adult Tobacco Survey in Turkey, we construct respondents' life-course smoking histories back to 1925 and model initiation and cessation decisions taken 1925-2008. To measure information, we count articles published in Milliyet, one of Turkey's major newspapers. Results from linear probability models show that people who have seen more smoking-health risk articles know more about the smoking-health relationship. Holding constant each individual's information stock, education, place of residence, and the price of cigarettes, we find that, as new information arrives, male and female smokers in all cohorts are significantly more likely to quit and women are less likely to start. Our analysis is one of the first that examines how new information affects smoking decisions while controlling for each individual's existing stock of information.
我们研究了在报纸上刊登更多关于吸烟危害健康的文章的年份，个人是否更有可能减少吸烟或戒烟。我们利用土耳其 2008 年全球成人烟草调查中 9030 名受访者的数据，构建了受访者从 1925 年到 2008 年的生命历程吸烟史，并对 1925 年至 2008 年的吸烟开始和戒烟决定进行了建模。为了衡量信息，我们对土耳其主要报纸之一的《 Milliyet 》上发表的文章进行了计数。线性概率模型的结果表明，看到更多吸烟与健康风险相关文章的人对吸烟与健康的关系了解得更多。在固定每个个体的信息存量、教育程度、居住地和香烟价格的情况下，我们发现，随着新信息的出现，所有年龄段的男性和女性吸烟者戒烟的可能性显著增加，而女性开始吸烟的可能性降低。我们的分析是首批在控制每个个体现有信息存量的情况下，考察新信息如何影响吸烟决策的分析之一。
