The Impact of Smoking-Related Advertisements on Smoking Cessation: An Econometric Analysis
» Project Details
Smoking is the leading preventable cause of death in the United States, contributing to an estimated 435,000 deaths annually. While policy debates have often focused on preventing youths from starting to smoke, recent analyses stress the need to encourage current smokers to quit.
The broad goal of this project is to examine whether smoking cessation decisions are influenced by smoking cessation product advertisements in magazines, cigarette advertisements in magazines, and anti-smoking public service advertisements (PSAs) in magazines. To achieve this broad goal, the project has three specific aims:
1. To further develop an existing media archive of smoking-related advertisements in magazines. We will expand the database by conducting content analysis to measure various features of these advertisements, including whether cigarette advertisements make claims about low tar and nicotine, whether cigarette advertisements contain features of sales promotions, and the specific warning labels that appear in cigarette advertisements.
2. To merge data from the existing media archive with individual-level survey data on smoking cessation behavior. Because of the unique advantages of the individual-level survey data, we will be able to use information on respondents magazine reading habits to create measures of their potential exposure to smoking-related advertisements.
3. To estimate an econometric model of the impact of smoking-related advertisements on smoking cessation behavior, both in general and by gender and race/ethnicity. The results will inform the policy debate on how to best regulate smoking cessation product advertisements, recognizing that these advertisements may help achieve some of the same public health goals as anti-smoking campaigns. The results will inform regulation of cigarette advertisements by documenting the impact the advertisements have on the size of the cigarette market by discouraging smoking cessation.
This project will estimate an econometric model of the impact of smoking-related advertisements on individual smoking cessation behavior. We will use individual-level data on smoking cessation behavior from multiple waves from 1995 to 1999 from the Simmons National Consumer Survey (NCS). Based on preliminary analysis of the data, we estimate that our combined sample will consist of approximately 140,000 adults with approximately 35,000 current smokers. These data include our research designs key elements: data on cessation behaviors, magazine reading habits, and rich data on respondents demographic and other characteristics.
We will use data on magazine advertisements from our Smoking Cessation Advertisements (SCADS) Archive. Previously constructed by the research team, these data will provide measures of all advertisements for smoking cessation products, smoking products, and smoking-related PSAs that appeared in any issue of 27 widely read magazines between October 1994 and September 1998. For this project, we will extend the SCADS Archive in several dimensions by conducting content analysis of the magazine advertisements. We will merge the individual-level NCS data with the extended SCADS data. Based on their magazine reading practices, we will measure each respondents potential exposure to the advertisements in the SCADS data. Using the merged data, we will estimate probit models of the probability an individual smoker will quit or attempt to quit smoking. Our probit models will control for other influences and will provide estimates of the impact of smoking-related advertisements on smoking cessation. By estimating specifications that include interaction terms between measures of race/ethnicity and potential exposure to advertisements, we will also be able to determine whether advertisements have differential impacts on smoking cessation decisions of racial and ethnic minorities.