U.S. flag An official website of the United States government
  1. Home
  2. Tobacco Products
  3. Tobacco Science & Research
  4. Research
  5. Extending a School-Based Cohort to Improve Longitudinal Modeling
  1. Tobacco Science & Research

Extending a School-Based Cohort to Improve Longitudinal Modeling

Extending a School-Based Cohort to Improve Longitudinal Modeling

Principal Investigator: Thomas Valente

Funding Mechanism: National Institutes of Health- Grant

ID number: 3R01CA157577-02S1

Award Date: 4/1/2013

Institution: University of Southern California


Social influences are an important factor that drives tobacco use or nonuse in adolescents. In this project, investigators will re-interview an existing cohort of approximately 1,260 12th grade students (approximately 18 years of age) in five southern California schools with a majority Hispanic/Latino population; many of these students had previously been interviewed twice in 10th grade (in five schools) and once in 11th grade (in four schools). In this fourth survey, data on multiple types of relationships will again be collected along with tobacco use behaviors; investigators will use the data to analyze the social influences that account for both changes in behavior and changes in social networks over time. In addition, investigators will collect new data assessing two dimensions of tobacco attitudes (affective and utilitarian) needed to empirically calibrate an agent-based model of smoking initiation; the affective component consists of mostly socially constructed imagery (e.g., smoking is cool, smoking is trashy/low class), and the utilitarian component measures concepts such as weight control, stress relief, and health risks. Specific aims are: (1): to collect social network and tobacco use data from these 12th grade students so that the previously-developed dynamic diffusion network/stochastic actor model can be applied to these data; and (2) to measure tobacco use attitudes among this cohort of students so that agent-based models of tobacco use dynamics may be accurately calibrated with empirical data.


Related Resources