Tobacco Use Supplement to the Current Population Survey (TUS-CPS) is commonly used as one of the standards for determining smoking prevalence in the U.S. adult population. However, the quality of TUS-CPS data has not been yet investigated. Given the high level of complexity of modern design of TUS-CPS and sensitive nature of smoking-related questions, establishing data quality is very important. Indeed, even if cutting-edge statistical methodology is used to analyze the data, ambiguous data can result in misleading findings about the incidence of smoking behaviors and, therefore, can affect policies targeted at preventing cancer and other smoking-attributable diseases and conditions. The proposed study is targeted at assessing data reliability of several smoking-related measures in the TUS-CPS. The following aims will be achieved: (1) several smoking and non-smoking measures will be defined;(2) reliability of these measures will be examined, and levels of reliability for smoking and non-smoking measures will be compared;and (3) the relationship between the strict data agreement and several respondent and survey attributes (socio-demographic characteristics, smoking habits, respondent mode (self, proxy) and survey method (phone, in-person), and their interactions) will be identified. It is anticipated that the results f this research will help to confirm either the high level of the TUS-CPS data quality or necessity of improvements in the design and administration of national surveys used to assess sensitive information such as smoking, sexual activity, drug and alcohol use, and other behaviors. Importance of the proposed investigation is also confirmed by preliminary research findings.
The proposed project will investigate the reliability of several smoking-related measures in the Tobacco Use Supplement to the Current Population Survey (TUS-CPS). Given that the TUS-CPS is commonly used as one of the standards for determining smoking prevalence in the U.S. adult population, assessing the quality of TUS-CPS data obtained is vital for accurately informing tobacco researchers and policymakers. It is anticipated that the investigation will either confirm a high level of data quality, or results will play an important role in improving the design and administration of national surveys to minimize the chance of obtaining ambiguous data and false findings.
| Bright, Brianna C; Soulakova, Julia N (2016) Evaluating the performance of simultaneous stepwise confidence intervals for the difference between two Poisson rates. Commun Stat Simul Comput 45:2338-2361 |
| Soulakova, Julia N; Huang, Huang; Crockett, Lisa J (2015) Racial/Ethnic Disparities in Consistent Reporting of Smoking-Related Behaviors. J Addict Behav Ther Rehabil 4: |
| Soulakova, Julia N; Bright, Brianna C; Crockett, Lisa J (2015) Perception of Time Since Smoking Cessation: Time in Memory Can Elapse Faster. J Addict Behav Ther Rehabil 4: |
| Soulakova, Julia N; Crockett, Lisa J (2014) Consistency and Recanting of Ever-Smoking Status Reported by Self and Proxy Respondents One Year Apart. J Addict Behav Ther Rehabil 3:1000113 |
| Bright, Brianna C; Soulakova, Julia N (2014) Evidence of telescoping in regular smoking onset age. Nicotine Tob Res 16:717-24 |
| Soulakova, Julia N; Bright, Brianna C; Crockett, Lisa J (2013) On Consistency of Self- and Proxy-reported Regular Smoking Initiation Age. J Subst Abus Alcohol 1:1001 |
