Population Assessment of Tobacco and Health (PATH) Study [United States] Restricted-Use Files (ICPSR 36231)

1. The PATH Study Data User Forum allows researchers using any PATH Study data files to communicate with each other to ask and answer questions. Announcements, data releases and updates, new publications, upcoming events, and other information for PATH Study data users will also be posted to the forum.
2. Data are provided via ICPSR's Virtual Data Enclave (VDE) where researchers will work with data stored on secure ICPSR servers. Researchers will not possess actual physical copies of the data; however, they may request permission to access selected output outside the virtual environment after review by ICPSR. See the Access Notes to apply for access. Researchers are also encouraged to read the VDE Guide.
3. The data files contain person-level (PERSONID) and household-level identification (R0#_HHID) variables allowing linkage of people within a file, between Adult and Youth/Parent files, and across waves of data collection. The values in these two variables are random and contain no direct or indirect personally identifiable information. Please review Appendix D in the Restricted-Use Files User Guide for information and programming code on linking files together. The files are sorted by the variable PERSONID.
4. ICPSR attempted to duplicate all information contained in the questionnaires into the question text used in the codebooks. Some of the longer programming instructions were not incorporated into the question text. In these cases, the question text includes a note for the user to read the full programming instructions in the corresponding section of the questionnaire. Derived and imputed variables contain the algorithms used in the creation of these variables. Users are advised to refer to the Restricted-Use Files User Guide and annotated questionnaires when reviewing the codebooks.
5. Some variables were withheld to limit the release of information that is a potential risk for disclosure. These variables are listed in Appendix B in the Restricted-Use Files User Guide.

6. The Youth Interview and Parent Interview questionnaires were distinct and separate questionnaires used in data collection. However, for each wave, both instruments have been combined into a single document since the responses to these instruments are also combined into a single file.

7. Both the Adult and Youth questionnaires in each wave include several questions about tobacco brands and products the respondent usually uses and most recently used. For each question, a list of response options was displayed on the computer screen for the respondent to select. For many major brands and products, the displayed list included both a text label and a thumbnail image of the brand logo or product package. The displayed list was different for each of the tobacco product types with the brands and products listed being those that were known to exist for the specific tobacco product type. Because these lists are long, they are not provided in a frequency table for each variable in the codebook or in the annotated instrument. For convenience, both the Adult and Youth/Parent codebooks contain an appendix with a frequency table of the top 20 responses for each variable. The PATH Study Master Tobacco Brand and Product Code Guide is available as an Excel workbook file [Documentation.xlsx (Tobacco_Brand)]. The spreadsheets in this Excel workbook file are protected and may not be edited. However, the last spreadsheet contains filters to narrow the complete list. This spreadsheet is the master file of all brand and product responses for these questions from all three waves, including any responses that were not in the list of options displayed to the respondent.

8. In the Youth/Parent files for Wave 1 and Wave 2 (DS1002 and DS2002), the last section of the questionnaire contains demographic and health history questions. A few of the questions were asked of all youth. However, most questions were only asked of emancipated youth. The responses to these questions for non-emancipated youth were coded as "Inapplicable". The questionnaire and codebook note which variables were asked only of emancipated youth. Conversely, in the Parent Interview section the same questions were asked of parents of all sampled youth except for the emancipated youth. In this section the cases for emancipated youth were coded as "Inapplicable". There are a small number of emancipated youth in Wave 3, but there are no individual questions asked exclusively of emancipated youth.

9. In both the Adult and Youth/Parent data files, several groups of variables contain the word "RANDOM" in both the variable name and label. This indicates computerized randomization of the question order. These "RANDOM" variables detail the order in which the questions were asked of a particular respondent.

10. The Wave 1 data files for both Adults and Youth contain a section about tobacco advertising. There are 20 variable triplets contained in this section. The computer randomly selected 20 advertisements and then asked the respondents whether they had seen the ad and whether they liked the ad. The Image ID variable (_AD) identifies the advertisement that was displayed to the respondent to characterize the ad, e.g., the tobacco product and brand. However, vendors did not grant permission to publicly release the actual .jpg and .bmp files containing the images seen by respondents.

11. Derived and imputed demographic variables (age, sex, Hispanic ethnicity, and race) are included near the end of each data file. The Adult file also contains education. An accompanying imputation flag variable is also included. These variables are distinguished by the variable name starting with "R0#R" and contain the word "DERIVED" or "IMPUTED" in the variable label. Imputed variables are only available on the Wave 1 data files.

12. Within the "Derived and Imputed Variables" section of the codebooks of the Adult and Youth/Parent files for Wave 1 only are two geographic variables - Census Region and Census Division. There are additional variables to designate urban areas and Census Block characteristics.

13. All Adult and Youth/Parent data files contain additional derived variables. These variables can be distinguished by the variable name starting with "R0#R" and contain the word "DERIVED" in the variable label. There are several variables for each tobacco category to identify certain classes of current and former tobacco users.

14. In accordance with the study's informed consent, information is suppressed about individuals who withdrew from the PATH Study. Their information was recoded to a special missing value, designated as -97777.

15. The current release contains the restricted-use versions of Wave 1, Wave 2, and Wave 3 data files. Wave 4 data files are tentatively planned to be released in 2019.

16. The documentation file named Documentation.pdf (Informed_Consent) includes the six consent forms provided to and signed by the respondents for the various types of interviews conducted and biological samples collected. Participants provide consent at their initial interview and biological sample collection; consents remain in effect for all subsequent waves. Aged-up adults who previously responded to the Youth interview at one wave need to re-consent as an adult at the time of the subsequent wave.
17. The Nonresponse Bias Analysis Report for Wave 1 details the response rates and the potential for bias from nonresponse. There is also a Nonresponse Bias Analysis Report for Wave 2 and a Nonresponse Bias Analysis Report for Wave 3.

18. The Informed Consent Document and Nonresponse Bias Analysis Reports are specific to each wave they are listed for. They are listed for both the Adult File and the Youth/Parent File, but they are the same files.

19. The questionnaires in this collection are updated versions of the fielded questionnaires that were annotated for analytic purposes. Spanish versions are also available.

20. The PATH Study's documentation is available for your use and may be reproduced in whole or in part without permission from NIH's National Institute on Drug Abuse or FDA's Center for Tobacco Products. Citation of the source is appreciated.

21. Additional background information including answers to frequently asked questions for study participants and researchers can be found in the Researchers section of the PATH Study series page.

22. The Restricted-Use Files User Guide provides an overview of the entire PATH Study. The guide covers topics such as sample design, data collection, weighting, response rates, and programming syntax to run common statistics and link the files together. Researchers should feel free to use the information in the User Guide for their publication and the guide should be cited as follows:

    • United States Department of Health and Human Services. National Institutes of Health. National Institute on Drug Abuse, and United States Department of Health and Human Services. Food and Drug Administration. Center for Tobacco Products. Population Assessment of Tobacco and Health (PATH) Study [United States] Restricted-Use Files, User Guide. ICPSR36231-v14 Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2018-05-01. http://doi.org/10.3886/ICPSR36231.userguide

23. The data for the PATH Study was collected and prepared by Westat. The contract number under which they performed their work is HHSN271201100027C.

2019-02-01 Wave 1 Adult and Youth State Identifier data files were added to the collection. Data and documentation for the Master linkage file was updated.

2018-10-01 2018-09-28 Data and documentation for the Master linkage file was updated. The Nonresponse Bias Analysis Report is now included for Wave 3.

Wave 3 Adult and Youth data files were added to the collection. Wave 1 and Wave 2 Adult and Youth data files were updated to improve the clarity and consistency of variable labels, especially in the Nicotine Dependence section.

A new variable was added to Wave 1 and Wave 2 Adult data - R0#_ND_DATA_ROUTE. A second variable was added to the Wave 2 Adult data - R02R_A_P12M_BLUNTONLY_GRILLO. An additional 18 derived variables in the Wave 2 Adult data were revised and replaced the original variables. The newly named variables possess the original name, but also contain "_REV" at the end of the variable name.

A skip error was identified in the Wave 2 Adult instrument, which resulted in some respondents being asked two questions when they should not have been. Therefore, the affected items, R02_AG0100CG and R02_AG0100FC, contain some extra data. Notes were added to the annotated instrument and codebook to describe the issue.

The User Guide and Questionnaires were also updated to improve understanding of the data files. A Nonresponse Bias Analysis report is now included for Wave 2.

• United States Department of Health and Human Services. National Institutes of Health. National Institute on Drug Abuse, and United States Department of Health and Human Services. Food and Drug Administration. Center for Tobacco Products. Population Assessment of Tobacco and Health (PATH) Study [United States] Restricted-Use Files. ICPSR36231-v16. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2019-02-01. http://doi.org/10.3886/ICPSR36231.v16

2017-06-19 The Wave 1 and Wave 2 data files, for both Adults and Youth, were updated to correct minor errors along with the questionnaires to correct minor typos and clarify specifications.

2017-04-27 A minor revision was made to the both the English and Spanish versions of the Wave 1 Adult questionnaire. The User Guide was also updated. Two Excel crosswalks, one for Adults and one for Youth, were added to the available documentation to highlight the differences between the Wave 1 and Wave 2 files.

2017-04-03 An update was made to internal files to correct an issue with how missing values are displayed online through ICPSR's variables database.

2017-03-23 Minor revisions were made to the Missing Values Code table within the User Guide and both Codebooks for Wave 2.

2017-03-15 Data from Wave 2 of the study were added to the collection. The User Guide and Master Tobacco Brand and Product Code Guide were expanded to include information for Wave 2.

2017-01-31 The variable R01X_CB_REGION in both the Wave 1 Adult and Youth/Parent files was updated to correct an error in the value labels. The values for codes 2 and 3 had been inadvertently swapped. The data did not change; only the value labels for codes 2 and 3 have been corrected.

2016-11-28 An additional 40 derived variables were added to the end of the Wave 1 Youth / Parent file that are similar to those already in the Wave 1 Adult file. Information for individuals who withdrew from the study is denoted in the datasets by the special missing value -97777. Spanish versions of the annotated instruments are also now available.

2016-05-24 The study's title changed with the removal of the year range. The Informed Consent Document and Non-Response Bias Analysis Report were changed from being study level files to being a part of the Wave 1 (DS1001 and DS1002) specific documentation.

2016-04-22 An additional documentation file (Non-response_Report) was added to the collection.

2016-04-20 Updated the file names only for the two study level documentation files (Informed_Consent and Tobacco_Brand) so that each file was easily identified and distinguishable from the other. No change was made to the content in either file.

2016-04-18 Coding was updated for the sexual attraction variables. The questionnaires were revised to enhance the clarity of the ASK statements. The PDF codebooks now contain full question text from the questionnaires. Lastly, the PATH Study Master Tobacco Brand and Product Code Guide and an Informed Consent Document were also released.

2016-01-13 PDF codebooks were released without question text. The codebooks will be updated in the near future that includes question text.

2015-12-19 ICPSR data undergo a confidentiality review and are altered when necessary to limit the risk of disclosure. ICPSR also routinely creates ready-to-go data files along with setups in the major statistical software formats as well as standard codebooks to accompany the data. In addition to these procedures, ICPSR performed the following processing steps for this data collection:

• Checked for undocumented or out-of-range codes.