OMB Justification Memo

FDA DOCUMENTATION FOR THE GENERIC CLEARANCE
OF PRETESTING COMMUNICATIONS ON TOBACCO PRODUCTS
(0910-0674)
Retailer interviews are designed to elicit information from tobacco retailers in order to support FDA communication
and education activities. Retailers who work in retail outlets that sell tobacco will be asked to discuss their
experiences and thoughts around FDA regulations, reducing tobacco use among youth, decision-making processes
and behaviors, communicating with the larger retail community, and motivating retailers to support FDA regulation
and compliance. They will represent a mix of retail outlets and will be demographically and geographically diverse.
As retailer interviews are designed to elicit qualitative information and do not yield meaningful quantitative
findings, they cannot be used to drive the development of policies, programs, and services.

TITLE OF INFORMATION COLLECTION: Interviews with Tobacco Retailers to Inform
Retailer Education; OMB Control Number 0910-0674
DESCRIPTION OF THIS SPECIFIC COLLECTION
1. Statement of need:
The Food and Drug Administration (FDA), Center for Tobacco Products (CTP), is
seeking OMB approval under the generic clearance 0910-0674 to conduct a series of
interviews, “Interviews with Tobacco Retailers to Inform Retailer Education,” to inform
future research and communication activities for CTP’s retailer education campaign. It is
critical to understand the communication needs of retailers and what influences their
decision-making processes. This will help CTP provide targeted communications to
retailers that assists in their implementation of and compliance with regulations under the
Act.
2. Intended use of information:
This study will be used to help FDA better understand how to communicate with retailers
about FDA regulations through FDA’s Break the Chain of Tobacco Addiction retailer
education campaign and other retailer communications strategies.
This qualitative study will (1) assess the primary influences on clerks' and managers'
decision making and behavior with regards to compliance with the tobacco point of sale
age verification and selling requirements; (2) assess the role and influence of
communications, including FDA's Break the Chain of Tobacco Addiction campaign, on
the perceptions, decisions and behavior of clerks and managers; (3) identify appropriate
channels through which to communicate with retailers.
Findings from this study will be paired with FDA’s previous research with key
stakeholders that was conducted to understand the communication needs and decisionmaking challenges in the retail environment from an expert perspective1. Experts
included those from academic institutions, federal and state government, retailer
associations, and public health associations. The information gathered from these
stakeholders was used to inform the development of the retailer interview protocol for the
current study. The expert stakeholder interviews and interviews with retailers will be
used to develop a framework—or mental model—to better understand retailer decision
making. The mental model will be used to identify potential points in the decision1

The Stakeholder Interviews for Break the Chain Retailer Education Campaign research was completed in October 2012. This
study received OMB approval June 6, 2012.

1

making process that can be supported through FDA education and communication
strategies. For more information about mental models research and the expert model
developed based on the Stakeholder Interviews for Break the Chain Retailer Education
Campaign research, please see Appendix A.
3. Description of respondents:
Understanding the communication needs of retailers requires eliciting qualitative
information from a variety of retailers. For the purposes of this research, tobacco retailers
will be defined as “clerks” and “managers” who work in retail outlets that sell tobacco.
The contractor (Decision Partners) is planning to conduct interviews with no more than
90 retailers. These interviews will be conducted individually with clerks (45 interviews)
and managers (45 interviews).
Retailers will be recruited from a mix of retail locations (e.g. convenience stores, grocery
stores, gas stations, mass merchandisers) within a 90-mile radius of three cities: Seattle,
WA, Cleveland, OH, and Raleigh, NC. These case study cities were selected because
they were all above the national mean in Synar-reported state-level retailer violation rates
for 2009; they also reflect a variation in geographic and demographic characteristics of
the U.S.; and they have a mix of urban, suburban and rural communities within a 40-mile
radius of the downtown area.
Eligibility Requirements
Given the diversity of the target population of tobacco retailers, sample development and
recruitment methods will require a combination approach, including coordinating with
retailer organizations, community organizations, or other key stakeholders in the study
locations and those beyond the study locations who have connections to retailers in the
study locations. To be eligible to participate, retailers must be at least 18 years of age and
be currently employed by (or own) a retail outlet that sells cigarettes, cigarette tobacco, or
smokeless tobacco. Types of retail outlets will include: convenience stores (without gas),
gas stations (with or without convenience stores), grocery stores, and massmerchandisers. Research has identified these types of outlets sell 98% of the tobacco sold
in the United States (Ribisl, 2011). Given the limited scope of this project, people who
work for pharmacies, restaurants, liquor stores, outlets on Native American reservations,
or Internet retailers will be excluded from the study. Interviewees must have a direct role
in the sale of tobacco, or in supervising or managing those who make sales. They must be
willing and able to participate in a phone or in-person interview, conducted in English.
An effort will be made to recruit a diverse group of participants (e.g. gender, age,
race/ethnicity, type of retailer outlet, location of retail outlet). Stores will be ineligible to
participate if there is an open legal investigation or pending litigation between the store
and FDA; we will work with senior management in CTP’s Office of Compliance and
Enforcement to identify these stores. Retailers will not be recruited from identified stores.
Sampling, Recruitment, and Screening Procedures
The sampling frames will be created through lists of retail outlet locations that were
provided to FDA through the state inspection contracts. Participants will be randomly
selected from stores within defined zip codes to allow for inclusion of urban, suburban
and rural locations (as determined by the 2010 U.S. Census data). The number and
location of zip codes will be defined based on the final determination of the most
appropriate mix of sample development and recruitment methods in each case study
2

location. In each case study, we will target sample sizes of 30 (15 salespeople, 15
managers) for a total of 90 interviews across all three cities. To protect privacy and
encourage open and honest discussions, clerks and managers will not be recruited from
the same store.
After the store has been selected through the abovementioned sampling method, the
contractor will use a combination of the following methods to aid in the recruitment of
participants.
 Partnering with a retail chain: for example, working with a particular chain to
identify and recruit managers and clerks from stores in various locations.
 Partnering with retail or other industry associations, and chambers of commerce,
to which retailers may be affiliated.
 Partnering with community-based associations, which may be particularly helpful
in recruiting harder-to-reach retailers, such as independent retailers, or retailers
located in marginalized or disadvantaged neighborhoods.
 Public advertisements, if needed, inviting people who work for retail
establishments that sell cigarettes, cigarette tobacco, or smokeless tobacco to
participate in a research interview.
In all the above partnerships, the level of coordination may include:
 Permission to recruit through organizations' listservs, websites, newsletters, or other
store-level communication
 Permission to post in-store recruitment flyers (e.g., in worker lounges)
 Permission to release member contact information so that researchers can contact
directly for recruitment
 Permission to recruit clerks and managers directly at the store location (recruitment
information would be handed out in the store but actual screening would be done via
phone to ensure privacy)
A combination of recruitment techniques is necessary because of the difficulty in
recruiting this population. The project team is working closely with Dr. Kurt Ribisl, an
expert in retailer education and research from the University of North Carolina Chapel
Hill, who advised that we take a broad and diverse approach to recruitment to ensure
adequate participation.
Recruitment materials have been developed that can be adapted for inclusion in the
abovementioned recruitment techniques (see Appendix B).
Clerks and managers who are interested in participating in the interviews will be screened
via telephone to ensure they are eligible for participation (see Appendices C-D). The
screener includes questions related to type of retail outlet, position/job title, years of
employment, and age. The recruitment screener also includes questions related to English
proficiency. This is intended to recruit individuals who are both capable of and
comfortable participating in an interview conducted in English. Race/ethnicity and sex
will be captured in the screener but will be used descriptively when reporting findings
rather than as inclusion/exclusion criteria. Participants will be recruited and screened
until approximately 30 interviews (15 clerks, 15 managers) per location are conducted.

3

4. Date(s) to be Conducted:
The retailer interview research will begin in April 2013 and be completed by February
2014. The interviews will be held in-person and via telephone.
5. How the Information is being collected:
IDIs will last approximately 45 minutes. Approximately one-third of the interviews will
be held in-person at a neutral, private location within the community. The remaining
interviews will be conducted via telephone. The questions will be designed to elicit
information based on participants’ professional experiences, thoughts, beliefs, and
perceptions (see Appendices E-F).
Prior to the start of the interview, the interviewer will ask for verbal consent to audio
record the interview. If verbal consent is given from the participant, the interviews will be
voice recorded. If consent for audio-recording is not received, detailed hand-written
notes will be taken. The audio recording will later be transcribed and will assist the
interviewer in analyzing the data. The audio recording will be destroyed upon completion
of the transcription. One paper and one electronic copy of the interview transcripts will
be supplied to FDA upon completion of the data collection. To protect participant
privacy, the contractor will redact any potentially identifiable information about the
participant in the transcripts.
Interviews will be led by a trained interviewer who will use a script to guide the
discussion (see Interviewer’s Guides). The discussion will cover the following topics: (a)
work and store responsibilities; (b) perceptions, beliefs, and decisions at the point of sale;
(c) experiences with customers who come into the store to purchase cigarettes and other
tobacco products; (d) communication and training within the store; (e) store rules and
federal laws related to the sale of tobacco products, including perceptions and attitudes
toward these; (f) awareness and perceptions of FDA; and (g) reactions to FDA’s Break
the Chain campaign and other communication and education strategies. Additionally,
managers will be asked to discuss the communication of rules and policies to employees.
The interviewer will emphasize the purpose of the interview is not to discuss specific
compliance behaviors or violations and that this research activity is independent from
CTP OCE compliance and enforcement activities.
Participation is voluntary and participants can choose not to answer any question. The
data collected will be used to develop a more sophisticated and robust understanding of
the communication needs of retailers.
6. Privacy of Respondents:
Decision Partners is conducting retailer interviews as a contractor to FDA. Decision
Partners is familiar with federal confidentiality and privacy provisions and conducted the
Stakeholder Interviews for Break the Chain Retailer Education Campaign discussed in
Section 2.
Prior to the start of the interview, the interviewer will ask for verbal consent to audio
record the interview. If verbal consent is given from the participant, the interviews will be
voice recorded. If consent for audio-recording is not received, detailed hand-written
notes will be taken. The audio recording will later be transcribed and will assist the
4

interviewer in analyzing the data. The audio recording will be destroyed upon completion
of the transcription. One paper and one electronic copy of the interview transcripts will
be supplied to FDA upon completion of the data collection. To protect participant
privacy, the contractor will redact any potentially identifiable information about the
participant in the transcripts.
All data will be collected with an assurance that the respondents' discussions will remain
private to the extent provided by the law. The store location represented by each
participant will not be disclosed to management or third-parties. The recruitment
materials and study interview guides emphasize to participants that no one will be able to
link the respondent’s identity to their responses. Identifying information will not be
included in the data files delivered by contractors to the agency.
The contractor will not share personal information regarding participants with any third
party without the participant’s permission unless it is required by law to protect their
rights or to comply with judicial proceedings, court order, or other legal process.
Identifying information will not be included in the transcripts and digital recordings
delivered to the agency. All data received by FDA will remain in a secured area. No data
will contain identifying information. Upon final analysis by FDA, transcripts and digital
recordings will be destroyed.
The following is documentation of Decision Partner’s privacy procedures:
Decision Partners adheres to the highest standards of research ethics in accordance with
our Human Research Protection Plan (HRRP), which is available upon request. Per our
standard practices, the following are the key components of our privacy plan for this
project:

A statement informing clerks and managers participants of their rights to privacy
is included in the interview discussion guides (see Appendices E and F).

FDA may be aware which store locations are being invited to participate in an
interview but data will be reported in aggregate so as not to identify specific
individual or store responses.

The identity of the research participants will be kept private within the Decision
Partners and FDA project team. Specific stores or individuals who participate in
an interview will not be identified in any summaries or reports, unless permission
is granted. The Summary Report and any future reports will describe participants
in general terms (e.g., clerk from urban convenience store).

The interview will be conducted in person or by phone by one or two senior
researchers from Decision Partners. FDA will not participate in any way in the
interviews.

Researchers will take detailed handwritten notes of the interview. With
permission, the phone interviews will be audio recorded to produce transcriptions
to supplement the interview notes. The recording will be securely stored by
Decision Partners and will be destroyed at the completion of the project as per
our HRPP.

A Summary Report will be prepared from the interview notes. No identifying
information will be included in the Report. The Summary Report and any future
reports will consolidate the responses of all participants, and no quotes will be
attributed to any individual.
5

7. Amount and justification for any proposed incentive
As participants often have competing demands for time, incentives are used to encourage
participation in qualitative research studies, including interviews. Incentives must be high
enough to cover travel costs, time, and to provide enough incentive to attend the focus
group rather than another activity. If the incentive is not adequate enough, participants
may agree to participate and then not show up for the group. Low participation may result
in inadequate data collection, or, in the worst cases, cancellation of groups and loss of
costs associated with recruitment, travel costs, and interviewer time (Morgan and Scanell,
1998).
At the conclusion of the interview session, participating clerks will be offered a cash
incentive of $30.00 and managers will receive $50.00. This payment was set at
approximately three times the average hourly wage for these positions (Bureau of Labor
Statistics, 2011). In addition to having a higher average hourly wage, managers are being
incentivized at slightly higher rates than clerks because there are fewer of them than
clerks and it is anticipated this will be a difficult group to recruit. Having the perspectives
of both clerks and managers is critical to our understanding of decision making and
communication within the retail environment and ultimately to the ability of FDA to
effectively and appropriately communicate its regulations around compliance and
enforcement.
8. Questions of a Sensitive Nature
Questions around decision-making and compliance behaviors may be potentially
sensitive to retailers since they may believe that discussion of non-compliant behaviors
may result in law enforcement actions. To address this concern, the questions in this
study are not designed to elicit discussion of specific non-compliant behaviors exhibited
or witnessed. Rather, the questions in this study are designed to encourage discussion of
general influences on decision-making, knowledge, awareness, and perceptions of
existing regulations, and communication preferences. The interviewer will emphasize the
purpose of interview is not to discuss specific compliance behaviors or violations and that
even if they do mention (spontaneously) such activity that this research activity is
independent from CTP OCE compliance/enforcement activities. Participants will be
informed about their right to privacy and the steps being taken to protect their privacy
and are told they are free to refrain from answering any questions they do not wish to
answer. In addition, information gathered during the interviews will be reported in
aggregate in a final report, thus not linking any mentions of non-compliant behaviors
back to a particular retail outlet or employee.
If the participant expresses discomfort or appears to be uncomfortable, interviewers will
be instructed to remind the participant that the information collected will be kept private,
that it will not be shared with FDA, store management, or other third parties, that there
are no right or wrong answers, that all comments are important to the research, that the
recording device can be switched off if the participant prefers, that their participation is
voluntary, and that they may refuse any question they wish. In addition, information
gathered during the interviews will be reported in aggregate in a final report, thus not
linking any non-compliant behaviors back to a particular retail outlet or employee.

6

This study was submitted to CTP’s Research Involving Human Subject Committee
(RIHSC) for review on December 20, 2012. It was submitted to FDA’s RIHSC as a
request for exemption on March 11, 2013.
9. Description of Statistical Methods
Interviews rely on qualitative methods and are not intended to yield results that are
statistically projectable. This qualitative research employs the mental models approach
and associated analyses. Mental models research is a robust qualitative research method
(see Appendix A for more information). Standard analytical tools and techniques used by
mental models researchers to ensure that coding and analysis is conducted to a high
research standard will be employed. This will include coding and analyzing the interview
data against an expert model (a depiction of influences on retailer behavior developed in
the previous research with expert stakeholders); summarizing the key research findings of
each cohort; and conducting an analysis of the gaps and alignments between the mental
models research findings and the expert model.
A coding guide will be developed to define and describe codes aligned with the expert
model. The coding process begins by parsing participant responses into discrete
segments. Each segment is assigned one or more codes from the codebook that best
aligns with the various expert model nodes. Additional tags or keywords may be applied
to the segments to capture finer details such as tone.
At the beginning of the coding phase, several interviews will be coded independently by
two coders. Differences will be resolved by discussion, with final determination by the
research lead. Segments that express new concepts not anticipated in the research design
will be assigned new codes, which will be reviewed and incorporated into the coding
guide. Once the coding procedure is deemed stable, each remaining interview will be
coded by one coder. To ensure data quality the research lead will review and perform
checks for drift in coding throughout the coding process. The research lead will conduct
regular briefings with the coding team to review and make any necessary clarifications or
changes to coded interviews or the coding guide as needed.
At the completion of the coding process, more detailed analyses will be conducted on
segments using interview question numbers, tags, codes, keywords, or any combination
thereof, in order to explore and subjectively assess the results and possible associations
among concepts. Summaries of these analyses will be included in the research reports. In
addition, aggregate mental models diagrams will be constructed, which provide a
summary of the relative “weighting” of the degree of influence of variables represented
by the various nodes within the expert model on the participants’ collective thinking.
Nodes in the expert model will also be qualitatively weighted based on the subjective
assessment of the prevalence and degree of strength of the themes captured within each
node. Themes that emerge in the research findings that were not previously identified in
the expert model will be depicted as new nodes and highlighted in the mental model as
necessary. Similarly, nodes that appear on the mental model but that were not found to be
significant influences through the mental models research will be highlighted in the
expert model.
Where appropriate, we may be able to draw on quantitative tools and techniques to
explore possible associations (e.g., suggested differences in the thinking of managers vs.
7

salespeople, men vs. women, etc.). However, this research method generally precludes
statistical analyses that can be generalized as representative of the broader U.S.
population. As retailer interviews are designed to elicit qualitative information, they do
not yield meaningful quantitative findings. As such, they cannot be used to drive the
development of policies, programs, and services.

BURDEN HOUR COMPUTATION (Number of responses (X) estimated response or
participation time in minutes (/60) = annual burden hours):
Type/Category
of Respondent

No. of Respondents

Clerk Interviews
Manager
Interviews
Total

45
45

Participation
Time Per
Interview
(minutes)
55
55

90

Burden
(hours)
41.25
41.25
82.5

REQUESTED APPROVAL DATE: April 17, 2013
NAME OF PRA ANALYST & PROGRAM CONTACT:
PRA Analyst

Daniel Gittleson
301-796-5156
daniel.gittleson@fda.hhs.gov

Program Contact

Greta Tessman
301-796-6722
Greta.Tessman@fda.hhs.gov

FDA CENTER: Center for Tobacco Products (FDA/CTP)
ATTACHMENTS:
 Appendix A: Background Information about Mental Models Research and Expert
Model Developed from Stakeholder Interviews Research [Note: this attachment is
not an instrument, it merely supplements and illustrates information provided in
the memo above]
 Appendix B: Recruitment Materials
 Appendix C: Clerk Screener
 Appendix D: Manager Screener
 Appendix E: Clerk Interview Guide
 Appendix F: Manager Interview Guide

8