3170-XXXX Supporting Statement part B OMB

30-day Federal Register Notice Version

BUREAU OF CONSUMER FINANCIAL PROTECTION
PAPERWORK REDUCTION ACT SUBMISSION
INFORMATION COLLECTION REQUEST
SUPPORTING STATEMENT PART B
DEBT COLLECTION QUANTITATIVE DISCLOSURE TESTING
(OMB CONTROL NUMBER: 3170-XXXX)

1. Respondent Universe and Selection Methods
Population and Sampling Frame
This research study is not intended to be representative of the U.S. population, but rather be an
experimental design that tests differences in form comprehension between groups. The study
sample will be chosen by selecting a sample from GfK’s KnowledgePanel®, an online panel
whose participants were recruited through probability sampling. KnowledgePanel currently
includes about 55,000 adult members.
Since its inception in 1999, KnowledgePanel participants have been recruited based on industry
standards for selecting general population surveys, first by random-digit dialing and currently by
address-based sampling (ABS). The ABS methodology is a random sample of addresses from the
U.S. Postal Service’s Delivery Sequence File. Individuals residing at randomly sampled
addresses are invited to join KnowledgePanel through a series of mailings (in English and
Spanish); non-responders are phoned when a telephone number can be matched to the sampled
address. Household members who were randomly selected can indicate their willingness to join
the panel by returning a completed acceptance form in a postage-paid envelope, calling a tollfree hotline and speaking to a bilingual recruitment agent, or accessing a dedicated recruitment
website. Non-internet households are provided a web-enabled computer and free internet access.
Historical recruitment rates for participation in the panel are approximately 15-20%.
The expected survey completion rate for those selected within the panel is 60–70%, based on
historical completion rates for KnowledgePanel participants. The survey will be conducted in
English.
Sample Size
The target sample size is 8,000 completed surveys, approximately 5,330 (2/3) of which will be
completed by panelists who have “experienced” debt collection in the past 24 months. We will
Page 1 of 5

30-day Federal Register Notice Version

select the sample in two phases. First, we will select a random sample of adults and screen the
panelists to determine whether they have experienced debt collections in the past 24 months—
estimated to be 30–35% of adults with a credit file based on research by the Urban Institute1 and
the BCFP2. In the second phase, we will select a sample of panelists for the full survey. All of
those who have experienced debt collection in the past 24 months will be invited to participate,
along with a random subsample of those who indicate they have not experienced debt collection
in the previous 24 months. The total sample distribution is expected to be 2/3 “experienced” and
1/3 without debt collection experience in the past 24 months. We designed the sample in this
way because we found during qualitative testing that people with experience and without
experience with debt collection responded to the form different. In this study, we wanted to
ensure that the form works for both groups (because both groups could have a debt collection
experience in the future). The study over-represents people who had experience with debt
collection, to make sure that the form worked well for this population of consumers.

2. Describe the procedures for the collection of information.
KnowledgePanel participants will be recruited by email to take the survey. The email invitation
will describe the purpose and importance of the research, identify BCFP as the sponsor, and
include an email address and toll-free telephone number for helpdesk support. The recruitment
email will include a personalized URL (e.g., www.researchsurvey/123456) for the web survey
that features a unique, non-sequential identifier for secure login. When panelists click on the
URL, they will be directed to the screening survey hosted by our contractor. The screening
survey will determine whether the respondent has experienced debt collection in the previous 24
months. If selected to continue by an automated program to ensure balanced demographic
characteristics and debt collection experience, the respondent will be directed to the full survey.
Panelists who do not respond to the email invitation will receive e-mail reminders starting on
Day 3 and will continue to receive follow-up correspondence once per week, thereafter.
The fielding of the survey will begin with a small pilot test of up to 200 completed surveys to
determine whether recently added questions are effective, followed by soft launch, targeting up
to 1,000 completed surveys to ensure that the instrument is functioning as intended. Upon
completing the pilot, we will pause collection, review the data, and refine or drop ineffective
new questions. Following the soft launch, we will pause the survey as well, review results, and
identify any technical changes that need to be made before fully launching the survey. After
incorporating any technical changes, we will then re-open the survey and invite the full set of
participants (additional 7,000). We expect the full survey will be in the field for three weeks.
Regarding the pilot test described above, this is to account for the fact that some aspects of the
rule changed in the past year. To obtain evidence of sufficient quality to support the rulemaking,
the testing must include new questions that were not included in past qualitative testing. It is not
Page 2 of 5

30-day Federal Register Notice Version

consistent with research best practice to include new items in a large survey without first
evaluating them either some kind of pilot testing. Given this, in addition to and preceding the
“soft launch” the Bureau intends to pilot new questions on a small group of 200 respondents
from the GfK panel, evaluate these questions for effectiveness, and decide whether to include
them in the final survey instrument. This is consistent with the spirit of PRA in that ineffective
questions can be removed or refined in order to decrease the burden to the remaining
respondents. The Bureau does not anticipate any changes made during the pilot will have PRA
implications, as changes will consist of refining wording or excluding ineffective items, and not
any substantive changes.
1

Ratcliffe, C., et al. (2014) Delinquent Debt in America. Urban Institute Report, July 30, 2014. Available at:
http://www.urban.org/research/publication/delinquent-debt-america.
2

CFPB (2014) Consumer credit reports: A study of medical and non-medical collections. Consumer Financial Protection Report, December 2014.
Available at: http://files.consumerfinance.gov/f/201412_cfpb_reports_consumer-credit-medical-and-non-medical-collections.pdf

Page 3 of 5

30-day Federal Register Notice Version

3. Describe methods to maximize response rates and to deal with issues of non-response.
The accuracy and reliability of information collected must be shown to be adequate for
intended uses. For collections based on sampling a special justification must be provided
for any collection that will not yield “reliable” data that can be generalized to the universe
studied.
Increasing Response Rate
To facilitate survey response, we will provide e-mail and telephone help desk support to
respondents. All help desk staff will receive training that includes a handbook containing copies
of relevant materials, such as the survey, an outline of the purpose and rationale of the survey,
the process for addressing participant issues, and a summary of common questions.
GfK offers incentives to encourage panel participation. In addition, we will offer an extra $5
incentive to ensure that we recruit our population of interest (ex: oversample people with debt
collection experience) to complete the survey. This incentive will be offered toward the end of
the period the survey is in the field. and is expected to boost the completion rate by 10%.
Overall we expect the completion rate to be 60-70%, based on GfK’s historical completion
rates for surveys of similar length.
Weighting
We will not publicly release non-response analysis for the survey respondents, or treat the data as
nationally representative in publicly released reports. For this study, we plan to conduct an
experimental research design and compare differences between groups to learn which debt
collection disclosure forms work best to improve comprehension and how decisions around debt
collection may vary based on the particular form respondents receive.

4. Describe any tests of procedures or methods to be undertaken.
The survey instrument underwent cognitive testing before the original 30 day PRA notice. New
questions have been added since then and additional pre-testing will be used to assess the
effectiveness of these new questions, as well as consultation with Bureau research experts.
As mentioned previously, we will include a soft launch to test the functionality of the technical
administration of the survey and ensure data quality.

Page 4 of 5

30-day Federal Register Notice Version

5. Provide the name and telephone number of individuals consulted on statistical aspects of
the design and the name of the agency unit, contractor(s), grantee(s), or other person(s)
who will actually collect and/or analyze the information for the agency.

BCFP

Contractors

Melissa Knoll
Decision Scientist; Section Chief
202-435-7282

Randal ZuWallack
Project Director
(802) 264-3724

Erik Durbin
Senior Economist
(202) 435-9243

Michael Long
Principal
(301) 572-0945

Brianna Middlewood
Decision Scientist; Project Lead
202-435-9356

Dr. Robert Tortora
Chief Methodologist
(301) 572-0351

Mick Couper
Visiting Scholar and Survey Expert

Melissa Cidade
Survey Methodologist
(301) 572-0867
Larry Luskin Vice President
(301) 572-0334
Andrew Dyer Principal
(802) 264-3726

###

Page 5 of 5