Consumer Fraud Survey - SS Part A - 2017 - FINAL

Paperwork Reduction Act Supporting Statement for a
FTC Survey on Consumer Fraud
(Requested Reinstatement OMB Control No. 3084-0125)
The Federal Trade Commission (“FTC” or “Commission”) proposes to conduct a fourth
survey about consumer experiences with consumer fraud. The Commission has conducted
three prior surveys designed to estimate the prevalence of consumer fraud among U.S. adults.
The most recent survey was conducted between November 2011 and February 2012. A report
describing the findings of that survey – Consumer Fraud in the United States, 2011: The Third
FTC Survey – was released in April 2013 and can be found at
https://www.ftc.gov/sites/default/files/documents/reports/consumer-fraud-united-states-2011-thir
d-ftc-survey/130419fraudsurvey_0.pdf.1 As with the prior surveys, the results of the proposed
survey will be used by the FTC to estimate the incidence of various types of fraud in the general
population and among various demographic groups. The survey will also help the FTC target
its enforcement and consumer education efforts concerning consumer fraud.
After the 2011 survey was conducted, at the FTC’s request, the Office of Management
and Budget (“OMB”) granted discontinuation of the prior clearance extended to this survey.
The FTC now requests, however, that OMB reinstate that clearance with changes, to conduct the
follow-up survey.
A.

JUSTIFICATION
1. & 2.

Necessity for Information Collection/How the Data Will Be Used

The FTC proposes to survey 3,700 consumers in order to gather specific information on
the incidence of consumer fraud in the general population.2 All information will be collected
on a voluntary basis and no identifying information about the participants will be collected.
Subject to OMB approval for the survey, the FTC has contracted with a consumer research firm
to identify consumers and conduct the survey. The results will assist the FTC in determining
the incidence of consumer fraud in the general population and whether the type and frequency of
consumer fraud is changing, and will inform the FTC about how best to combat consumer fraud.

1

The first consumer fraud survey was conducted in May and June of 2003. The results of that survey are reported in
“Consumer Fraud in the United States: An FTC Survey”
(https://www.ftc.gov/sites/default/files/documents/reports/consumer-fraud-united-states-ftc-survey/040805confraudr
pt.pdf). The 2005 survey was conducted in November and December of 2005 and the findings of that survey are
reported in “Consumer Fraud in the United States: The Second FTC Survey,”
(https://www.ftc.gov/sites/default/files/documents/reports/consumer-fraud-united-states-second-federal-trade-commi
ssion-survey-staff-report-federal-trade/fraud.pdf).
2

An additional 100 non-respondents to the main survey (those who refuse to participate in the survey or with whom
we did not have contact) will be asked to answer a small number of questions to permit us to see whether
non-respondents differ from respondents in their responses. See Part B section 3 (“Methods to Maximize Response
Rates / Reliability of Sample Data”) below.

As was done in the earlier surveys, the proposed survey will oversample demographic
groups that the earlier surveys found to be at an elevated risk of becoming victims of consumer
fraud, including Hispanics and African Americans. The purpose of the oversampling is to
acquire information on what additional factors affect victimization within those demographic
groups and which frauds they are most likely to experience.
The questions will be similar to those in the 2011 survey. All of the frauds in the 2011
survey are included in the current one. However, there have been minor changes in the
questions related to a few of the frauds to enhance the clarity of the questions and the findings of
the survey. In addition, questions have been added to explore four additional types of fraud that
were not included in the 2011 survey. To make room for these added questions, questions
related to some of the characteristics explored in the 2011 survey have been removed.3
3.

Information Technology

Though use of electronic media to conduct the survey is theoretically possible, it would
be infeasible. One frequently used method of conducting a survey via electronic media
involves the use of the Internet and participants in a contractor’s Internet panel. This, however,
is unsatisfactory for the proposed survey for at least two reasons. First, changing the mode used
to collect data could lead to differences in the resulting estimates that do not reflect actual
differences in the underlying things being measured. Second, questions remain about the
appropriateness of using the Internet to collect data where the goal of the project is to estimate
the prevalence of fraud victimization in the population as a whole.4
Before conducting the prior survey in 2011, the Commission explored the feasibility of
using an automated interactive technology – either interactive voice recognition (“IVR”) or
telephone dial entry (“TDR”) – to conduct the survey. Rather than reducing respondent burden,
one reason to use IVR or TDR technology would be to reduce the cost of administering the
survey. Also, there have been some suggestions in the literature that consumers may provide
3

The changes being made are discussed in more detail in response to A. 15 (“Program Changes or Adjustments”)
below.
4

See, e.g., Dillman, Don A., Glenn Phelps, Robert Tortora, Karen Swift, Julie Kohrell, Jodi Berck, and Benjamin L.
Messer, “Response rate and measurement differences in mixed-mode surveys using mail, telephone, interactive
voice response (IVR) and the Internet,” Social Science Research, 38 (2009), p. 2. (“Internet access in the US has
been increasing with about 67% of American adults (18 and older) having access to the Internet access from home in
March 2007, but this coverage is not sufficient for general public surveys.”) By 2015, the percentage of U.S.
adults who reported using the Internet had risen to 84 percent. However, the percentage of U.S. adults using the
Internet is lower for older consumers, with only 58 percent of those 65 and older estimated to use the Internet.
Usage rates are also lower among consumers with less education – only 66 of those who have not graduated from
high school were estimated to use the Internet – and in households with lower incomes – 74 percent of those with
incomes below $30,000 were users. (Andrew Perrin and Maeve Duggan, “Americans’ Internet Access: 2000 –
2015,” Pew Research Center, June 2015)

2

more honest answers when the questions are asked by an impersonal computer than when the
questions are asked by a real person as in a traditional telephone interview.5 Studies, however,
have also found that response rates are lower when automated systems, rather than people, are
used to administer surveys.6 Ultimately, the Commission concluded that the benefits of using
IVR or TDR technology to administer the fraud survey were more than offset by the reduced
response rates that would likely result from using these technologies.
While actual persons will be used to conduct the interviews, the contractor will use CATI
(Computer Assisted Telephone Interviewing) technology to assist the process.
4.

Efforts to Identify Duplication/Availability of Similar Information

Since the early 1990s, numerous studies have attempted to estimate the prevalence of
consumer fraud.7 However, none of these studies provides current data covering the types of
fraud to be covered in the proposed FTC survey.8 There is no current information available
elsewhere that can be used to explore and compare consumers’ fraud experiences with the types
of fraud covered by the proposed survey.
5.

Efforts to Minimize Small Organization Burden

Not applicable.
6.

Only individual consumers will be interviewed.

Consequences to Federal Program and Policy Activities/Obstacles to
Reducing Burden

If this information is not collected, the FTC will lack information to address important
issues and to more effectively target future law enforcement and consumer education actions.
The survey scope and burden has been reduced as much as possible short of sacrificing the
statistical value of the information to be collected.
5

See, e.g., Kreuter, Frauke, Stanley Presser, and Roger Tourangeau, “Social Desirability Bias in CATI, IVR, and
Web Surveys,” Public Opinion Quarterly, 72 (Special Issue 2008), pp. 847-865.

6

See, e.g., Kreuter, Presser, and Tourangeau; Dillman, Don A., Glenn Phelps, Robert Tortora, Karen Swift, Julie
Kohrell, Jodi Berck, and Benjamin L. Messer, “Response Rate and Measurement Differences in Mixed-Mode
Surveys Using Mail, Telephone, Interactive Voice Response (IVR) and the Internet,” Social Science Research 38
(March 2009), pp. 1-18; and Tourangeau, Roger, Darby Miller Steiger, and David Wilson, “Self-Administered
Questions by Telephone,” Public Opinion Quarterly, 66 (Summer 2002), pp. 265-278.
7

For a review of existing studies, see Keith B. Anderson, “Mass-Market Consumer Fraud: Who is Most Susceptible
to Becoming a Victim?,” Federal Trade Commission Bureau of Economics, Working Paper 332, September 2016.
8

The most recent survey was conducted in 2012 and looked at financial frauds. (See FINRA Investor Education
Foundation, “Financial Fraud and Fraud Susceptibility in the United States: Research Report from a 2012 National
Survey,” September 2013.

3

7.

Circumstances Requiring Collection Inconsistent with Guidelines

The collection of information in the proposed survey is consistent with all applicable
guidelines contained in 5 C.F.R. § 1320.5(d)(2).
8.

Public Comments/Consultation Outside the Agency
a.

Public Comments

As required by 5 C.F.R. § 1320.8(d), the FTC published a notice seeking public comment
on the proposed collections of information. See 81 Fed. Reg. 18,628 (March 31, 2016). Three
comments were received.9
A comment from Douglas M. Brooks, dated June 13, 2016, suggested that the survey
should be designed to “reflect the incidence of fraud in vulnerable populations, including
non-English speaking persons and undocumented residents.” As in prior surveys, the proposed
survey will be conducted in both English and Spanish. In addition, questions specifically
developed by the U.S. Census Bureau are being added to the survey to determine whether the
person being interviewed is a U.S. citizen or has recently immigrated to the United States.
A comment from Vin Dwyer, dated March 26, 2016, asked that the survey be structured
to gather information about fraud that is the result of telemarketing calls to landline phones. As
have prior surveys, the proposed survey will collect information about experiences of fraud that
are promoted by a wide variety of communication methods, including landline telephone.
A comment from William Keep, John Breyault, and Peter Vander Nat (“Keep et al.”),
dated June 11, 2016, requested greater clarity about changes in the data regarding victims of
pyramid schemes. The commenters noted that a change was made in the definition of who was
counted as being a victim of a pyramid scheme between the first FTC fraud survey, which was
conducted in 2003, and the second survey, conducted in 2005. As they noted, this change and a
discussion of the impact on the resulting estimates was discussed in the 2007 report that
described the findings of the second survey. However, this change was not noted in the
subsequent report on the 2011 survey, and the commenters were concerned that someone who
looked only at the first and third reports might miss the change. While it seems somewhat
unlikely that someone reading any report from the currently proposed survey would simply
compare those figures to the first survey, which was conducted more than a dozen years before
the proposed one, the Commission will seek to make clear that changes have been made.

9

The comments are available at https://www.ftc.gov/policy/public-comments/initiative-658.

4

Keep et al. also point out that a significant percentage of consumers only have cell
phones and that it may be necessary to include calls to cell phones in conducting a telephone
survey. Indeed, the 2011 survey included cell phones in the sample design. Moreover, in the
proposed survey, 70 percent of interview calls will be to consumers using cell phones.
Keep et al. also express concern that some researchers who have used the information in
the prior surveys have failed to note when differences between survey results are not statistically
significant. When the FTC has made such comparisons, it has indicated whether the differences
are significant or not and it will endeavor to do so for any future reports.
Finally, Keep et al. express concern about the FTC’s discontinuing, after the initial 2003
survey, to collect and report on whether victims of the frauds covered by the surveys have
complained about their experiences. In fact, FTC staff have collected data on complaining
behavior in each of the later surveys and will do so for the proposed survey. Unfortunately,
FTC staff have not had sufficient time to analyze these data since the initial report but hope to do
such an analysis in the future.
Pursuant to the OMB regulations that implement the PRA (5 CFR Part 1320), the FTC is
providing a second opportunity for public comment while seeking OMB approval for the
proposed study.
b.

Consultation Outside the Agency

The design of the prior surveys, on which the current questionnaire is based, were
reviewed by FTC’s survey consultant, Manoj Hastak, Ph.D., Professor of Marketing at American
University’s Kogod College of Business Administration in Washington, DC.
Moreover, the methodology of the earlier surveys, which, again, forms the basis of the
current survey, have been well received by academia and government agencies. For example,
the results of the 2005 survey were presented at a meeting on “Elder Mistreatment and Abuse
and Financial Fraud,” held at the National Academy of Sciences in June of 2010, where it was
favorably reviewed by Professor Roger Tourangeau, Director of the Joint Program in Survey
Methodology, at the University of Maryland. The 2003 survey was cited in a commentary in
the Fall 2004 issue of the University of Maryland Law Journal of Race, Religion, Gender &
Class. Also, the study was a reference for research by the Canadian Competition Bureau and it
was mentioned in a British report for the Organisation for Economic Co-operation and
Development. The 2003 survey was also reviewed by officials with the U.S. Department of
Justice and the AARP.
Comments from the Consumer Financial Protection Bureau and the Federal
Communications Commission were carefully considered. The FTC incorporated several
suggestions to capture more useful data from the survey instrument.

5

9.

Payments or Gifts to Respondents

Not applicable.
10. & 11.

Assurances of Confidentiality/Matters of a Sensitive Nature

Survey participants will not be asked to provide any personally identifying information.
The FTC’s contract provides that Gallup, the survey contractor, will not deliver to the FTC any
information about the identity of individual respondents (e.g., their phone numbers). The FTC
will receive only de-identified demographic data about survey respondents.
The FTC’s Chief Privacy Officer has reviewed the draft survey and, in accordance with
the applicable Privacy Impact Assessment (PIA) for surveys of this type, see
https://www.ftc.gov/system/files/attachments/privacy-impact-assessments/consumer_survey_pia
09152016.pdf, the contractor shall maintain any identifiable data or identifiers separately from
any other data collected from individual survey respondents. Additional restrictions the FTC
places on collection, use, disclosure, and retention of data in order to mitigate privacy risks are
discussed in the PIA. For example, the survey is designed to limit the collection of data to basic
demographic information (such as age, income, state, and race/ethnicity), and to avoid the
collection of any highly sensitive personal data, which is not necessary to fulfill the FTC’s
survey needs. The FTC’s survey contract prohibits the contractor from using the data for any
other purpose. The FTC’s survey contract also requires that the contractor comply with
specified information technology security and data handling requirements.
Gallup operates in a proactive security posture using a variety of preventative, detective,
and administrative information security controls. Gallup bases its information security
framework on industry and Federal standards: the International Standards Organization’s
IEC27001/27002 and the U.S. Government’s Special Publication 800-53 of the National Institute
of Standards and Technology.
Because the FTC’s survey contract does not contemplate the maintenance or retrieval of
any records about individuals by name or other personally assigned identifier by either the
contractor or the FTC, the Privacy Act of 1974 is not expected to apply and no System of
Records Notice applies. Nevertheless, the FTC’s survey contract requires that the contractor
provide all survey respondents with appropriate notice of the authority, purpose, and use of their
information and, pursuant to the nondisclosure agreement and other restrictions set forth in the
FTC’s survey contract, the data shall remain protected from public disclosure to the extent
permitted by law. See, e.g., Freedom of Information Act Exemption 6, 5 U.S.C. 552(b)(6).
All data received by the FTC will be maintained on a Commission drive that is backed-up
nightly with access limited to those who are directly working on the survey project.

6

12.

Estimated Annual Hours Burden

The FTC will pretest the survey on approximately 100 respondents to ensure that all
questions are easily understood. This pretest will take approximately 20 minutes per person on
average and 33 hours as a whole (100 respondents x 20 minutes each).10 Answering the final
survey will require approximately 15 minutes per respondent on average and 900 hours as a
whole (3,600 respondents x 15 minutes each).11 Additionally, 100 interviews will be conducted
with non-respondents – those who refused to participate or those with whom we did not have
contact. (See B.3 below “Methods to Maximize Response Rates/Reliability of Sample Data.”)
These interviews will take approximately 5 minutes each, for a total burden of 8 hours. Thus,
cumulative total burden will be approximately 941 hours.
13.

Estimated Annual Cost Burden

The cost per respondent should be negligible. Participation is voluntary, and will not
require any labor expenditures by respondents. There are no capital, start-up, operation,
maintenance, or other similar costs to the respondents.
14.

Estimated Cost to the Federal Government

The total cost to the Federal government for the information collection will be
approximately $263,500. To obtain a contractor to review the survey questionnaire, identify the
consumers, conduct the surveys, and provide the resulting data to FTC staff will cost $244,400.
In addition, the Commission staff time necessary to identify a contractor and to assist the
contractor in completing its duties is estimated to require approximately 300 staff hours and cost
approximately $19,100. The cost of FTC staff time is necessarily an estimate because several
factors in this calculation may vary, including the number of staff involved and the actual
amount of time required. Clerical and other support services and costs of conducting the study
are included in this estimate.
15.

Program Changes or Adjustments

Questions are being added to permit the Commission to learn about four types of fraud
that were not included in the prior surveys. These are (i) being victimized by a mystery shopper
scam where someone promises you work as an undercover shopper who is supposed to report to
a retailer’s headquarters about your experience shopping in one of their stores, (ii) being
victimized by someone who falsely claimed that there were problems with your computer and
10

The pre-test is expected to take five minutes longer on average than the final survey because the pre-test has several
places where open-ended probes have been added to assist in refining the survey instrument.

11

The contract for the present survey calls for a survey that will not, on average, exceed 15 minutes. The FTC and
Gallup, the contractor, will verify this during the pre-test interview process and will shorten the interview if necessary
to meet the 15-minute average time.

7

that they could fix them remotely, (iii) being victimized by someone who falsely claimed that
you owed back taxes or were the subject of a court case, and (iv) being charged for something
for your cell phone – such as ringtones or an app – that you had not agreed to purchase. In
recent years, the Commission has brought numerous law enforcement actions against persons
who are operating the first two types of fraud. Further, a large number of complaints about
several of these types of frauds have been received in the FTC’s Consumer Sentinel database of
consumer complaints.12 None of the frauds that were included in the 2011 survey are being
eliminated from the proposed questionnaire.
In addition to adding questions about additional frauds, a few other changes have been
made. The questions about how the consumer paid for the fraudulent product or service have
been revised to provide additional detail about payment methods. An additional question was
added to the questions about prize promotions to learn more about what the victim had been
promised. The questions about bad checks have been reorganized, and the wording of the
questions about items for which the consumer may have been billed though they had not agreed
to purchase have been revised slightly to correct a problem that arose in the prior survey.
Several changes were also made in the sections of the survey that ask about
demographics and other characteristics that may be related to being a victim of a mass market
consumer fraud. Two sets of questions about consumer characteristics – questions about how
the participant perceived themselves on a series of personality attributes and questions designed
to measure basic mathematical skills – have been eliminated. There is more interest at the
Federal Trade Commission in having additional information about the prevalence of different
types of frauds than in further exploring the consumer characteristics that may be related to
becoming a victim. And, reductions in other questions are needed to accommodate the
questions about additional types of fraud discussed above.13 Furthermore, these variables had
limited explanatory power in analyzing the results of the prior survey.

12

A total of 228,066 complaints received in the Consumer Sentinel database during 2015 were classified in the
category “Imposter: Government.” This was 7.4 percent of all complaints received in Consumer Sentinel during
2015. A review of a sample of these complaints suggests that the vast majority involve claims that the caller is
from the Internal Revenue Service and the person being called owes back taxes or that the person being called is
being sued for other reasons. Complaints in the category “Tech Support Scams” totaled 39,924, 1.3 percent of
complaints received during 2015. Consumer Sentinel also shows 1,174 complaints in the category “Mobile
Telephone: Applications / Other Downloads” and 1,913 in the category “Mobile Telephone: Unauthorized Charges
or Debits” during 2015. (Federal Trade Commission, “Consumer Sentinel Network Data Book for January December 2015,” February 2016, Appendix B-3.) The Consumer Sentinel system does not include a separate
category for mystery shopper scams, and therefore, it is not possible to determine how many complaints have been
received about this type of fraud.

13

The “accommodation” is a recognition of both practical time limits on how long consumers will be willing to spend
on a survey such as this and contractual ones regarding the terms of the contract that have been entered with the
contractor who will conduct the survey if OMB clearance is granted.

8

Questions that asked about marital and employment status were also deleted because they
had not been particularly useful in the analysis of the results from prior surveys. It was also not
clear that these questions were measuring anything meaningful given changes in consumers’ life
styles and work situations.
Two demographic questions were added. A question about the ZIP code of the
participant’s residence was added because of some recent preliminary work at the FTC that
raises the possibility of geographic differences in victimization rates. The ZIP code data will
allow for the possibility of performing similar analysis with the fraud survey data. A question
was also added asking participants about their self-perception of their socio-economic status
relative to other members of society. The role of self-perception of socio-economic status has
been explored in at least one recent study of fraud victimization.14
Because the total number of types of fraud included in the current survey will be greater
than what was included in the earlier survey, it will not be possible to directly compare the total
victimization figures for the two surveys. However, because none of the frauds from the earlier
survey are being eliminated, it will be possible to construct a measure of total victimization
based just on the frauds included in the 2011 survey and compare that to the 2011 figures. It
will also be possible to compare the findings for specific frauds that are included in both
surveys.15
16.

Plans for Tabulation and Publication

The results of the surveys will be used to inform the FTC about the incidence of
consumer fraud in the general population. The summary of findings will be released to the
public to inform the efforts of other consumer protection agencies, including other federal
agencies. The collection of the information would begin subject to OMB approval after it
completes its review of this clearance request.
17.

Display of Expiration Date for OMB Approval

Not applicable.
18.

Exceptions to Certification

Not applicable.

14

See Michaela Beals, “The Impact of Survey Context on Self-Reported Rates of Fraud Victimization,” Financial
Fraud Research Center, Stanford Center on Longevity, October 2014.

15

While the minor changes in the questions about some of the types of fraud will mean that some caution will be
needed in making comparisons, we expect those effects to be fairly minor.

9