2017 SFS OMB Supporting Statement Part A_FINAL

Supporting Statement – 2017 National Crime Victimization Survey (NCVS) Supplemental Fraud Survey (SFS)

A. Justification

1. Necessity of the Information Collection

The Bureau of Justice Statistics (BJS) requests clearance to conduct the 2017 Supplemental Fraud Survey (SFS) to the National Crime Victimization Survey (NCVS). The BJS is authorized to collect statistics on victimization under Title 42, United States Code, Section 3732 of the Justice Systems Improvement Act of 1979 (see Attachment 1). The NCVS and all related contacts and protocols for the 2017 collection year were approved separately by OMB (OMB NO: 1121-0111). This request is specifically for a supplemental data collection instrument that will be added to the approved NCVS core from October through December of 2017 (see Attachment 2). The SFS is primarily an effort to measure the prevalence of personal financial fraud victimization among persons ages 18 or older, the prevalence of a range of different types of fraud, characteristics of fraud victims, and patterns of reporting to the police and consumer protection agencies. The SFS was also designed to collect important information on the consequences of fraud, including financial losses, and victims’ reactions to specific fraud victimization experiences. The 2017 administration will be the first for this data collection.

Research suggests that financial fraud is a major problem for individuals and for society. Estimates suggest that annual losses associated with fraud total more than $50 billion and that fraud prevalence rates may be as high as 17%.^a However, our understanding of the scope of the problem has been hampered by a lack of valid, national statistics. Key sources of crime statistics in the United States, including the NCVS and the Federal Bureau of Investigation's Uniform Crime Reports (UCR), have historically focused on traditional violent and property crimes and have not previously attempted to measure the prevalence of financial fraud. Some organizations (e.g., Better Business Bureau, Internet Crime Complaint Center, Federal Trade Commission, and Consumer Financial Protection Bureau) have collected data on financial fraud for the purposes of consumer protection, criminal prosecution, or financial restitution, but, these data are either not representative of the U.S. population, or key subgroups within the population, or do not cover all major types of fraud. These data cannot be used to generate prevalence estimates of financial fraud in the United States.

Researchers and practitioners working in fraud prevention, awareness, and consumer protection have highlighted the need for national estimates of the prevalence and consequences of fraud. Researchers from the Financial Fraud Research Center (FFRC), a joint project of the Stanford Center on Longevity and the FINRA Investor Education Foundation (FINRA Foundation), noted that the lack of consistency in defining fraud and lack of data measuring fraud “has led to a proliferation of overlapping and often confusing definitions and categorizations that affect fraud prevalence estimates as well as our understanding of the mechanisms and consequences of fraud.”^b Other groups have focused on the need for data on financial impact of fraud on certain high risk subgroups. For instance, the Elder Justice Initiative (EJI), which coordinates elder abuse, neglect and financial exploitation programmatic efforts within the U.S. Department of Justice (USDOJ), has expressed a need for data on financial fraud victimization committed against persons 65 or older.

In 2016, the National Academy of Sciences (NAS) released a report recommending that BJS turn a focus toward measuring new and emerging crimes and not just the current street crimes already included on the NCVS.^c One strength of the NCVS is its ability to capture hard-to-measure crimes. A self-report survey like the NCVS is a key tool for capturing personally sensitive crimes such as domestic violence and rape or sexual assault, as well as crimes like fraud that have a low likelihood of being reported to police or other agencies.^d Measuring financial fraud victimization via a NCVS supplemental survey will allow for assessment of the level and types of fraud that are occurring annually, associated financial impacts, and the proportion that come to the attention of police and other agencies that serve to prevent future occurrences.

BJS is requesting a one-year OMB clearance from September of 2017 through September of 2018 to collect data on the prevalence of financial fraud victimization in the United States. Data collection will be conducted from October through December of 2017. During the 3-month data collection period, the supplement will be administered to all NCVS respondents age 18 or older, following the completion of the NCVS screener (NCVS-1) and the NCVS crime incident report (NCVS-2; if applicable NCVS crimes were reported).

2. Needs and Uses

One of the impediments to the inclusion of fraud in national data collections on crime has been the lack of a clear definition for the term “fraud.” Without a way to systematically categorize fraud, the relatively limited prior research on fraud resorted to classifying on different characteristics of incidents, including communication method (e.g., cyber fraud, mail fraud), product marketed (e.g., lottery fraud, securities fraud), group targeted (e.g., elder fraud), and/or offender relationship to the victim. This resulted in overlapping and often confusing definitions of fraud that created a muddled understanding of the level and consequences of fraud.

To address the need for a fraud classification system, the Financial Fraud Research Center (FFRC) collaborated with BJS to develop a standardized fraud classification scheme. The purpose was to group and organize fraud types meaningfully and systematically into a definitional framework that could be translated into survey questions that could be administered as a supplement to the NCVS.

Using the taxonomy as the basis for instrument development, BJS, working in collaboration with the FFRC, developed an instrument to measure the key categories and attributes of financial fraud. The resulting instrument was designed to measure the annual prevalence of seven types of financial fraud – consumer investment fraud, consumer products and services fraud, employment fraud, prize and grant fraud, phantom debt fraud, charity fraud, and relationship and trust fraud – and to capture more detailed information about the fraud incident experienced most recently, including –

• Information needed for coding detailed fraud types based on the taxonomy

• Mode of initial contact

• Method used for transferring funds

• Monetary losses

• Victim reporting behaviors

The SFS will provide national-level data on the prevalence and nature of personal financial fraud victimization. In addition, due to the expected prevalence rate and redesigned state sampling plan, it is likely that state estimates of fraud victimization may be produced for the largest 22 states. The data being collected through the SFS are needed to more fully understand financial fraud in the United States and to obtain a clearer picture of its impact on society and consequences suffered by victims. Most importantly, the SFS will capture both financial fraud reported and not reported to the police or other authorities. Understanding this “dark figure” helps to inform victim outreach efforts, resource allocation, and to gain a better understanding of victim decision-making and the resulting consequences. For example, research has demonstrated an association between reporting crimes to the police, receiving victim services, and being involved in the criminal justice process.^e The findings from the SFS will not only be beneficial to the general public by increasing awareness of this crime but they also will have significance for legislators, policymakers, and law enforcement in making sound decisions regarding these criminal acts and providing assistance to its victims.

Uses of SFS data

Table 1 below details the estimates that will be produced with the 2017 SFS data. With the 2017 SFS data, BJS will be able to examine prevalence estimates of personal financial fraud victimization; characteristics of fraud victims; emotional and financial impacts on victims; reporting fraud victimization to police and consumer protection agencies; and reasons for not reporting fraud victimization to the police and consumer protection agencies.

Table 1. Types of estimates that can be generated from the 2017 SFS

Estimates that can be generated from the 2017 SFS	Relevant questions
Percent of persons age 18 or older who experienced financial fraud victimization in the past 12 months	S1a, S1b, S1c, S2, S3, S4c, S4c1, S4c2, S4d, S4d1, S4d2, S5c1, S5cx, S5c2, S6c2, S6d, S6d1, S6d2, S7
Percent of persons age 18 or older who experienced specific types of financial fraud in the past 12 months, including
Prize or grant fraud Phantom debt collection fraud Charity fraud Employment fraud Consumer investment fraud Consumer products or services fraud Relationship or trust fraud	S1a, S1b, S1c, S2, S3, S4c, S4c1, S4c2, S4d, S4d1, S4d2, S5c1, S5cx, S5c2, S6c2, S6d, S6d1, S6d2, S7
Demographic characteristics of persons age 18 or older who experienced fraud victimization in the past 12 months	NCVS core + S1a, S1b, S1c, S2, S3, S4c, S4c1, S4c2, S4d, S4d1, S4d2, S5c1, S5cx, S5c2, S6c2, S6d, S6d1, S6d2, S7
Amount of money victim lost in the fraudulent transaction, by type of fraud victimization	S1B3, S1B5, S2B2, S2B4, S3B3, S3B5, S4B4, S4B6, S5B2, S5B4, S6B6, S6B8, S7B4, S7B6
Percent of fraud victimizations reported to police, by type of fraud victimization	S1B6, S2B5, S3B6, S4B7, S5B5, S6B9, S7B7
Reason(s) for not reporting fraud victimization to police, by type of fraud victimization	S1B7, S2B6, S3B7, S4B8, S5B6, S6B10, S7B8
Percent of fraud victimizations reported to consumer protection agency, by type of fraud victimization	S1B8e, S1B8f, S1B8g, S1B8h, S1B8i. S2B7e, S2B7f, S2B7g, S2B7h, S2B7i, S2B7j, S3B8e, S3B8f, S3B8g, S3B8h, S3B8i, S4B9e, S4B9f, S4B9g, S4B9h, S4B9i, S5B7e, B5B7f, S5B7g, S5B7h, S5B7i, S5B7j, S6B11e, S6B11f, S6B11g, S6B11h, S6B11i, S7B9d, S7B9e, S7B9f, S7B9g, S7B9h
Reason(s) for not reporting fraud victimization to consumer protection agency, by type of fraud victimization	S1B8k, S2B7l, S3B8k, S4B9k, S5B7l, S6B11l S7B9j
Financial and work or school loss (direct and indirect) attributed to the fraud victimization, by type of fraud victimization	S1B9, S1B10, S1B11, S1B12, S2B8, S2B9, S2B10, S2B11, S3B9, S3B10, S3B11, S3B12, S4B10, S4B11, S4B12, S4B13, S5B8, S5B9, S5B10, S5B11, S6B12, S6B13, S6B14, S6B15, S7B10, S7B11, S7B12, S7B13

The estimates that can be generated through the SFS are needed by a wide range of government agencies, consumer protection bureaus, and victim advocacy groups, as well as to provide the general public with reliable data on the prevalence and characteristics of personal financial fraud. The paragraphs below provide examples of potential users and uses of the 2017 SFS statistics.

Government agencies

Bureau of Justice Statistics. The 2017 SFS data will enable BJS to report on “new and emerging” crime types and to expand the array of crime types against persons that are counted as part of national crime statistics (beyond the traditional crime types reported by the FBI). BJS will use the data from the 2017 SFS to produce a report on financial fraud victimization. We anticipate the report covering topics such as the percentage of persons age 18 or older who had experienced any type of financial fraud victimization during the prior year; the percentage of persons age 18 or older who had experienced specific types of financial fraud victimization during the prior year; demographic characteristics of fraud victims; amount of money lost in the fraudulent transaction; and the percentage of fraud victimizations that went unreported to police and other consumer protection agencies.

BJS will disseminate the report through the BJS website and alert the public through a statistical press release. Through AskBJS, the BJS email account that allows data users to ask statisticians specific data questions, BJS will respond to any external requests from the public and media regarding the report findings. BJS will also make the 2017 SFS data available for public use and download through the archives at the University of Michigan’s Inter-University Consortium for Political and Social Research (ICPSR). The 2017 SFS restricted-use data files will be available through the U.S. Census Bureau’s Federal Statistical Research Data Centers (FSRDC).

Other federal agencies. The 2017 SFS data on the prevalence of financial fraud and the characteristics of victims can assist agencies like the Federal Trade Commission (FTC), Securities and Exchange Commission (SEC), Internal Revenue Service (IRS), and U.S. Postal Inspection Service (USPIS) in identifying populations that may be particularly vulnerable, and appropriately targeting awareness and prevention campaigns. The Federal Government would also benefit from data assessing the magnitude of financial fraud in the United States.

Other federal agencies have expressed interest in and uses for the 2017 SFS data. The Elder Justice Initiative within the U.S. Department of Justice has expressed interest in estimates of financial fraud victimization committed against persons 65 or older that can be produced with data from the 2017 SFS.

Victim advocates

The 2017 SFS data may be of interest to victim advocates that are interested in the financial, social, and emotional impacts of financial fraud on victims. These data may assist the Office for Victims of Crime (OVC) and other victim advocacy groups in understanding the impact, seriousness, and harms associated with financial fraud victimization, as well as the specific needs of fraud victims. The information is needed for making decisions regarding the allocation of victim assistance funds and resources to various types of crime victims. Since data from the NCVS and the UCR focus on traditional street crimes, the SFS statistics also provide victim advocates with a more complete picture of the range of victims that may require assistance.

Media outlets and the general public

Findings from the 2017 SFS data may be reported by various news and advocacy organizations. In addition, the resulting report will be the basis for BJS’s responses to public and press inquiries concerning fraud victimization.

3. Use of Information Technology

The 2017 SFS will be conducted in a fully automated interviewing environment using computer-assisted personal interviewing (CAPI) methods whereby field representatives (FRs) use a laptop computer to read questions and record answers. The use of CAPI technologies reduces data collection costs as well as respondent and interviewer burden. Furthermore, automated instruments afford the opportunity to implement inter-data item integrity constraints which minimize the amount of data inconsistency. More consistent data, in turn, reduces the need for extensive post-data collection editing and imputation processes which will significantly reduce the time needed to release the data for public consumption. The use of technology results in more accurate data products that are delivered in a more timely fashion giving data users access to information while it is still relevant.

4. Efforts to Identify Duplication

One contemporary survey currently collects information about financial fraud victimization. The FTC Survey on Consumer Fraud, sponsored by the Federal Trade Commission (FTC), is a nationally based data collection that asks a random sample of the U.S. population about their experiences with consumer fraud, specifically, whether the type and frequency of consumer fraud is changing. The FTC survey is focused on certain types of consumer fraud of which financial fraud is a part while the 2017 SFS aims to collect more detailed data on specific types of financial fraud and the reporting of financial fraud to authorities. The next administration of the FTC survey is expected to be in the field later in 2017. The FTC survey and SFS are both nationally-based collections but there are substantial methodological and substantive differences.

The FTC survey is a national random digit dial (RDD) telephone survey of approximately 3,700 U.S. adults age 18 or older residing in U.S. households. The response rate for the previous administration of the FTC survey (2011/2012 administration) was 14%. The 2017 SFS is a nationally representative sample of U.S. households and anticipates interviewing approximately 79,000 persons age 18 or older. As previously mentioned, this is the first administration of the SFS, therefore, response rates from previous administrations cannot be examined. However, response rates for other recent NCVS supplements (i.e., the 2016 Identity Theft Supplement and 2016 Supplemental Victimization Survey) were approximately 77%.

The SFS is designed to capture information on the seven primary types of personal financial fraud outlined in the FFRC taxonomy. Experts determined that the seven types of personal financial fraud were comprehensive and exhaustive and could be summed to estimate the total prevalence of personal financial fraud. On the other hand, the FTC survey asks respondents about a selection of specific types of fraud, such as government job offers fraud which is one type of employment fraud. The specific types of fraud included in the FTC survey do not sum to a comprehensive estimate of the prevalence of personal financial fraud in the United States. The most recent FTC survey report (2011/2012 administration) includes prevalence estimates for each of the specific types of fraud included in the survey. BJS plans to include prevalence estimates for each specific type of fraud included in the SFS survey as well, but will also provide an overall estimate of the prevalence of personal financial fraud.

Both surveys ask respondents if the fraudulent incidents were reported to police or consumer protection agencies, such as the Better Business Bureau (BBB) or State Attorneys General Office. If the fraud victimization was not reported, the SFS questionnaire includes a question on the reasons for not reporting. This will be useful in determining if respondents didn’t report because they didn’t think it was a crime or didn’t know the type of agency that was best suited for reporting this type of victimization. The FTC survey does not probe on reasons for not reporting to the police or consumer protection agencies.

Going beyond what the FTC survey collects, the SFS also includes questions related to the consequences of the fraud victimization on the victim. Questions focus on the social, emotional, and financial distress that the fraud caused the victim. These questions will help to get a clearer pictures of the different types of harm that impact victims.

5. Efforts to Minimize Burden

The NCVS is a household-based sample and does not impact small businesses or small entities. To minimize the burden for individual respondents and reduce nonresponse rates, supplemental questionnaires like the SFS are designed to take no longer than 15 minutes to administer. Field representatives (FR) will alert respondents to the additional burden from the supplement at the beginning of the SFS interview. The field representatives will be instructed to inform respondents that “From time to time, the Bureau of Justice Statistics of the U.S. Department of Justice collects information on special topics. We are now going to ask you about experiences in which someone convinced you to pay, invest, or donate money, by tricking or lying to you, hiding information, or promising you something that you never received. We estimate these questions will take between 5 to 15 minutes depending on your circumstances.”

The 2017 SFS will be conducted using CAPI. Approximately 45% of the NCVS interviews are conducted face-to-face in the sampled households (including all first interviews, all replacement households and all households requiring personal contact to obtain a response). The remaining 55% of NCVS interviews are collected via telephone.

To help minimize burden, the SFS consists of a screener and an incident interview. Only respondents that report an eligible fraud victimization will receive the incident interview. The screener is approximately 5 minutes and captures information on whether the respondent had experienced behaviors consistent with specific types of personal financial fraud victimization in the past 12 months, how many times they experienced each behavior, and whether they received their money back. The incident instrument asks detailed questions about the amount of money lost in the transaction, whether it was reported to police or a consumer protection agency, the social and emotional distress associated with the victimization, and the financial impact of the fraud victimization on the victim. The screener plus the incident interview is expected to take less than 15 minutes.

6. Consequences of Less Frequent Collection

The 2017 SFS is the first time BJS is collecting information on personal financial fraud victimization. With the development of this questionnaire for 2017, BJS is now in the position to continue to field a consistent recurring supplement on financial fraud. Other supplements to the NCVS, such as the School Crime Supplement (OMB NO: 1121-0184) and the Police-Public Contact Survey (OMB NO: 1121-0260), are typically conducted on a biennial basis. BJS will assess the need and feasibility of conducting the SFS every two years.

7. Special Circumstances

No special circumstances.

8. Adherence to 5 CFR 1320.8(d) and Outside Consultations

The research under this clearance is consistent with the guidelines in 5 CFR 1320.6. Comments on this data collection effort were solicited in the Federal Register, Vol. 82, No. 84, pages 20636-20637 on May 3, 2017 and Vol. 82, No. 130, pages 31785-31786 on July 10, 2017. Public comments were received during the 60 day period. BJS replied via email to each comment. Table 2 details who the comment was from, the nature of the comment, and whether BJS implemented the comment.

Table 2: Public comments on the 2017 SFS instrument

Comment from:	Comment:	Implementation of public comment and BJS response:
Jan Chaiken	Although the topics are explored at length, the burden on respondents is unlikely to be high, as few would experience multiple types of fraud. To minimize burden in those few instances, you could establish priorities for the different types of fraud (or generate random priorities) and then ask details only about the two types (say) of highest priority experienced by the respondent.	Not implementing comment. BJS learned from cognitive testing that the number of people experiencing multiple types of fraud was relatively low. BJS also learned from cognitive testing that the questions on the individual incident forms are straightforward and don’t take much time to answer so the burden is not expected to be high if a respondent endorses multiple types of fraud and has to complete multiple incident forms.
Jan Chaiken	In Question S2 about debt collection, did you consider separating out the incidents where the money was paid to someone the respondent knew (family member, coworker, ex-spouse, etc.)?	Not implementing comment. Because the screener was designed to align with the fraud taxonomy (http://longevity3.stanford.edu/wp-content/uploads/2015/11/Full-Taxonomy-report.pdf), we didn’t consider separating out incidents where the money was paid to someone the respondent knew. In cognitive testing we had a number of reports of incidents involving lending money to family and friends but most of these would not be classified as fraud so we are trying to avoid those false positive responses.
Jan Chaiken	In Question S2: "you were being tricked or lied to and the debt was not real or not yours” should say "or the debt was not real or was not yours”, since the respondent may have spent money that did not result in paying down the debt.	Not implementing comment. Because the screener was designed to align with the taxonomy, we need both of these components of the question in order for the incident to be classified as phantom debt collection fraud. According to the taxonomy, phantom debt collection fraud involves debt that was not real: “expected benefit is avoiding the consequences of failing to pay debts that the victim did not previously know were owed (and that turn out to be fake). Examples include government debt collection scams (court impersonation scam, IRS back taxes owed scam), and other scams in which fraudsters demand payment for false debt owed to lenders or businesses (obituary scam, fake medical debt, fake loan debt collection).” It sounds like you’re describing a situation in which a person attempted to pay down an actual debt but the payments were not put towards that debt. That seems like an example of consumer services fraud (the debt payment service was not provided as promised).
Jan Chaiken	I suggest moving Questions S4 (Job Fraud) after S5 (investment fraud), since the term "business opportunity” may bring to mind investment fraud, but that possibility will be avoided if investment fraud comes first.	Not implementing comment. Based on the early rounds of cognitive testing, we shifted the order of the screeners to move from the more specific fraud types to the broader categories of consumer investment fraud and consumer products and services fraud. While we appreciate the suggestion, we hesitate to change the order without further testing because the investment fraud category could also pick up incidents that should be classified as employment fraud.
Jan Chaiken	Questions S6 and S6z — I suggest changing “lost money by paying” to “paid”, since later questions find out if the respondent got the money back. I’m assuming if the respondent got all the money back without unusual effort, the incident will not be counted as a crime.	Comment implemented. Yes, you are correct in assuming that if the respondent got all their money back without unusual effort the incident would not be considered fraud. BJS made this suggested change.
Jan Chaiken	In S1B7 and elsewhere, there is a list of possible reasons why something was not reported. I suggest adding “DID NOT WANT TO HAVE ANY CONTACT WITH THE POLICE.”	Comment implemented. This is a useful addition and that could be easily implemented without additional testing. BJS will incorporate this as a response option.
Shirley Pribble	A database would be useful in the sense it may make their [police] job a little bit easier with the goal of them [police] actually pursing a proper investigation.	Not implementing comment. With regard to our Supplemental Fraud Survey (SFS), this is not a database that will be used to properly investigate cases of financial fraud. Instead, it is a household survey that will ask respondents a series of questions to determine if they are a victim of financial fraud. BJS will then analyze these data and publish statistics on the amount of financial fraud that is occurring in the United States. A respondent’s answers to the SFS are confidential and will not be tied back to the respondent.

The BJS, U.S. Census Bureau, and outside experts collaborated to develop the final version of the questionnaire and procedures used to collect this supplemental information. For the 2017 SFS, principal consultants from the BJS were Dr. Lynn Langton and Dr. Rachel Morgan. Principal persons consulted from the Census Bureau included Ms. Meagan Meuchel, Ms. Jill Harbison, Mr. David Hornick, Ms. Emily Mohn, and Mr. Steve Bittner. Outside experts included Dr. Marguerite DeLiema from the Financial Fraud Research Center and Dr. Gary Mottola from the FINRA Investor Education Foundation. Dr. Christopher Krebs and Ms. Sarah Cook from RTI International’s Center for Justice, Safety, and Resilience conducted the iterative cognitive testing that shaped the final SFS questionnaire.

The FFRC taxonomy was developed by the Taxonomy of Fraud working group which consisted of fraud and measurement experts representing the government, academic and nonprofit organizations. After the initial taxonomy was developed it was reviewed by an extended review panel consisting of a wider scope of fraud and measurement researchers and practitioners. Input from the extended review panel helped refine the taxonomy by addressing potential areas of overlap or confusion and also reviewed the final report. Working group members included –

• Keith Anderson, Federal Trade Commission

• Robert Anguizola, Federal Communications Commission

• Michaela Beals, Financial Fraud Research Center at the Stanford Center on Longevity

• Martha Deevy, Financial Fraud Research Center at the Stanford Center on Longevity

• Marguerite DeLiema, Financial Fraud Research Center at the Stanford Center on Longevity

• Kristy Holtfreter, Arizona State University

• Dominika Jaworski, Financial Security Division at the Stanford Center on Longevity

• Sara Kern, Gonzaga University

• Christine Kieffer, FINRA Investor Education Foundation

• Lynn Langton, Bureau of Justice Statistics

• Gary Mottola, FINRA Investor Education Foundation

• Michael Planty, formerly of the Bureau of Justice Statistics

• Patricia Poss, Federal Trade Commission

• Michael Reisig, Arizona State University

• Richard Riley, Institute for Fraud Prevention at West Virginia University

• Doug Shadel, AARP Washington

• Roger Tourangeau, Westat

Extended review panel members included –

• Debbie Deem, Federal Bureau of Investigation

• Owen Donley, Securities and Exchange Commission

• Lois Greisman, Federal Trade Commission

• Dana Hermanson, Kennesaw State University

• Michael Herndon, Commodity Futures Trading Commission

• Andi McNeal, Association of Certified Fraud Examiners

• Jerry O’Farrell, U.S. Postal Inspection Service

• Nicole Piquero, University of Texas at Dallas

• Lori Schock, Securities and Exchange Commission

• Terry Thome, U.S. Postal Inspection Service

• Richard Titus, formerly of the National Institute of Justice

• John Warren, Association of Certified Fraud Examiners

• Johan van Wilsem, Leiden University

• Judy van Wyk, University of Rhode Island

9. Paying Respondents

No payment or gifts are provided to respondents in return for participation in the supplement.

10. Assurance of Confidentiality

All NCVS information about individuals or households is confidential by law – Title 42, United States Code, Sections 3789g and 3735 (formerly Section 3771) and Title 13, United States Code, Section 9. Only Census Bureau employees sworn to preserve this confidentiality may see the survey responses. Even BJS, as the sponsor of the survey, is not authorized to see or handle the data in its raw form. All unique and identifying information is scrambled or suppressed before it is provided to BJS statisticians. Data are maintained in secure environments and in restricted access locations within the Census Bureau. All data provided to BJS must meet the confidentiality requirements set forth by the Disclosure Review Board at the Census Bureau.

In a letter signed by the Director of the Census Bureau, sent to all participants in the survey, respondents are informed of this law and assured that it requires the Census Bureau to keep all information provided by the respondent confidential. The letter also informs respondents that this is a voluntary survey. Furthermore, in addition to the legal authority and voluntary nature of the survey, the letter informs respondents of the public reporting burden for this collection of information, the principal purposes for collecting the information, and the various uses for the data after it is collected which satisfies the requirements of the Privacy Act of 1974.

11. Justification for Sensitive Questions

The SFS asks about financial fraud victimization experiences that may be sensitive for some respondents. Given the objective of the SFS – to estimate the amount of financial fraud victimization in the Nation – this is inevitable. The SFS does not ask questions relating to sexual behaviors, drug use, religious beliefs, or other matters commonly considered private or of a sensitive nature. SFS interviewers receive training and guidance on how to ask sensitive questions. The importance of estimating crime levels, as well as the potential value of detailed information about victimization for designing crime prevention strategies, is explained to any respondent who seems hesitant to answer. All respondents have the option of refusing to answer any question.

12. Estimate of Respondent Burden

Only respondents age 18 or older that complete the NCVS-1 and NCVS-2 (if applicable) are eligible to receive the SFS instrument. We estimate that 103,678 NCVS respondents age 18 or older will be eligible to be interviewed for the SFS between October and December of 2017. We estimate each SFS screening interview will take, on average, 0.08 hours (5 minutes) and each full interview (screener plus incident interview) for persons experiencing fraud victimization will take, on average, 0.25 hours (15 minutes) to complete. We expect that about 12% of respondents will be victims of fraud victimization in the 2017 SFS. This prevalence estimate is based on results from the 2011/2012 FTC survey which reported 11% of U.S. adults experienced one or more types of frauds covered in the survey.^f This survey was conducted more than five years ago and is limited in the types of fraud covered compared to the 2017 SFS so we expect that 12% is a conservative prevalence estimate.

The final burden estimate assumes that the total NCVS sample from October through December of 2017 will be approximately 57,599 households yielding approximately 103,678 persons age 18 or older. Based on the response rates for the 2016 NCVS Identity Theft Supplement (ITS) and 2016 NCVS Supplemental Victimization Survey (SVS), we expect that about 77%, or 79,832 of the 103,678 eligible respondents will be interviewed.^g Based on prevalence estimates from the 2011/2012 FTC survey, it is expected that 12% of the 79,832 interviewed respondents will be fraud victims and therefore follow the long interview path in the questionnaire. The remaining 88% will not be victims of fraud and, as such, will follow the short interview path. As stated above, our assumption is that the short interview path will take about 5 minutes and the long interview path will take 15 minutes. Total expected respondent burden is therefore estimated to be 8,015 hours (see Table 3 for calculation).

Table 3. SFS estimated burden hours

	Number of SFS Persons (A)	Time per interview (hours) (B)	Burden hours (AxB)
Total Expected SFS Eligible Persons	103,678
Expected SFS Interviews	79,832
Expected SFS Short Interviews	70,252	.08	5,620
Expected SFS Long Interviews	9,580	.25	2,395
Expected SFS Noninterviews	23,846
2017 SFS Burden Hours Estimate			8,015

13. Estimate of Respondent’s Cost Burden

There are no costs to respondents other than their time to respond.

14. Costs to Federal Government

There are no capital or start-up costs associated with this data collection.

BJS Cost Estimate Summary

The estimated cost for BJS staff is $64,178 for the 2017 Supplemental Fraud Survey, and covers overall program management, data analysis, publication review, and dissemination activities.

Table 4. Estimated BJS costs for the 2017 SFS

Staff salaries	Estimated Cost
GS-13 Statistician, BJS (20%)	$21,487
GS-15 Supervisory Statistician, BJS (10%)	$14,934
GS-13 Technical Editor, BJS (3%)	$3,233
GS-12 Production Editor, BJS (2%)	$1,807
GS-13 Digital Information Specialist, BJS (2%)	$2,149
Subtotal salaries	$43,599
Fringe benefits (28% of salaries)	$12,208
Subtotal: Salary and fringe	$55,807
Other administrative costs of salary and fringe (15%)	$8,371
Total estimated costs	$64,178

The U.S. Census Bureau will act as the data collection agent for the 2017 SFS. Census will review the final 2017 SFS survey instrument, develop all data collection support and training materials, train interviewers and support staff, and collect, process, and disseminate the 2017 SFS data.

Table 5. Estimated U.S. Census Bureau costs for the 2017 SFS

Division	Estimated Cost
CSM (Expert Review and Cognitive Testing)	$28,358
DSMD (Sample Design and Estimation)	$131,485
ADSD (Instrument Development)	$66,746
DSD (Data Processing)	$158,498
FLD (Data Collection)	$394,461
ADDP-SO (Survey Operations and Project Management)	$260,804
Total estimated costs	$1,040,352







The total estimated cost to the Federal Government for the 2017 SFS activities is $1,104,530. The Office for Victims of Crime (OVC) is supporting the entire cost of the data collection.

15. Reasons for Change in Burden

There is no change in total burden because the 2017 SFS is the first administration of this data collection.

16. Project Schedule and Publication Plans

Through August of 2017, Census will develop and test the CAPI instrument to ensure that it functions as designed and that all survey skip patterns have been properly programmed. This testing will be done in consultation with BJS. By early September of 2017, Census will develop and distribute all training materials to their Field Representatives (FRs). Interviewing for the 2017 SFS will be conducted from October through December of 2017 by the Census Bureau FRs. Processing of the data will take place on an ongoing basis between November 2017 and April 2018. The computer processing, editing, imputation, and weighting of the data will be completed by the end of May 2018. The Census Bureau will prepare and deliver a 2017 NCVS/SFS micro-data user file and accompanying file documentation including a nonresponse bias report to BJS by June of 2018.

The BJS will be responsible for the statistical analysis and publication of the data from the 2017 SFS. BJS will produce a report examining the prevalence and nature of fraud victimization by the fourth quarter of 2018. Key estimates to be presented in the report include—

• the prevalence of fraud victimization;

• the percent of persons who experienced specific types of financial fraud;

• amount of money lost in the fraudulent transaction;

• if the fraud victimization was reported to police or a consumer protection agency;

• fraud victimization by victim demographic characteristics

Due to the expected 12% prevalence rate and the redesigned state sampling plan, it is likely that in addition to national estimates, state estimates of fraud victimization can be produced for the largest 22 states. Given the expected variability by state, the type and number of estimates will be determined based on realized sample cases and acceptable measures of precision.

The data will be archived for public download and use at the University of Michigan’s ICPSR immediately following the publication of the BJS report. The 2017 SFS restricted-use data files will be available through the U.S. Census Bureau’s Federal Statistical Research Data Centers (FSRDC).

17. Display of Expiration Date

N/A.

18. Exceptions to the Certificate Statement

N/A. There are no exceptions to Certification for Paperwork Reduction Act Submissions. Collection is consistent with the guidelines in 5 CFR 1320.9.

a Deevy, M., & Beals, M. (2013). The scope of the problem: An overview of fraud prevalence measurement. Retrieved from the Financial Fraud Research Center at: http://longevity.stanford.edu/wp-content/uploads/2016/07/Scope-of-the-Problem-FINAL_corrected2.pdf.

b Beals, M., DeLiema, M., and Deevy, M. (2015) Framework for a Taxonomy of Fraud. Retrieved from the Financial Fraud Research Center at: http://162.144.124.243/~longevl0/wp-content/uploads/2016/03/Full-Taxonomy-report.pdf.

c National Academies of Sciences, Engineering, and Medicine. (2016). Modernizing crime statistics – report 1: Defining and classifying crime. Washington, DC: The National Academies Press.

d National Research Council. (2008). Surveying victims: Options for conducting the National Crime Victimization Survey. Panel to Review the Programs of the Bureau of Justice Statistics. Robert M. Groves and Daniel L. Cork, eds. Committee on National Statistics and Committee on Law and Justice, Division of Behavioral and Social Sciences and Education. Washington, DC: The National Academies Press.

e Langton, L. (2011). Use of Victim Service Agencies by Victims of Serious Violent Crime, 1993-2009. U.S. Department of Justice Special Report (NCJ 234212). Retrieved from http://www.bjs.gov/content/pub/pdf/uvsavsvc9309.pdf.

f Anderson, K.B. (2013). Consumer fraud in the United States, 2011: The third FTC survey. Federal Trade Commission staff report. Retrieved from https://www.ftc.gov/sites/default/files/documents/reports/consumer-fraud-united-states-2011-third-ftc-survey/130419fraudsurvey_0.pdf.

g These two surveys are used as comparable rates since both are administered as NCVS supplements.

15