SVS 2019 OMB Supporting Statement Part B

Supporting Statement – 2019 National Crime Victimization Survey (NCVS) Supplemental Victimization Survey (SVS)

B. Collection of Information Employing Statistical Methods

1. Universe and Respondent Selection

The sample universe for the NCVS Supplemental Victimization Survey (SVS) is all persons age 16 or older living in NCVS interviewed households. The NCVS sample of households is drawn from the more than 120 million U.S. households and excludes military barracks and institutionalized populations. In 2019, the annual national sample is planned to be approximately 240,000 designated addresses located in 542 stratified Primary Sampling Units (PSUs) throughout the United States.

Frame

The Master Address File (MAF) contains all addresses from the most recent decennial census plus updates from the United States Postal Service, state and local address lists, and other address listing operations. The MAF is the frame for the target NCVS population. Every ten years, the Census Bureau redesigns the samples for all of their continuing demographic surveys, including the NCVS. In 2015, the 2000 sample design started to phase out and the 2010 sample design started to be phased in. Beginning in 2016, some PSUs were removed from the sample, some new PSUs were added to the sample, and some continuing PSUs that were selected for both the 2000 and 2010 designs remained in the sample. The phase-in and phase-out of the sample designs started in January 2015 and continued through December 2017. The new sample sizes are larger than in previous years to support state-level estimates in 22 most populous states. (See State Samples.)

Rotating Panel Design

The NCVS uses a rotating panel design. The sample consists of seven groups for each month of enumeration. Each of these groups stays in the sample for an initial interview and six subsequent interviews, for a total of seven interviews for the typical household. During the course of a 6-month period, a full sample of seven rotation groups is interviewed (one-sixth each month). One rotation group enters the sample for its first interview each month.

SAMPLE SELECTION

The sample design for the NCVS is a stratified, multi-stage cluster sample. Sample selection for the NCVS, and by default the SVS, has three stages: the selection of PSUs, the selection of address units within sample PSUs, and the selection of persons and households from those addresses to be included in the sample.

Stage 1. Defining and Selecting PSUs

Defining PSUs – Formation of PSUs begins with listing counties and independent cities in the target area. The PSUs comprising the first stage of the sample are formed from counties or groups of adjacent counties based upon data from the decennial census and the American Community Survey (ACS). For the NCVS, the target area is all 50 states and the District of Columbia. Counties are either grouped with one or more contiguous counties to form PSUs or are PSUs all by themselves. The groupings are based on certain characteristics such as total land area, current and projected population counts, large metropolitan areas, and potential natural barriers such as rivers and mountains. For the NCVS, decennial census counts, ACS estimates, and administrative crime data drawn from the FBI’s Uniform Crime Reporting (UCR) Program are also used to stratify the PSUs. The resulting county groupings are called PSUs.

After the PSUs are formed, the larger PSUs are included in the sample with certainty and are considered to be self-representing (SR). The remaining PSUs, called non self-representing (NSR) because only a subset of them are selected, are combined into strata by grouping PSUs with similar geographic and demographic characteristics.

Stratifying PSUs – For the 2010 design, the NSR PSUs are grouped with similar NSR PSUs within states to form strata. Each SR PSU forms its own stratum. The data used for grouping the PSUs is based on decennial census demographic data, ACS data, and administrative crime data. NSR PSUs are grouped to be as similar or homogeneous as possible. Just as the SR PSUs must be large enough to support a full workload so must each NSR strata. The most efficient stratification scheme is determined by minimizing the between PSU variance and the within PSU variance.

Selecting PSUs – The SR PSUs are automatically selected for sample or “selected with certainty.” NSR PSUs are sampled with probability proportional to the population size using a linear programming algorithm. One PSU is selected from each NSR stratum. The 2010 design NCVS sample includes 339 SR PSUs and 203 NSR PSUs. PSUs are defined, stratified, and selected once every ten years. The 2010 design sample PSUs were sampled using population data from the 2010 census.

Stage 2. Preparing Frames and Sampling within PSUs

Frame Determination – The 2010 sample design selects its sample from two dynamic address-based sampling frames, one for housing units and one for group quarters (GQs). Both frames are based upon the MAF, which is a national inventory of addresses. The MAF is continually updated by various Census Bureau programs and external sources. New housing units are added to the MAF, and therefore the NCVS sampling frame, through semiannual updates from a variety of address sources, including the U.S. Postal Service Delivery Sequence File, local government files, and field listing operations.

In the 2010 design, each address in the country was assigned to the housing unit or GQ frame based on the type of living quarter. Two types of living quarters are defined in the decennial census. The first type is a housing unit (HU). An HU is a group of rooms or a single room occupied as separate living quarters or intended for occupancy as separate living quarters. An HU may be occupied by a family or one person, as well as by two or more unrelated persons who share the living quarters. The second type of living quarters is GQ. GQs are living quarters where residents share common facilities or receive formally authorized care. About 3% of the population counted in the 2010 Census resided in GQs. Of those, less than half resided in non-institutionalized GQs. About 97% of the population counted in the 2010 Census lived in HUs.

Within-PSU Sampling – All of the Census Bureau’s continuing demographic surveys, such as the NCVS, are sampled together in order to minimize respondent burden and maximize response rates. This procedure takes advantage of updates from the January MAF delivery and ACS data. This within-PSU selection occurs every year for housing units and every three years for GQs. The NCVS sample changes each year based on three factors: (1) to accurately account for population changes within states; (2) to account for changes in expected response rates; and (3) to be large enough to produce annual national violent crime estimates with a coefficient of variation of no more than five percent.

Selection of samples is done one survey at a time (sequentially). Each survey determines how the unit addresses within the frame should be sorted prior to sampling. For the NCVS, each frame is sorted by geographic variables. A systematic sampling procedure is used to select addresses from each frame. A skeleton sample is also selected in every PSU. Every six months new addresses on the MAF are matched to the skeleton frame. The skeleton frame allows the sample to be refreshed with new addresses and thereby reduces the risk of under-coverage errors due to an outdated frame.

Addresses selected for a survey are removed from the frames, leaving an unbiased or clean universe behind for the next survey that is subsequently sampled. By leaving a clean universe for the next survey, duplication of addresses across surveys is avoided. This is done to help preserve response rates by insuring that no unit falls into more than one survey sample.

Stage 3. Persons within Sample Addresses

The last stage of sampling is done during the initial contact of the sample address during the data collection phase. For the SVS, if the address is a residence and the occupants agree to participate, then an attempt is made to interview every person age 16 or older who lives at the resident address and completes the NCVS-1 (crime screener). The NCVS has procedures to determine who lives in the sample unit and a household roster is completed with names and other demographic information of all persons who live there. If someone moves out (in) of the household during the interviewing cycle, he or she is removed from (added to) the roster.

Based on the total 2019 NCVS sample, approximately 26,066 persons a month, age 16 or older, in NCVS sampled households will be eligible to be interviewed for the supplement during July to December 2019 for a total of 156,394 possible interviews. Based on data from the 2016 SVS, interviewers are typically able to obtain SVS interviews with approximately 76.4% of the SVS eligible household members in occupied units in sample in any given month. A total of 119,526 persons age 16 or older are expected to be interviewed for the SVS during the 6-month collection period.

State Samples

Beginning in January 2016, BJS and Census increased and reallocated the existing national sample in the 22 largest states. The states receiving a sample boost include Arizona, California, Colorado, Florida, Georgia, Illinois, Indiana, Maryland, Massachusetts, Michigan, Minnesota, Missouri, New Jersey, New York, North Carolina, Ohio, Pennsylvania, Tennessee, Texas, Virginia, Washington, and Wisconsin. In 2017, each of these 22 states had a population greater than 5 million persons and in total these 22 states comprised 79% of the U.S. population.^a In each of the 22 states, enough sample was selected to achieve a 10% relative standard error (RSE) for a three year average violent victimization rate of 0.02. The underlying assumption of the subnational sample design is that three years of data will be needed to produce precise estimates of violent crime, which is experienced by about 1% of the population. Sample sizes in the remaining 28 states and the District of Columbia were determined to ensure full representation and unbiased estimates at the national level. For the 2010 design, unlike the 2000 sample design, no strata cross state boundaries and all 50 states and the District of Columbia have at least one sampled PSU.

It may be possible to produce state-level estimates using the 2019 SVS, including estimates of stalking victimization, given the prevalence of stalking during the 2006 and 2016 SVS administrations.

Weighting and Estimation

The purpose of the SVS is to be able to make inferences about stalking victimizations for the population of persons age 16 or older in the United States. Before such inferences can be drawn, it is necessary to adjust, or weight, the sample of people to ensure it is similar to the entire population in this age group. The SVS weights are a combination of household-level and person-level adjustment factors. Household and person respondents from the NCVS sample are adjusted on a bi-annual basis to represent the U.S. population age 12 or older. For the SVS, the population is restricted to persons age 16 or older.

NCVS household and person weights are first adjusted to account for any subsampling that occurs within large GQs. The NCVS nonresponse weighting adjustment then allocates the sampling weights of nonresponding households and persons to respondents with similar characteristics. Additional factors are then applied to correct for the differences between the sample distributions of age, race and Hispanic origin, and sex and the population distributions of these characteristics. The resulting weights were assigned to all interviewed households and persons in the NCVS file.

SVS weighting begins with the NCVS final person weight, which is the then multiplied by a SVS noninterview adjustment factor. SVS noninterview adjustment factors were computed by distributing the weights of SVS noninterviews to the weights of the SVS interviews, with adjustment cells determined by age, race and Hispanic origin, and sex. The result is a SVS person-level weight that can be used for producing estimates from the SVS variables.

Variance Estimates

The NCVS and SVS estimates come from a sample, so they may differ from figures from an enumeration of the entire population using the same questionnaires, instructions, and enumerators. The difference between the sample estimate and true population parameter is known as sampling error. The sampling error quantifies the amount of uncertainty and bias in an estimate as a result of selecting a sample.

Variance estimates can be derived using direct estimation or generalized variance functions (GVFs). GVFs for the NCVS are created by the Census Bureau for the BJS. The Census Bureau produces parameters for GVFs that estimate the variance of any crime count estimate based on the value of the estimate. To do this, estimates and their relative variance are fit to a regression model using an iterative weighted least squares procedure where the weight is the inverse of the square of the predicted relative variance.

2. Procedures for Collecting Information

The SVS is designed to calculate national and state-level estimates (for the 22 most populous states) of stalking victimization for the target population – the noninstitutionalized resident population age 16 years or older. The SVS is administered to all age-eligible NCVS respondents during the 6-month period from July through December of 2019.

DATA COLLECTION

For the six month period, July through December 2019, the SVS will be administered to approximately 126,635 designated households. Each housing unit selected for the NCVS remains in the sample for three years, with each of seven interviews taking place at 6-month intervals.

The NCVS-500 (control card) is used to complete a household roster with names and other demographic information of the household members. For some demographic questions that are asked directly of respondents, flashcards are used, including for education, race, Hispanic origin, sexual orientation, gender identity, employment, and household income. Respondents are asked to report victimization experiences occurring in the six months preceding the month of interview. The NCVS crime screener instrument (NCVS-1) is asked of all respondents age 12 or older in the household and is used to ascertain whether the respondent has experienced a personal crime victimization during the prior six months and is therefore eligible to be administered the NCVS crime incident report instrument (NCVS-2). The NCVS-1 collects the basic information needed to determine whether the respondent experienced a crime victimization (rape or sexual assault, robbery, aggravated or simple assault, personal larceny, burglary, motor vehicle theft, or other household theft). When a respondent reports an eligible personal victimization, the NCVS-2 is then administered to collect detailed information about the crime incident. The NCVS-2 is administered for each incident the respondent reports. For each victimization incident, the NCVS-2 collects information about the offender (e.g. sex, race, Hispanic origin, age, and victim-offender relationship), characteristics of the crime (including time and place of occurrence, use of weapons, nature of injury, and economic consequences), whether the crime was reported to police, reasons the crime was or was not reported, and victim experiences with the criminal justice system. Clearance for the core NCVS forms and materials including the NCVS-500, NCVS-1 and NCVS-2 are requested through a separate OMB request and number (OMB NO: 1121-0111).

Each interview period, the interviewer completes or updates the household composition component of the NCVS interview and asks the crime screener questions (NCVS-1) for each household member age 12 or older. The interviewer then completes a crime incident report (NCVS-2) for each reported crime incident identified in the crime screener. Once the NCVS interview is completed (i.e. nonvictims responded to all NCVS-1 screening questions or victims completed all necessary NCVS-2 incident reports), the interviewer administers the SVS questionnaire to persons age 16 or older.

The first contact with a household is by personal visit and subsequent contacts may be by telephone. For the second through seventh visits, interviews are done by telephone whenever possible. Approximately half of all interviews conducted each month are by telephone.

SVS collection

The SVS is designed to calculate national and some subnational (i.e., 22 most populous states) estimates of stalking victimization for the target population – all persons age 16 or older living in NCVS households.

The SVS instrument includes a screener and incident survey. Based on the VAWA definition of stalking on the screener, respondents are asked to report if they have experienced repeated unwanted contacts or behaviors that caused them to experience substantial emotional distress, or to fear for their safety or the safety of someone else, or that would have caused a reasonable person to fear for their safety or the safety of someone they knew in the 12-months prior to the interview. Each eligible person age 16 or older will be asked the screener questions. The screener questions collect the following information: (1) unwanted contacts or behaviors; (2) repeated course of conduct (i.e. experiencing the same behavior or contact more than once, or experiencing two or more different behaviors one time); (3) actual fear; (4) substantial emotional distress; and (5) reasonable fear. When a respondent reports an eligible stalking victimization, the SVS incident instrument is then administered to collect detailed information about this victimization to learn more about the nature and consequences of the victimization.

The SVS incident instrument covers eight areas including (1) offender information, (2) duration of stalking, (3) frequency of stalking, (4) motive for stalking, (5) other threats or attacks the victim may have experienced, (6) help-seeking, (7) self-protective actions, and (8) cost to victim. The offender section asks about the offender characteristics including number of offenders, sex, age, race, Hispanic origin, and the victim-offender relationship. Section two on duration of stalking asks about how long the behaviors had been happening and how the victim found out about the behaviors. Section three, frequency of stalking, asks about how many times the behaviors occurred in the past 12 months. The fourth section, motive for stalking, asks why the victim thought the offender was stalking them. Section five is a follow-up to the reasonable fear questions in the screener, and asks how the victim was attacked or threatened, whether a weapon was involved, and if someone close to them was threatened or attacked. Section six, help-seeking, asks about reporting to police, and seeking/receipt of victim services. Section seven, self-protective actions, asks respondents about actions they took to protect themselves or stop the behaviors from continuing. Finally, section eight, cost to victim, asks respondents about how the stalking may have affected them, any related distress or socio-emotional problems, and the impact of victimization on school or work.

3. Methods to Maximize Response Rates

Contact Strategy

Contact materials focus on the NCVS in general and do not specifically reference the SVS or other supplemental surveys. The Census Bureau mails notifications to households prior to data collection, interviewers contact households for the first time in-person, and interviewers conduct nonresponse follow-up. The Census Bureau mails an introductory letter (NCVS-572(L)) or continuing household letter (NCVS-573(L)) explaining the NCVS to the household before the interviewer's visit or call (Attachments 4 and 5). The introductory letters are sent to households before their first NCVS interview, or time-in-sample 1 (TIS-1) interviews, and the continuing household letters are sent to households in time-in-sample 2 through 7 interviews. When they go to a household, the interviewers carry cards identifying them as Census Bureau employees. Potential respondents are assured that their answers will be held in confidence and are only used for statistical purposes. For respondents who have questions about the NCVS, interviewers provide a brochure (NCVS-110), and can also reference information in their Information Card Booklet (NCVS-554) that contains information such as uses of NCVS data and frequently asked questions and answers. At the FRs discretion, thank you letters are sent to the household (NCVS-593(L) or NCVS-594(L)). All forms and materials used for contact with the household have been previously approved by OMB (OMB NO: 1121-0111).

The Census Bureau trains interviewers to obtain respondent cooperation and instructs them to make repeated attempts to contact respondents and complete all interviews. The interviewer obtains demographic characteristics of noninterview persons for use in the adjustment for nonresponse. SVS response rates are monitored on a monthly basis and compared to the previous month’s average to ensure their reasonableness.

As part of their job, interviewers are instructed to keep noninterviews, or nonresponse from a household or persons within a household, to a minimum. Household nonresponse occurs when an interviewer finds an eligible household but obtains no interviews. Person nonresponse occurs when an interview is obtained from at least one household member, but an interview is not obtained from one or more other eligible persons in that household. Maintaining a high response rate involves the interviewer’s ability to enlist cooperation from all kinds of people and to contact households when people are most likely to be home. As part of their initial training, interviewers are exposed to ways in which they can persuade respondents to participate as well as strategies to use to avoid refusals. Furthermore, the office staff makes every effort to help interviewers maintain high participation by suggesting ways to obtain an interview, and by making sure that sample units reported as noninterviews are in fact noninterviews. Also, survey procedures permit sending a letter to a reluctant respondent as soon as a new refusal is reported by the interviewer to encourage their participation and to reiterate the importance of the survey and their response.

Interviewer Training

Training for NCVS interviewers consists of classroom and on-the-job training. Initial training for interviewers consists of a full day pre-classroom self-study, four-day classroom training, post-classroom self-study, and on-the-job observation and training. Initial training includes topics such as protecting respondent confidentiality, gaining respondent cooperation, answering respondent questions, proper survey administration, use of systems to collect and transmit survey data, NCVS concepts and definitions, and completing simulated practice NCVS interviews. The NCVS procedures and concepts taught in initial training are also regularly reinforced for experienced NCVS interviewers. This information is received via monthly written communications, ongoing feedback from observations of interviews by supervisors, and monthly performance and data quality feedback reports.

NCVS interviewers also receive specific training on the SVS including eligibility, the organization of the SVS interview, content of the survey questionnaire, addressing potential respondent questions, and internal check items that are in place to help the interviewer ensure that the respondent is being asked the appropriate questions and follow-up when clarification is needed. The SVS training materials are distributed to interviewers approximately a month before the supplement goes into the field.

Monitoring Interviewers

In addition to the above procedures used to ensure high participation rates, the Census Bureau implements additional performance measures for interviewers based on data quality standards. Interviewers are trained and assessed on administering the NCVS-1 (crime screener), NCVS-2 (crime incident report), and SVS exactly as worded to ensure the uniformity of data collection, completing interviews in an appropriate amount of time (not rushing through them), and keeping item nonresponse and “don’t know” responses to a minimum. The Census Bureau also uses quality control methods to ensure that accurate data are collected. Interviewers are continually monitored by their Regional Office to assess whether performance and response rate standards are being met and corrective action is taken to assist and discipline interviewers who are not meeting the standards.

Another component of the data quality program is monthly feedback. In 2011, the Census Bureau implemented a series of field performance and data quality indicators. Previously, high response rates were the primary measure of interviewer performance. The data quality indicators are tracked through the Census Bureau’s expanded Performance and Data Analysis (Giant PANDA) tool, and monthly reports provided to the field. Under the revised performance structure, interviewers are monitored on the following –

• response rates (household, person, and the current supplement in the field);

• time stamps (the time it takes to administer the screener questions on the NCVS-1 or the crime incident questions on the NCVS-2);

• overnight starts (interviews conducted very late at night or very early in the morning);

• late starts (cases not started until the 15th or later in the interview month);

• absence of contact history records (cases missing records of contact attempts with the household and/or persons within the household); and

• quality of crime incidents (changes made to the location, presence, or theft data items on the NCVS-2 during post-processing coding operations).

Noncompliance with these indicators results in supervisor notification and follow-up with the interviewer. The follow-up activity may include simple points of clarification (e.g., the respondent works nights and is only available in the early morning for an interview), additional interviewer training, or removal of the interviewer from the survey.

Every effort has been made to make the survey materials clear and straightforward. The SVS instrument has been designed to make collection of the data as concise and easy for the respondent as possible. The SVS questions have been cognitively tested to ensure that they are easily understood by most respondents.

Nonresponse and Response Rates

Interviewers are able to obtain interviews with about 84% of household members in 76% of the occupied units in sample in a given month.^b Beginning with 2018 and following data collection years, the Census Bureau plans to report nonresponse and response rates, respondent and nonrespondent distribution estimates, and proxy nonresponse bias estimates for various subgroups. Should the analyses reveal evidence of nonresponse bias, BJS will work with the Census Bureau to assess the impact to estimates and ways to adjust the weights accordingly. The interviewers obtain demographic characteristics of noninterview persons for use in the adjustment for nonresponse.

In 2016, the Census Bureau found evidence of potential bias in the SVS estimates because the overall response rate was low (58%). Analysis indicated that respondent and nonrespondent distributions were significantly different for race and Hispanic origin, sex, age, Census region and metropolitan statistical area (MSA) status subgroups. However, after applying weights adjusted for person nonresponse, there was no evidence that these response differences introduced nonresponse bias in the final victimization estimates.

4. Final Testing of Procedures

The majority of the survey instrument questions have not changed substantially, if at all, from the 2016 administration. Therefore, there was no need for cognitive testing of the 2019 SVS instrument. The instrument is in an automated CAPI instrument form and has been fully tested and performed without incident in 2016.

5. Contacts for Statistical Aspects and Data Collection

The Victimization Statistics Unit at BJS takes responsibility for the overall design and management of the activities described in this submission, including developing study protocols, sampling procedures, and questionnaires and overseeing the conduct of the studies and analysis of the data by contractors.

The Census Bureau is responsible for the testing of interview materials and the collection of all data. Ms. Meagan Meuchel is the NCVS Survey Director and manages and coordinates the NCVS and its supplements. Mr. David Hornick of the Demographic Statistical Methods Division of the Census Bureau oversees the statistical aspects of the supplement. BJS and Census Bureau staff responsible for the SVS include –

BJS Staff: all staff located at- 810 7th Street NW Washington, DC 20531	Census Staff: all staff located at- 4600 Silver Hill Road Suitland, MD 20746
Jeffrey H. Anderson, Ph.D. Director	Meagan Meuchel NCVS Survey Director Associate Directorate for Demographic Programs – Survey Operations
Devon B. Adams Acting Principal Deputy Director	Jill Harbison NCVS Assistant Survey Director Associate Directorate for Demographic Programs – Survey Operations
Lauren Glaze Acting Unit Chief Victimization Statistics Unit	Megan Ruhnke NCVS Assistant Survey Director Associate Directorate for Demographic Programs – Survey Operations
Rachel Morgan, Ph.D. Statistician Victimization Statistics Unit	Scott Raudabaugh Chief, Crime Surveys Programming & Population Support Branch Chief Demographic Surveys Divisions
Jennifer Truman, Ph.D. Statistician Victimization Statistics Unit	David Hornick Lead Scientist Demographic Statistical Methods Division

a Table 1. Annual Estimates of the Resident Population for the United States, Regions, States, and Puerto Rico: April 1, 2010 to July 1, 2017 (NST-EST2017-01). Source: U.S. Census Bureau, Population Division. Release Date: December 2017.

b Morgan, R.E., & Truman, J.L. 2018. Criminal Victimization, 2017. U.S. Department of Justice Bulletin (NCJ 252472), available at https://www.bjs.gov/content/pub/pdf/cv17.pdf.

7