Memo to OMB on Cognitive and Proxy Interview Protocol

U.S. Department of Justice

Office of Justice Programs

Bureau of Justice Statistics

Washington, D.C. 20531

MEMORANDUM

To: Robert Sivinski

Office of Statistical and Science Policy

Office of Management and Budget

Through: Melody Braswell, Clearance Officer, Justice Management Division

Jeffrey H. Anderson, Director, Bureau of Justice Statistics

Allen J. Beck, Senior Statistical Advisor, Bureau of Justice Statistics

Devon B. Adams, Acting Deputy Director, Bureau of Justice Statistics

Guy Burnett, Senior Advisor, Bureau of Justice Statistics

From: Heather Brotsos

Chief, Victimization Statistics Unit

Bureau of Justice Statistics

Date: May 18, 2020

Re: Bureau of Justice Statistics (BJS) Request for Office of Management and Budget (OMB) Generic Clearance for Cognitive Interviewing and Proxy Interview Field Test under the Generic Clearance for Cognitive, Pilot and Field Studies for Bureau of Justice Statistics Data Collection Activities, OMB Number 1121-0339

The Bureau of Justice Statistics (BJS) requests clearance for cognitive interviewing and testing designed to inform improvements to measurement of youth victimization in the National Crime Victimization Survey (NCVS). The objective of this clearance request, submitted under the BJS OMB generic clearance agreement (OMB Number 1121-0339), is to supplement research conducted under the BJS’s NCVS Instrument Redesign and Testing Project (OMB #1121-0368). Under this project, the NCVS instrument was redesigned and cognitively tested with adults and a small number of youth ages 12-17. The testing under the current clearance request will include a larger cognitive test with youth to more fully assess the revisions to the instrument for this population, along with other related testing components. The proposed data collection activities are exploratory in nature and are informed by a review of the literature on interviewing juveniles and secondary NCVS data analyses.

As BJS prepares to assess results from the full field test of the redesigned NVCS instrument, the results from the proposed testing effort are envisioned as a supplement to those efforts, with a focus on youth respondents. The proposed testing is intended to address known issues with interviewing juveniles in three main areas:

1. Evaluating youth comprehension of the NCVS instrument and the quality of data collected from youth;

2. Increasing youth participation in the NCVS; and

3. Assessing the efficacy of proxy reporting for youth.

These issue areas will be described further in the sections below.

This memo first provides background on issues related to juvenile participation and interviewing. Next is a description of the proposed testing procedures, followed by a description of language, burden hours, reporting, protection of human subjects, informed consent, data confidentiality and security.

1. Description and Purpose of Overall Project

1. Background on Juvenile Participation and Interviewing

Youth Comprehension and Data Quality. The NCVS currently interviews youth ages 12–17 using the same protocol and instruments that are used with adults ages 18 or older. Interviewing youth about crime and victimization poses a variety of potential measurement challenges. By virtue of their cognitive and psychosocial development, juveniles may not interpret survey questions in the same way as adults. Youth may also have limited cognitive ability and experience and thus may not understand critical items and concepts. Victimization is also a sensitive topic and youth are at a different life stage than adults, thus, age-appropriate adaptations may help ensure sufficient validity of youth interview data.

Comprehensive cognitive testing of NCVS items with youth is critical for understanding the manner in which youth of different ages (i.e., 12-13-year-olds, 14-15-year-olds, and 16-17-year-olds) understand and interpret questions and formulate their responses, and for identifying what adaptations may be necessary to improve the validity of data provided by youth. The proposed youth cognitive interview component will result in recommendations to NCVS items for youth (including specific age groups) that will improve the quality of data collected for youth.

Youth Participation. Response rates for youth ages 12-17 eligible for the NCVS have lagged behind those for other age groups. In 2018, overall person-level response rates were just below 60% for youth ages 12-17, compared to the overall response rate of 82% for all persons age 12 or older. Excluding proxy interviews across all age groups, 2009-2018 NCVS data indicate that persons ages 18 or older had a much higher response rate (84%) than did those ages 12-13 (50%) and those ages 14-17 (60%).¹ This finding also held when adults were restricted to those 18-25 years old (71.3%). Low response rates for youth in the NCVS are a key concern because they introduce the potential for nonresponse error and bias. These issues could threaten the accuracy and statistical precision of crime estimates produced from the NCVS for youth ages 12-17. Therefore, it is important to better understand the reasons for nonresponse and develop strategies for increasing participation among this age group.

Nonresponse among the youth NCVS population appears to stem from three primary factors: (1) parental refusal, (2) lack of youth availability to participate in the interview, and (less commonly) (3) youth refusal. Each of these factors will be explored in the proposed study.

Proxy Interviews. Proxy interviews are triggered when parents refuse to provide consent for their 12-13 year-olds to participate and in cases where the respondent is away for a prolonged period, mentally incapacitated, or otherwise unable to respond to the survey. Based on 2009-2018 NCVS data, the use of proxy interviews for 12-13-year-olds is high (39%); in comparison, 6% of NCVS data is gathered via proxy overall. Further, analyses also indicate that proxy interviews among youth aged 12-13 are associated with significantly lower victimization rates than self-interviews, which could indicate bias. Therefore, additional study to assess and quantify the bias associated with the use of proxy interviewing and evaluate this bias relative to the benefits of allowing proxy interviews is needed. The results of the proposed proxy interview study will inform BJS decision-making regarding the use of proxy interviews for youth of various age groups.

2. Project Description and Purpose

The current request for OMB generic clearance is for a project consisting of two inter-related activities: (1) cognitive interviews and (2) a proxy interview field test.

Cognitive Interviews. Cognitive interviews will be conducted with up to 130 youth ages 12-17 and up to 80 parents (or guardians) of youth.

1. Parent cognitive interviews. During the parent cognitive interviews, interviewers will give parents a detailed review of the current NCVS recruitment and interview processes. They will—

   1. Query parents on how they would decide whether to allow their child to participate in a national crime survey such as the NCVS, including what their primary concerns might be among factors including time commitment, confidentiality, and topic sensitivity.

   2. Discuss parents’ willingness to allow direct communication between the field representative and their child(ren) (via cell phone, email, or other means) to schedule an interview if they are not at home during the time of the parent interview.

   3. Discuss parents’ thoughts on whether their child(ren) would participate in an interview and whether the interview mode would influence the likelihood of participation.

   4. Elicit parents’ opinions about various NCVS recruitment materials and how these materials might impact their decision-making regarding their child’s participation.

2. Youth cognitive interviews. The youth cognitive interviews will focus on adolescents’ comprehension of the survey items. The interviews will include the following components—

   1. A think-aloud exercise at the beginning to help orient youth to the cognitive interview.

   2. Probes including questions to help address bounding, such as how the youth knows how long they have lived at their current address, and whether youth understand concepts such as threats, attempts, attacks, physical force, and series victimizations.

   3. Probes to elicit examples that will be relevant to youth, such as the types of items that are stolen from youth and other types of crimes or victimizations youth might experience that are not covered in the field test instrument.

   4. Questions about challenges youth think they might face in responding to a survey like the NCVS, including availability, and whether other options such as a self-administered web version would help.

The youth cognitive interviews will be conducted in an iterative, two-phase approach. In the first phase, interviewers will conduct in-depth interviews with juveniles using the redesigned NCVS instrument currently being field tested. Interview and project staff will confer regularly about issues that are surfacing during the interviewing and, in conjunction with BJS, propose revisions to any parts of the instrument that may benefit from youth-specific adaptations. In addition, BJS and RTI are working to identify potential youth-specific revisions to consider for the instrument, including clarifications that may improve comprehension of certain items (e.g., questions about other household members, employment status, certain crime screener questions, and incident location), transition language before items that could be particularly sensitive (e.g., the sexual violence crime incident report items), and recommendations for potential skip patterns for items that may be especially difficult for youth to answer (e.g., estimating out-of-pocket health care expenses or lost pay, whether the incident was a hate crime, police involvement, and attitudes toward police). Any areas identified for modification will be cognitively tested in a second round, with a focus on any items/sections that have been changed.

Proxy Interview Field Test. The proxy interview field test will be conducted with up to 200 parent-child pairs, with an emphasis on recruiting victims. It will entail conducting NCVS interviews (using the redesigned instrument currently being field tested or the slightly modified version based on initial cognitive interview findings) with both youth and parents (reporting on the youth’s experiences) to assess the impact of proxy vs. self-interviews on youth victimization estimates. The full instrument will be administered to both youth and parents (focusing on the youth’s experiences) following procedures that simulate existing NCVS administration procedures as closely as possible. Within-pair analyses will then examine the level of agreement between youth and parents on the youth’s victimization experience. Data from the NCVS can only provide overall victimization rates for youth respondents vs. proxy respondents. This test will enhance BJS’ knowledge on this issue by providing victimization levels and detailed data from parents and youth for the same incidents.

Results may vary by youth age and crime type and may reflect a complex set of issues, including parent willingness to disclose more serious incidents to interviewers, social desirability bias, and the extent of parents’ knowledge about certain victimizations. Informed by the results of this proxy test, BJS could potentially consider expanded use of proxy interviews (currently allowable for youth aged 12-13). The proposed proxy interview study will evaluate the benefits (increased youth participation) and tradeoffs (potential introduction of bias) associated with proxy interviews for younger (ages 12-13) and older (ages 14-15 and 16-17) youth.

The purpose of the overall project is to inform improvements to NCVS procedures that will increase participation and data quality for the youth NCVS population (see Figure 1). Each data collection activity will result in recommendations designed to inform improvements to NCVS survey validity (by guiding potential revisions to question wording for youth and quantifying any bias associated with proxy interviewing) and response rates (by directly addressing any concerns that parents have about their child’s participation). High-level procedures for each data collection activity are included below; additional details are available in Attachment A.

Figure 1: Youth and Parent Recruitment Process

2. Current Request for Cognitive Interviewing

2.1 Recruitment and Screening

Planned recruitment and data collection efforts for both the cognitive interviews and the proxy interview field test reflect current COVID-19 pandemic conditions and federal, state, and local responses in place throughout the United States—specifically, restrictions on geographic mobility, the closure of nonessential businesses (including RTI offices), social distancing recommendations, and school closures in most states that will remain in place until the end of the 2019-2020 school year. Because the majority of interviews will need to take place via video-interviewing from the families’ and interviewers’ respective homes, online recruitment at a national level will be used to identify eligible families. The mechanisms to identify families for the data collection efforts will include Mechanical Turk, an online workforce consisting of hundreds of thousands of adults, as well as advertisements placed in Craigslist, Facebook, Instagram, and Reddit (see recruitment flyer in Attachment B). In addition, snowball sampling will be employed on an as-needed basis. See Attachments A and C for additional details on planned recruitment strategies and the online screening form.

All recruitment efforts will direct interested parents to the online screening form. The information provided by parents in the screening form will be used to identify and prioritize likely youth victims, but some non-victims will be recruited as well. Based on screening results, a total of 410 eligible parents and youth will be identified for the study.

2.2 Consent/Assent Procedures

For parents and youth who are recruited for the cognitive interview activity, a recruiter will hold an initial telephone call with the parent to go over study procedures, including verifying the technological and privacy requirements for the video interviews as well as reviewing the parent permission form for the youth’s interview and obtaining the parent’s permission. The recruiter will then set up appointments for the parent’s and youth’s cognitive interviews, which need not take place simultaneously. For parents and youth who decline to participate, a single question will be asked about their reasons for declining. The goal of soliciting this feedback is to better understand the factors that influence nonparticipants in their decision-making.

During the parent interview appointment, before beginning the interview, the interviewer will review the parent consent form (for the parent’s interview) with the parent. The interviewer will document the parent’s decision to participate, signing the consent form as a witness. The form will include separate consent for allowing the interview to be recorded for those who decide to proceed.

For youth whose parents consent for them to participate, during the interview appointment and before beginning the interview, the interviewer will first go over the youth assent form with the youth. The interviewer will note that the youth’s parent has provided permission for him or her to be invited to participate but that the decision to participate is completely his or hers. The interviewer will read the entire form and answer any questions that the youth has. The interviewer will document the youth’s decision on participation, signing the youth assent form as a witness. The form will include separate assent for allowing the interview to be recorded.

Copies of all consent and assent forms are included in Attachment D. Each form includes a list of local and national helpline numbers for youth victims of crime on the last page. The consent procedures described here follow the procedures used in the current NCVS field study. In addition, Attachment D includes an overview of the topics contained in the instrument to aid in the decision of whether to let their child participate in the study.

All youth and parent interviews will be conducted by experienced RTI staff who will complete a training on the respective cognitive interview guides (see Attachments E and F) and all study protocols. The interviews will be conducted via video-interviewing through a secure, online videoconference platform (e.g., Doxy.me, Zoom). Youth and their parents will each receive $40 for participating in the 45-minute interview. Payment will be made in the form of an electronic gift card to an online vendor (e.g., Amazon.com). RTI will track the distribution of all incentives.

3. Youth Cognitive Interviews

The cognitive interview will assess youth’s understanding of the redesigned NCVS instrument currently being field tested. Comprehension of the instructions, screening, and incident form in the redesigned NCVS instrument will be assessed. After confirming the privacy of respondents’ interview setting (and asking respondents to communicate if the setting is no longer private at any time during the interview), interviewers will begin the interview with a practice “think aloud” exercise, to develop rapport with youth and demonstrate what types of thoughts and reflections are of interest throughout the interview. Concurrent probes will be built into the interview guide to assess youth’s comprehension of the survey questions, including both vocabulary comprehension and question meaning. Interviewers will also use spontaneous probes during the interview, based on how respondents react to particular questions. The full NCVS screener will be administered to all youth. For those who are identified as victims of crime based on the screener, the most serious incident will be selected; for this incident, as many of the crime incident report (CIR) modules as possible will be covered during the 45-minute interview (some respondents will receive select CIR modules to keep the interview within the 45-minute time slot). Non-victims will receive the police ask-all items, community measures, and additional personal characteristics. Concluding questions will request additional thoughts about the questions covered and overall impressions of the interview.

Both privacy and emotional distress will be carefully monitored during the youth cognitive interviews. Interviewers will be trained in the study’s privacy requirements (which pertain to both the interviewer’s setting as well as the respondent’s) (Attachment G) and the emotional distress protocol developed for the study (Attachment H).

As noted above, the youth cognitive interviews will be conducted in an iterative, two-phase approach. In the first phase, interviewers will conduct in-depth cognitive interviews with juveniles using the redesigned NCVS instrument that is currently being field tested. Interview and project staff will confer regularly about issues that are surfacing during the interviewing and, in conjunction with BJS, propose revisions to any parts of the instrument that may benefit from youth-specific adaptations. In addition, BJS and RTI are working to identify potential youth-specific revisions to consider for the instrument. Any areas identified for modification will be tested in a second round, with a focus on the items/sections that have been changed.

3. Parent Cognitive Interviews

The first set of interviews (i.e., those conducted during the first few weeks of data collection) will be “formative interviews” that will identify the types of questions or concerns parents are likely to have about their child’s participation in a national crime survey. During these interviews, interviewers will also discuss the NCVS field procedures with parents and obtain input on additional strategies that might be feasible for NCVS field representatives to use to contact older youth who are not at home during the initial household enumeration (e.g., whether parents would provide cell phone numbers or email addresses for their children at this stage of enumeration and allow them to be contacted directly by field representatives). A formative parent interview guide is included as Attachment F.

Based on the information learned during the formative interviews, additional materials will be developed and tested in the remaining interviews, including a Question and Answer brochure regarding the importance of youth participation (which could be included in the household mailing or used as needed by field representatives). This brochure will be designed for parents and household gatekeepers and will address questions that reluctant gatekeepers might have. For these interviews, interviewers will screen share (via the videoconferencing platform) different recruitment materials with parents and ask structured questions regarding the materials as well as other procedures used in the NCVS. Detailed feedback will be obtained on which materials/protocols would make parents more (or less) likely to allow their child to participate in the NCVS.

3. Data Analysis

For the cognitive interviews, interviewers will enter notes from each interview conducted in a master spreadsheet designed to facilitate item-specific analysis as well as identify broader issues. The interview recordings will be used to facilitate the completion of this spreadsheet. The information will be reviewed periodically throughout data collection and synthesized. The youth cognitive interview data will be analyzed to determine how adequately the questions and response options capture youth experiences, including younger (ages 12-13) and older (ages 14-17) youth. Age-specific modifications that may improve youth comprehension of questions will be identified and iteratively tested during the cognitive interview data collection period. Parent interview data will be synthesized and analyzed to identify strategies and specific materials that appear most likely to increase youth participation in the NCVS.

Upon completion of the cognitive interviews, a draft report will be prepared for BJS that will include recommended modifications to NCVS items (including age-specific modifications for younger and older youth) to improve data quality among youth respondents in the NCVS and changes to the NCVS recruitment materials and protocols that are most likely to increase youth participation in the study (based on decision-making by parents).

3. Current Request for Conducting a Proxy Interview Field Test

1. Recruitment and Screening

The same recruitment and screening procedures used to identify parents and youth for the cognitive interviews will be used in recruitment for the proxy interview field test. Based on the screening forms completed by parents, youth who have experienced victimization will be prioritized for the proxy interview field test.

2. Consent/Assent Procedures

The same consent and assent procedures described above for the cognitive interviews will be implemented in the proxy interview field test, including an initial telephone appointment with parents to go over study procedures and obtain parent permission for youth to be interviewed. The only exceptions are that the parent and youth proxy interviews will, ideally, take place during the same interview slot. This is important for ensuring the integrity of the data that are collected and the independence of the two interviews; staggered interviews could result in respondents conveying to one another how they answered the questions. The consent and assent procedures will be the same as those described for the cognitive interviews, although the forms themselves differ slightly to reflect the purpose and procedures of the proxy interview field test (see Attachment D).

3. Youth and Parent Interview Procedures

All interviews will be conducted by experienced RTI staff who will complete a training on the proxy field test instrument (see Attachment I) and all study protocols. The interviews will be conducted via video-interviewing through a secure, online videoconference platform (e.g., Doxy.me, Zoom). Youth and their parents will each receive $40 for participating in the 45-minute interview. Payment will be made in the form of an electronic gift card to an online vendor (e.g., Amazon.com). RTI will track the distribution of all incentives.

A two-person interview team will be assigned to each parent-child pair, and the interviews will take place in private locations in the respondents’ home, with youth and parents interviewed concurrently, but separately, and each in a private location. Both the youth and parent interviews will be administered by RTI interviewers via computer-assisted personal interviewing (CAPI). Before beginning, interviewers will ask respondents to confirm that they are doing the interview from a private location and request that respondents communicate if the setting is no longer private at any time during the interview.

Parents will be asked to report on youth’s experiences to the best of their knowledge. Youth will be asked to report on their own experiences. No probes will be asked and the interviews will approximate the actual NCVS administration procedures as closely as possible (including covering all screener questions and the CIR for all incidents youth experienced). Interviewers will be trained to follow the privacy and distressed respondent protocols developed for the study, including monitoring respondents for signs of distress and implementing a graduated response to signs of distress.

The interviews are anticipated to take around 45 minutes, on average, to complete, including a short debriefing at the end of the interviews. The debriefing questions will assess the extent to which parents have the knowledge to answer victimization questions for youth in their household and what considerations affect parents’ willingness to answer questions about criminal victimization for their child. The parent interview debrief will assess how confident parents were in reporting on their child’s experiences and which sections/questions were problematic for them to answer. The youth interview debrief will assess whether adolescents think that their parents know about the experiences reported in the interview and, if so, their perceptions of the extent of parents’ knowledge regarding the incident characteristics covered in the CIR modules.

4. Data Analysis

For the proxy interview field test, data analysis will use youth’s self-reported data as the gold standard. Congruence indicators for adolescents’ status as a victim of each crime type (e.g., theft, physical attack, sexual violence) will be generated for each parent-child pair (e.g., non-victim [youth] – non-victim [parent], victim [youth] – victim [parent], non-victim [youth] – victim [parent], victim [youth] – non-victim [parent]). RTI will develop syntax to define each type of crime based on the instrument and will submit the proposed syntax to BJS for approval before conducting the analysis. McNemar’s Test will be used to estimate the extent of discordance (bias) associated with proxy (i.e., parent) reports for younger (ages 12-13) and older (ages 14-17) youth on each victimization indicator. The analyses will directly estimate proxy bias by comparing proxy reports to self-reports within parent-child pairs. Additional analyses will explore the level of congruence with regard to incident characteristics among victims. Specifically, for pairs in which both youth and parents indicated that youth had experienced a particular crime type (and therefore received the CIR modules for that particular crime type), incident-specific comparisons will be conducted to assess the degree of agreement on key incident characteristics (e.g., month of incident, location, perpetrator). An indicator of the level of agreement on key incident characteristics for each crime youth experienced will be developed. Analyses using this indicator will explore which crime types appear to be associated with the greatest (and lowest) level of agreement and whether variation exists in the level of agreement for younger and older youth.

Upon completion of the proxy test analyses, a draft report will be prepared for BJS that will discuss the extent of any bias associated with proxy interviews, weigh this bias against the benefits of proxy interviews (i.e., allowing more youth’s experiences to be included than would be realized with self-interviews only, greater overall participation in the NCVS, lower data collection costs), and include associated recommendations regarding the expansion of proxy interviewing to older age groups.

4. Timeline

Key milestones for the project, and their associated start and end dates, are shown below.

Table 1: Key Milestones

Milestone	Start Date	End Date
Obtain OMB generic clearance	5/11/20	5/25/20
Recruit and screen parents and youth for cognitive interviews	5/26/20	7/20/20
Conduct cognitive interviews with parents and youth	6/1/20	7/24/20
Recruit and screen parent-child pairs for proxy interview field test	7/27/20	8/31/20
Conduct proxy interviews with parent-child pairs	8/3/20	9/4/20
Analyze data	6/1/20	9/15/20
Develop final report with recommendations	9/15/20	9/30/20

5. Language

All interviews will be conducted in English.

6. Burden Hours for Testing

The burden for all tasks consists of participants being screened for and subsequently participating in video interviews. The burden associated with these activities is presented in the table below.

Table 2: Burden Estimates

Data Collection Type	Total Respondents	Average Administration Time (minutes)	Burden (hours)
Cognitive Interview – Youth	130	45	978
Cognitive Interview – Parents	80	55*	73
Proxy Interview Field Test - Youth	100	45	75
Proxy Interview Field Test – Parents	100	55*	92
TOTAL	410	50	342

  *includes screening and scheduling time

7. Justification of Respondent Burden

Every effort has been made to minimize respondent burden and government cost in the conduct of the proposed interviews. The sample size for the cognitive interviews reflects a need to recruit youth victims of different types of crimes in order to gather feedback on different items within the CIR (to ensure that as many items are cognitively tested by youth as possible) as well as non-victims to document comprehension of the police and community ask-all items. A sufficient number of parents is needed for the cognitive interview component to evaluate different types of recruitment materials (e.g., assess different versions of a potential brochure) and capture different perspectives of parents. Therefore, each cognitive interview conducted via video interview will not be able to cover the full set of survey questions. To ensure that every question is sufficiently tested (with each respondent receiving a subset of questions), a larger number of participants will be needed for the video interviews. For the proxy interview field test, a sufficient number of parent-child pairs is needed to compare agreement in youth victimization status for various types of crime and to make such comparisons for younger (ages 12-13) and older (ages 14-17) youth.

8. Cost to the Federal Government

The cost of using MTurk and social media platforms for recruitment will be approximately $1,380.  The cost of stipends will be $8,400 ($40*210 respondents) for the cognitive interviews and $8,000 ($40*200 respondents) for the proxy interviews.

9. Protection of Human Subjects

There is some risk of emotional distress for respondents, given the sensitive nature of the topics covered in the NCVS and the questions that are asked. This risk is possible for youth who participate in the cognitive interviews and both youth and parents who participate in the proxy field test. However, appropriate safeguards are in place to mitigate this risk to the extent possible, including

• ensuring that all interviews are conducted by RTI staff experienced in conducting interviews on sensitive topics and who have undergone a study-specific training on recognizing signs of emotional distress among youth and activating the study’s distressed respondent protocol;

• having study leaders hold check-ins with the full interview team on a regular basis to monitor the frequency with which the distressed respondent protocol is being activated and the actions taken;

• ensuring that the recruitment, consent, and assent processes (described above) make youth and parents fully aware of the sensitive nature of the questions before making the decision about whether to participate, and notifying them that they may skip any question or stop the interview at any time;

• building into the cognitive interview protocols instructions for interviewers to check regularly on youth’s emotional state and remind them that they can skip any question they want or stop the interview at any time; and

• providing all youth and parents with a list of national resources for youth crime victims and encouraging them to consult these resources if needed.

10. Use of Information Technology to Reduce Burden

Technology will be utilized to facilitate recruitment and the scheduling process for both the cognitive interview and proxy interview field test activities, which reduces participant burden and controls study costs. Recruitment efforts will largely take place via electronic means. Potentially eligible families will be notified about the study using an electronic mechanism (Mechanical Turk). If interested, parents will click on a link that takes them to an online screening form (hosted in SurveyGizmo) for more information. The screening form will be programmed with skip and fill patterns to maximize the efficiency of this data collection. Parent permission will also be documented electronically, through a SurveyGizmo survey.

The use of video-interviewing will also reduce burden on participants, allowing them to avoid the time and expense of traveling to an RTI office or other interview location. Further, this mode will enable nationwide recruitment (as opposed to concentrating recruitment in select geographic areas, which is necessary for in-person interviews) and allow families from across the country to be included.

The proxy interview field test activity will involve additional technology, with all interviews conducted via the interviewer’s laptop, using the instrument programmed for CAPI. This programmed version of the NCVS instrument includes response-driven skip patterns that avoid the need to ask questions not relevant to the types of crimes reported by respondents. For example, for respondents who report no victimizations, the CAPI instruments will skip them completely out of the detailed CIR.

11. Data Confidentiality and Security

The BJS is authorized to conduct this data collection under 34 U.S.C. § 10132. The BJS will protect and maintain the confidentiality of personally identifiable information (PII) to the fullest extent under federal law. The BJS, its employees, and its contractors (RTI staff) will only use the information provided for statistical or research purposes pursuant to 34 U.S.C. § 10134 and will not disclose respondent information in identifiable form to anyone outside of the BJS project team. All PII collected under the BJS’s authority is protected under the confidentiality provisions of 34 U.S.C. § 10231. Any person who violates these provisions may be punished by a fine up to $10,000, in addition to any other penalties imposed by law. Further, per the Cybersecurity Enhancement Act of 2015 (6 U.S.C. § 151), federal information systems are protected from malicious activities through cybersecurity screening of transmitted data.

Participation in the cognitive interviews and proxy interview field test is voluntary. A number of steps will be taken to ensure participants’ privacy during data collection. These include—

• only obtaining the minimum amount of private identifiable information from parents that is absolutely necessary to schedule interviews, using case identifiers instead of personal identifiers for youth, and storing the private identifiable information securely (e.g., on the restricted access project share drive behind RTI’s firewall);

• conveying in advance the need for a private interview setting and having interviewers confirm at the beginning of the interview that respondents feel confident they are participating in the interview in a setting where no one can overhear either interviewers or respondents;

• using a secure, encrypted videoconferencing platform (e.g., Doxy.me, Zoom) for the interviews;

• securely handling and storing all hard copy materials (e.g., any hard copy notes);

• deleting electronic files as soon as possible; and

• ensuring that the final report any information that could be used to identify study participants.

The procedures proposed for these study activities have been reviewed by RTI’s institutional review board (IRB) and classified as not research involving human subjects as defined by 28 CFR 46. Therefore, RTI IRB oversight of these activities is not required.

1 Data were calculated using the NCVS public-use data file (PUF); 2016 response rates are not available on the PUF.

11