NISMART-4 Supporting Statement A – Justification

SUPPORTING STATEMENT

Fourth National Incidence Studies of Missing, Abducted, Runaway and Thrownaway Children (NISMART-4)

Overview

The National Institute of Justice (NIJ), in partnership with the Office of Juvenile Justice and Delinquency Prevention (OJJDP), request clearance for the Fourth National Incidence Studies of Missing, Abducted, Runaway and Thrownaway Children (NISMART-4), which includes a pilot test and national data collection on child victims of stranger abductions (i.e., “stereotypical kidnappings”) known to law enforcement agencies, and pilot tests to assess instruments and methods to collect information from law enforcement agencies on parental abductions and other types of missing episodes involving children.

NISMART was designed to respond to the requirements of the 1984 Missing Children’s Assistance Act (Pub. L. 98–473) amended in 2013 (P.L. 113-38), to conduct national incidence studies on missing children. It has been carried out three times in the past, in 1988, 1999, and 2011. Each study involved multiple components, including a national household survey and a national law enforcement survey. Upon the completion of NISMART-3, it was evident that the household survey methodology, which had served as the principal data source for NISMART’s estimates of missing children was no longer an efficient and cost effective method for obtaining the data needed to respond to the Act’s reporting requirements. The increasingly lower numbers of missing children and the larger samples of households needed to identify adequate numbers of qualifying cases, combined with the declining response rates to household surveys and the concomitant greater cost of achieving acceptable response rates, undermine the feasibility of estimating numbers of episode children and missing children on the basis of data from household surveys of parents (and youth).

At the same time, the law enforcement survey for collecting data on child victims of stranger abductions has been a successful element of NISMART. It has achieved a high participation rate at reasonable cost even in an era of declining participation rates. The NISMART-3 law enforcement survey yielded a response rate of 86 percent on the screener and 91 percent on the detailed case-level child survey.¹ The survey has produced estimates of stereotypical stranger abductions that comport with other sources of information about serious non-family kidnappings, like data from the National Center for Missing and Exploited Children (NCMEC), has been widely cited in the media, and is the topic for which OJJDP receives the most queries for updated data from policymakers, service practitioners, and the public.

The present NISMART is intended to continue to support studies consistent with legislative requirements, but it requires some methodological work to revise and update the past designs. The methodology needs revision because of both the changing environment of household survey research, with its increasing costs and decreasing response rates and because the amended legislation requires more frequent (triennial) reporting of rates, which necessitates developing the most cost-effective approach to be feasible as a repeating survey. These two concerns have framed the goals of the current project—to create reliable and cost-effective study designs that will allow for more frequent replication, using exclusively police agency data, the collection of which has been consistently one of the most successful elements of past NISMARTs.

Specifically, this package seeks clearance for three pilot tests and one national study:

1. Pilot test an efficient and cost-effective methodology for collecting national data on the child victims of stereotypical kidnappings (i.e., stranger abductions) known to law enforcement, referred to hereafter as the Law Enforcement Survey – Stereotypical Kidnappings (LES-SK);

2. Implement the redesigned LES-SK to produce national estimates;

3. Pilot test a new instrument and sampling method to collect information from law enforcement agencies on family abductions (LES-FA) in preparation for a future national survey; and

4. Pilot test a new instrument and sampling method to collect information from law enforcement agencies on other types of missing children (LES-MC) and returned children in preparation for a future national survey.

This submission describes each of the four data collection efforts: the three pilot studies to test new approaches to collecting information on stereotypical kidnappings, family abductions and other types of missing children reported to law enforcement and the full survey administration designed to provide national estimates of stereotypical kidnappings. The LES-SK pilot test will inform the design and implementation of the national administration of the LES-SK. While the current request seeks approval for all four studies in this submission, NIJ will submit an amendment to OMB for approval of any changes to the national LES-SK based on pilot findings. Likewise, NIJ will submit amendments for changes to the LES-FA and LES-MC if experiences from the LES-SK pilot and national survey administration suggest that improvements to the methodologies for the other two pilots are needed.

A. Justification

1. Necessity of Information Collection

OJJDP is authorized to conduct national incidence studies of missing children under the Missing Children's Assistance Act (see 34 U.S.C. § 11293(c), Appendix 1). In fiscal year 2019, the Department of Justice transferred OJJDP’s research, evaluation, and statistical functions and activities to NIJ, including the management of these national incidence studies. NIJ will work in collaboration with OJJDP and its data collection agent, to advance this work consistent with the Act.

The Fourth National Incidence Studies of Missing, Abducted, Runaway and Thrownaway Children (NISMART-4) will respond to the 1984 Missing Children’s Assistance Act (Pub. L. 98–473), which requires OJJDP to conduct national incidence studies “to determine for a given year the actual number of children reported missing each year, the number of children who are victims of abduction by strangers, the number of children who are the victims of parental kidnappings, and the number of children who are recovered each year.” The Act was amended in 2013 to require the collection of this information every 3 years (Pub. L. 113–38).

The legislation was passed as a response to concerns among the general public and law enforcement agencies about the problem of missing children. Children going missing from the custody of parents have been a widespread source of parental fear, especially cases where a stranger abducts a child from a public setting. Missing children cases generate a great deal of media interest, but also pose considerable challenges for law enforcement, policymakers, and missing children’s support organizations. In the absence of reliable statistics and quality social science, many misperceptions and misunderstandings about the nature of the missing children in the United States have persisted. Congress, under the Missing Children’s Assistance Act, has tried to improve law enforcement response as well as increase useful knowledge about the diversity and complexity of missing children cases. Tracking the incidence of missing children episodes is also important for knowing whether public policies, including law enforcement strategies and tactics, are succeeding in reducing its incidence or its negative outcomes.

NISMART addresses the absence of other sources of detailed information about missing children that can be used for tracking trends and gaining insights about the problem. While there are other federal agencies and federally-supported efforts that collect information on missing children, their goals for and use of their data do not meet NISMART’s requirements. The National Crime Information Center (NCIC) archives reports of missing persons from local police, but this resource is restricted to law enforcement agencies.² Moreover, the NCIC does not obtain details about the circumstances of many cases or apply the definitional categories and distinctions that NISMART uses. The National Center for Missing and Exploited Children (NCMEC) maintains a national database of missing children for whom help has been sought from NCMEC. But this resource does not capture the cases for which help is not sought from the NCMEC, including reasons such as quick resolution or adequate resources in the local agency. Since some missing children cases are not crimes (e.g., runaways and lost/stranded children), there are a number of problems with using crime databases such as the Uniform Crime Reporting (UCR) system or the National Incident-Based Reporting System (NIBRS) to generate national estimates on missing children, including that they do not collect many elements crucial to understanding the nature of a missing child episode, such as the length of time children have been missing or the case outcomes. In short, a national data collection effort specifically about missing children is very important to meet the legislative requirements and to guide public policy on this issue.

The NISMART program has identified five categories of episodes that can cause children to become missing. The episode types and their definitions are:

• Family abduction where a member of the child’s family or someone acting on behalf of a family member takes or fails to return a child in violation of a custody order or other legitimate custodial rights and—

  • Conceals the child or

  • Transports the child out of state with the intent to prevent contact, or

  • Expresses the intent to deprive the caretaker of custodial rights permanently or indefinitely.

For children who are age 15 or older and mentally competent, use of physical force or threat of bodily harm is required.

• Nonfamily abduction where a nonfamily perpetrator, without lawful authority or parental permission, uses force or threat—

  • To take a child (at least 20 feet or into a vehicle or building), or

  • To detain a child in a place where the child cannot leave or appeal for help for at least 1 hour.

For children who are younger than age 15 or mentally incompetent, force or threat is not needed if the perpetrator—

• Conceals the child’s whereabouts, or

• Demands ransom, or

• Expresses the intent to keep the child permanently.

Stereotypical kidnapping is a nonfamily abduction subtype perpetrated by a stranger, person of unknown identity, or slight acquaintance in which the perpetrator—

• Kills the child, or

• Detains the child overnight, or

• Transports the child at least 50 miles, or

• Demands ransom, or

• Expresses the intent to keep the child permanently.

• Runaway/thrownaway.

Runaway:

• A child leaves home without permission and stays away overnight, or

• A child who is away with permission but chooses not to come home and stays away for—

  • One night (if age 14 or younger or mentally incompetent), or

  • Two or more nights (if ages 15-17)

Thrownaway: A child whom an adult household member tells to leave or prevents from returning home, and—

• Does not arrange for adequate alternative care, and

• The child is gone overnight.

• Missing involuntary, lost, stranded, or injured: A child whose whereabouts are unknown to the caretaker, causing the caretaker to—

  • Contact law enforcement or a missing children’s’ agency to locate the missing child, or

  • Become alarmed for at last 1 hour and try to locate the child, and the child—

    • Was trying to get home or make contact but was unable to do so because he or she was lost, stranded, or injured or

    • Was too young to know how to return home or contact the caretaker.

• Missing, benign explanation: A child whose whereabouts are unknown to the caretaker, causing the caretaker to—

  • Become alarmed for at least 1 hour, and

  • Try to locate the child, and

  • Contact the police about the episode for any reason, as long as the child did not fit one of the above episode types.

NISMART has had a profound impact on the understanding of the missing children problem among policy makers, practitioners and the public at large. The important insights include:

• That the ways in which children go “missing” comprise a number of distinctly different components that are not at all similar in their rates, dynamics, and intervention requirements. A typology was developed that has been widely accepted by the field as guide to this diversity.³

• That the most-feared type of nonfamily abduction, stranger kidnapping, is extremely infrequent and outcomes of these cases have improved. In 2011, police recovered more of these children and fewer children were killed than in 1997.⁴

• That there is an important distinction between children deemed missing by parents and those who are actually reported as missing to police or a missing children’s agency.⁵

• That some of the children reported as “missing” to police, actually are in known locations, but police become involved to retrieve the children.⁵

• That most missing children episodes resolve fairly quickly; for example, 93 percent of runaways returned within one month and less than 1 percent were gone 6 months or longer.⁶

• That the number of missing children episodes declined between 1999 and 2013.⁷

• That the advent of cell phone technology may have helped to resolve many missing children episodes quickly, so parents were not concerned or alarmed long enough to call the police.⁸

NISMART has provided national estimates to agencies like the NCMEC and Missing Children’s Clearinghouses (MCCs) for use in their educational and training programs, uniform definitions, and statistics they can use to illustrate the scope and diversity of the problem of missing children.

The studies proposed under this submission will focus attention on the capacity of law enforcement agencies to provide information on the recovery of missing children. Much is still unknown about how law enforcement follows up on and tracks the resolution of missing child episodes, and this revised approach will gather more detail about the process. The redesign will also enable OJJDP to meet the new mandate that requires triennial data collection through expedited future data collection methods and enhanced collaboration with law enforcement agencies by:

• Improving the ability to map the data onto the practice and legal elements that are most important to law enforcement.

• Putting into place systems and mechanisms to routinize regular data collection, reduce its cost, and increase its accuracy.

• Revisiting some of the definitional and conceptual questions behind the missing children issues that may need to be updated because of changes in society, technology, and law.

• Improving the accuracy of missing children estimates and the sensitivity of trend analyses.

• Exploring the possibility of integrating NISMART data collection into other law enforcement data collection systems such as NIBRS at some future point.

In 2017, Westat and subgrantee, the University of New Hampshire (UNH), were competitively selected to serve as OJJDP’s data collection agent (via a cooperative agreement) to implement the data collection and study design efforts outlined in this submission. The Westat Institutional Review Board (IRB) has reviewed the NISMART-4 (referred to as NISMCRLE⁹ in the IRB submission) study design, data collection instruments and respondent correspondence. The IRB determined that the study involved minimal risks to human subjects. A copy of the IRB approval notice is attached (Appendix 2). As noted above, NIJ social scientists will manage all aspects of the proposed pilot tests and national data collection, in collaboration with OJJDP program staff and the data collection agent.

2. Needs and Uses

The data collected in the three pilots (LES-FA, LES-MC, and LES-SK) and the national LES-SK will be used to refine definitions and typologies of missing children from previous NISMART collections. The national LES-SK and future national studies using the new LES-MC and LES-FA instruments and methods will offer findings that can improve practice and advocacy as well as research. NISMART categorizations have been adopted by most of the advocacy and investigative agencies in the field and have increased recognition of the complexity and diversity of the issue and the need for differentiated responses. This includes the recognition that types of missing children have less and more serious forms, like endangered runaways. NISMART findings have helped to dispel mistaken beliefs and misunderstandings about missing children, such as the idea that stranger abduction is the most frequent type of missing child. The results have moderated public anxiety about the problem of stranger abduction with the understanding that these events, although serious, are relatively rare.

Like previous NISMARTs, data collected on the national LES-SK will be used to track trends in the incidence of kidnapped children, showing where there are increases and decreases in the types of missing children episodes over time. This means, among other things, that it will be important to retain definitions and operationalizations of concepts used in the previous waves of NISMART so as to be able to do trend analyses. The pilot studies of the LES-FA and LES-MC will be used to test the feasibility of new methodologies for collecting information on family abductions and other missing children solely from law enforcement in preparation for the planning of future data collections to develop national estimates. These pilots are critical to ensure that future national studies can collect the data required under the Missing Children’s Assistance Act on the number of children who are the victims of parental abductions, the number who go missing and the number of missing children who are recovered each year.

In the past, law enforcement received criticism for reluctance to act expeditiously on missing children reports.¹⁰ Although much has changed over recent decades, families of missing children point to improvements still needed, especially for cases that involve teens^11,12 or children abducted by family.^13,14 Under the National Child Search Assistance Act of 1990, agencies are required to report missing children to the NCIC without any waiting period, and to update that record within 60 days. Including questions about NCIC reporting in the pilot tests will also help us understand whether and how agencies are fulfilling this requirement. Knowledge gained from the LES-FA and LES-MC pilot studies will be crucial to designing subsequent studies that can fill gaps and obtain information that currently may go unreported. One goal for the pilot study is to test out and refine the questions that will assess this information about police response.

The data collected through the LES-SK pilot will be used to inform the design of the national LES-SK. Data collected from the LES-FA and LES-MC pilots will be used to inform the instrumentation and design of national data collection efforts on family abductions and other missing children. Only data collected from the national LES-SK will be used to produce national estimates for published OJJDP reports, such as Child victims of stereotypical kidnappings known to law enforcement in 2019.

OJJDP Needs and Uses

The NISMART collections have provided foundational information on the incidence and characteristics of different categories of missing children that is used to inform the development and implementation of OJJDP’s Missing and Exploited Children’s (MEC) program. The MEC program seeks to prevent cases of missing, exploited, and abducted children and to support communities in responding to such cases. Historically, the MEC program has funded research, demonstration projects, and training and technical assistance efforts for practitioners at the state and local levels. Examples of current efforts funded by OJJDP under the MEC program that support secondary distribution of NISMART publications, materials, and related information through their training and technical assistance efforts include:¹⁵

• The National Center for Missing & Exploited Children (NCMEC), serves as an information clearinghouse and national resource center on issues related to victims, missing and exploited children and operates a national toll-free hotline. NCMEC partners with OJJDP to provide resources, technical assistance, and prevention services to victims, families and the public, as well as support to law enforcement agencies at the federal, state, and local levels in cases involving missing and exploited children. NCMEC's 24-hour hotline, 800-THE-LOST has received more than 4.6 million calls through September 2017. Additionally, NCMEC operates the CyberTipline the nation’s centralized reporting system for suspected child sexual exploitation.

• The Internet Crimes Against Children Task Force Program (ICAC program) helps state and local law enforcement agencies develop an effective response to technology-facilitated child sexual exploitation and Internet crimes against children. This help encompasses forensic and investigative components, training and technical assistance, victim services, and community education.

• The AMBER Alert (America’s Missing: Broadcast Emergency Response) Training and Technical Assistance program supports a grantee that provides training and technical assistance to states and tribal communities in developing and improving their AMBER Alert plans to be more efficient in alerting the public when children are abducted and believed to be in imminent danger.

• The MEC program also provides other grants for training and technical assistance on child victimization and related supports topics, including assisting communities in developing a multi-disciplinary response to domestic minor sex trafficking (recognizing that missing, abducted or runaway children are particularly vulnerable to be trafficked).

Use of the NISMART Data by Others

The findings from NISMART have been highly sought after and widely quoted by policymakers and the media. For example, the estimates of prevalence for stereotypical kidnapping cases appear in the press frequently in two main contexts: (a) when high profile missing child cases make the news and (b) when parenting experts write columns about children’s safety with guidance for parents. There is concern among the public in recent years about whether children are being abducted to be sold into sex trafficking.^16,17 Data from NISMART can help law enforcement address these concerns.¹⁸ The estimates for all types of missing children also are widely cited in the criminology community in guides for policing,^19,20 understanding of crime and crime perceptions,^{21,22,23,24,25} and in the child welfare community,^26,27 which recognizes that missing child problems like running away and family abduction require resources related to child protection.

Some of the places that the NISMART findings get regularly posted and cited include:

• On the website and in the publications of the NCMEC, the most publicly visible agency on this topic;

• In numerous publications from OJJDP. For example, OJJDP published two fact sheets and eight bulletins following NISMART-2 and two additional bulletins following NISMART-3;²⁸

• In publications and websites for other missing children’s agencies like the Polly Klaas Foundation and the Global Missing Children’s Network;

• In magazine articles directed particularly to parents, such as Parenting Magazine, educating the general public on scientific assessments of the scope of the problem;

• At OJJDP-funded trainings on missing and exploited children for law enforcement and other professionals;

• In information briefs on various state child welfare websites regarding child sex trafficking, risks to juveniles who run away from placements; and

• On fact sheets for family court professionals concerning the characteristics of family abductions (children at risk, circumstances, and types of perpetrators).

A list of NISMART citations is found in Appendix 3.

Anticipated Products

The studies proposed under this submission will lead to the following reports and products:

• A report on the methodology and results of the LES-SK pilot with proposed revisions for the national LES-SK.

• A report on the methodology and results of the national LES-SK, including national estimates and characteristics of stereotypical kidnappings (i.e., stranger abductions) reported to law enforcement.

• A report on the goals, methodologies and results of the LES-FA and LES-MC pilots.

• A report on the administration plans for the national LES-FA and LES-MC studies.²⁹

• A final methodology report that provides an integrated and detailed description of the three pilots and the national LES-SK, including the sample design, data collection procedures and outcomes, response rate, editing and coding procedures, assessment of reliability and validity, non-response bias assessment, weighting for the national survey, and methods to be used to generate standard errors and documentation of constructed variables.

• An NIJ-OJJDP Bulletin on the results of the national LES-SK survey, which will contain tables similar to those in previous OJJDP reports on stereotypical kidnapping.³⁰ The report will examine changes in victimization rates compared to previous reports by victim demographics and type of incidents.

• One or more scholarly articles for publication in a peer-reviewed journal.

In addition, the project will submit the final cleaned dataset and codebook for the LES-SK national study to the National Archive of Criminal Justice Data (NACJD) so that the data are available to other researchers for secondary analyses.

3. Use of Information Technology

1. LES-SK Pilot and National LES-SK

Following the conclusion of the 2011 NISMART-3 survey, OJJDP worked with the Bureau of Justice Statistics (BJS) and the project team to develop a more efficient way to administer the LES-SK, which would save time and cost given the requirement for more frequent (triennial) reporting of rates. For NISMART-3, the LES-SK obtained a sample of stereotypical kidnapping cases through a survey of all law enforcement agencies (n=4,644) located in a nationally representative sample of 433 counties. First, agencies in the sample received a mail survey that asked whether the agency investigated any stereotypical kidnappings in their jurisdiction between October 1, 2010 and September 30, 2011. When agencies reported such investigations, the contractor conducted extensive telephone interviews with investigating officers to obtain details of the episodes. The response rate for the Phase 1 mail survey was 86 percent. In the Phase 2 telephone component, interviews were completed for 91 percent of the targeted cases.

The new design for the LES-SK will maintain the cost-efficient and effective Phase 1 mail screener to law enforcement agencies but will substitute a self-administered online questionnaire for most of the telephone interviews with investigating officers. ³¹ Telephone interviews will be provided for officers who do not respond online and whose online responses require clarification. Previous experience on NISMART-3 demonstrated that while investigators were interested and motivated to talk about their child kidnapping cases, the many demands on their time and time spent out of the office made scheduling of the interviews difficult and time-consuming with much effort spent on repeated rescheduling. The online self-administered questionnaire will allow investigators to respond at their convenience, without having to schedule a phone conversation.

Westat’s Survey Framework will be used to develop the LES-SK web survey. The Survey Framework is programmed in .Net, JavaScript, and Bootstrap, with a SQLServer 2016 database engine. The following lists the features of the proposed web survey and describes how it will function:

• The web survey allows use of login, passwords, and Transport Layer Security (TLS) to ensure limited access and data security. The framework allows creation of respondent-unique URLs that directs a respondent to their survey without the need to enter login/password credentials. It is not possible to enter the application through a saved or bookmarked web page.

• Uses Microsoft Windows technologies (ASP.NET, SQL Server 2016) and JavaScript for client side data validation. It does not utilize persistent cookies and is not dependent on JavaScript for its functionality. This means that all respondents, independent of the security settings on their browser, are able to access and complete the online survey. Further, it does not make use of any third-party software.

• Programmable conditional and skip logics are built in. Respondents are automatically navigated to the correct location on the survey based on their responses.

• Validations and edits can be customized for each survey, including both “soft” and “hard” edits. Cross validations between questions are available, as well as summation and range type validations both within a question screen and between previously answered questions.

• The framework can prompt for missing values, either at the point of data entry or upon final submission of the survey.

• Respondents can save and return to the next unanswered question at any time before the survey is completed.

• The respondent is able to navigate back through the survey and change prior responses without data loss.

• The framework allows for the use of data grids as an online data collection method.

• The interface design is simple and uncluttered, with easy-to-identify navigational buttons and prompts.

In addition, the following are some specific interactive features that will help to keep respondents engaged in completing the web-based survey:

• Respondents who cannot complete the survey in a single session are allowed to save their responses, log off, and return to complete the survey at a later time. When they log back in, they are taken to their next unanswered question.

• After the last question is answered, respondents will be taken to a final submission page that prompts them to review their survey responses before formally submitting the survey. Finally, once respondents have submitted their answers, they may print out their completed surveys.

The benefits of the new methodology for the LES-SK will allow for

• Reduction in the survey cost, since interviewers will not need to spend time making appointments with respondents and collecting the information.

• Reduction in the burden to respondents since self-administration of surveys is much faster than interviewer-administration.

• Reduction in length of the data collection period. The web survey may be completed more quickly when respondents complete it at their convenience, without needing to wait to be contacted by an interviewer. As all cases identified in Phase I can be released simultaneously, this may result in less time in the field compared to when interviewers are involved.

• Maintaining the historically high response rate in a time of declining telephone survey participation. Given the option to fill out a questionnaire at their own convenience rather than committing to a lengthy telephone interview at a specific time, busy law enforcement investigators, who would otherwise not participate, may be more willing to respond.

However, the new online methodology does not entirely remove interviewers from the data collection. Telephone interviewers will follow up with nonrespondents to the online survey to encourage them to complete the survey and to offer to conduct the survey over the telephone, if the officer desires. Moreover, all completed online surveys will be examined by senior staff at Westat for ambiguities and missing information and, where appropriate, flagged for telephone follow up for data retrieval and clarification.

2. LES-FA Pilot

The LES-FA will take a new approach to measuring family abductions that leverages information technology, provides estimates more efficiently and obtains valuable information that were not available in previous NISMART designs. Previous NISMART estimates of family abduction and other missing child episodes were based on a household survey that had limitations because of declining response rates. As such, these estimates offered limited insight into important issues surrounding police management of missing child cases.

Under plans devised for NISMART-4, the information about family abduction and other missing child episodes will be harvested from law enforcement agencies using the power of their electronic management information systems. To develop the NISMART-4 plans, the project staff has consulted with experts on police management information systems, both developers and marketers of these systems as well as data management officials connected to law enforcement agencies, to leverage access to data that is already routinely collected by law enforcement.

The existence of sophisticated management information system software used by agencies means that agencies can search efficiently and comprehensively for episodes. In the case of family abduction, experts on police management information systems have indicated that most agencies have record codes to identify abduction episodes as well as missing persons episodes. Thus asking agencies to search their electronic systems for such cases is a relatively low cost and efficient operation. Many also have codes for family offenses. A search using these codes can also be supplemented with text searches for terms such as custody, which may flag additional cases. Police officials are familiar with searches of this sort since they do them routinely to provide information to other agencies such as the Federal Bureau of Investigation (FBI) as well as to independent inquirers such as the news media. Guidance about conducting these searches will be provided to law enforcement agencies who are asked to participate in the pilot studies. The pilot studies are intended to test and refine these instructions.

The other use of information technology that will be critical to the new design is to invite investigators to provide information on cases through an online self-administered questionnaire (as opposed to telephone interviews). The design, rationale and benefits are similar to the design of the LES-SK web survey described in the previous section.

3. LES-MC Pilot

The pilot for the LES-MC will utilize the same information technologies described for the LES-FA pilot.

4. Efforts to Avoid Duplication

Some data on missing children episodes currently reside in state MCCs, in the records of the NCIC, in the reports to the NCMEC, and in the UCR Summary Reporting System and the NIBRS for kidnapping incidents. However, these sources do not have detailed data that adequately comply with the requirements under the Missing Children’s Assistance Act because they lack completeness, definitional rigor, or national scope.

The following describes the type of information collected by these other sources and their limitations.

NCIC:

• The FBI’s National Crime Information Center (NCIC) is a centralized information system that facilitates information flow between the numerous law enforcement branches. NCIC collects data on missing persons including children who have been reported missing to law enforcement and there is a reasonable concern for their safety. NCIC collects counts of missing-child events and data include the age, gender and race of the missing person as well as indications of circumstances under which the person is reported missing, e.g., victim’s disappearance was not voluntary, or victim’s physical safety may be in danger, or proven physical or mental disability of victim, therefore subjecting victim or others to personal and immediate danger. However, only very crude summaries of aggregate data are provided to the public. In addition, many of the elements that are required by the Missing Children’s Act, such as the number recovered are not included. If all that were needed is a simple count of missing-child events known to law enforcement, the NCIC data that are posted each year could in principle provide a crude approximation. However, NCIC data are not unduplicated to the child level, so a given child might be entered multiple times depending on the number of their episodes during the year. Also, circumstances of the missing event are only entered for slightly less than one-half of the cases submitted.³² The NCIC information is therefore limited by (1) the fact that for more than half of the cases entered the circumstances are not recorded in the database, and (2) the circumstances that are indicated are rudimentary. Moreover, NCIC case level data are available only to authorized law enforcement agencies.³³

MCCs:

• State Missing Child Clearinghouses (MCCs) operate an information and referral resource for the public, local law enforcement, and other state clearinghouses. They provide information and, where possible, guidance to these entities, and resources on missing children to their families and the professionals who serve them. The clearinghouse oversees and ensures that law enforcement agencies follow the law as it relates to the immediate entry of missing children,³⁴ ensuring that parents’ reports of a missing child, runaway, parentally abducted or otherwise missing or exploited child are immediately available to other law enforcement agencies around the state. Like other sources, the MCC records are incomplete and are not unduplicated. Moreover, MCCs are organized differently and have different functions in different states. Consequently, they may not include all reported cases of missing children.³⁵

NCMEC:

• The National Center for Missing and Exploited Children (NCMEC) is the nation's clearinghouse and comprehensive reporting center for all issues related to the prevention of and recovery from child victimization. It takes reports of missing and exploited children through a hotline and online portal. Information includes case identifiers, law enforcement agencies involved and counties where children went missing or were recovered. NCMEC also provides technical assistance to locate abductors and recover missing children, and help with identifying child victims of sexual exploitation, in addition to providing support in the investigation of cases. NCMEC maintains a liaison with each MCC and helps ensure they are familiar with the many resources available through NCMEC.

• Limitations of NCMEC data are that NCMEC receives information only in cases in which local agencies or parents are seeking assistance. Cases that are resolved quickly tend not to be reported. Moreover, NCMEC data reflects reports of missing children episodes and that means if a child runs away multiple times in a year, each instance would be entered into the system separately and counted in the yearly total of reports. In addition, records often do not provide all the information surrounding the circumstances of a missing child.

UCR Summary Reporting System:

• The FBI’s national Uniform Crime Reporting (UCR) Summary Reporting System collects aggregate data on the number of offenses known to law enforcement. UCR provides data on an aggregate tally of crimes including violent crime, murder and non-negligent manslaughter, forcible rape, robbery, aggravated assault, property crime, larceny/theft, motor vehicle theft, and arson. No category exists for abduction or kidnapping.

NIBRS:

• The National Incident-Based Reporting System (NIBRS) is used by law enforcement agencies to collect and report data on crimes. It captures details on each single crime incident—as well as on separate offenses within the same incident—including information on child victims, known offenders, relationships between victims and offenders, arrestees, and property involved in the crimes. NIBRS’ inclusion of specific data on kidnapping offers an opportunity to learn more about the nature and extent of this crime, about which so few data have been available in the past. NIBRS data are however limited for purposes of research because it is not complete for all states and counties and has limited categories of information collected. This incompleteness means that standard approaches that involve computing totals by summing the auxiliary data across the country do not work. Furthermore, the missing data in the files are not due to controlled sampling, which means it is not feasible to compute precise national estimates using probability weighting (instead of using known totals).

In summary, none of the above sources provide data that are both nationally representative and suitable for identifying cases that meet the NISMART definitions and the requirements of the Missing Children’s Assistance Act. Besides providing national estimates specific to missing children, NISMART efforts will:

• Provide data on missing children unduplicated to the child level; unlike the other data sources, which may include multiple entries of the same child for different episodes in a given year.

• Provide specific information on the numbers and categories of episodes that can cause children to become “missing.” It will provide detailed definitional categories and distinctions of children reported as “missing” using the NISMART definitions.

• Provide details about the law enforcement response to and management of missing children reports that can be used for monitoring trends or gaining insights about the issue. Since some missing children (e.g., runaways, lost children) are not missing due to crimes, there are problems with using crime databases such as the UCR or NIBRS to obtain information about these episodes. Also, these sources do not include many important missing child elements crucial to understanding the issue, such as the length of time children have been missing.

• Represent the dimensions and characteristics of the issue most relevant to practitioners and policy makers, which allow them to improve their prevention and response efforts and approaches.

NISMART-4 will, however, try to make use of information from these other sources to facilitate its process of case identification and ensure completeness. For the national LES-SK, the MCCs in sampled counties as well NCMEC will be queried to determine if there are any child kidnapping cases that were not identified by law enforcement in the screeners. The MCCs and NCMEC will be asked to provide files with information about non-family abductions and children reported lost, injured or missing during the NISMART-4 LES timeframe in the sampled counties. Information will include case identifiers, law enforcement agencies involved and counties where children went missing or were recovered. These cases will be cross-referenced with mail screener responses to screen out duplicate cases. Where non-duplicate cases are found, the investigator will be asked to provide the detailed information required by the online survey.

In addition, part of the pilot process for the LES-FA and LES-MC will include comparing cases of family abductions and other missing children reported by law enforcement agencies in the search component of the pilot with reports from the MCCs and NCMEC on all cases reported to them by the law enforcement agencies in the sample. NCMEC and the MCCs will be asked to provide a file with cases of family abductions and other missing children and will compare these cases to those identified in the law enforcement searches. Cases reported by NCMEC or the MCCs that are missing from the law enforcement searches will be discussed with investigators in the debriefing interviews to determine if there were deficiencies in the search strategy or alternate approaches that would lead to identification of otherwise missing cases.

5. Efforts to Minimize Burden

Several efforts have been made in the redesign of the LES-SK to minimize burden. The LES-SK instrument is significantly shorter than the interviewer-administered version used in previous NISMARTs. The total number of questions has decreased from 236 in the NISMART-3 LES-SK to 89 for the NISMART-4 LES-SK. The instrument has been redesigned so that it can be self-administered as an online survey rather than a telephone interview. This permits investigators to complete it at their convenience and in multiple sittings, where necessary. It is important to note that the investigators (and not data analysts) are the target audience for the survey as the investigators are the ones who will be able to provide the most accurate information about the case. Because investigators are frequently out of the office, experiences on NISMART-3 demonstrated that the scheduling of appointments was one of the most time-consuming aspects of the data collection. A self-administered survey eliminates the burden of multiple attempts to schedule and reschedule appointments between the interviewer and the respondent. While a self-administered paper instrument would also eliminate the scheduling burden, web surveys are less burdensome to respondents as skip patterns are automatically programmed so respondents see only questions that are applicable, thereby reducing the time for administration.

One of the key goals of the pilot studies for LES-FA and LES-MC is to test out burden-minimizing strategies to obtain relevant information from agencies. Among the most important of these is to test out search strategies with law enforcement records management systems that will easily and efficiently yield the universe of records that need to be checked for cases relevant to NISMART.

6. Consequences of Less Frequent Data Collection

The frequency of data collection for this study is specified in the Missing Children’s Assistance Act as every three years. Less frequent data collection would be inconsistent with the Act. Moreover, the three pilot tests are important to maximizing the quality of the data collection for the national studies. If the pilot test for the LES-SK, LES-FA and LES-MC are not conducted, it will be unknown if the new proposed methodologies for implementation are effective in obtaining high response rates, minimizing burden and bias, and collecting high quality data that can produce timely and valid national estimates.

7. Special Circumstances Influencing Collection

The three pilot surveys (LES-SK, LES-FA and LES-MC) and the national LES-SK will be conducted according to the guidelines specified in 5 CFR § 1320.6. No special circumstances are known that would cause inconsistency with these guidelines.

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

The research under this clearance is consistent with the guidelines in 5 CFR 1320.6. The 60-day Federal Register notice was published on November 8, 2018 (Federal Register, Vol. 83, No. 217, pages 55913-55914). The 30-day Federal Register notice was published on February 4, 2019 (Federal Register, Vol. 84, No. 23, pages 1512-1514). No public comments have been received in response to this notice.

In addressing issues identified in NISMART-3 and planning for alternate approaches in NISMART-4, OJJDP social scientists (now NIJ social scientists) consulted extensively with the Bureau of Justice Statistics including Dr. Howard Snyder, then Deputy Director of the Statistical Programs Division, and Dr. Allen Beck, Senior Statistician. Drs. Snyder and Beck met regularly with OJJDP staff to discuss possible design benefits and limitations and to provide feedback on OJJDP’s FY17 solicitation for a data collection agent.

Once a competitive award was made to Westat and University of New Hampshire (UNH), the project staff convened three expert panels. The first panel provided input on the design of the LES-SK and was composed of the following experts:

LES-SK Panel Experts
Ben Adams Social Science Analyst National Institute of Justice (previously with OJJDP)	Thomas Simon, PhD Acting Branch Chief National Center for Injury Prevention and Control (NCIPC)
Jennifer Bronson Statistician Bureau of Justice Statistics	James Walters Program Administrator National Criminal Justice Training Center
Brecht Donoghue Senior Social Science Analyst National Institute of Justice (previously Deputy Associate Administrator with OJJDP)	Janet Warren Professor of Psychiatry and Neurobehavioral Sciences University of Virginia
Stacy Jeleniewski Analyst National Center for Missing and Exploited Children	Tyesha Wood Detective (Navajo) of the Gila River Indian Community (AZ) Police Department
Kenneth Lanning Consultant and former FBI agent specializing in investigation of missing and exploited children

The first panel was a webinar convened and moderated by co-PI Dr. David Finkelhor of UNH. The consultation focused on gathering expert feedback on the LES-SK instrument’s content, the formulation and wording of questions and definitions of key terms, and the recruitment of the agencies. The panel members indicated that data collected in previous LES-SKs was relied on by the field as the standard for information on the incidence and characteristics of stranger abductions of children. The panel meeting included considerable discussion about the criteria for what constitutes a stranger abduction, including issues related to the age of the victims, luring or grooming behaviors vs. use or threat of force, and the role and use of technology.

The panel suggested strategies for publicizing the survey in advance through the national AMBER Alert newsletter which goes to 50,000 law enforcement agencies, as well as through email blasts. They also suggested the team consider hosting a webinar on the survey, either in advance or on the results, which could satisfy some law enforcement training requirements.

Changes made to the instrument based on the panel’s recommendations included:

• Changing the term “slight acquaintance” to “person with limited previous contact.”

• Adding questions about whether the perpetrator was listed on the National Sex Offender Registry or National Sex Offender Public Website.

• Providing a free-text box near the end of the instrument to allow respondents to clarify any answers to the closed-ended questions.

• Providing a free-text box at the end of the survey to allow respondents to provide a narrative description of the case if they felt this would facilitate greater understanding of the case.

The panel also suggested that the team consult with some of the larger information technology providers to learn more about how the systems are structured and how searches for cases may be done most efficiently.

The second panel focused on conducting searches in law enforcement management information systems to identify cases of family abductions and missing children. It convened these experts:

LES-FA and -MC Search Component Panel Experts
Brian Acken Senior Law Enforcement Policy Analyst RTI International	Brecht Donoghue Senior Social Science Analyst National Institute of Justice (previously Deputy Associate Administrator with OJJDP)
Ben Adams Social Science Analyst National Institute of Justice (previously with OJJDP)	Shelley Hyland Statistician Bureau of Justice Statistics
Jennifer Bronson Statistician Bureau of Justice Statistics	Tanea Parmenter ISP BCI Auditing and Training Specialist/Missing Persons/UCR Bureau of Criminal Identification Idaho State Police
Mike Carter Project Management Services IJIS Institute	Paul Wormeli Innovation Strategist Wormeli Consulting, LLC
Alexia Cooper Statistician Bureau of Justice Statistics

This panel was conducted as a webinar and was also moderated by Dr. Finkelhor. The panel drew on the expertise of professionals familiar with the content and operation of computer-aided dispatch systems, records management systems, and other databases that are maintained by local law enforcement about missing children. It was designed for the project staff to learn about what information is available and how searches might be done. In addition, the panel was intended to discuss possible limitations and barriers to the collection and strategies to assist local officials, reduce their burden, and achieve high rates of response.

The panel noted that that while there are about 150 companies that provide law enforcement with off-the-shelf software systems, members felt that all the systems would have the capacity to search for specific incidents of missing children. Moreover, they noted that law enforcement agencies would have experience conducting such searches, as requests for these types of searches are common. Searches for kidnappings or family abduction cases will likely be much easier for the agencies to conduct, but other types of missing children may be categorized in less standardized ways and may require more expansive definitions. The panel recommended that the team learn more (1) about agencies’ practice in retaining records for those systems that purge cases, and (2) whether their electronic records include information on the recovery of missing children or whether those case details are found only in paper files. The panel also recommended that the national LES-FA and LES-MC might gain greater compliance by having a letter of support from state MCCs and by reassuring the law enforcement agencies that the study is not auditing them or seeking to determine non-compliance with reporting to NCIC, but rather that the effort is only for research.

The third expert panel focused on revising data elements, definitions, and questions in the draft LES-FA and LES-MC instruments to collect information on family abductions and other missing children. It included these experts:

LES-FA and -MC Data Elements Panel Experts
Ben Adams Social Science Analyst National Institute of Justice (previously with OJJDP)	Michelle Jeanis, PhD Assistant Professor University of Louisiana at Lafayette
Jennifer Bronson Statistician Bureau of Justice Statistics	Stacy Jeleniewski Analyst National Center for Missing and Exploited Children
Brecht Donoghue Senior Social Science Analyst National Institute of Justice (previously Deputy Associate Administrator with OJJDP)	Norweeta Milburn, PhD Director of Research and Evaluation, Nathanson Family Resilience Center, Professor-in-Residence, Department of Psychiatry and Biobehavioral Sciences University of California, Los Angeles
Chris Holloway Program Manager Family and Youth Services Bureau Administration for Children and Families Department of Health and Human Services	Peggy Plass, PhD Professor James Madison University

Like the other panels, this too was conducted as a webinar and moderated by Dr. Finkelhor. The focus of this panel was to draw on the expertise of individuals knowledgeable about child abductions and missing children, youth homelessness and runaways, and issues related to custody and visitation orders. The panel discussed how to revise data elements, definitions, and questions in the draft instruments for collecting information on family abductions and other missing children.

The panel recommended that in defining “missing,” the study include all episodes where a caretaker is calling to locate a child, and where the caretaker includes a foster parent or agency. The panel members noted that information on the recovery of missing children, particularly runaways, was likely to be scarce as it often relies on a caretaker calling law enforcement to report that that his/her child has returned home, which happens infrequently.

In addition, members of the project staff consulted individually with these experts on police management information systems and search capabilities to learn more about how these systems could be used to search for relevant cases for all three studies:

Bob Koenig

Vice President of Sales

TriTech Software Systems

Erica Mathis

Vice President of Sales

TriTech Software Systems

Dan Twohig

Vice President

Smart Public Safety Solutions Division

Motorola

9. Explanation of any payment or gift to respondents

No payments or gifts will be given to respondents for participating in in the LES-SK, LES-FA, or LES-MC pilots or the national LES-SK.

10. Assurances of privacy provided to respondents

All personally identifiable data collected under are protected under the confidentiality provisions of Title 34, United States Code, Section 10231. A copy of this section is included in this submission (Appendix 4). Regulations implementing this legislation require that OJP staff and its data collection agents maintain the confidentiality of the information and specify necessary procedures for guarding this confidentiality. A copy of these regulations (28 CFR Part 22) is included in this submission (Appendix 5). The cover letters that accompany the LES-SK, LES-FA, and LES-MC pilots and the national LES-SK will notify persons responsible for providing these data that their response is voluntary and that the identity of all participants and victims will be held confidential as required (Appendix 6).

Westat and UNH project staff will take the following precautions to ensure the anonymity and confidentiality of all data collected:

• All Westat and UNH project staff working on the project will be instructed in the privacy requirements of the survey and will be required to sign statements affirming their obligation to maintain privacy;

• Personally identifiable information will be stored separately from the participants' responses. Hard-copy personally identifiable information will be stored in locked file cabinets in a locked fieldroom with limited access to the key. Names and contact information will be stored separately from responses; and

• Data files that are delivered will contain no personal identifiers.

Westat and UNH have extensive experience protecting and maintaining the privacy of respondent data collected from surveys. Westat has implemented several procedures to protect privacy of survey participants:

• Electronic personally identifiable information data will be stored on secure, password protected network servers. All data will be accessible only to key Westat and UNH study staff working on the project (e.g., the PIs, project director, data collectors, and data analysts).

• Westat operates web servers, database servers, and other specialized application servers for hosting project-related web sites and other Internet-supported services. A comprehensive firewall system with redundant firewalls, routers, and other devices are configured and actively managed to provide maximum security between the public Internet and Westat systems. Internet access is permitted only to public facing servers that host web sites. These systems must conform to Westat IT security policies and procedures.

• Communications between web servers and users is encrypted using transport layer security (TLS) protocol. This widely available security protocol is frequently used to support e-commerce transactions and encryption of data in motion using digital certificates.

• Operating systems are maintained with the latest approved releases and updated regularly with applicable security and feature patches as they are made available by the vendors. Westat subscribes to several security alert services to stay abreast of emerging security issues or product vulnerabilities. Westat also conducts weekly scans of servers for newly identified vulnerabilities. Procedures are in place to respond to early warnings about security threats whether they occur during or outside regular business hours. Our response protocol includes immediate action to protect systems, inform users, gather information, and apply additional protective measures such as newly released security software updates, when appropriate.

11. Justification for sensitive questions

The information being collected is deidentified data on missing and abducted children episodes from police records. This information is typical of all the widely used crime and victimization data that are collected and published from law enforcement, including the UCR, the NIBRS, and many local law enforcement publications. While the topic area might be considered sensitive, the questions being posed to law enforcement officers about cases known to or investigated by them are not sensitive.

11. Estimate of respondent burden including annualized hourly cost

There are an estimated 1,712 total burden hours (1,259 hours for law enforcement investigators and 453 for NCMEC and MCC database administrators) requested for the three pilots and the national LES-SK.

Table A-1 presents the number of respondents, frequency of response, and annual hour burden for law enforcement investigators and NCMEC and MCC database administrators. The assumptions used to estimate burden hours are based on experience with previous NISMARTs, expert testing of instruments using missing child cases, discussion with NCMEC, MCC and law enforcement staff, and the study team’s professional experience with similar data collections:

• LES-SK Pilot: Based on preliminary testing of the LES-SK using records on old missing child cases provided by NCMEC, NIJ/OJJDP expect the 20 law enforcement investigators to spend an average of 40 minutes completing the online case detail survey, including time to read the advance letter and provide comments on problem questions in the instrument (20 x 40 minutes = 13.3 hours). NIJ/OJJDP expect the 20 law enforcement investigators to spend 20 minutes completing the telephone debriefing about the online survey, based on prior experience with debreifings about new instruments (20 x 20 minutes = 6.7 hours). Given the salience of these cases to the investigators, it is expected that all will respond. In addition, NIJ/OJJDP expect the NCMEC database administrator to spend 5 hours pulling 20 stereotypical kidnappings cases from the NCMEC database for use in testing the online survey, based on discussions with NCMEC (1 x 5 hours = 5 hours). The total amount of time for the LES-SK pilot is 25 hours.

• National LES-SK: A total of 4,727 law enforcement agencies are included in the national stratified cluster sample of 400 PSUs (Primary Sampling Units). All of these agencies will receive the mail screener. NIJ/OJJDP estimate that 2,836 (60 percent) of the law enforcement agencies will complete the screener by mail, based on the response rate for the mail screener obtained for NISMART-3. The great majority of these will have no stereotypical kidnapping cases during the 1-year timeframe of the survey and it is estimated the average time to complete the mail screener to be 15 minutes, based on the experience of previous NISMART data collections (2,836 x 15 minutes = 709.05 hours), and that 1,891 will not respond (1,891 x 3 minutes = 94.54 hours). NIJ/OJJDP estimate that 1,229 (26 percent) of the law enforcement agencies will complete the mail screener by telephone, based on the percentage of mail screeners completed by telephone in NIMSART-3. NIJ/OJJDP estimate that the time to complete the screener by telephone will be 4 minutes. Estimate of 4 minutes per complete are based on NISMART-3 findings that 4.7 percent of screeners (1,229 x 4.7 percent = 58) will identify a stereotypical kidnapping case and will take 30 minutes to complete the telephone survey as previous experience shows telephone administration takes twice as long as self-administration and 95.3 percent (1,229 x 95.3 percent = 1,171) will have no stereotypical kidnapping cases and take 3 minutes to complete with time estimate based on expert testing [((58 x 30 minutes)+(1,173 completes x 3 minutes))/1,229 = 4 minutes]. Total time to complete screener via telephone is estimated at 81.9 hours (1,229 x 4 minutes), and 662 will not respond (662 x 3 minutes = 33.1 hours).

• NIJ/OJJDP estimate that 204 cases will be identified that appear to meet the definition of a qualifying stereotypical kidnapping case. The estimate is based on the number of cases identified in NISMART-3 for telephone followup from the screener and searches of other databases.³⁶ Investigators of these cases will be asked to complete the online survey about case details. NIJ/OJJDP estimate that 161 (79 percent) of the law enforcement officers will complete the case detail instrument (estimate again based on the percentage of investigators who completed this component for NISMART-3) with 145 (90 percent) completing online, based on the ease of this method and assumptions of its preference over telephone administration. NIJ/OJJDP estimate that the online instrument will take an average of 40 minutes to complete based on the finding of 30 minutes to complete the detailed case survey from expert testing using closed missing child cases, 7 minutes to read invitation letters and FAQs and 3 minutes for data retrieval (145 x 40 minutes = 96.6 hours), 59 will not respond online (59 x 3 minutes = 2.95 hours). NIJ/OJJDP estimate that 16 (10 percent) of the 161 law enforcement officers who complete the detailed case survey will do it via telephone interview and that the interview will take 60 minutes to complete based on previous experience showing telephone administration takes twice as long as self-administration and assuming the survey invitation letters and FAQs are not read and no data retrieval is needed because of telephone administration (16 x 60 minutes = 16 hours), and that 43 will not respond (43 x 3 minutes = 2.15 hours).

• NIJ/OJJDP estimate the time for NCMEC and state MCCs database administrators to conduct a database search of any stereotypical kidnapping cases in their states to be 4 hours based on discussions with NCMEC and three state MCCs and expect that NCMEC and all the state MCCs in the 49 states where the sampled PSUs are located will participate. The total amount of time for the National LES-SK study is 1,236.5 hours.

• LES-FA Pilot: The sample size for Components 1 and 2 of the pilot is 30 law enforcement investigators who will be asked to search their database for cases of family abductions occurring in a 1-year period. NIJ/OJJDP estimate the search will take an average of 3 hours, based on prior experience with agency database searches, and that 28 will comply, given the salience of the topic (28 x 3 hours = 84 hours), 2 will decline (2 x 3 minutes = 6 minutes). NIJ/OJJDP estimate that database administrators for NCMEC and the MCCs associated with states in the sample will all agree to conduct database searches for the agencies in Component 1 and Component 2 and that these searches will take an average of 4 hours, based on discussions with NCMEC and three MCCs (31 x 4 hours = 124 hours). NIJ/OJJDP estimate that all 5 agencies selected for telephone debriefing in Component 1 will participate and the interviews will take an average of 20 minutes to complete based on prior experience with debriefing interviews on new instruments (5 x 20 minutes = 1.67 hours). Given the importance of the topic, NIJ/OJJDP expect that all 10 of the law enforcement investigators selected to complete the Component 3 case detail telephone survey will participate and that the interview will take 30 minutes (10 x 30 minutes = 5 hours). The estimate of 30 minutes is based on project staff and LEA expert review on the time to complete the self-administered survey (15 minutes) multiplied by 2 for telephone administration. NIJ/OJJDP estimate that 18 of the 20 investigators selected to complete the Component 3 case detail online survey will comply and that the instrument will take an average of 15 minutes, based on project staff and LEA expert review (18 x 15 minutes = 4.5 hours), and that 2 will decline (2 x 3 minutes = 6 minutes). NIJ/OJJDP expect that all 18 investigators who complete the online survey will agree to participate in the 20-minute debriefing telephone interview (18 x 20 minutes = 6 hours). The total amount of time for the LES-FA pilot is 225.4 hours.

• LES-MC Pilot: The sample size for Components 1 and 2 of the pilot is 30 law enforcement investigators who will be asked to search for case of missing children occurring in a 1-month period. NIJ/OJJDP estimate the search will take an average of 3 hours, based on prior experience with agency database searches, and that 28 will comply, given the salience of the topic (28 x 3 hours = 84 hours), 2 will decline (2 x 3 minutes = 6 minutes). NIJ/OJJDP estimate that database administrators for NCMEC and the MCCs associated with states in the sample will all agree to conduct database searches for the agencies in Component 1 and Component 2 and that these searches will take an average of 4 hours, based on discussions with NCMEC and three MCCs (31 x 4 hours = 124 hours). NIJ/OJJDP estimate that all 5 agencies selected for telephone debriefing in Component 1 will participate and the interviews will take an average of 20 minutes to complete based on prior experience with debriefing interviews on new instruments (5 x 20 minutes = 1.67 hours). Given the importance of the topic, NIJ/OJJDP expect that all 10 of the law enforcement investigators selected to complete the Component 3 case detail telephone survey will participate and that the interview will take 30 minutes (10 x 30 minutes = 5 hours). The estimate of 30 minutes is based on project staff and LEA expert review on the time to complete the self-administered survey (15 minutes) multiplied by 2 for telephone administration. NIJ/OJJDP estimate that 18 of the 20 investigators selected to complete the Component 3 case detail online survey will comply and that the instrument will take an average of 15 minutes, based on project staff and LEA expert review (18 x 15 minutes = 4.5 hours), and that 2 will decline (2 x 3 minutes = 6 minutes). NIJ/OJJDP expect that all 18 investigators who complete the online survey will agree to participate in the 20-minute debriefing telephone interview (18 x 20 minutes = 6 hours). The total amount of time for the LES-MC pilot is 225.4 hours.

Table A-1: Estimates of annualized burden hours and costs

Respondent	Number of Respondents	Number of Responses per Respondent	Total Annual Responses	Average burden per response (in hours)	Total burden (in hours)	Hourly rate ($)	Total annualized cost ($)
Pilot LES-SK Pilot
Law Enforcement Survey
Completed (a)	20	1	20	0.6667	13.3340	40.06	534.16
Attempted	0	0	0	0	0.0000	40.06	0.00
Telephone Debriefing
Completed (b)	20	1	20	0.3333	6.6660	40.06	267.04
Attempted	0	0	0	0	0.0000	40.06	0.00
NCMEC Identification of Cases
Completed (c )	1	1	1	5	5.0000	42.81	214.05
Attempted	0	0	0	0	0.0000	42.81	0.00
Total	41		41		25.0000		1,015
National LES-SK
Mail Screener Completed by Mail – Law Enforcement
Completed (d)	2836	1	2836	0.25	709.0500	40.06	28404.54
Attempted	1891	1	1891	0.05	94.5400	40.06	3787.27
Mail Screener Completed by Telephone - Law Enforcement
Completed (e)	1229	1	1229	0.0667	81.9756	40.06	3283.94
Attempted	662	1	662	0.05	33.0890	40.06	1325.55
Web Survey
Completed (f)	145	1	145	0.6667	96.6048	40.06	3869.99
Attempted	59	1	59	0.05	2.9550	40.06	118.38
Telephone interviews for web survey
Completed (g)	16	1	16	1	16	40.06	644.97
Attempted	43	1	43	0.05	2.1500	40.06	86.13
NCMEC and MCC Search of Records and Identification of Cases
Completed (h)	50	1	50	4	200.0000	42.81	8562.00
Attempted	0	0	0	0	0.0000	42.81	0.00
Total	6,931		6,931		1,236.464		50,083
Pilot LES-FA
Law Enforcement Search of Records for Components 1 and 2
Completed (i)	28	1	28	3	84.0000	40.06	3365.04
Attempted	2	1	2	0.05	0.1000	40.06	4.01
NCMEC and MCC Search of Records for Components 1 and 2
Completed (h)	31	1	31	4	124.0000	42.81	5308.44
Attempted	0	0	0	0	0.0000	42.81	0.00
Law Enforcement Telephone Debriefing (Component 1 only)
Completed (b)	5	1	5	0.3333	1.6665	40.06	66.76
Attempted	0	0	0	0	0.0000	40.06	0.00
Law Enforcement Case Detail Telephone Survey for Component 3
Completed (j)	10	1	10	0.5	5.0000	40.06	200.30
Attempted	0	0	0	0	0.0000	40.06	0.00
Law Enforcement Case Detail Online Survey Component 3
Completed (k)	18	1	18	0.25	4.5000	40.06	180.27
Attempted	2	1	2	0.05	0.1000	40.06	4.01
Law Enforcement Telephone Debriefing for Component 3
Completed (b)	18	1	18	0.3333	5.9994	40.06	240.34
Attempted	0	0	0	0	0.0000	40.06	0.00
Total	114		114		225.3659		9,369
Pilot LES-MC
Law Enforcement Search of Records for Components 1 and 2
Completed (i)	28	1	28	3	84.0000	40.06	3365.04
Attempted	2	1	2	0.05	0.1000	40.06	4.01
NCMEC and MCC Search of Records for Components 1 and 2
Completed (h)	31	1	31	4	124.0000	42.81	5308.44
Attempted	0	0	0	0	0.0000	42.81	0.00
Law Enforcement Telephone Debriefing (Component 1 only)
Completed (b)	5	1	5	0.3333	1.6665	40.06	66.76
Attempted	0	0	0	0	0.0000	40.06	0.00
Law Enforcement Case Detail Telephone Survey for Component 3
Completed (j)	10	1	10	0.5	5.0000	40.06	200.30
Attempted	0	0	0	0	0.0000	40.06	0.00
Law Enforcement Case Detail Online Survey Component 3
Completed (k)	18	1	18	0.25	4.5000	40.06	180.27
Attempted	2	1	2	0.05	0.1000	40.06	4.01
Law Enforcement Telephone Debriefing for Component 3
Completed (b)	18	1	18	0.3333	5.9994	40.06	240.34
Attempted	0	0	0	0	0.0000	40.06	0.00
Total	114		114		225.3659		9,369
Grand Total	7,200		7,200		1712.1963		69,836

(a) Estimate of 40 minutes includes 5 minutes to read invitation letter and FAQs, 30 minutes to complete survey based on expert testing using closed missing child cases, and 5 minutes for respondents to provide written comments.

(b) Estimate of 20 minutes based on prior experience with debriefing on new instruments

(c ) Estimate of hours to identify cases based on discussion with NCMEC

(d) Estimate of number of screeners completed by mail based on NISMART-3 results of 60% of screeners completed by mail (4,727*.60) = 2,836. Estimate of 15 minutes based on NISMART-3 time to complete.

(e) Estimate of number of screeners complete by telephone based on NISMART-3 results of 26% of screener completed by telephone (4,727*.26)=1,229. Estimate of 4 minutes per complete based on NISMART-3 findings that 4.7% of screeners will identify a case and will take 30 minutes to complete telephone survey (previous experience showing telephone administration takes twice as long as self-administration) and 95.3% will have no cases and take 3 minutes to complete (based on expert testing) (1,229*.047)=58; (1,229*.953)=1,171. ((58*.5)+(1,171*.05)/1,229 = 0.07 hours or 4 minutes

(f) Estimate total of 204 possible SK cases identified by screeners, internet searches and NCMEC based on number of cases identified in NISMART-3. Assume full survey will be completed by 79% of respondents (using NISMART-3 rates) and that web will be greatly preferred over telephone (based on prior survey experience) so 90% will be completed by web. (204*.79)= 161. 161*.90=145.

Estimate 40 minutes for full survey based on 30 minutes to complete survey based on expert testing using closed missing child cases, 7 minutes to read invitation letters and FAQs and 3 minutes for data retrieval.

(g) Estimate 10% of surveys completed will be complete by telephone 161*.10 =16; Estimate 60 minutes to complete based on previous experience showing telephone administration takes twice as long as self-administration. Assume invitation letters and FAQs not read and no data retrieval needed.

(h) Estimate of hours to search of records to identify cases based on discussions with NCMEC and 3 MCCs

(i) Estimate of hours per search based on prior experience with agency database searches

(j) Estimate of 30 minutes for telephone survey based on project staff and LEA expert review x 2 based on prior experience with telephone vs. self-administration.

(k) Estimate of 15 minutes to complete web survey based on project staff and LEA expert review

The estimated annualized cost for this collection is $69,836. This includes $50,443 for law enforcement officers at $40.06 per hour (job category “Detectives and Criminal Investigators” code #33-3021); and $19,393 for NCMEC and MCC database administrators at $42.81 per hour (job category “Database Administrators” code #15-1141). The estimate of costs is based on the burden estimates and utilizes the U.S. Department of Labor, Bureau of Labor Statistics, May 2017 National Occupational and Wage Statistics (https://www.bls.gov/oes/current/oes_nat.htm).

11. Estimates of Other Total Annualized Cost Burden.

There are no capital/start-up or ongoing operation/maintenance costs associated with this information collection.

11. Cost to the Federal Government

The total estimated cost to the Federal Government for the three pilot tests and the national LES-SK is $1,083,328. This includes costs of $999,968, over a period of 36 months, for work conducted by the data collection agent (Westat) and its subcontractor (UNH) for planning, developing survey instruments and methodologies, preparation of materials, collecting the data, evaluating the results, and generating deliverables and reports. In addition to the costs for data collection, an NIJ GS-Level 13 social scientist will be responsible for overseeing the data collection agent’s work at 15-percent time for 36 months ($49,500) plus a fringe benefit rate of 28-percent ($13,860). The NIJ cost to produce and print a statistical bulletin on results of the national LES-SK survey are estimated at $20,000.

11. Reason for Changes in Burden

Not applicable as this is a new data collection.

11. Project Schedule and Publication Plans

The anticipated schedule for major project activities and deliverables are found in Table A-2.

Table A-2: Major Project Activities and Deliverables

Activity	Dates	Major Deliverables	Due Date
LES-SK Pilot	Apr. – Aug. 2019	Pilot Test Results LES-SK Final Administration Plan	July 2019 Aug. 2019
LES-SK National Data Collection	Aug. 2019 – Aug. 2020	Draft Methodology Report	Aug. 2020
LES-FA Pilot	Mar. – Aug. 2020	Draft Report on FA Pilot Findings Final Report on FA Pilot Findings	July 2020 Aug. 2020
LES-MC Pilot	Oct. 2020 – June 2021	Draft Report on MC Pilot Findings Final Report on MC Pilot Findings Draft Administration Plan for National FA and MC Data Collections Final Administration Plan for National FA and MC Data Collections	Mar. 2021 Apr. 2021 May 2021 June 2021
Dissemination of Findings	Aug. 2020 – June 2021	Final Study Methodology Report OJJDP Bulletin on LES-SK Scholarly article(s) for publication	Oct. 2020 Dec. 2020 June 2021

11. Display of Expiration date

All data collection instruments will display the OMB control number and expiration date.

11. Exceptions to certification for paperwork reduction act. Submissions

There are no exceptions to the Certification for Paperwork Reduction Act (5 CFR 1320.9) for this study.

1 The detailed case-level survey was completed by investigators in a telephone interview for 78.6% of identified cases. In another 12.6%, the case-level survey was completed from media reports that were sufficiently detailed to answer most interview questions.

2 https://fas.org/irp/agency/doj/fbi/is/ncic.htm; https://legalbeagle.com/7643538-access-ncic-database.html

3 Sedlak, A.J., Finkelhor, D., Hammer, H., and Schultz, D. (2002). National estimates of missing children: An overview (NISMART Bulletin Series, National Incidence Studies of Missing, Abducted, Runaway, and Thrownaway Children, NCJ 196465). Washington, DC: Office of Juvenile Justice and Delinquency Prevention, U.S. Department of Justice.

4 Wolak, J., Finkelhor, D., and Sedlak, A.J. (2016). Child victims of stereotypical kidnappings known to law enforcement in 2011 (Juvenile Justice Bulletin, NCJ 249249). Washington, DC: Office of Juvenile Justice and Delinquency Prevention, U.S. Department of Justice.

5 Sedlak et al. (2002), op. cit.

6 Hammer, H., Finkelhor, D., and Sedlak, A.J. (2002). Runaway/thrownaway children: National estimates and characteristics (NISMART Bulletin Series, National Incidence Studies of Missing, Abducted, Runaway, and Thrownaway Children, NCJ 196469). Washington, DC: Office of Juvenile Justice and Delinquency Prevention, U.S. Department of Justice.

7 Sedlak, A.J., Finkelhor, D., and Brick, J.M. (2017). National estimates of missing children: Updated findings from the Survey of Parents and Other Primary Caretakers (Juvenile Justice Bulletin). Washington, DC: Office of Juvenile Justice and Delinquency Prevention, U.S. Department of Justice.

8 Ibid.

9 National Incidence Studies of Missing Children Reported to Law Enforcement

10 Greenblatt, A. (2012). The face that changed the search for missing kids. National Public Radio. Available: https://www.npr.org/2012/05/24/153623769/the-face-that-changed-the-search-for-missing-kids

11 Luthern, A. (2018). Frustrated with police, Milwaukee families with missing relatives turn to social media, activists. Journal Sentinel. Available: https://www.jsonline.com/story/news/crime/2018/07/26/milwaukee-missing-persons-families-rely-social-media-not-police/796516002/

12 Brown, S. (2014). Why police can’t always do more when it comes to runaway teens. CBS 6 Noon News, A Tribune Broadcasting Station. Available: https://wtvr.com/2014/11/12/why-police-cant-always-do-more-when-it-comes-to-runaway-teens/

13 Stoever, J.K. (2017). Parental abduction and the state intervention paradox. Washington Law Review, 92, 861-936. Available
https://digital.lib.washington.edu/dspace-law/bitstream/handle/1773.1/1690/92WLR0861.pdf?sequence=1

14 Stoever, J.K. (2017). Most kidnapped children are taken by a parent. That doesn’t mean they’re safe. The Washington Post. Available:
https://www.washingtonpost.com/outlook/most-kidnapped-children-are-taken-by-a-parent-that-doesnt-mean-theyre-safe/2017/07/21/8340cefe-6bc9-11e7-b9e2-2056e768a7e5_story.html?noredirect=on&utm_term=.fc8da31cf884

15 Department of Justice, Office of Justice Programs. (2018). OJJDP Awards More Than $104 Million To Protect Youth, Find Missing Children, And Prosecute Child Exploitation. Available: https://ojp.gov/newsroom/pressreleases/2018/ojp-news-10152018_a.pdf .

16 In 2017, the National Human Trafficking Hotline received reports on nearly 2,500 cases in the U.S. involving minors. See https://humantraffickinghotline.org/states. The NHTH produces a number of resources for public outreach campaigns to communities, educators, and healthcare workers.

17 The issue has received specific attention from the media (cf. http://www.baltimoresun.com/news/opinion/oped/bs-ed-sex-trafficking-20170328-story.html), organizations such the AAUW (cf. https://www.aauw.org/what-we-do/public-policy/aauw-issues/human-trafficking/), foundations (https://mbfchildsafetymatters.org/2017/01/10/concerned-human-trafficking/ https://sharedhope.org/the-problem/faqs/ ), and government agencies (https://www.cdc.gov/violenceprevention/sexualviolence/trafficking.html https://www.acf.hhs.gov/otip )

18 Shutt, J.E., Miller, J.M., Schreck, C.J., and Brown, N.K. (2004). Reconsidering the leading myths of stranger child abduction. Journal of Criminal Justice Studies: A Critical Journal of Crime, Law and Society. 17, 127-134. Available: https://www.tandfonline.com/doi/abs/10.1080/0888431042000217688

19 Quinet, K. (2012). Missing Persons. Problem-oriented guides for police, Problem-specific guides series, Guide No. 66. Center for Problem-Oriented Policing, Inc. Available: http://blogs.lexisnexis.com/public-safety/wp-content/uploads/2014/12/missing_persons.pdf

20 Dedel, K. (2010) Juvenile Runaways. Problem-oriented guides for police, Problem-specific guides series, Guide No. 37. Center for Problem-Oriented Policing, Inc. Available: http://www.justiceacademy.org/iShare/Library-COPS/cops-p095-pub.pdf

21 Lampinen, J.M. (2017). Missing and exploited children. In C.J. Schreck, M.J. Leiber, H.V. Miller, and K. Welch (Eds.), The encyclopedia of juvenile delinquency and justice. Hoboken, NJ: John Wiley & Sons. Available: https://onlinelibrary.wiley.com/doi/abs/10.1002/9781118524275.ejdj0145

22 Kennedy, M.A. (2015). Missing and exploited children. In W.G. Jennings (Ed.), The Encyclopedia of Crime and Punishment. Hoboken, NJ: John Wiley & Sons, Inc. Available: https://onlinelibrary.wiley.com/doi/abs/10.1002/9781118519639.wbecpx259

23 Plass, P.S. (2007). Secondary victimizations in missing child events. American Journal of Criminal Justice, 32, 30-44. Available: https://link.springer.com/article/10.1007/s12103-007-9008-9

24 Kappeler, V.E. and Potter, G.W. (2018). The Mythology of Crime and Criminal Justice, 5^th Ed. Long Grove, IL: Waveland Press.

25 Howell, J.C. (2003). Preventing and reducing juvenile delinquency: A comprehensive framework. Thousand Oaks, CA: Sage.

26 Crossland, K. and Dunlap, G. (2015). Running away from foster care: What do we know and what do we do? Journal of Child and Family Studies, 24, 1697-1706.

27 Ben-Arieh, A. (2010). From Child Welfare to Children Well-Being: The Child Indicators Perspective. Springer.

28 OJJDP NISMART publication series: https://www.ojjdp.gov/Publications/PubResults.asp?sei=85.

29 The national LES-FA and LES-MC will be implemented pending future funding.

30 Wolak, J., Finkelhor, D., and Sedlak, A.J. (2016). Child Victims of Stereotypical Kidnappings Known to Law Enforcement in 2011. Juvenile Justice Bulletin - NCJ 249249, 1-20.

31 Sedlak, A., Finkelhor, D., Brick, M. and Wolak, J. (2016). Law Enforcement Survey (LES) Redesign: Planning Papers and Draft Instruments. Rockville Institute: Rockville, MD.

32 https://www.fbi.gov/file-repository/2016-ncic-missing-person-and-unidentified-person-statistics.pdf/view. The official statistics note on page 2 that the Missing Person Circumstances (MPC) field is optional. Of the more than 647K records entered in 2016, just shy of 317K had information in the MPC field, i.e., only 48.8% of the records entered had any circumstances coded.

33 https://fas.org/irp/agency/doj/fbi/is/ncic.htm; https://legalbeagle.com/7643538-access-ncic-database.html

34 National Child Search Assistance Act (42, U.S.C. 5779. 5780)

35 Sedlak, A., Finkelhor, D., Brick, M. and Wolak, J. (2016). Law Enforcement Survey (LES) Redesign: Planning Papers and Draft Instruments. Rockville Institute: Rockville, MD.

36 Lounsbury, K., Wolak, J., Broene, P. (2016). NISMART-3: Law Enforcement Study (LES-3) Technical Report.  Office of Juvenile Justice and Delinquency Prevention (OJJDP), U.S. Department of Justice, Washington, DC.

19