Background

Before NISMART 1, there was no single reliable source of information about episodes involving missing children. NISMART 1 provided the first nationally representative, comprehensive data on the incidence of missing children events. It also made other important contributions to the scientific measurement of the problem. For instance, it provided clear, multilayered definitions of the missing children problem (e.g., how to define episodes of various types and how to distinguish serious from nonserious cases) resulting in five major categories and provided detailed estimates of the numbers for each type of episode.

No single research strategy can provide all the data needed to generate a comprehensive picture of the problem of missing children. Therefore, NISMART 1 included seven distinct data collection and data analysis efforts: a large telephone survey of households, a study of family networks, a survey of youth residential facilities, a study of returned runaways, a survey of police records, a reanalysis of Federal Bureau of Investigation data on child homicides, and a reanalysis of data from the Study of the National Incidence and Prevalence of Child Abuse and Neglect, conducted by Westat in 1986. The results of this work are discussed in many of the documents in the list of references at the end of this Bulletin.

After NISMART 1 was completed, OJJDP began to plan for NISMART 2 by reviewing and revising the research methods, definitions, and concepts that were used in NISMART 1. The major planning activities for NISMART 2 included:

• A thorough review and evaluation of all aspects of NISMART 1.

• A survey of key respondents to understand what information should be gathered in NISMART 2 and what options exist for obtaining this information.

• A planning symposium of law enforcement, research, and government experts to make recommendations for NISMART 2.

• An exploration of additional data sources and research methods that could enhance NISMART 1.

• Development of draft definitions, screening questions, and survey questions for a new household survey.

This planning activity revised and extended the approach, methodology, and concepts used by NISMART 1 to improve NISMART 2. For example, NISMART 1 provided data on five main categories of missing children episodes—runaways, thrownaways, children missing due to nonfamily abductions, children missing due to family abductions, and lost or otherwise missing children. These five categories of missing children were revised and expanded to eight in NISMART 2. The researchers combined the runaway/thrownaway category; preserved the nonfamily abduction category, including stereotypical child kidnaping; distinguished custodial interference situations from more serious family abductions; distinguished episodes that result from a child being lost from those in which a child is injured; and added a category of missing episodes that result from a simple miscommunication in which the child was not, in fact, in any danger. In order to capture information on sexual exploitation, researchers studied children who were sexually assaulted. While children in this eighth category are not necessarily missing, they share important risk factors with missing children.

Study Categories in NISMART 1 and 2
NISMART 1	NISMART 2
Runaway Thrownaway Nonfamily abduction Family abduction Lost or otherwise missing	Runaway/Thrownaway Nonfamily abduction Family abduction Custodial interference Lost and involuntarily missing Missing due to injury Missing due to false alarm situations Sexually assaulted

Who Is a Missing Child?

To the general public, the definition of a missing child may seem relatively simple: a child who is missing from home. However, the researchers conducting these studies realized that in order to measure "missingness" as accurately and as comprehensively as possible, they needed to define the concept more specifically. Based on the NISMART 1 experience and consultations with experts on the problem of missing children, the researchers developed the following working definitions for the eight main categories of children being studied.