NISMART I had six major components. The Household Survey collected incidence and other information about each of the five missing child case types. The Police Records Study focused on nonfamily abductions. The Returned Runaway Study interviewed children who had returned home after running away to determine how well children's accounts about their experiences matched caretaker reports. The Juvenile Facilities Study collected information about children who ran away from these facilities. The FBI's Supplementary Homicide Report reanalysis developed a range estimate of how many children were murdered in conjunction with possible abductions by strangers. The Community Professionals Study analyzed data from a nation sample of child welfare agencies to develop an estimate of thrownaway (abandoned) children. One recommendation for NISMART II is that a telephone survey of approximately 78,000 households be conducted, using youth aged 12 to 17 as primary respondents and proxy respondents (parents or caretakers) for children younger than age 12. The second recommendation is that juvenile facilities be surveyed in a stratified random sample of counties; a modified police records study could be conducted in the same counties. The third recommendation is that data from the Third National Incidence Study of Child Abuse and Neglect be used to estimate the number of thrownaway children. Appended supplementary information and 17 references