Endnotes

2 A designated household respondent reports on household experiences with crimes such as burglary, vandalism, and theft from the property.

3 NCVS measures attempted and completed violent victimization using a series of survey questions. For questionnaire details, see Bureau of Justice Statistics (2000).

4 Information from the decennial census is used to describe the social and economic characteristics of communities throughout the United States. The research described in this Bulletin began before the release of data from the 2000 census. Although the data for each community were collected 5 years before the victimization interviews, they are highly reliable indicators of area characteristics because communities change slowly in most places (for more detail on the area measures, see Lauritsen, 2001).

5 The “relatives and nonrelatives” of the adult head of household include a wide variety of persons, such as brothers, sisters, parents, and boyfriends and girlfriends. Moreover, in many of the families with children and other relatives or nonrelatives, the youth was that relative or nonrelative (e.g., a youth living with an aunt and cousins or a youth living with another family). Unfortunately, it is not possible to use these data to make finer distinctions about various family types because information about the relationship of each child to each member of the household is not available. For instance, it is impossible to determine whether children living with their mother and grandmother are at lower risk for violence than children living with their mother and her sister. Also, children living with their mother and her boyfriend cannot be compared with those living with their mother and some other nonrelative because this information is not available. Although the data cannot address these important questions, they do permit a more detailed analysis of youth victimization than has been possible in the past.

6 The overall reliability of this factor is very high (alpha=.92). See Lauritsen (2001) for additional details on the results of this factor analysis and for information about the process of merging the area-identified NCVS data with census tract data.

7 Because of the small number of cases, conducting reliable analyses of more specific kinds of nonstranger violence (e.g., friends versus family members) is not possible.

8 NCVS uses 14 categories of nonequal income brackets to measure household income. For this reason, household income and group differences in household income can only be estimated. On the 14-point scale, the average single-parent household income is 6.9 (where 6=$17,500 and 7=$20,000), whereas the average two-parent household income is 10.6 (where 10=$30,000 and 11=$35,000).

9 The upturn shown in figure 1 (between community disadvantage and youth victimization) is similar to that found in analyses of adults (Lauritsen and White, 2001).

10 Additional analyses found that residential stability (an index representing the percent of persons who have lived in their home for more than 5 years and the percent of housing units that are occupied) is unrelated to youth victimization once disadvantage is taken into account. Similarly, no significant relationship was found between the percent of foreign-born persons or the percent of Latino residents and victimization risk (see Lauritsen, 2001).

11 See Finkelhor (1997) for a discussion of developmental processes and youth victimization.

12 The 1995 NCVS data show that victimization rates vary considerably with length of time in the current home. Among all respondents, victimization rates were 113.6 (per 1,000) for persons in the home 6 months or less, 72.8 for 6 months to 1 year, 57.9 for 1–2 years, 46.5 for 2–3 years, 41.9 for 3–4 years, 44.5 for 4–5 years, and 29.1 for 5 years or more. Rates for the period 6 months or less are believed to be overestimates because many of those interviews are unbounded (Bureau of Justice Statistics, 2000). Nonetheless, for length of residence beyond 6 months, the declines persist.

How Families and Communities Influence Youth Victimization	OJJDP Bulletin • November 2003