12.07.2015 Views

Fourth National Incidence Study of Child Abuse and Neglect (NIS–4)

Fourth National Incidence Study of Child Abuse and Neglect (NIS–4)

Fourth National Incidence Study of Child Abuse and Neglect (NIS–4)

SHOW MORE
SHOW LESS
  • No tags were found...

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

design. The measure <strong>of</strong> size was the population <strong>of</strong> children under 18 years old in Census 2000.Large counties with a child population over 400,000 were selected with certainty. Undersizedcounties (those with fewer than 4,000 children) were combined with contiguous counties to formgeographically compact PSUs with sufficiently large populations to justify data collectionefforts. Constructing multi-county PSUs considered local CPS agency jurisdictions (whichusually, but not always, follow county boundaries) <strong>and</strong> limited the number <strong>of</strong> counties in a PSUto no more than six.Under this scheme, the 14 largest counties were selected with certainty, leaving alisting <strong>of</strong> 2,282 noncertainty PSUs nationwide, 1,685 <strong>of</strong> which were single-county <strong>and</strong> restmultiple-county PSUs. Statisticians allocated these noncertainty PSUs to 48 strata, based onCensus region, metro status, the NCANDS child victim substantiation rate, FBI crime rate, <strong>and</strong>percentage <strong>of</strong> households headed by single females with children. (Choice <strong>of</strong> these factorsderived from prior analyses on predictors <strong>of</strong> child maltreatment rates.) Two PSUs were selectedfrom each stratum, using probability-proportionate-to-size (PPS) <strong>of</strong> the Census 2000 population<strong>of</strong> children, which identified 96 noncertainty PSUs: 88 single-county PSUs, 6 two-county PSUs,1 three-county PSU <strong>and</strong> 1 five-county PSU. Thus, the final NIS–4 sample consisted <strong>of</strong> 110PSUs (14 certainty <strong>and</strong> 96 noncertainty), which included 122 counties—14 certainty <strong>and</strong> 108noncertainty counties.CPS agencies <strong>and</strong> cases. The CPS agency sample comprised all 126 CPS agenciesserving all or part <strong>of</strong> a sampled county. Most were county-level agencies serving individualcounties. Eligible CPS cases were those reported to these CPS agencies during the 13-weekstudy reference period <strong>and</strong> assigned for investigation. CPS case sampling followed the approachused in the NIS–3. Fatality cases were included with certainty. An approximately equalprobabilitysample was r<strong>and</strong>omly selected from the remaining eligible cases. Sampled caseswere assigned to receive CPS Maltreatment data forms, while CPS Summary data forms wereassigned to the remaining eligible cases for use during unduplication <strong>and</strong> weighting, as describedin those sections. (The Data Collection section describes the different data forms.) Theseprocedures resulted in listings <strong>of</strong> 140,206 CPS case investigations during the study referenceperiods, <strong>and</strong> identified a sample <strong>of</strong> 11,930 <strong>of</strong> these for CPS Maltreatment data forms. To supportspecial analyses examining whether any <strong>of</strong> the children not investigated during the study periodwere reported to CPS <strong>and</strong> investigated later, the NIS–4 also obtained Summary data forms on allcases reported <strong>and</strong> assigned for investigation during the month following the reference period.The NIS–4 gathered an additional 38,398 Summary data forms for this component.A-4

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!