|
Activity Number:
|
330
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #308818 |
|
Title:
|
A Case Study in Dual-Frame Estimation Methods
|
|
Author(s):
|
Keith Rust*+ and Inho Park and Marianne Winglee and Gary Shapiro
|
|
Companies:
|
Westat and Westat and Westat and Westat
|
|
Address:
|
1650 Research Blvd, Rockville, MD, 20850,
|
|
Keywords:
|
Multi-stage sampling ; Sampling weights ; Rare populations ; Domain estimation ; Single-frame (SF) estimator ; Pseudo-maximum likelihood (PML) estimator
|
|
Abstract:
|
In the National Incidence Study of Child Abuse and Neglect (NIS), the survey uses a multistage and multiframe design to broaden coverage of possible reporting sources in order to produce more accurate and efficient estimates. Within sampled counties the NIS collects lists of maltreated children investigated by Child Protective Services (CPS) agencies and those seen by samples of professionals in agencies such as police departments, hospitals, and schools. While the list frame of children investigated is known for CPS agencies, only sampled data are available for the other agency categories. As a result duplication of cases across agencies cannot be fully identified. We present a modified single-frame estimator designed to accommodate this situation. This estimator is compared with the classic single-frame estimator and the pseudo-maximum likelihood estimator in a simulation study.
|
