|
Activity Number:
|
478
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #305442 |
|
Title:
|
PPS Network Sampling with Partial PSU Replacement
|
|
Author(s):
|
Monroe Sirken*+
|
|
Companies:
|
National Center for Health Statistics
|
|
Address:
|
3311 Toledp Rd., Hyattsville, MD, 20782,
|
|
Keywords:
|
finite population corrections ; psu size measures ; sampling small populations
|
|
Abstract:
|
The original version of the pps network sampling with partial replacement scheme (NS1)using the original psu size measures is more efficient than single-stage conventional pps sampling with replacement scheme (CS). However, the NS1 precision improvement is small if either the psu sampling rate is small or the average psu size measure is large. Two modified pps network sampling schemes are introduced that seek to improve NS1 precision by reducing the original psu size measures. This paper compares the precision of the modified network sampling schemes relative to the NC1 and CS schemes.
|
