Activity Number:
|
627
- Estimation with Nonprobability Samples
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 3, 2017 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Survey Research Methods Section
|
Abstract #324914
|
View Presentation
|
Title:
|
Correcting for the Multiplicity Issue in a Probability Sample of Homeless Youth
|
Author(s):
|
Daniela Golinelli* and Joan Tucker and William Shadel
|
Companies:
|
Mathematica Policy Research and RAND and RAND
|
Keywords:
|
hard to reach population ;
homeless youth ;
location sampling ;
multiplicity ;
sampling weights
|
Abstract:
|
Obtaining probability samples of homeless youth is challenging as it is difficult to generate a complete sampling frame listing all the members of the population of interest. Therefore, statisticians resort to alternative sampling methods such as location sampling. Location sampling consists of developing a list of locations where homeless youth are known to hang out. Location sampling introduces other challenges such as the multiplicity issue. The multiplicity issue arises because homeless youth can enter the sample in multiple ways. When the population of interest is homeless youth and not homeless youth-visits it is important to devise corrections to the sampling weights so that a sample of homeless youth is obtained. We will analyze the effects of these corrections on the estimates of smoking outcomes and background characteristics of homeless youth. We will compare those homeless youth who visit the sites in the frame frequently versus those who don't. The results of this analysis will speak to the importance of correcting for the multiplicity issue and how sensitive the results are when these corrections are computed in different ways.
|
Authors who are presenting talks have a * after their name.