2008 Massachusetts
Health Insurance Survey Methodology Report
By
Sharon K. Long, Timothy Triplett
Publication
Date: December 18, 2008
The text below is an excerpt from the complete document. Read the full
report in PDF format.
Abstract
The Urban Institute, along with its subcontractor, International
Communications Research, conducted the 2008 Massachusetts Health Insurance
Survey (HIS) for the Massachusetts Division of Health Care Finance and Policy
to obtain information on health insurance coverage and access to and use of
health care for the non-institutionalized population in Massachusetts. This report provides
information about the methods used to collect and analyze the 2008 HIS data.
Introduction
The Massachusetts Division of Health Care Finance and Policy (DHCFP)
contracted with the Urban Institute, and its subcontractor, International
Communications Research/ICR, to conduct the 2008 Massachusetts Health Insurance
Survey (HIS). The goal of the 2008 HIS is to document health insurance coverage
and access to and use of health care for the non-institutionalized population
in Massachusetts.
This report provides information about the methods used to collect and analyze
the 2008 HIS data.
The report is organized in seven sections. The first section (Section I)
describes the design of the sample used for the 2008 survey. Section II
discusses the design of the survey instrument. The next two sections describe
our data collection strategy (Section III) and data processing and preparation
(Section IV), respectively. Section V addresses the response rate to the
survey. Section VI describes the survey weights and variance estimation. The
final section (Section VII) presents estimates of the uninsurance
rate in Massachusetts
in 2008.
I. Sample Design
The 2008 HIS has a dual sample frame design that combines a
random-digit-dial (RDD) telephone sample and an address-based (AB) household
sample. The decision to rely on the dual-frame sample for the 2008 HIS reflects
the changing telephone environment in the United States. Historically, RDD
telephone interviewing has been the method of choice for many survey data
collection efforts given the strength of its randomization method (random digit
dialing), ease of administering complex questionnaires using computerized
interviewing systems, excellent coverage of the overall population (given that
less than 2% of Americans live in a household without telephone service), and
relatively low cost. Survey coverage refers to the extent to which the sample
frame for a survey includes all members of the target population. A survey
design with a gap in coverage raises the possibility of bias if the individuals
missing from the sample frame (e.g., households without landline telephones)
differ from those in the sample frame. Unfortunately, the coverage of the
overall population in RDD surveys is changing as more and more households are
relying on cell phones and giving up their landline telephones. Cell phone
numbers are typically not called in RDD surveys.
Reliable estimates of the number of cell-phone-only households are not
available at the state level; however, data from the National Health Interview
Survey (NHIS) found that, nationally, 15.8% of households were cell-phone-only
households in the July-December 2007 period (Blumberg and Luke 2008). This is
up from 12.8% over the same period in the prior year (Blumberg and Luke, 2007),
and is expected to continue to increase over time. In fact, NHIS data from 2003
to 2007 show a steady increase in the share of cell-phone-only households in
the United States
(Blumberg and Luke, 2008).
(End of excerpt. The entire
report is available in PDF format.)
http://www.urban.org/publications/411814.html
