Methodology
The California Employer Health Benefits Survey is a joint product of the
California Health Care Foundation (CHCF) and the National Opinion
Research Center (NORC). The survey was designed and analyzed by
researchers at NORC and administered by National Research (NR). The
findings are based on a random sample of 705 interviews with employee
benefit managers in private firms in California. NR conducted interviews
from June to October 2016. As with prior years, the sample of firms was
drawn from the Dun & Bradstreet list of private employers with three or
more workers. The margin of error for responses among all employers is
+/– 3.8%, for responses among employers with 3 to 199 workers it is +/–
5.0%, and among employers with 200 or more workers it is +/– 5.9%.
Some exhibits do not sum to 100% due to rounding effects.
The Kaiser Family Foundation sponsored this survey of California
employers from 2000 to 2003. A similar employer survey was also
conducted in 1999 in California, in conjunction with the Center for Health
and Public Policy Studies at the University of California, Berkeley. The
Health Research and Educational Trust (HRET) collaborated on these
surveys from 1999 to 2004. The Center for Studying Health System Change
collaborated on these surveys from 2005 to 2006.
This survey instrument is similar to a national employer survey conducted
annually by the Kaiser Family Foundation and HRET.
The US results in this study are either from the published reports,
or in a few cases, from author calculations from the survey’s public use
files. A full analysis of the US data set is available on the foundation’s
website at www.kff.org. Both the California and US surveys asked questions
about health maintenance organizations (HMOs), preferred provider
organizations (PPOs), point-of-service (POS) plans, and high-deductible
health plans with a savings option (HDHP/SOs). Conventional (fee-for-
service) plans are generally excluded from the plan type analyses because
they compose such a small share of the California market
Many variables with missing information were identified as needing
complete information within the database. To control for item
nonresponse bias, missing values within these variables were imputed
using a hot-deck approach. Calculation of the weights follows a common
approach. First, the basic weight is determined, followed by a survey
nonresponse adjustment. Next, the weights are trimmed in order to
reduce the influence of weight outliers. Finally, a post-stratification
adjustment is applied.
All statistical tests in this chart pack compare either changes over time, a
plan-specific estimate with an overall estimate, or subcategories versus all
other firms (e.g., firms with 3 to 9 workers vs. all other firms). Tests include
t-tests and chi-square tests, and significance was determined at p < .05
level. Due to the complex nature of the design, standard errors are
calculated in SUDAAN.
An important note about the methodology: Rates of change for total
premiums, for worker or employer contributions to premiums, and other
variables calculated by comparing dollar values in this report to data
reported in past CHCF or KFF publications should be used with caution
due to both the survey’s sampling design and the way in which plan
information is collected. Rates calculated in this fashion not only reflect a
change in the dollar values but also a change in enrollment distribution,
thus creating a variable enrollment estimate. However, rates of change in
premiums are collected directly as a question in the California survey. This
rate of change holds enrollment constant between the current year and
the previous year, thus creating a fixed enrollment estimate. Because the
survey does not collect information on the rate of change in other
variables, additional rates are not reported. The national survey conducted
by Kaiser/HRET, however, stopped collecting directly rates of change in
premiums in its 2008 survey. Therefore, the rate of change in total
premiums in the US provided in this report uses a variable enrollment
estimate.
California Employer Health Benefits
AB O U T T H I S SERI ES
The California Health Care Almanac is an
online clearinghouse for data and analysis
examining the state’s health care system. It
focuses on issues of quality, affordability,
insurance coverage and the uninsured, and
the financial health of the system with the
goal of supporting thoughtful planning and
effective decisionmaking. Learn more at
www.chcf.org/almanac.
AU T H O R S
Heidi Whitmore, Principal Research Scientist
Jon Gabel, Senior Fellow
NORC at the University of Chicago
FOR MORE INFORMAT IO N
California Health Care Foundation
1438 Webster Street, Suite 400
Oakland, CA 94612
510.238.1040
www.chcf.org
41