Over several years the Bureau of Labor Statistics (BLS) has been integrating some of its business establishment surveys on compensation into one consolidated program, the National Compensation Survey. This has entailed developing new data collection and processing systems, methods and concepts.
During the first fully-integrated period of data collection, BLS was unable to obtain information from employers concerning type of health insurance plan for March 2003 for approximately 8 percent of employees. This resulted in the use of a new category "plan type not available" in table 1 of the news release "Employee Benefits in Private Industry, 2003" issued by BLS on September 17, 2003, and in table 1 of the BLS Summary "National Compensation Survey: Employee Benefits in Private Industry in the United States, March 2003" issued in April 2004. The use of the new category in those tables meant that the accompanying estimates shown for the percent of workers with medical care, dental care, and vision care were not comparable to data published for prior years.
For the March 2004 data issued on November 9, 2004, BLS implemented a statistical imputation process for estimating the missing information on health insurance plan types. Imputation is a widely-used statistical procedure by which missing data are inferred from other information provided by the survey respondent and other respondents.
In order to make the March 2003 estimates comparable with those for March 2004 and earlier years, BLS has recalculated the 2003 data on health care plan access and participation using the imputation technique. The recalculated estimates for 2003 are available from the tables listed below. The recalculated participation data are also available from the Employee Benefits historical database.
Last Modified Date: December 22, 2004