stages of ratio estimation. For some items, the reduction is substantial. The resultant gains in reliability are greatest in estimates of month-to-month change, although gains are also usually obtained for estimates of level in a given month, change from year to year, and change over other intervals of time.

Seasonal Adjustment

Over the course of a year, the size of the Nation's labor force, the levels of employment and unemployment, and other measures of labor market activity undergo sharp fluctuations due to such seasonal events as changes in weather, reduced or expanded production, harvests, major holidays, and the opening and closing of schools. Because these seasonal events follow a more or less regular pattern each year, their influence on statistical trends can be eliminated by adjusting the statistics from month to month. These adjustments make it easier to observe the cyclical and other nonseasonal movements in the series. In evaluating changes in a seasonally adjusted series, it is important to note that seasonal adjustment is merely an approximation based on past experience. Seasonally adjusted estimates have a broader margin of possible error than the original data on which they are based, because they are subject not only to sampling and other errors but are also affected by the uncertainties of the seasonal adjustment process itself.

Since January 1980, national labor force data have been seasonally adjusted with a procedure called X-11 ARIMA (Auto-Regressive Integrated Moving Average), which was developed at Statistics Canada as an extension of the standard X-11 method.

At the beginning of each calendar year, projected seasonal adjustment factors are calculated for use during the January-June period. In July of each year, BLS calculates and publishes in Employment and Earnings projected seasonal adjustment factors for use in the second half, based on the experience through June. Revisions of historical data, usually for the most recent 5 years, are made only at the beginning of each calendar year. However, as a result of revisions to the estimates for 1970-81 based on 1980 census population counts, revisions to seasonally adjusted series in early 1982 were carried back to 1970. In 1994, data were revised only for that year because of the major redesign and 1990 census-based population controls, adjusted for the estimated undercount, introduced into the Current Population Survey. In 1996, 1990-93 data also were revised to incorporate these 1990 census-based population controls and seasonally adjusted series were revised back to 1990.

Presentation and Uses

The CPS provides comprehensive information on the social, demographic, and economic characteristics of the

civilian noninstitutional population 16 years of age and over, the employed, and the unemployed.

Each month, the employment and unemployment data are published initially in The Employment Situation news release about 2 weeks after they are collected. The release includes a narrative summary and analysis of the major employment and unemployment developments together with tables containing statistics for the principal data series. The news release is available on the Internet and can be accessed via the World Wide Web. The Universal Resource Locator is: http://stats.bls.gov/cpshome.htm. The news release also is available on the BLS fax-on-demand service.

More detailed statistics are subsequently published in Employment and Earnings, a monthly periodical. The detailed tables provide information on the labor force, employment, and unemployment by a number of characteristics, such as age, sex, race, marital status, industry, and occupation. In addition, the January issue of Employment and Earnings provides annual averages on employment and earnings by detailed occupational categories, union affiliation, and employee absences.

About 32,000 of the monthly labor force data series plus quarterly and annual averages are maintained in LABSTAT, the Bureau's public database, on the Internet. They can be accessed from http:/stats.bls.gov/cgi-bin/dsrv?lf. In most cases, these data are available from the inception of the series through the current month.

The CPS is used also for a program of special inquiries to obtain detailed information from particular segments, or for particular characteristics of the population and labor force. About four such special surveys are made each year. The inquiries are repeated annually in the same month for some topics, including the earnings and total incomes of individuals and families (published by the Bureau of the Census); the extent of work experience of the population during the calendar year; and the employment of schoolage youth, high school graduates and dropouts, and recent college graduates. Surveys also are made periodically on subjects such as contingent work, job mobility, job tenure, displaced workers, and disabled veterans.

Generally, the persons who provide information for the monthly CPS questions also answer the supplemental questions. Occasionally, the kind of information sought in the special survey requires the respondent to be the person about whom the questions are asked. The results of these special surveys are usually first published as news releases and subsequently in the Monthly Labor Review or BLS reports. In addition to the regularly tabulated statistics described above, special data can be generated through the use of the CPS individual record (micro) files. These files contain records of the responses to the survey questionnaire for all individuals in the survey. While the files can be used simply to create additional cross-sectional detail, an important feature of their use is the ability to match the

records of specific individuals at different points in time during their participation in the survey. By matching these records, data files can be created which lend themselves to some limited longitudinal analysis and the investigation of short-run labor market dynamics. An example is the statistics on gross labor force flows, which indicate how many persons move among the labor force status categories each month. Although a number of technical difficulties lie in the path of more complete use of these data files for the purposes of longitudinal analysis, this area is continually being investigated and holds considerable promise. Microdata files are available for all months since January 1976 and for various months in prior years. These data are made available on magnetic tape, CD-Rom, or diskette. Address inquires regarding these files to: Division of Data Development, Bureau of Labor Statistics, Room 4965, 2 Massachusetts Ave., NE, Washington, DC 20212-0001, phone 202-606-6345.

Limitations

Geographic. Although the present CPS sample is a Statebased design, the CPS produces reliable national monthly estimates. The sample does not permit the production of reliable monthly estimates for the States. However, demographic, social, and economic detail are published annually for the census regions and divisions, all States and the District of Columbia, 50 large metropolitan areas, and selected central cities. The production of subnational labor force and unemployment estimates is discussed in more detail in chapter 4 of this bulletin.

Sources of errors in the survey estimates. There are two types of errors possible in an estimate based on a sample survey-sampling and nonsampling. The mathematical discipline of sampling theory provides methods for estimating standard errors when the probability of selection of each member of a population can be specified. The standard error, a measure of sampling variability, can be used to compute confidence intervals that indicate a range of differences from true population values that can be anticipated because only a sample of the population has been surveyed. Nonsampling errors such as response variability, response bias, and other types of bias occur in complete censuses as well as sample surveys. In some instances, nonsampling error may be more tightly controlled in a well-conducted survey where it is feasible to collect and process the data more skillfully. Reinterview programs are often used to measure response variability and response bias. Estimation of other types of bias is one of the most difficult aspects of survey work, and often adequate measures of bias cannot be made. Nonsampling error. The full extent of nonsampling error is unknown, but special studies have been conducted to quantify some sources of nonsampling error in the CPS. The

effect of nonsampling error should be small on estimates of relative change, such as month-to-month change. Estimates of monthly levels would be more severely affected by nonsampling error.

Nonsampling errors in surveys can be attributed to many sources, e.g., the inability to obtain information about all persons in the sample; differences in the interpretation of questions; inability or unwillingness of respondents to provide correct information; inability to recall information; errors made in collecting and processing the data; errors made in estimating values for missing data; and failure to represent all sample households and all persons within sample households (undercoverage).

Nonsampling errors occurring in the interview phase of the survey are studied by means of a reinterview program. This program is used to estimate various sources of error as well as to evaluate and control the work of the interviewers. A random sample of each interviewer's work is inspected through reinterview at regular intervals. The results indicate, among other things, that the data published from the CPS are subject to moderate systematic biases.

The effects of some components of nonsampling error in the CPS data can be examined as a result of the rotation plan used for the sample, since the level of the estimates varies by rotation group. For example, unemployment estimates from a rotation group tend to be higher in the first and fifth months of interviewing.

Undercoverage in the CPS results from missed housing units and missed persons within sample households. The CPS covers about 92 percent of the decennial census population (adjusted for the undercount). It is known that the CPS undercoverage varies with age, sex, race, and Hispanic origin. Generally, undercoverage is larger for men than for women and larger for blacks, Hispanics, and other races than for whites. Ratio adjustment to independent age-sexrace-origin population controls, as described previously, partially corrects for the biases due to survey undercoverage. However, biases exist in the estimates to the extent that missed persons in missed households or missed persons in interviewed households have different characteristics than interviewed persons in the same age-sex-race-origin group. The independent population estimates used in the estimation procedure may be a source of error although, on balance, their use substantially improves the statistical reliability of many of the figures. Errors may arise in the independent population estimates because of underenumeration of certain population groups or errors in age reporting in the last census (which serves as the base for the estimates) or similar problems in the components of population change (mortality, immigration, etc.) since that date.

Sampling error. When a sample rather than the entire population is surveyed, estimates differ from the true population values that they represent. This difference, or sampling error, occurs by chance, and its variability is measured by

the standard error of the estimate. Sample estimates from a given survey design are unbiased when an average of the estimates from all possible samples would yield, hypothetically, the true population value. In this case, the sample estimate and its standard error can be used to construct approximate confidence intervals, or ranges of values, that include the true population value with known probabilities. If the process of selecting a sample from the population were repeated many times and an estimate and its standard error calculated for each sample then:

1. Approximately 68 percent of the intervals from one standard error below the estimate to one standard error above the estimate would include the true population value.

2. Approximately 90 percent of the intervals from 1.6 standard errors below the estimate to 1.6 standard errors above the estimate would include the true population value.

3. Approximately 95 percent of the intervals from two standard errors below the estimate to two standard errors

above the estimates would include the true population value.

Although the estimating methods used in the CPS do not produce unbiased estimates, biases for most estimates are believed to be small enough so that these confidence interval statements are approximately true.

Standard error estimates computed using generalized variance functions are provided in Employment and Earnings and other publications. Using replicate variance techniques, standard error estimates are generated. As computed, these standard error estimates reflect contributions not only from sampling error, but from some types of nonsampling error, particularly response variability. Because replicate variance techniques are somewhat cumbersome, simplified formulas called generalized variance functions (GVF's) have been developed for various types of labor force characteristics. The GVF can be used to approximate an estimate's standard error, but this only indicates the general magnitude of its standard error rather than a precise value.

Butz, W. P. and Plewes, T. J. (1989) "A Current Population Survey for the 21st Century," Proceedings of the Fifth Annual Research Conference, U.S. Department of Commerce, Bureau of the Census: Washington, DC, pp. 3-13.

Bregger, John E. "The Current Population Survey: A Historical View and BLS' Role," Monthly Labor Review, June 1984.

Cohany, Sharon R., Polivka, Anne E., and Rothgeb, Jennifer M. "Revisions in the Current Population Survey Effective January 1994," Employment and Earnings, February 1994. Dagum, Estela Bee. The X-11 ARIMA Seasonal Adjustment Method. Ottawa, Statistics Canada, January 1983 (Statistics Canada Catalogue No. 12-564E).

National Commission on Employment and Unemployment Statistics. Counting the Labor Force, 1979.

A comprehensive review of the entire labor market data system; includes an appraisal of current concepts and methodology and recommendations for further research and improvements.

"Overhauling the Current Population Survey," Monthly Labor Review, September 1993.

A trilogy of articles that provides additional information on the background of the CPS redesign.

President's Committee to Appraise Employment and Unemployment Statistics. Measuring Employment and Unemployment, 1962.

A review of all Federal statistical series on employment and unemployment and a comparison of the sources and uses of each series; includes a brief history of the develop

ment of labor force statistics, an evaluation of concepts and techniques, and recommendations for further research and improvements.

U.S. Department of Commerce, Office of Federal Statistical Policy and Standards. "An Error Profile: Employment as Measured by the Current Population Survey," Statistical Policy Working Paper 3, 1978.

A description of the potential sources of error in the CPS as they affect the national employment statistics. U.S. Department of Labor, Bureau of Labor Statistics, Employment and Earnings, Explanatory Notes and Estimates of Error, monthly.

An up-to-date, concise description of the concepts and methods used in the labor force data from the Current Population Survey. Provides tables which present sampling errors for various labor force series.

U.S. Department of Labor, Bureau of Labor Statistics, Geographic Profile of Employment and Unemployment, 1994, Bulletin 2469, 1995.

A tabular presentation of 1994 annual average data on the labor force characteristics of the population residing in the Census regions, the 50 States and the District of Columbia, 50 large metropolitan areas, and 17 central cities.

U.S. Department of Labor, Bureau of Labor Statistics, How the Government Measures Unemployment, Report 864, 1994.

A short, nontechnical discussion of the concepts and methods used in obtaining labor force statistics from the Current Population Survey.

Chapter 2.

Employment, Hours, and Earnings from the Establishment Survey

B

LS cooperates with State employment security agen

cies in the Current Employment Statistics (CES) survey to collect data each month on employment, hours, and earnings from a sample of nonfarm establishments (including government). In early 1996, this sample included over 390,000 reporting units. From these data, a large number of employment, hours, and earnings series in considerable industry and geographic detail are prepared and published each month. The employment data include series on all employees, women workers, and production or nonsupervisory workers. Hours and earnings data include average weekly hours, average weekly overtime hours, and average hourly and weekly earnings. For many series, seasonally adjusted data also are published.

Background

The first monthly studies of employment and payrolls by BLS began in 1915 and covered four manufacturing industries. Before 1915, the principal sources of employment data in the United States were the census surveys-the decennial Census of Population and the quinquennial Census of Manufactures. No regular employment data were compiled between the censuses.

In 1916, the BLS survey was expanded to cover employment and payrolls in 13 manufacturing industries; by 1923, the number had increased to 52, and by 1932, 91 manufacturing and 15 nonmanufacturing industries were covered by a monthly employment survey.

With the deepening economic crisis in 1930, President Hoover appointed an Advisory Committee on Employment Statistics which recommended extension of the Bureau's program to include the development of hours and earnings series. In 1932, Congress granted an increase in the BLS appropriation for the survey. In 1933, average hourly earnings and average weekly hours were published for the first time for total manufacturing, for 90 manufacturing industries, and for 14 nonmanufacturing categories.

During the Great Depression, there was controversy concerning the actual number of unemployed people; no reli

able measures of employment or unemployment existed. This confusion stimulated efforts to develop comprehensive estimates of total wage and salary employment in nonfarm industries, and BLS survey data produced such a figure for the first time in 1936.

Interest in employment statistics for States and areas also grew. Even before BLS entered the field in 1915, Massachusetts, New York, and New Jersey were preparing employment statistics. In 1915, New York and Wisconsin entered into cooperative agreements with BLS, whereby sample data collected from employers by a State agency would be used jointly with BLS to prepare State and national series. By 1928, five other States had entered into

such compacts, and another five were added by 1936. By 1940, estimates of total nonfarm employment for all 48 States and the District of Columbia were available. In 1996, cooperative arrangements were in effect with all 50 States, the District of Columbia, Puerto Rico, and the Virgin Islands.

Since 1949, the Current Employment Statistics (CES) program has been a fully integrated Federal-State project which provides employment, hours, and earnings information by industry on a national, State, and area basis. Between 1991 and 1994, BLS expanded the published industry detail for the service-producing sector, bringing it closer to parity with the goods-producing sector. During this period, BLS also more than doubled the number of specific industries for which data were published at the first release to include all 2-digit-level industries under the Standard Industrial Classification System.

In 1996, BLS completed implementation of computerassisted reporting through telephone interviews, touch-tone self-reporting, and voice recognition systems, and introduced electronic data interchange. In June 1996, improvements to seasonal adjustment techniques were introduced. Comprehensive research is being conducted for a new survey design, to be phased in over several years.

Establishment

Concepts

An establishment is an economic unit, such as a factory, mine, or store, which produces goods or services. It is generally at a single location and engaged predominantly in one type of economic activity. Where a single location encompasses two or more distinct activities, these are treated as separate establishments, provided that separate payroll records are available and certain other criteria are met.

Employment

Employment is the total number of persons employed full or part time in nonfarm establishments during a specified payroll period. Temporary employees are included. In general, data refer to persons who worked during, or received pay for, any part of the pay period that includes the 12th of the month, which is standard for all Federal agencies collecting employment data from business establishments. National employment figures for Federal Government establishments, however, represent the number of persons who were paid for the last full pay period of the calendar month except for the Department of Defense, which reports the number of civilian employees on the payroll the last day of the month; intermittent Federal Government workers are counted if they performed any service during the month.

Workers on an establishment payroll who are on paid

sick leave (when pay is received directly from the employer); on paid holiday or vacation; or who work during only a part of the specified pay period, even though they are unemployed or on strike during the rest of the pay period, are all counted as employed. Persons on the payroll of more than one establishment during the pay period are counted in each establishment which reports them, whether the duplication is due to turnover or dual jobholding. Persons are considered employed if they receive pay for any part of the specified pay period, but they are not considered employed if they receive no pay at all for the pay period. Since proprietors, the self-employed, and unpaid family workers do not have the status of paid employees, they are not included. Also excluded from the employed are domestic workers in households; persons who are on layoff, on leave without pay, or on strike for the entire pay period; and persons who were hired but have not yet started work during the pay period. The employment statistics for government refer to civilian employees only. All persons who meet these specifications are included in the designation "all employees," regardless of industry.

In addition to employment data for all employees, the survey also collects data on a major category of workers in each industry, differentiated primarily to ensure the expeditious collection of current statistics on hours and earnings. These groups of employees are designated production workers, construction workers, or nonsupervisory workers, depending upon the industry.

Data are collected for production workers in manufacturing and mining industries. In manufacturing, this group covers employees, up through the level of working supervisors, who engage directly in the manufacture of the establishment's product. Among those excluded from this category are persons in executive and managerial positions and persons engaged in activities such as accounting, sales, advertising, routine office work, professional and technical functions, and force-account construction. (Force-account construction is construction work performed by an establishment, primarily engaged in some business other than construction, for its own account and for use by its own employees.) Production workers in mining are defined in a similar manner. A more detailed description of the classes of employees included in the production and nonproduction worker categories in manufacturing is shown on the facsimile of the BLS 790 C schedule at the end of this chapter.

In construction, the term "construction workers" covers workers, up through the level of working supervisors, who are engaged directly on the construction project either at the site or in shops or yards at jobs ordinarily performed by members of construction trades. Excluded from this category are executive and managerial personnel, professional and technical employees, and workers in routine office jobs.

In the remaining private sector industries (transportation, communications, and public utilities; wholesale and