OVERVIEW

OVERVIEW

Current Population Survey



Introduction

The Current Population Survey (CPS) is the source of the official Government statistics on employment and unemployment. The CPS has been conducted monthly for over 50 years. Currently, we interview about 48,000 households monthly, scientifically selected on the basis of area of residence to represent the Nation as a whole, individual States, and other specified areas. Each household is interviewed once a month for four consecutive months one year, and again for the corresponding time period a year later. This technique enables us to obtain month-to-month and year-to-year comparisons at a reasonable cost while minimizing the inconvenience to any one household.

Although the main purpose of the survey is to collect information on the employment situation, a very important secondary purpose is to collect information on the demographic status of the population, information such as age, sex, race, marital status, educational attainment, and family structure. From time to time additional questions are included on such important subjects as health, education, income, and previous work experience. The statistics resulting from these questions serve to update similar information collected once every 10 years through the decennial census, and are used by Government policymakers and legislators as important indicators of our Nation's economic situation and for planning and evaluating many Government programs.

The CPS provides current estimates of the economic status and activities of the population of the United States. Because it is not possible to develop one or two overall figures (such as the number of unemployed) that would adequately describe the labor market, the CPS is designed to provide a large amount of detailed and supplementary data. Such data are made available to meet a wide variety of needs on the part of users of labor market information.

Thus, the CPS is the only source of monthly estimates of total employment (both farm and nonfarm); nonfarm self-employed persons, domestics, and unpaid workers in nonfarm family enterprises; wage and salary employees; and, finally, estimates of total unemployment.

It provides the only available distribution of workers by the number of hours worked (as distinguished from aggregate or average hours for an industry), permitting separate analyses of part-time workers, workers on overtime, etc. The survey is also the only comprehensive current source of information on the occupation of workers and the industries in which they work. Information is available from the survey not only for persons currently in the labor force but also for those who are outside the labor force. The characteristics of such persons -- whether married women with or without young children, disabled persons, students, older retired workers, etc., can be determined. Information on their current desire for work, their past work experience, and their intentions for job seeking are also available.

The March CPS, also known as the Annual Demographic File, contains the basic monthly demographic and labor force data described above, plus additional data on work experience, income, noncash benefits, and migration.

CPS Sample

The CPS sample is based on the civilian noninstitutional population of the United States. The sample is located in 792 sample areas comprising 2,007 counties and independent cities with coverage in every State and in the District of Columbia.

In all, some 60,000 housing units or other living quarters are assigned for interview each month; about 48,000 of them containing approximately 94,000 persons 15 years old and over are interviewed. Also included are demographic data for approximately 28,000 children 0-14 years old and 450 Armed Forces members living with civilians either on or off base within these households. The remainder of the assigned housing units are found to be vacant, converted to nonresidential use, contain persons with residence elsewhere, or are not interviewed because the residents are not found at home after repeated calls, are temporarily absent, or are unavailable for other reasons. Approximately 12,000 noninterview households are present each month. The resulting file size is approximately 136,000 records. In March of each year supplemental data are collected for Armed Forces members residing with their families in civilian housing units or on a military base. The Armed Forces members, however, are not asked the monthly labor force questions. In addition, the March CPS is supplemented with a sample of Hispanic households identified the previous November. This results in the addition of about 3,000 households (2,500 Interviewed) in the March CPS. The inclusion of the additional sample of Hispanic households began in 1976.

A more precise explanation regarding the CPS sample design is provided in Technical Paper 40, The Current Population Survey: Design and Methodology.

For a more detailed discussion about the basic labor force data gathered on a monthly basis in the CPS survey, see the Bureau of Labor Statistics Report No. 463 and the Current Population Report P-23, No. 62, issued jointly by the Bureau of Labor Statistics and the Bureau of the Census in October, 1976, and entitled Concepts and Methods Used In Labor Statistics derived from the Current Population Survey.

Questionnaire

Questionnaire facsimiles of the March 1997 CPS income supplement are shown in Appendix D in this documentation.

Revisions To The March CPS Processing System

Introduction. Between 1988 and 1989 a new computer processing system was introduced for the March Current Population Survey. This processing system rewrite was long overdue; the system in use before this year was first introduced in March 1976 and was never fully updated to reflect the numerous questionnaire changes that had taken place since that time. In addition, the programs used to process the CPS file were written in a computer language that is being phased out of use at the Census Bureau. While the March 1989 file is the first to reflect this new processing system, the March 1988 file was reprocessed based on these new procedures in order to: 1) better evaluate the new processing procedures, and 2) allow year-to-year comparisons to be made between income years 1987 and 1988 using a consistent processing system.

While the following section deals mainly with modifications to the March imputation procedures and their subsequent effect on income and poverty rates, it should be pointed out that all of the processing programs were rewritten in 1989, so that not only are the files from 1989 forward based on a somewhat different imputation system, but also reflect a rewritten weighting system, data acceptance program, family relationship edits, and new procedures to match income supplement records to the monthly CPS file. As a result, it is difficult to ascertain whether differences (especially those based on relatively small bases) are the result of imputation or other processing differences between the original and revised files.

Since the Census Bureau began imputing the missing income data on the CPS in 1962, there have been three major revisions to the processing system (in 1967, 1976, and 1989). Through all of these revisions, the basic strategy used in make imputations has remained the same. This approach, commonly referred to as ``hot deck'' imputation, assigns missing responses to sample persons with information from matched sample persons with similar demographic and economic information who answered these questions.

Under the new March processing system, there were three main modifications to the income imputation programs:

1. The edits and imputations were expanded to reflect the full detail of the March income questionnaire. The original processing system was still based on a less detailed CPS income questionnaire that had not been in use since 1980.

2. Under the revised processing procedures, entire sets of March income and noncash benefits data were imputed to supplement noninterviews from the same interviewed person. Under the original processing procedures, earnings, unearned income, and noncash benefits were imputed in separate stages during the processing system. Thus, the new processing system imputes noninterviews more efficiently and is better able to preserve the correlation between earnings, unearned income, and noncash benefits.

3. Both the old and new processing systems employ what are called ''statistical matches'' to link sample cases with reported data to those requiring imputation for missing responses. In the original processing system, under certain circumstances, reported data were removed and replaced with imputed data during the statistical matching process. Under the new system, reported data are never removed. This was accomplished by expanding the number of statistical matches in the income imputation system.

File Structure

There is a household record for each household or group quarters. The household record is followed by one of three possible structures:

A. If the household contains related persons and is not a group quarters household:

1. The family record appears next followed by person records for members of the family who are not also members of a related subfamily. The person records would be ordered: family householder, spouse of family householder, children in the family, and other relatives of the family householder.

2. The above records may be followed by one or more related subfamily records, each related subfamily record being followed immediately by person records for members of that related subfamily. The person records would be ordered: reference person of the related subfamily, spouse of subfamily reference person, and children of subfamily reference person.

3. The above records may be followed by one or more unrelated subfamily records, each unrelated subfamily record being followed immediately by person records for members of that unrelated subfamily. The person records would be ordered: unrelated subfamily reference person, spouse of subfamily reference person, and children of subfamily reference person.

4. The above records may be followed by one or more persons living with nonrelatives family records, each to be followed by the person record for the unrelated individual it represents. (See Figure 1, page 2-4.)

B. If the household contains a householder with no relatives and is not a group quarters household:

1. The family record for the nonfamily householder is followed immediately by the person record for that nonfamily householder.

2. These records may be followed by one or more unrelated subfamily records, each unrelated subfamily record being followed immediately by the person records for members of that unrelated subfamily.

3. These records may be followed by one or more family records for persons living with nonrelatives, each person living with nonrelatives family record being followed immediately by the person record for that person living with nonrelatives. (See Figure 2, page 2-5.)

C. If the household is a group quarters, each person is defined as a person living with nonrelatives. There will be a family record followed by a person record for each person in the group quarters.

Relationship of Current Population Survey Files to Publications

Each month, a significant amount of information about the labor force is published by the Bureau of Labor Statistics in the Employment and Earnings and Monthly Labor Review Reports.

As mentioned previously, the CPS also serves as a vehicle for supplemental inquiries on subjects other than employment which are periodically added to the questionnaire. From the basic and supplemental data the Bureau of the Census issues four series of publications under the general title Current Population Reports:

P-20 Population Characteristics
P-23 Special Studies
P-27 Farm Population
P-60 Consumer Income

Of particular interest to users of the March microdata file would be those reports based on information collected in March. These reports include the following titles:

P-20 Population Profile of the United States: (Year)
P-20 Household and Family Characteristics: March (Year)
P-20 Households, Families, Marital Status, and Living Arrangements: March (Year)
P-20 Geographical Mobility (Years)
P-20 Educational Attainment in the United States (Years)
P-20 Persons of Hispanic Origin in the United States (Year)

P-60 Money Income and Poverty Status of Families and Persons in the United States: (Year)
P-60 Characteristics of the Population Below the Poverty Level: (Year)
P-60 Characteristics of Households Receiving Selected Noncash Benefits: (Year)

All Current Population Reports may be obtained by subscription from the U.S. Government Printing Office. Subscriptions are available as follows: Population Characteristics, Special Studies, Farm

Population, and Consumer Income series (P-20,

P-23, P-27, P-60) combined, $71 per year (sold as a package only); Population Estimates and Projections (P-25), $25 per year. Single issues may be ordered separately; ordering information and prices are in the Bureau of the Census Catalog and Guide, in Census and You, and the Monthly Product Announcement (MPA). 

Figure 1.  Illustration of Record Sequence for Households Containing a Family.


    Household Record


        Family Record


          Person 1 (Householder) Record

          Person 2 (Spouse) Record

                   .

                   .

                   .

                   .

          Person n (Family Member)


        Family (Related Subfamily Record)


          Person 1 (Related Subfamily Reference Person) Record

          Person 2 (Spouse) Record

                   .

                   .

                   .

                   .

          Person n (Related Subfamily Member) Record


        Family (Unrelated Subfamily) Record


          Person 1 (Unrelated Subfamily Reference Person) Record

          Person 2 (Spouse) Record

                   .

                   .

                   .

                   .

          Person n (Unrelated Subfamily Member) Record


        Family (Persons Living With Nonrelatives) Record


          Person 1 (Person Living With Nonrelatives) Record


Figure 2.  Illustration of Record Sequence for Households Containing a Nonfamily Householder.

	


	    Household Record

	

	        Family (Nonfamily Householder) Record

	

	          Person (Nonfamily Householder) Record

	

	        Family (Unrelated Subfamily) Record

	

	          Person 1 (Unrelated Subfamily Reference Person) Record

	          Person 2 (Spouse) Record

	                   .

	                   .

	                   .

	                   .

	          Person n (Unrelated Subfamily Member) Record


	


	        Family (Person Living With Nonrelatives) Record

	

	          Person (Persons Living With Nonrelatives) Record

	

	

Figure 3.  Illustration of Record Sequence for Group Quarters.1

	

	    Household Record

	

	        Family (Persons Living With Nonrelatives) Record

	

	          Person (Persons Living With Nonrelatives) Record

	

	

	
1.  NOTE:  Each person in group quarters is by definition a person living with nonrelatives.


Geographic Limitations

One set of estimates that can be produced from CPS microdata files should be treated with caution. These are estimates for individual metropolitan areas. Although estimates for the larger areas such as New York, Los Angeles, and so forth, should be fairly accurate and valid for a multitude of uses, estimates for the smaller metropolitan areas (those with populations under 500,000) should be used with caution because of the relatively large sampling variability associated with these estimates. For these areas, estimates comparing percent distributions and ratios will provide data with less sampling variability than estimates of levels will.

It should be kept in mind that the sample design and methods of weighting CPS data are geared towards producing estimates for the entire Nation. Consequently, data for states are not as reliable as national data, and the file will lose some of its utility in certain applications. For further discussion of such considerations, the user should consult The Current Population Survey: Design and Methodology (Technical Paper 40, U.S. Bureau of the Census).

The nature of the work done by each individual investigator using the microdata file will determine to what extent his/her requirements for precision will allow using some of the smaller geographic areas identified on the file.

Weights

For all CPS data files a single weight is prepared and used to compute the monthly labor force status estimates. An additional weight was prepared for the earnings universe which roughly corresponds to wage and salary workers in the two outgoing rotations. This is explained below in the section on earnings data. However, the difference in content of the March CPS supplement requires the presentation of additional weights: a household weight, a family weight, and a March supplement weight. In this section we briefly describe the construction and use of these weights. Chapter 5 of Technical paper 40, The Current Population Survey: Design and Methodology provides documentation of the weighting procedures for the CPS both with and without supplement questions.

The final weight, which is the product of several adjustments, is used to produce population estimates for the various items covered in the regular monthly CPS. This weight is constructed from the basic weight for each person, which represents the probability of selection for the survey. The basic weight is adjusted for special sampling situations and failure to obtain interviews from eligible households (noninterview adjustment). A two-stage ratio estimation procedure adjusts the sample population to the known distribution of the entire population. This two-stage ratio estimation process produces factors which are applied to the basic weight (after the special weighting and noninterview adjustments are made) and results in the final weight associated with each record. In summary, the final weight is the product of: (1) the basic weight, (2) adjustments for special weighting, (3) noninterview adjustment, (4) first stage ratio adjustment factor, and (5) second stage ratio adjustment factor. This final weight should be used when producing estimates from the basic CPS data.

Differences in the questionnaire, sample and data uses for the March CPS supplement result in the need for additional adjustment procedures to produce what is called the March supplement weight. The sample for the March CPS supplement is expanded to include male members of the Armed Forces who are living in civilian housing or with the family on a military base, as well as additional Hispanic households which are not included in the monthly labor force estimates.

The expanded sample and the need to have a husband and wife receive the same weight has resulted in a weighting system which produces the March supplement weight. The March supplement weight should be used for producing estimates from March supplement data.

Finally, household and family weights are the weights assigned from the householder or reference person after all adjustments have been made and should be used when tabulating estimates of families-households.

Earnings Data

Beginning in 1982, usual hourly and weekly earnings data appear on the Annual Demographic File (ADF) for that portion of the population roughly corresponding to wage and salary workers (self-employed persons in incorporated businesses are excluded, although they are normally included with the wage and salary population). These data are now collected on a monthly basis in the two outgoing rotation groups as part of the basic CPS labor force interview.

Since the intent of the regular collection of earnings data was to initiate a family earnings data series, all persons in the two outgoing rotations receive an ''earnings weight,'' even if they are not eligible for the earnings item. The earnings weight is a simple ratio-estimation to the person's labor force status by age, race, and sex. When tabulating estimates of earnings based on basic CPS data, use the earnings weight.

Further information on this earnings series is contained in Technical Description of the Quarterly Data on Weekly Earnings from the Current Population Survey, BLS Report #601, July 1980. This report is available on request from the U.S. Department of Labor, Bureau of Labor Statistics, Washington, D.C. 20212. Attn: Office of Inquiries and Correspondence.

NOTE: For 1982 and 1983, usual weekly earnings are not present for individuals who were not paid on an hourly basis.

Contacts - -Table of Contents

Source: U.S. Census Bureau