known distribution of these population characteristics. This is accomplished through two stages of ratio estimates as follows:

a. First-stage ratio estimate. This is a procedure in which the sample proportions are weighted by the known 1970 Census data on the color-residence distribution of the population. This step takes into account the differences existing at the time of the 1970 Census between the color-residence distribution for the Nation and for the sample areas.

b. Second-stage ratio estimate. In this step, the sample proportions are applied to independent current estimates of the population by age, sex, and color. Prior to January 1974 these estimates were prepared by carrying forward the most recent census data (1970) after taking account of subsequent aging of the population, births, deaths, and migration between the United States and other countries.

Beginning in 1974, the "inflation-deflation" method of deriving independent population controls was introduced into the CPS estimation procedures. In this procedure, the most recent census population adjusted to include estimated net census undercount by age, sex, and color (i.e., "inflated") is carried forward to each subsequent month and later age by adding births, subtracting deaths, and adding net migration. These postcensal population estimates are then "deflated" to census level to reflect the pattern of net undercount in the most recent census by age, sex, and color. The actual percent change over time in the population in any age group is preserved. 3. Composite estimate procedure. In deriving statistics for a given month, a composite estimating procedure is used which takes account of net changes from the previous month for continuing parts of the sample (75 percent) as well as the sample results for the current month. This procedure reduces the sampling variability of month-to-month changes especially and of the levels for most items also.

Rounding of estimates

The sums of individual items may not always equal the totals shown in the same tables because of independent rounding of totals and components to the nearest thousand. Differences, however, are insignificant.

Unemployment

Table B. Standard error of level of monthly estimates [In thousands]

Reliability of the estimates

Since the estimates are based on a sample, they may differ from the figures that would have been obtained if it were possible to take a complete census using the same schedules and procedures.

The standard error is a measure of sampling variability, that is, the variations that might occur by chance because only a sample of the population is surveyed. The chances are about 2 out of 3 that an estimate from the sample would differ from a complete census by less than the standard error. The chances are about 19 out of 20 that the difference would be less than twice the standard error.

Table A shows the average standard error for the major employment status categories, by sex, computed from data for past months. Estimates of change derived from the survey are

The standard error of the change in an item from one month to the next month is more closely related to the standard error of the monthly level for that item than to the size of the specific month-to-month change itself. Thus, in order to use the approximations to the standard errors of month-to-month changes as presented in table C, it is first necessary to obtain the standard error of the monthly level of the item in table B, and then find the standard error of the month-to-month change in table C corresponding to this standard error of level. It should be noted that table C applies to estimates of change between 2 consecutive months. For changes between the current month and the same month last year, the standard errors of level shown in table B are acceptable approximations.

Illustration. Assume that the tables showed the total number of persons working a specific number of hours as 15,000,000, an increase of 500,000 over the previous month. Linear interpola. tion in the first column of table B shows that the standard error of 15,000,000 is about 133,000. Consequently, the chances are about 68 out of 100 that the sample estimate differs by less than 133,000 from the figure which would have been obtained from a complete count of the number of persons working the given number of hours. Using the 133,000 as the standard error of the monthly level in table C, it may be seen that the standard error of the 500,000 increase is about 126,000.

for percentages derived from the survey. Linear interpolation may be used for percentages and base figures not shown in table D. As a general rule, percentages will not be published when the monthly base is less than 75,000 or the annual base is less than 35,000. Table E shows the standard error of percentage of monthly levels and consecutive month change for frequently analyzed unemployment rate series. These errors are computed from data for recent months. Errors on change for nonconsecutive months are slightly greater (by roughly a factor of 1.1 times the month-to-month error).

Table E. Standard error of percentage for major unemployment rates

The reliability of an estimated percentage, computed by using sample data for both numerator and denominator, depends upon both the size of the percentage and the size of the total upon which the percentage is based. Where the numerator is a subclass of the denominator, estimated percentages are relatively more reliable than the corresponding absolute estimates of the numerator of the percentage, particularly if the percentage is large (50 percent or greater). Table D shows the standard errors Table D. Standard error of percentage

Estimated percentage

Blue-collar workers

Farm workers

Construction

.52

.66

50

2.6 2.9 3.2

1.8 2.1 2.3

1.2

1.3 1.5

1.6

1.0

1.1

.6

67432

86432

075322

1.2

1.3

.9 1.0

4

.4

.3 .3

.2 .3

6307433

.18

.22

1.23

Base of percentages

(thousands)

COLLECTION

Payroll reports provide current information on wage and salary employment, hours, earnings, and labor turnover in nonagricultural establishments, by industry and geographic location.

Federal-State cooperation

Under cooperative arrangements with State agencies, the respondent fills out a single employment or labor turnover reporting form, which is then used for national, State, and area estimates. This eliminates duplicate reporting on the part of respondents and, together with the use of identical techniques at the national and State levels, insures maximum comparability of estimates.

State agencies mail the forms to the establishments and examine the returns for consistency, accuracy, and completeness. The States use the information to prepare State and area series and then send the establishment data to the BLS for use in preparing the national series.

Shuttle schedules

Two types of data collection schedules are used: Form BLS 790-Monthly Report on Employment, Payroll, and Hours; and Form DL 1219-Monthly Report on Labor Turnover. These schedules are of the "shuttle" type, with space for each month of the calendar year. The collection agency returns the schedule to the respondent each month so that the next month's data can be entered. This procedure assures maximum comparability and accuracy of reporting, since the respondent can see the figures he has reported for previous months.

Form BLS 790 provides for entry of data on the number of full- and part-time workers on the payrolls of nonagricultural establishments and, for most industries, payroll and manhours of production and related workers or nonsupervisory workers for the pay period which includes the 12th of the month. Form DL 1219 provides for the collection of information on the total number of accessions and separations, by type, during the calendar month.

CONCEPTS

Industrial classification

Establishments reporting on Form BLS 790 and Form DL 1219 are classified into industries on the basis of their principal product or activity determined from information on annual sales volume. This information is collected each year on a supplement to the monthly 790 or 1219 report. For an establishment making more than one product or engaging in more than one activity, the entire employment of the establishment is included under the industry indicated by the most important product or activity.

All national, State, and area employment, hours, earnings, and labor turnover series are classified in accordance with the Standard Industrial Classification Manual, Bureau of the Budget, 1967.

Industry employment

Employment data, except those for the Federal Government, refer to persons on establishment payrolls who received pay for any part of the pay period which includes the 12th of the month. For Federal Government establishments, employment figures represent the number of persons who occupied positions on the last day of the calendar month. Intermittent workers are counted if they performed any service during the month.

The data exclude proprietors, the self employed, unpaid volunteer or family workers, farm workers, and domestic workers in households. Salaried officers of corporations are included. Government employment covers only civilian employees; military personnel are excluded.

Persons on establishment payrolls who are on paid sick leave (when pay is received directly from the firm), on paid holiday or paid vacation, or who work during a part of the pay period and are unemployed or on strike during the rest of the period, are counted as employed. Not counted as employed are persons who are laid off, on leave without pay, or on strike for the entire period or who are hired but have not been paid during the period.

Industry hours and earnings

Hours and earnings data are derived from reports of payrolls and man-hours for production and related workers in manufacturing and mining, construction workers in contract construction, and nonsupervisory employees in the remaining private nonagricultural components. For Federal Government, hours and earnings relate to all employees, both supervisory and nonsupervisory. Terms are defined below. When the pay period reported is longer than 1 week, figures are reduced to a weekly basis.

Production and related workers include working foreman and all nonsupervisory workers (including leadmen and trainees) engaged in fabricating, processing, assembling, inspection, receiving, storage, handling, packing, warehousing, shipping, maintenance, repair, janitorial and watchman services, product development, auxiliary production for plant's own use (e.g., power plant), and recordkeeping and other services closely associated with the above production operations.

Construction workers include the following employees in the contract construction division: Working foremen, journeymen, mechanic's apprentices, laborers, etc., whether working at the site of construction or in shops or yards, at jobs (such as precutting and preassembling) ordinarily performed by members of the construction trades.

Nonsupervisory employees include employees (not above the working supervisory level) such as office and clerical workers, repairmen, salespersons, operators, drivers, physicians, lawyers, accountants, nurses, social workers, research aids, teachers,

draftsmen, photographers, beauticians, musicians, restaurant workers, custodial workers, attendants, linemen, laborers, janitors, watchmen, and similar occupational levels, and other employees whose services are closely associated with those of the employees listed.

Payroll covers the payroll for full- and part-time production, construction, or nonsupervisory workers who received pay for any part of the pay period which includes the 12th of the month. The payroll is reported before deductions of any kind, e.g., for old age and unemployment insurance, group insurance, withholding tax, bonds or union dues; also included is pay for overtime, holidays, vacations, and sick leave paid directly by the firm. Bonuses (unless earned and paid regularly each pay period), other pay not earned in the pay period reported (e.g., retroactive pay), tips, and the value of free rent, fuel, meals, or other payment in kind are excluded. "Fringe benefits" (such as health and other types of insurance, contributions to retirement, etc. paid by the employer) are also excluded.

Man-hours cover man-hours paid for, during the pay period which includes the 12th of the month, for production, construction, or nonsupervisory workers. The man-hours include hours paid for holidays and vacations, and for sick leave when pay is received directly from the firm.

Overtime hours cover hours worked by production or related workers for which overtime premiums were paid because the hours were in excess of the number of hours of either the straight-time workday or the workweek during the pay period which includes the 12th of the month. Weekend and holiday hours are included only if overtime premiums were paid. Hours for which only shift differential, hazard, incentive, or other similar types of premiums were paid are excluded.

Gross average hourly and weekly earnings

Average hourly earnings are on a ''gross" basis, reflecting not only changes in basic hourly and incentive wage rates but also such variable factors as premium pay for overtime and late-shift work and changes in output of workers paid on an incentive plan. Shifts in the volume of employment between relatively high paid and low-paid work and changes in workers' earnings in Individual establishments also affect the general earnings averages. Averages for groups and divisions further reflect changes in average hourly earnings for individual industries.

Averages of hourly earnings differ from wage rates. Earnings are the actual return to the worker for a stated period of time; rates are the amounts stipulated for a given unit of work or time. The earnings series does not measure the level of total labor costs on the part of the employer since the following are excluded: Irregular bonuses, retroactive items, payments of various welfare benefits, payroll taxes paid by employers, and earnings for those employees not covered under the production-worker, construction-worker, or nonsupervisory employee definitions.

Gross average weekly earnings are derived by multiplying average weekly hours by average hourly earnings. Therefore, weekly earnings are affected not only by changes in gross average hourly earnings but also by changes in the length of the workweek. Monthly variations in such factors as proportion of part-time workers, stoppages for varying causes, labor turnover during the survey period, and absenteeism for which employees are not paid may cause the average workweek to fluctuate.

Long-term trends of gross average weekly earnings can be affected by structural changes in the makeup of the work force. For example, persistent long-term increases in the proportion of

part-time workers in retail trade and many of the service industries have reduced average workweeks in these industries and have affected the average weekly earnings series.

Average weekly hours

The workweek information relates to the average hours for which pay was received and is different from standard or scheduled hours. Such factors as unpaid absenteeism, labor turnover, part-time work, and stoppages cause average weekly hours to be lower than scheduled hours of work for an establishment. Group averages further reflect changes in the workweek of component industries.

Average overtime hours

The overtime hours represent the portion of the gross average weekly hours which were in excess of regular hours and for which overtime premiums were paid. If an employee worked on a paid holiday at regular rates, receiving as total compensation his holiday pay plus straight-time pay for hours worked that day, no overtime hours would be reported.

Since overtime hours are premium hours by definition, gross weekly hours and overtime hours do not necessarily move in the same direction from month-to-month; for example, overtime premiums may be paid for hours in excess of the straight-time workday although less than a full week is worked. Diverse trends at the industry-group level also may be caused by a marked change in gross hours for a component industry where little or no overtime was worked in both the previous and current months. In addition, such factors as stoppages, absenteeism, and labor turnover may not have the same influence on overtime hours as on gross hours.

Hours and earnings for total private nonagricultural industries

This series covers all nonagricultural industry divisions except government. The principal source of payroll data is Form BLS 790. Secondary source material such as the Bureau's Employ ment and Wages, County Business Patterns of the Bureau of the Census, and additional supporting information such as The Hospital Guide, Part II, of the American Hospital Association and special studies by the National Council of Churches supplement data for certain industry groups within the service division.

For a technical description of this series, see the article, "Hours and Earnings for Workers in Private Nonagricultural Industries," published in the May 1967 issue of Employment and Earnings and Monthly Report on the Labor Force. Reprints are available upon request.

Railroad hours and earnings

The figures for class I railroads (excluding switching and terminal companies) are based on monthly data summarized in the M-300 report of the Interstate Commerce Commission and relate to all employees except executives, officials, and staff assistants (ICC group 1) who received pay during the month Gross average hourly earnings are computed by dividing total compensation by total hours paid for. Average weekly hours are

to a weekly basis, by the number of employees, as defined above. Gross average weekly earnings are derived by multiplying average weekly hours by average hourly earnings.

aggregates are the product of manhour aggregates and average hourly earnings. At all higher levels of aggregation, man-hour and payroll aggregates are the sum of the component aggregates.

Spendable average weekly earnings

Spendable average weekly earnings in current dollars are obtained by deducting estimated Federal social security and income taxes from average weekly earnings. The amount of income tax liability depends on the number of dependents supported by the worker and his marital status, as well as on the level of his gross income. To reflect these variables, spendable earnings are computed for a worker with no dependents and a married worker with three dependents. The computations are based on gross average weekly earnings for all production or nonsupervisory workers in the industry division excluding other income and income earned by other family members.

The series reflects the spendable earnings of only those workers, with either none or three dependents, whose gross weekly pay approximates the average earnings indicated for all production and nonsupervisory workers. It does not reflect, for example, the average earnings of all workers with three dependents; such workers, in fact have higher gross average earnings than workers with no dependents.

Since part-time as well as full-time workers are included, and since the proportion of part-time workers has been rising, the series understates the increase in earnings for full-time workers. As noted, "fringe benefits" are not included in the earnings. For a more complete discussion of the uses and limitations of these series, see the article by Paul M. Schwab, "Two Measures of Purchasing Power Contrasted," in the Monthly Labor Review for April 1971. Reprints of this article are available from the Bureau of Labor Statistics.

"Real" earnings are computed by dividing the current Consumer Price Index into the earnings averages for the current month. This is done for gross average weekly earnings and for spendable average weekly earnings. The level of earnings is thus adjusted for changes in purchasing power since the base period (1967).

Average hourly earnings excluding overtime

Average hourly earnings excluding overtime premium pay are computed by dividing the total production-worker payroll for the industry group by the sum of total production-worker man-hours and one-half of total overtime man-hours. Prior to January 1956, these data were based on the application of adjustment factors to gross average hourly earnings (as described in the Monthly Labor Review, May 1950, pp. 537-540). Both methods eliminate only the earnings due to overtime paid for at 11⁄2 times the straight-time rates. No adjustment is made for other premium payment provisions, such as holiday work, late shift work and overtime rates other than time and one-half.

Indexes of aggregate weekly payrolls and man-hours

The indexes of aggregate weekly payrolls and man-hours are prepared by dividing the current month's aggregate by the monthly average for the 1967 period. The man-hour aggregates are the product of average weekly hours and production-worker

Labor turnover

Labor turnover is the gross movement of wage and salary workers into and out of employed status with respect to individual establishments. This movement, which relates to a calendar month, is divided into two broad types: Accessions (new hires and rehires) and separations (terminations of employment initiated by either employer or employee). Each type of action is cumulated for a calendar month and expressed as a rate per 100 employees. The data relate to all employees, whether full- or part-time, permanent or temporary, including executive, office, sales, other salaried personnel, and production workers. Transfers to another establishment of the company are included, beginning with January 1959.

Accessions are the total number of permanent and temporary additions to the employment roll, including both new and rehired employees.

New hires are temporary or permanent additions to the employment roll of persons who have never before been employed in the establishment (except employees transferring from another establishment of the same company) or of former employees not recalled by the employer.

Other accessions, which are not published separately but are included in total accessions, are all additions to the employment roll which are not classified as new hires, including transfers from other establishments of the company and employees recalled from layoff.

Separations are terminations of employment during the calendar month and are classified according to cause: Quits, layoffs, and other separations, are defined as follows:

Quits are terminations of employment initiated by employees, failure to report after being hired, and unauthorized absences, if on the last day of the month the person has been absent more than 7 consecutive calendar days.

Layoffs are suspensions without pay lasting or expected to last more than 7 consecutive calendar days, initiated by the employer without prejudice to the worker.

Other separations, which are not published separately but are included in total separations, are terminations of employment because of discharge, permanent disability, death, retirement, transfers to another establishment of the company, and entrance into the Armed Forces for a period expected to last more than 30 consecutive calendar days.

Relationship of labor turnover to employment series

Month-to-month changes in total employment in manufacturing industries reflected by labor turnover rates are not comparable with the changes shown in the Bureau's employment series for the following reasons: (1) Accessions and separations are computed for the entire calendar month; the employment reports refer to the pay period which includes the 12th of the month; and (2) employees on strike are not counted as turnover actions although such employees are excluded from the employment estimates if the work stoppage extends through the report period.