Technical Note The estimates in this release were obtained from the Current Pop- ulation Survey (CPS), which provides basic information on the labor force, employment, and unemployment. The survey is conducted monthly for the Bureau of Labor Statistics by the U.S. Census Bureau from a scientifically selected national sample of about 60,000 households, with coverage in all 50 states and the District of Columbia. The earn- ings data are collected from one-quarter of the CPS monthly sample and are limited to wage and salary workers (both incorporated and unincor- porated self-employed are excluded). The data, therefore, exclude self- employment income. Information in this release will be made available to sensory impaired individuals upon request. Voice phone: (202) 691-5200; Federal Relay Service: (800) 877-8339. Reliability Statistics based on the CPS are subject to both sampling and non- sampling error. When a sample rather than the entire population is surveyed, there is a chance that the sample estimates may differ from the "true" population values they represent. The exact difference, or sampling error, varies depending on the particular sample selected, and this variability is measured by the standard error of the estimate. There is about a 90-percent chance, or level of confidence, that an estimate based on a sample will differ by no more than 1.6 standard errors from the "true" population value because of sampling error. BLS analyses are generally conducted at the 90-percent level of confidence. The CPS data also are affected by nonsampling error. Nonsampling errors can occur for many reasons, including the failure to sample a segment of the population, inability to obtain information for all respondents in the sample, inability or unwillingness of respondents to provide correct information on a timely basis, mistakes made by respondents, and errors made in the collection or processing of the data. A full discussion of the reliability of data from the Current Pop- ulation Survey and information on estimating standard errors is avail- able on the BLS Web site www.bls.gov/cps/documentation.htm#reliability. Definitions The principal definitions used in connection with the earnings series are described briefly below. Usual weekly earnings. Data represent earnings before taxes and other deductions and include any overtime pay, commissions, or tips usually received (at the main job in the case of multiple jobholders). Prior to 1994, respondents were asked how much they usually earned per week. Since January 1994, respondents have been asked to identify the easiest way for them to report earnings (hourly, weekly, biweekly, twice monthly, monthly, annually, other) and how much they usually earn in the reported time period. Earnings reported on a basis other than weekly are converted to a weekly equivalent. The term "usual" is as perceived by the respondent. If the respondent asks for a definition of usual, interviewers are in- structed to define the term as more than half the weeks worked during the past 4 or 5 months. Medians (and other quantiles) of weekly earnings. The median (or upper limit of the second quartile) is the amount which divides a given earnings distribution into two equal groups, one having earnings above the median and the other having earnings below the median. Ten percent of a given distribution have earnings below the upper limit of the first decile (90 percent have higher earnings); 25 percent have earnings below the upper limit of the first quartile (75 percent have higher earnings); 75 percent have earnings below the upper limit of the third quartile (25 percent have higher earnings); and 90 percent have earnings below the upper limit of the ninth decile (10 percent have higher earnings). The estimation procedure places each reported or calculated weekly earnings value into $50-wide intervals which are centered around multiples of $50. The actual value is estimated through the linear interpolation of the interval in which the quantile boundary lies. Over-the-year changes in the medians (and other quantile boundaries) for specific groups may not necessarily be consistent with the movements estimated for the overall quantile boundary. The most common reasons for this possible anomaly are: (1) There could be a change in the rel- ative weights of the subgroups. For example, the medians of both 16- to-24 year olds and those 25 years and over may rise; but if the lower- earning 16-to-24 group accounts for a greatly increased share of the total, the overall median could actually fall. (2) There could be a large change in the shape of the distribution of reported earnings, particularly near a quantile boundary. This could be caused by survey observations that are clustered at rounded values, such as $250, $300, or $400. An estimate lying in a $50-wide centered interval containing such a cluster or "spike" tends to change more slowly than one in other intervals. Wage and salary workers. Workers who receive wages, salaries, com- missions, tips, payment in kind, or piece rates. The group includes em- ployees in both the private and public sectors but, for the purposes of the earnings series, excludes all self-employed persons, regardless of whether or not their businesses are incorporated. Full-time workers. Workers who usually work 35 hours or more per week at their sole or principal job. Part-time workers. Workers who usually work fewer than 35 hours per week at their sole or principal job. Constant dollars. The Consumer Price Index for All Urban Consumers (CPI-U) is used to convert current dollars to constant (1982-84) dollars. Hispanic or Latino ethnicity. Refers to persons who identified themselves in the enumeration process as being Spanish, Hispanic, or Latino. Persons whose ethnicity is identified as Hispanic or Latino may be of any race. Seasonal adjustment Over the course of a year, the size of the nation's labor force and other measures of labor market activity undergo regularly occurring fluctuations. These recurring events include seasonal changes in weather, major holidays, and the opening and closing of schools. The effect of such seasonal variations can be very large. Because seasonal events follow a more or less regular pattern each year, their influence on the level of a series can be tempered by adjust- ing for regular seasonal variation. These adjustments make nonseasonal developments easier to spot. The seasonally adjusted figures provide a more useful tool with which to analyze changes in quarter-to-quarter activity. At the end of each calendar year, the seasonally adjusted data are re- vised for the past 5 years when the seasonal adjustment factors are updated. More information on seasonal adjustment is available on the BLS Web site at www.bls.gov/cps/documentation.htm#sa.