Total money income. Total money income is the algebraic sum of money wages and salaries, net income from self-employment, and income other than earnings. The total income of a family is the algebraic sum of the amounts received by all income recipients in the family.

The income tables for families include in the lowest income group (under $2,500) those who were classified as having no income in the income year and those reporting a loss in net income from farm and nonfarm self-employment or in rental income. Some of these were living on income "in-kind," savings, or gifts, or were newly constituted families. However, other families or unrelated individuals who reported no income probably had some money income which was not recorded in the survey.

Total money earnings. Total money earnings is the algebraic sum of money wages or salary and net income from farm and nonfarm self-employment. For a detailed explanation, see Current Population Reports, Series P-60, No. 180, Money Income of Households, Families, and Persons in the United States: 1991.

Number of earners. This includes all persons in the household with $1 or more in wages and salaries or $1 or more or a loss in net income from farm or nonfarm self-employment.

Per capita income. Per capita income is the mean income computed for every man, woman, and child in a particular group. It is derived by dividing the total income of a particular group by the total population in that group (excluding patients or inmates in institutional quarters).

Poverty. Families and unrelated individuals are classified as being above or below the poverty level using the poverty index originated at the Social Security Administration in 1964 and revised by Federal Interagency Committees in 1969 and 1981. The poverty index is based soley on money income and does not reflect the fact that many low-income persons receive noncash benefits such as food stamps, Medicaid, and public

housing. The index is based on the Department of Agriculture's 1961 Economy Food Plan and reflects the different consumption requirements of families based on their size and composition. It was determined from the Department of Agriculture's 1955 Survey of Food Consumption that families of three or more persons spend approximately one-third of their income on food; the poverty level for these families was, therefore, set at three times the cost of the Economy Food Plan. For smaller families and persons living alone, the cost of the Economy Food Plan was multiplied by factors that were slightly higher in order to compensate for the relatively larger fixed expenses of these smaller households. The poverty thresholds are updated every year to reflect changes in the Consumer Price Index (CPI-U). The average poverty threshold for a family of four was $12,674 in 1989, but $13,359 in 1990. For a detailed explanation of the poverty definition, see Current Population Reports, Series P-60, No. 181, Poverty in the United States: 1991.

Median. The median is presented in connection with the data on age, income, and earnings. It is the value which divides the distribution into two equal parts, one-half of the cases exceeding this value. The median income for families is based on all families. The median income for persons is based on persons with income.

Mean. The mean (average) is presented in connection with data on number of persons per family, income of persons, and income of families. The mean number of persons per family is the value obtained by dividing the number of persons having the characteristic under consideration by the appropriate number of families. The mean income is the amount obtained by dividing the total income of a group by the number of units in that group. The mean income for families is based on all families. The mean income for persons is based on persons with income. Mean income in this report is calculated using grouped data and may vary from published mean income using ungrouped data obtained from individual records.

SOURCE OF DATA

Most estimates in this report come from data obtained from the Current Population Survey (CPS) conducted in March of years 1974 through 1992. The Bureau of the Census conducts the survey every month, although this report uses mostly March data for its estimates. Also, some estimates come from Decennial Census data for years 1960 through 1990. The March survey uses two sets of questions, the basic CPS and the supplements.

Basic CPS. The basic CPS collects primarily labor force data about the civilian noninstitutional population. Interviewers ask questions concerning labor force participation about each member 15 years old and over in every sample household.

The present CPS sample was selected from the 1980 Decennial Census files with coverage in all 50 states and the District of Columbia. The sample is continually updated to account for new residential construction. It is located in 729 areas and includes 1,973 counties, independent cities, and minor civil divisions. About 60,000 occupied housing units are eligible for interview every month. Interviewers are unable to obtain interviews at about 2,600 of these units because the occupants are not found at home after repeated calls or are unavailable for some other reason.

Since the introduction of the CPS, the Bureau of the Census has redesigned the CPS sample several times. These redesigns have improved the quality and reliability of the data and have satisfied changing data needs. The most recent changes were completely implemented in July 1985.

March Supplement. In addition to the basic CPS questions, interviewers asked supplementary questions in March about marital status, educational attainment, family composition, and about the economic situation of persons and families for the previous year.

To obtain more reliable data for the Hispanic population, the March CPS sample was increased by about 2,500 eligible housing units. These housing units were interviewed the previous November and contained at least one sample person of Hispanic origin. In addition, the sample included persons in the Armed Forces living off post or with their families on post.

Estimation Procedure. This survey's estimation procedure inflates weighted sample results to independent estimates of the civilian noninstitutional population of the United States by age, sex, race and Hispanic/nonHispanic categories. The independent estimates were based on statistics from decennial censuses of population; statistics on births, deaths, immigration and emigration; and statistics on the size of the Armed Forces. The independent population estimates used for 1981 to present were based on updates to controls established by the 1980 Decennial Census. Before 1981, independent population estimates from the most recent decennial census were used. For more details on the change in independent estimates, see the section entitled "Introduction of 1980 Census Population Controls" in an earlier report (Series P-60, No. 133). The estimation procedure for the March supplement included a further adjustment so husband and wife of a household received the same weight.

The estimates in this report for 1985 and later also employ a revised survey weighting procedure for persons of Hispanic origin. In previous years, weighted sample results were inflated to independent estimates of the noninstitutional population by age, sex, and race. There was no specific control of the survey estimates for the Hispanic population. Since then, the Bureau of the Census developed independent population controls for the Hispanic population by sex and detailed age groups. Revised weighting procedures incorporate these new controls. The independent population estimates include some, but not all, undocumented immigrants.

ACCURACY OF ESTIMATES

Since the CPS estimates come from a sample, they may differ from figures from a complete census using the same questionnaires, instructions, and enumerators. A sample survey estimate has two possible types of error: sampling and nonsampling. The accuracy of an estimate depends on both types of error, but the full extent of the nonsampling error is unknown. Consequently, one should be particularly careful when interpreting results based on a relatively small number of cases or on small differences between estimates. The standard errors for CPS estimates primarily indicate the magnitude of sampling error. They also partially measure the effect of some nonsampling errors in responses and enumeration, but do not measure systematic biases in the data. (Bias is the average over all possible samples of the differences between the sample estimates and the desired value.)

Nonsampling Variability. There are several sources of nonsampling errors including the following:

• Inability to get information about all sample cases.

• Definitional difficulties.

• Differences in interpretation of questions. Respondents' inability or unwillingness to provide correct information.

Respondents' inability to recall information.

• Errors made in data collection, such as recording and coding data.

• Errors made in processing the data.

• Errors made in estimating values for missing data. • Failure to represent all units with the sample (undercoverage).

CPS undercoverage results from missed housing units and missed persons within sample households. Compared to the level of the 1980 Decennial Census, overall CPS undercoverage is about 7 percent. CPS undercoverage varies with age, sex, and race. Generally, undercoverage is larger for males than for females and larger for Blacks and other races combined than for Whites. As described previously, ratio estimation to independent age-sex-race-Hispanic population controls partially corrects for the bias caused by 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 from those of interviewed persons in the same age-sex-race-Hispanic group. Furthermore, the independent population controls have not been adjusted for undercoverage in the 1980 census.

A common measure of survey coverage is the coverage ratio, the estimated population before ratio adjustment divided by the independent population control. Table B-2 shows CPS coverage ratios for age-sex-race groups for a recent month. The CPS coverage ratios can exhibit some variability from month to month, but these are a typical set of coverage ratios. Other Census Bureau household surveys experience similar coverage.

For additional information on nonsampling error including the possible impact on CPS data when known, refer to Statistical Policy Working Paper 3, An Error Profile: Employment as Measured by the Current Population Survey, Office of Federal Statistical Policy and Standards, U.S. Department of Commerce, 1978 and Technical Paper 40, The Current Population Survey: Design and Methodology, Bureau of the Census, U.S. Department of Commerce.

Comparability of Data. Data obtained from the CPS and other sources are not entirely comparable. This results from differences in interviewer training and experience and in differing survey processes. This is an example of nonsampling variability not reflected in the standard errors. Use caution when comparing results from different sources.

Caution should also be used when comparing estimates in this report (which reflect 1980 census-based population controls) with estimates for 1980 and earlier years (which reflect 1970 census-based population controls). This change in population controls had relatively little impact on summary measures such as means, medians, and percent distributions. It did have a significant impact on levels. For example, use of 1980-based population controls results in about a 2-percent increase in the civilian noninstitutional population and in the number of families and households. Thus, estimates of levels for data collected in 1981 and later years will differ from those for earlier years by more than what could be attributed to actual changes in the population. These differences could be disproportionately greater for certain subpopulation groups than for the total population.

Since no independent population control totals for persons of Hispanic origin were used before 1985, compare Hispanic estimates over time cautiously.

Note When Using Small Estimates. Summary measures (such as medians and percentage distributions) are shown only when the base is 75,000 or greater. Because of the large standard errors involved, summary measures would probably not reveal useful information when computed on a smaller base. However, estimated numbers are shown even though the relative standard errors of these numbers are larger than those for corresponding percentages. These smaller estimates permit combinations of the categories to suit data users' needs. These estimates may not be reliable for the interpretation of small differences. For instance, even a small amount of nonsampling error can cause a borderline difference to appear significant or not, thus distorting a seemingly valid hypothesis test.

Sampling Variability. Sampling variability is variation that occurred by chance because a sample was surveyed rather than the entire population. Standard errors, as calculated by methods described next, are primarily measures of sampling variability, although they may include some nonsampling errors.

Standard Errors and Their Use. A number of approximations are required to derive, at a moderate cost, standard errors applicable to all the estimates in this report. Instead of providing an individual standard error for each estimate, generalized sets of standard errors are provided for various types of characteristics. Thus, the tables show levels of magnitude of standard errors rather than the precise standard errors.

Table B-3 provides standard errors of estimated numbers. Table B-4 provides standard errors of estimated percentages. Table B-5 has standard error parameters for persons, families, households, householders,

and unrelated individuals. Table B-5 also provides factors to apply to the standard errors in tables B-3 and B-4.

The sample estimate and its standard error enable one to construct a confidence interval. A confidence interval is a range that would include the average result of all possible samples with a known probability. For example, if all possible samples were surveyed under essentially the same general conditions and using the same sample design, and if an estimate and its standard error were calculated from each sample, then approximately 90 percent of the intervals from 1.645 standard errors below the estimate to 1.645 standard errors above the estimate would include the average result of all possible samples.

A particular confidence interval may or may not contain the average estimate derived from all possible samples. However, one can say with specified confidence that the interval includes the average estimate calculated from all possible samples.

Some statements in the report may contain estimates followed by a number in parentheses. This number can be added to and subtracted from the estimate to calculate upper and lower bounds of the 90-percent confidence interval. For example, if a statement contains the phrase "grew by 1.7 percent (±1.0)," the 90percent confidence interval for the estimate, 1.7 percent, is from 0.7 percent to 2.7 percent.

Standard errors may be used to perform hypothesis testing. This is a procedure for distinguishing between population parameters using sample estimates. The most common type of hypothesis appearing in this report is that the population parameters are different. An example of this would be comparing Black families with White families.

Tests may be performed at various levels of significance. The significance level of a test is the probability of concluding that the characteristics are different when, in fact, they are the same. All statements of comparison in the text have passed a hypothesis test at the 0.10 level of significance or better. This means that the absolute value of the estimated difference between characteristics is greater than or equal to 1.645 times the standard error of the difference.

Here x is the size of the estimate and a and b are the parameters in table B-5 associated with the particular type of characteristic. When calculating standard errors for numbers from cross-tabulations involving different characteristics, use the factor or set of parameters for the characteristic that will give the largest standard error.

Illustration. Suppose there were 2,077,000 Black families in poverty. Use the appropriate parameters from table B-5 and formula (2) to get

Number, x

The alternate calculation of the standard error, using formula (1) with f= 0.68 from table B-5 and s=98,000 by interpolation from table B-3, is

0.68 x 98,000 = 67,000

Standard Errors of Estimated Percentages. The reliability of an estimated percentage, computed using sample data for both numerator and denominator, depends on the size of the percentage and its base. Estimated percentages are relatively more reliable than the corresponding estimates of the numerators of the percentages, particularly if the percentages are 50 percent or more. When the numerator and denominator of the percentage are in different categories, use the factor or parameter from table B-5 indicated by the numerator. The approximate standard error, Sx,p, of an estimated percentage can be obtained by use of the formula

2,077,000