20
Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include not only Federal statistical bureaus and other orga- nizations that collect and issue statistics as their principal activity, but also govern- mental administrative and regulatory agencies, private research bodies, trade associations, insurance companies, health associations, and private organizations such as the National Education Associa- tion and philanthropic foundations. Con- sequently, the data vary considerably as to reference periods, definitions of terms and, for ongoing series, the number and frequency of time periods for which data are available. The statistics presented were obtained and tabulated by various means. Some statistics are based on complete enumera- tions or censuses while others are based on samples. Some information is ex- tracted from records kept for administra- tive or regulatory purposes (school enroll- ment, hospital records, securities registration, financial accounts, social se- curity records, income tax returns, etc.), while other information is obtained ex- plicitly for statistical purposes through interviews or by mail. The estimation pro- cedures used vary from highly sophisti- cated scientific techniques, to crude ‘‘in- formed guesses.’’ Each set of data relates to a group of indi- viduals or units of interest referred to as the target universe or target population, or simply as the universe or population. Prior to data collection the target universe should be clearly defined. For example, if data are to be collected for the universe of households in the United States, it is necessary to define a ‘‘household.’’ The target universe may not be completely tractable. Cost and other considerations may restrict data collection to a survey universe based on some available list, such list may be inaccurate or out of date. This list is called a survey frame or sam- pling frame. The data in many tables are based on data obtained for all population units, a census, or on data obtained for only a portion, or sample, of the population units. When the data presented are based on a sample, the sample is usually a sci- entifically selected probability sample. This is a sample selected from a list or sampling frame in such a way that every possible sample has a known chance of selection and usually each unit selected can be assigned a number, greater than zero and less than or equal to one, repre- senting its likelihood or probability of se- lection. For large-scale sample surveys, the prob- ability sample of units is often selected as a multistage sample. The first stage of a multistage sample is the selection of a probability sample of large groups of population members, referred to as pri- mary sampling units (PSUs). For example, in a national multistage household sample, PSUs are often counties or groups of counties. The second stage of a multi- stage sample is the selection, within each PSU selected at the first stage, of smaller groups of population units, referred to as secondary sampling units. In subsequent stages of selection, smaller and smaller nested groups are chosen until the ulti- mate sample of population units is ob- tained. To qualify a multistage sample as a probability sample, all stages of sam- pling must be carried out using probabil- ity sampling methods. Prior to selection at each stage of a multi- stage (or a single-stage) sample, a list of the sampling units or sampling frame for that stage must be obtained. For ex- ample, for the first stage of selection of a national household sample, a list of the counties and county groups that form the PSUs must be obtained. For the final stage of selection, lists of households, and sometimes persons within the house- holds, have to be compiled in the field. For surveys of economic entities and for the economic censuses the Census Bureau Appendix III 941 U.S. Census Bureau, Statistical Abstract of the United States: 2006

Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

  • Upload
    others

  • View
    6

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

Appendix III

Limitations of the Data

Introduction The data presented in thisStatistical Abstract came from manysources. The sources include not onlyFederal statistical bureaus and other orga-nizations that collect and issue statisticsas their principal activity, but also govern-mental administrative and regulatoryagencies, private research bodies, tradeassociations, insurance companies, healthassociations, and private organizationssuch as the National Education Associa-tion and philanthropic foundations. Con-sequently, the data vary considerably asto reference periods, definitions of termsand, for ongoing series, the number andfrequency of time periods for which dataare available.

The statistics presented were obtainedand tabulated by various means. Somestatistics are based on complete enumera-tions or censuses while others are basedon samples. Some information is ex-tracted from records kept for administra-tive or regulatory purposes (school enroll-ment, hospital records, securitiesregistration, financial accounts, social se-curity records, income tax returns, etc.),while other information is obtained ex-plicitly for statistical purposes throughinterviews or by mail. The estimation pro-cedures used vary from highly sophisti-cated scientific techniques, to crude ‘‘in-formed guesses.’’

Each set of data relates to a group of indi-viduals or units of interest referred to asthe target universe or target population,or simply as the universe or population.Prior to data collection the target universeshould be clearly defined. For example, ifdata are to be collected for the universeof households in the United States, it isnecessary to define a ‘‘household.’’ Thetarget universe may not be completelytractable. Cost and other considerationsmay restrict data collection to a surveyuniverse based on some available list,such list may be inaccurate or out of date.This list is called a survey frame or sam-pling frame.

The data in many tables are based ondata obtained for all population units, acensus, or on data obtained for only aportion, or sample, of the populationunits. When the data presented are basedon a sample, the sample is usually a sci-entifically selected probability sample.This is a sample selected from a list orsampling frame in such a way that everypossible sample has a known chance ofselection and usually each unit selectedcan be assigned a number, greater thanzero and less than or equal to one, repre-senting its likelihood or probability of se-lection.

For large-scale sample surveys, the prob-ability sample of units is often selected asa multistage sample. The first stage of amultistage sample is the selection of aprobability sample of large groups ofpopulation members, referred to as pri-mary sampling units (PSUs). For example,in a national multistage householdsample, PSUs are often counties or groupsof counties. The second stage of a multi-stage sample is the selection, within eachPSU selected at the first stage, of smallergroups of population units, referred to assecondary sampling units. In subsequentstages of selection, smaller and smallernested groups are chosen until the ulti-mate sample of population units is ob-tained. To qualify a multistage sample asa probability sample, all stages of sam-pling must be carried out using probabil-ity sampling methods.

Prior to selection at each stage of a multi-stage (or a single-stage) sample, a list ofthe sampling units or sampling frame forthat stage must be obtained. For ex-ample, for the first stage of selection of anational household sample, a list of thecounties and county groups that form thePSUs must be obtained. For the final stageof selection, lists of households, andsometimes persons within the house-holds, have to be compiled in the field.For surveys of economic entities and forthe economic censuses the Census Bureau

Appendix III 941

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 2: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

generally uses a frame constructed fromthe Census Bureau’s Business Register.The Business Register contains all estab-lishments with payroll in the United Statesincluding small single establishment firmsas well as large multiestablishment firms.

Wherever the quantities in a table refer toan entire universe, but are constructedfrom data collected in a sample survey,the table quantities are referred to assample estimates. In constructing asample estimate, an attempt is made tocome as close as is feasible to the corre-sponding universe quantity that would beobtained from a complete census of theuniverse. Estimates based on a samplewill, however, generally differ from thehypothetical census figures. Two classifi-cations of errors are associated with esti-mates based on sample surveys: (1) sam-pling error—the error arising from the useof a sample, rather than a census, to esti-mate population quantities and (2) non-sampling error—those errors arising fromnonsampling sources. As discussed be-low, the magnitude of the sampling errorfor an estimate can usually be estimatedfrom the sample data. However, the mag-nitude of the nonsampling error for anestimate can rarely be estimated. Conse-quently, actual error in an estimate ex-ceeds the error that can be estimated.

The particular sample used in a survey isonly one of a large number of possiblesamples of the same size which couldhave been selected using the same sam-pling procedure. Estimates derived fromthe different samples would, in general,differ from each other. The standard error(SE) is a measure of the variation amongthe estimates derived from all possiblesamples. The standard error is the mostcommonly used measure of the samplingerror of an estimate. Valid estimates ofthe standard errors of survey estimatescan usually be calculated from the datacollected in a probability sample. For con-venience, the standard error is sometimesexpressed as a percent of the estimateand is called the relative standard error orcoefficient of variation (CV). For example,an estimate of 200 units with an esti-mated standard error of 10 units has anestimated CV of 5 percent.

A sample estimate and an estimate of itsstandard error or CV can be used to con-struct interval estimates that have a pre-scribed confidence that the interval in-cludes the average of the estimatesderived from all possible samples with aknown probability. To illustrate, if all pos-sible samples were selected under essen-tially the same general conditions, andusing the same sample design, and if anestimate and its estimated standard errorwere calculated from each sample, then:1) approximately 68 percent of the inter-vals from one standard error below theestimate to one standard error above theestimate would include the average esti-mate derived from all possible samples;2) approximately 90 percent of the inter-vals from 1.6 standard errors below theestimate to 1.6 standard errors above theestimate would include the average esti-mate derived from all possible samples;and 3) approximately 95 percent of theintervals from two standard errors belowthe estimate to two standard errors abovethe estimate would include the averageestimate derived from all possiblesamples.

Thus, for a particular sample, one can saywith the appropriate level of confidence(e.g., 90 percent or 95 percent) that theaverage of all possible samples is in-cluded in the constructed interval. Ex-ample of a confidence interval: An esti-mate is 200 units with a standard error of10 units. An approximately 90 percentconfidence interval (plus or minus 1.6standard errors) is from 184 to 216.

All surveys and censuses are subject tononsampling errors. Nonsampling errorsare of two kinds—random and nonran-dom. Random nonsampling errors arisebecause of the varying interpretation ofquestions (by respondents or interview-ers) and varying actions of coders, key-ers, and other processors. Some random-ness is also introduced when respondentsmust estimate. Nonrandom nonsamplingerrors result from total nonresponse (nousable data obtained for a sampled unit),partial or item nonresponse (only a por-tion of a response may be usable), inabil-ity or unwillingness on the part of respon-dents to provide correct information,difficulty interpreting questions, mistakes

942 Appendix III

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 3: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

in recording or keying data, errors of col-lection or processing, and coverage prob-lems (overcoverage and undercoverage ofthe target universe). Random nonre-sponse errors usually, but not always, re-sult in an understatement of sampling er-rors and thus an overstatement of theprecision of survey estimates. Estimatingthe magnitude of nonsampling errorswould require special experiments or ac-cess to independent data and, conse-quently, the magnitudes are seldom avail-able.

Nearly all types of nonsampling errorsthat affect surveys also occur in completecensuses. Since surveys can be conductedon a smaller scale than censuses, non-sampling errors can presumably be con-trolled more tightly. Relatively more fundsand effort can perhaps be expended to-ward eliciting responses, detecting andcorrecting response error, and reducingprocessing errors. As a result, survey re-sults can sometimes be more accuratethan census results.

To compensate for suspected nonrandomerrors, adjustments of the sample esti-mates are often made. For example, ad-justments are frequently made for nonre-sponse, both total and partial. Adjust-ments made for either type of nonre-sponse are often referred to as imputa-tions. Imputation for total nonresponse isusually made by substituting for thequestionnaire responses of the nonre-spondents the ‘‘average’’ questionnaireresponses of the respondents. These im-putations usually are made separatelywithin various groups of sample mem-bers, formed by attempting to place re-spondents and nonrespondents togetherthat have ‘‘similar’’ design or ancillarycharacteristics. Imputation for item nonre-sponse is usually made by substitutingfor a missing item the response to thatitem of a respondent having characteris-tics that are ‘‘similar’’ to those of the non-respondent.

For an estimate calculated from a samplesurvey, the total error in the estimate iscomposed of the sampling error, whichcan usually be estimated from thesample, and the nonsampling error, whichusually cannot be estimated from the

sample. The total error present in a popu-lation quantity obtained from a completecensus is composed of only nonsamplingerrors. Ideally, estimates of the total errorassociated with data given in the Statisti-cal Abstract tables should be given. How-ever, due to the unavailability of esti-mates of nonsampling errors, onlyestimates of the levels of sampling errors,in terms of estimated standard errors orcoefficients of variation, are available. Toobtain estimates of the estimated stand-ard errors from the sample of interest,obtain a copy of the referenced reportwhich appears at the end of each table.

Source of Additional Material: The FederalCommittee on Statistical Methodology(FCSM) is an interagency committee dedi-cated to improving the quality of federalstatistics <http://fcsm.ssd.census.gov>.

Principal databases—Beginning beloware brief descriptions of 35 of the samplesurveys and censuses that provide a sub-stantial portion of the data contained inthis Abstract.

U.S. DEPARTMENT OF AGRICUL-TURE, National Agriculture Statis-tics Service

Basic Area Frame SampleUniverse, Frequency, and Types of Data:

June agricultural survey collects data onplanted acreage and livestock invento-ries. The survey also serves to measurelist incompleteness and is subsampledfor multiple frame surveys.

Type of Data Collection Operation: Strati-fied probability sample of about 11,000land area units of about 1 sq. mile(range from 0.1 sq. mile in cities to sev-eral sq. miles in open grazing areas).Sample includes 42,000 parcels of agri-cultural land. About 20 percent of thesample replaced annually.

Data Collection and Imputation Proce-dures: Data collection is by personalenumeration. Imputation is based onenumerator observation or datareported by respondents having similaragricultural characteristics.

Estimates of Sampling Error: EstimatedCVs range from 1 percent to 2 percentfor regional estimates to 3 percent to 6percent for state estimates of majorcrop acres and livestock inventories.

Appendix III 943

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 4: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

Other (nonsampling) Errors: Minimizedthrough rigid quality controls on the col-lection process and careful review of allreported data.

Sources of Additional Material: U.S.Department of Agriculture, NationalAgricultural Statistics Service, USDA’sNational Agricultural Statistics Service:The Fact Finders of Agriculture, Septem-ber 1994.

Census of AgricultureUniverse, Frequency, and Types of Data:

Complete count of U.S. farms andranches conducted once every 5 yearswith data at the national, state, andcounty level. Data published on farmnumbers and related items/characteristics.

Type of Data Collection Operation: Com-plete census for number of farms; landin farms; agriculture products sold; totalcropland; irrigated land; farm operatorcharacteristics; livestock and poultryinventory and sales; and selected cropsharvested. Market value of land andbuildings, total farm productionexpenses, machinery and equipment,fertilizer and chemicals, and farm laborare estimated from a sample of farms.

Data Collection and Imputation Proce-dures: Data collection takes place bymailing questionnaires to all farmersand ranchers. Nonrespondents are con-tacted by telephone and correspondencefollow-ups. Imputations were made forall nonresponse items/characteristics.Coverage adjustments were made toaccount for missed farms and ranches.

Estimates of Sampling Error: Variability inthe estimates is due to the sample selec-tion and estimation for items collectedby sample and census nonresponse andcoverage estimation procedures. TheCVs for national and state estimates aregenerally very small. The response rateis approximately 81 percent.

Other (nonsampling) Errors: Nonsamplingerrors are due to incompleteness of thecensus mailing list, duplications on thelist, respondent reporting errors, errorsin editing reported data, and in imputa-tion for missing data. Evaluation studiesare conducted to measure certain non-sampling errors such as list coverageand classification error. Results from the

evaluation program for the 2002 censusindicate the net under-coverageamounted to about 18 percent of thenation’s total farms.

Sources of Additional Material: U.S.Department of Agriculture (NASS), 2002Census of Agriculture, Volume 1, SubjectSeries C Part 1, Agriculture Atlas of theU.S.; Part 2, Coverage Evaluation; Part 3,Rankings of States and Counties; Part 4,History; Part 5, ZIP Code Tabulation ofSelected Items; and Volume 3 SpecialStudies, Part 1, Farm and Ranch Irriga-tion Survey; Part 2, Census of Horticul-tural Specialties; Part 3, Census ofAquaculture.

Multiple Frame Surveys

Universe, Frequency, and Types of Data:Surveys of U.S. farm operators are takento obtain data on major livestock inven-tories, selected crop acreage and pro-duction, grain stocks, and farm laborcharacteristics; farm economic data andchemical use data.

Type of Data Collection Operation: Pri-mary frame is obtained from general orspecial purpose lists, supplemented by aprobability sample of land areas used toestimate for list incompleteness.

Data Collection and Imputation Proce-dures: Mail, telephone, or personal inter-views used for initial data collection.Mail nonrespondent follow-up by phoneand personal interviews. Imputationbased on average of respondents.

Estimates of Sampling Error: EstimatedCV for number of hired farm workers isabout 3 percent. Estimated CVs rangefrom 1 percent to 2 percent for regionalestimates to 3 percent to 6 percent forstate estimates of livestock inventoriesand crop acreage.

Other (nonsampling) Errors: In addition toabove, replicated sampling proceduresused to monitor effects of changes insurvey procedures.

Sources of Additional Material: U.S.Department of Agriculture, NationalAgricultural Statistics Service), USDA’sNational Agricultural Statistics Service:The Fact Finders of Agriculture, Septem-ber 1994.

944 Appendix III

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 5: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

Objective Yield SurveysUniverse, Frequency, and Types of Data:

Surveys for data on corn, cotton, pota-toes, soybeans, and wheat, to forecastand estimate yields.

Type of Data Collection Operation: Ran-dom location of plots in probabilitysample. Corn, cotton, soybeans, springwheat, and durum wheat selected inJune from Basic Area Frame Sample (seeabove). Winter wheat and potatoesselected from March and June multipleframe surveys, respectively.

Data Collection and Imputation Proce-dures: Enumerators count and measureplant characteristics in sample fields.Production measured from plots at har-vest. Harvest loss measured from postharvest gleanings.

Estimates of Sampling Error: CVs fornational estimates of production areabout 2−3 percent.

Other (nonsampling) Errors: In addition toabove, replicated sampling proceduresused to monitor effects of changes insurvey procedures.

Sources of Additional Material: U.S.Department of Agriculture, NationalAgricultural Statistics Service), USDA’sNational Agricultural Statistics Service:The Fact Finders of Agriculture, Septem-ber 1994.

U.S. BUREAU OF JUSTICE STATIS-TICS (BJS)

National Crime VictimizationSurvey

Universe, Frequency, and Types of Data:Monthly survey of individuals andhouseholds in the United States toobtain data on criminal victimization ofthose units for compilation of annualestimates.

Type of Data Collection Operation:National probability sample survey ofabout 42,000 interviewed households in203 PSUs selected from a list ofaddresses from the 1990 census,supplemented by new construction per-mits and an area sample where permitsare not required.

Data Collection and Imputation Proce-dures: Interviews are conducted every 6months for 3 years for each household

in the sample; 7,000 households areinterviewed monthly. Personal inter-views are used in the first interview; theintervening interviews are conducted bytelephone whenever possible.

Estimates of Sampling Error: CVs for2003 estimates are: 3.6 percent for per-sonal crimes (includes all crimes of vio-lence plus purse-snatching crimes), 3.6percent for crimes of violence; 14.0 per-cent for estimate of rape/sexual assaultcounts; 8.7 percent for robbery counts;3.9 percent for assault counts; 14.5 per-cent for purse snatching (it refers topurse snatching and pocket picking);1.9 percent for property crimes; 3.8 per-cent for burglary counts; 2.1 percent fortheft (of property); and 5.9 percent formotor vehicle theft counts.

Other (nonsampling) Errors: Respondentrecall errors which may include report-ing incidents for other than the refer-ence period; interviewer coding andprocessing errors; and possible mis-taken reporting or classifying of events.Adjustment is made for a householdnoninterview rate of about 8 percentand for a within-household noninterviewrate of 14 percent.

Sources of Additional Material: U.S.Bureau of Justice Statistics, Criminal Vic-timization in the United States, annual.

U.S. Bureau of Labor Statistics

Consumer Expenditure Survey(CE)

Universe, Frequency and Types of Data:Consists of two continuous compo-nents: a quarterly Interview survey anda weekly diary or record-keeping survey.They are nationwide surveys that collectdata on consumer expenditures,income, characteristics, and assets andliabilities. Samples are national probabil-ity samples of households that are rep-resentative of the civilian noninstitu-tionalized population. The surveys havebeen ongoing since 1980.

Type of Data Collection Operation: TheInterview Survey is a panel rotation sur-vey. Each panel is interviewed for fivequarters and then dropped from the sur-vey. About 7,500 consumer units are

Appendix III 945

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 6: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

interviewed each quarter. The Diary Sur-vey sample is new each year and con-sists of about 7,500 consumer units.Data have been collected on an ongoingbasis in 105 PSUs since 1996.

Data Collection and Imputation Proce-dures: For the Interview Survey, data arecollected by personal interview witheach consumer unit interviewed onceper quarter for five consecutive quar-ters. Designed to collect informationthat respondents can recall for 3 monthsor longer, such as large or recurringexpenditures. For the Diary Survey,respondents record all their expendi-tures in a self-reporting diary for twoconsecutive one-week periods. Designedto pick up items difficult to recall over along period, such as detailed foodexpenditures. Missing or invalidattributes or expenditures are imputed.Income, assets, and liabilities are notimputed. The U.S. Census Bureau col-lects the data for the Bureau of LaborStatistics.

Estimates of Sampling Error: Standarderror tables have been available since2000.

Other (nonsampling) Errors: Includesincorrect information given by respon-dents, data processing errors, inter-viewer errors, and so on. They occurregardless of whether data are collectedfrom a sample or from the entire popu-lation.

Sources of Additional Material: Bureau ofLabor Statistics, Internet site<http://www.bls.gov/cex>.

Consumer Price Index (CPI)Universe, Frequency, and Types of Data:

Monthly survey of price changes of alltypes of consumer goods and servicespurchased by urban wage earners andclerical workers prior to 1978, andurban consumers thereafter. Bothindexes continue to be published.

Type of Data Collection Operation: Priorto 1978, and since 1998, sample of vari-ous consumer items in 87 urban areas;from 1978−1997, in 85 PSUs, exceptfrom January 1987 through March 1988,when 91 areas were sampled.

Data Collection and Imputation Proce-dures: Prices of consumer items areobtained from about 50,000 housing

units, and 23,000 other reporters in 87areas. Prices of food, fuel, and a fewother items are obtained monthly; pricesof most other commodities and servicesare collected every month in the threelargest geographic areas and everyother month in others.

Estimates of Sampling Error: Estimates ofstandard errors are available.

Other (nonsampling) Errors: Errors resultfrom inaccurate reporting, difficulties indefining concepts and their operationalimplementation, and introduction ofproduct quality changes and new prod-ucts.

Sources of Additional Material: U.S.Bureau of Labor Statistics, Internet site<http://www.stats.bls.gov/cpi/home.htm>and BLS Handbook of Methods, Chapter17, Bulletin 2490.

Current Employment Statis-tics (CES) Program

Universe, Frequency, and Types of Data:Monthly survey drawn from a samplingframe of over 8 million unemploymentinsurance tax accounts in order obtaindata by industry on employment, hours,and earnings.

Type of Data Collection Operation: In2004, the CES sample included about160,000 businesses and governmentagencies, which represent approxi-mately 400,000 individual worksites.

Data Collection and Imputation Proce-dures: Each month, the state agenciescooperating with BLS, as well as BLSData Collection Centers, collect datathrough various automated collectionmodes and mail. BLS-Washington staffprepares national estimates of employ-ment, hours, and earnings while statesuse the data to develop state and areaestimates.

Estimates of Sampling Errors: The relativestandard error for total nonfarm employ-ment is 0.1 percent.

Other (nonsampling) Errors: Estimates ofemployment adjusted annually to reflectcomplete universe. Average adjustmentis 0.3 percent over the last decade, withan absolute range from less than 0.05percent to 0.5 percent.

946 Appendix III

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 7: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

Sources of Additional Material: U.S.Bureau of Labor Statistics, Employmentand Earnings, monthly, ExplanatoryNotes and Estimates of Errors, Tables2-A through 2-F.

National Compensation Sur-vey

Universe, Frequency, and Types of Data:Nationwide sample survey of establish-ments of all employment size classes,stratified by geographic area, in privateindustry and state and local govern-ment. Data collected include wages andsalaries, and employer costs foremployee compensation, employmentcost index, and employee benefits. Dataproduced include percent changes in thecost of employment cited in the Employ-ment Cost Index (ECI) and costs perhour worked for individual benefitscited in the Employer Costs forEmployee Compensation (ECEC). Thesurvey provides data by ownership (pri-vate industry and state and local gov-ernment), industry sector, majorindustry divisions, major occupationalgroups, bargaining status, metropolitanarea status, and census region. ECECalso provides data by establishment sizeclass.

Type of Data Collection Operation: Prob-ability proportionate to size sample ofestablishments. The sample is replacedon a continual basis. Establishments arein the survey for approximately 5 years.

Data Collection and Imputation Proce-dures: For the initial visit, data are pri-marily collected in a personal visit to theestablishment. Quarterly updates areobtained primarily by mail, fax, and tele-phone. Imputation is done for individualbenefits.

Estimates of Sampling Error: Becausestandard errors vary from quarter toquarter, the ECI uses a 5-year movingaverage of standard errors to evaluatepublished series. These standard errorsare available at <http://www.bls.gov/ncs/ect/home.htm>.

Other (nonsampling) Errors: Nonsamplingerrors have a number of potentialsources. The primary sources are (1)survey nonresponse and (2) data collec-tion and processing errors. Nonsam-pling errors are not measured.

Procedures have been implemented forreducing nonsampling errors, primarilythrough quality assurance programs.These programs include the use of datacollection reinterviews, observed inter-views, computer edits of the data, andsystematic professional review of thereports on which the data are recorded.The programs also serve as a trainingdevice to provide feedback to the fieldeconomists, or data collectors, onerrors. They also provide information onthe sources of error which can be rem-edied by improved collection instruc-tions or computer processing edits.Extensive training of field economists isalso conducted to maintain high stan-dards in data collection.

Sources of Additional Material: Bureau ofLabor Statistics, BLS Handbook of Meth-ods, Chapter 8 (Bulletin 2490) and<http://www.bls.gov/ncs>.

Producer Price Index (PPI)

Universe, Frequency, and Types of Data:Monthly survey of producing companiesto determine price changes of all com-modities and services produced in theUnited States for sale in commercialtransactions. Data on agriculture, for-estry, fishing, manufacturing, mining,gas, electricity, public utilities, whole-sale trade, retail trade, transportation,healthcare, and other services.

Type of Data Collection Operation: Prob-ability sample of approximately 30,000establishments that result in about100,000 price quotations per month.

Data Collection and Imputation Proce-dures: Data are collected by mail andfacsimile. If transaction prices are notsupplied, list prices are used. Someprices are obtained from trade publica-tions, organized exchanges, and govern-ment agencies. To calculate index, pricechanges are multiplied by their relativeweights taken from the Census Bureau’s1997 shipment values from their Censusof Industries.

Estimates of Sampling Error: Not appli-cable.

Other (nonsampling) Errors: Not availableat present.

Appendix III 947

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 8: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

Sources of Additional Material: U.S.Bureau of Labor Statistics, BLS Handbookof Methods, Chapter 14, Bulletin 2490.U.S. Bureau of Labor Statistics Internetsites <http://stats.bls.gov/ppi>.

BOARD OF GOVERNORS OF THEFEDERAL RESERVE SYSTEM

Survey of Consumer Finances

Universe, Frequency, and Types of Data:Periodic sample survey of families. Inthis survey a given household is dividedinto a primary economic unit and othereconomic units. The primary economicunit, which may be a single individual,is generally chosen as the unit that con-tains the person who either holds thetitle to the home or is the first personlisted on the lease. The primary unit isused as the reference family. The surveycollects detailed data on the composi-tion of family balance sheets, the termsof loans, and relationships with financialinstitutions. It also gathered informationon the employment history and pensionrights of the survey respondent and thespouse or partner of the respondent.

Type of Data Collection Operation: Thesurvey employs a two-part strategy forsampling families. Some families wereselected by standard multistage areaprobability sampling methods applied toall 50 states. The remaining families inthe survey were selected using statisti-cal records derived from tax returns,under the strict rules governing confi-dentiality and the rights of potentialrespondents to refuse participation.

Data Collection and Imputation Proce-dures: NORC at the University of Chi-cago has collected data for the surveysince 1992. Since 1995, the survey hasused computer-assisted personal inter-viewing. Adjustments for nonresponseare made through multiple imputationof unanswered questions and throughweighting adjustments based on dataused in the sample design for familiesthat refused participation.

Estimates of Sampling Error: Because ofthe complex design of the survey, theestimation of potential sampling errorsis not straightforward. A replicate-basedprocedure is available.

Other (nonsampling) Errors: The surveyaims to complete 4,500 interviews, withabout two thirds of that number deriv-ing from the area-probability sample.The response rate is typically about 70percent for the area-probability sampleand about 35 percent over all strata inthe tax-data sample. Proper training andmonitoring of interviewers, carefuldesign of questionnaires, and system-atic editing of the resulting data wereused to control inaccurate surveyresponses.

Sources of Additional Material: Board ofGovernors of the Federal Reserve Sys-tem, ‘‘Recent Changes in U.S. FamilyFinances: Evidence from the 1998 and2001 Survey of Consumer Finances,’’Federal Reserve Bulletin, January 2003.

U.S. CENSUS BUREAU

2002 Economic Census(Industry Series, GeographicArea Series and SubjectSeries Reports) (for NAICSsectors 22, 42, 44-45, 48-49,and 51-81)

Universe, Frequency, and Types of Data:Conducted every 5 years to obtain dataon number of establishments, numberof employees, total payroll size, totalsales/receipts/revenue, and otherindustry-specific statistics. In 2002, theuniverse was all employer and nonem-ployer establishments primarily engagedin wholesale, retail, utilities, finance &insurance, real estate, transportation &warehousing, information, education,health care, and other service industries.

Type of Data Collection Operation: Alllarge employer firms were surveyed(i.e., all employer firms above payrollsize cutoffs established to separatelarge from small employers) plus a 5percent to 25 percent sample of thesmall employer firms. Firms with noemployees were not required to file acensus return.

Data Collection and Imputation Proce-dures: Mail questionnaires were usedwith both mail and telephone follow-upsfor nonrespondents. Data for nonre-spondents and for small employer firmsnot mailed a questionnaire wereobtained from administrative records of

948 Appendix III

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 9: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

other federal agencies or imputed. Non-employer data were obtained exclu-sively from IRS 2002 income taxreturns.

Estimates of Sampling Error: Not appli-cable for basic data such as sales, rev-enue, receipts, payroll, etc.

Other (nonsampling) Errors: Establish-ment response rates by NAICS sector in2002 ranged from 80 percent to 89 per-cent. Item response rates generallyranged from 50 percent to 90 percentwith lower rates for the more detailedquestions. Nonsampling errors mayoccur during the collection, reporting,and keying of data, and due to industrymisclassification.

Sources of Additional Material: U.S. Cen-sus Bureau, 2002 Economic Census:Industry Series, Geographic Area Seriesand Subject Series Reports (by NAICSsector), Appendix C and <http://www.census.gov/econ/census02/guide/index.html>.

American Community SurveyUniverse, Frequency, and Types of Data:

Nationwide survey to obtain data aboutdemographic, social, economic, andhousing characteristics of people,households and housing units. Covershousehold population and excludes thepopulation living in institutions, collegedormitories, and other group quarters.

Type of Data Collection Operation: Two-stage stratified annual sample ofapproximately 829,000 housing units.The ACS samples housing units from theMaster Address File (MAF). The firststage of sampling involves dividing theUnited States into primary samplingunits (PSUs) most of which comprise ametropolitan area, a large county, or agroup of smaller counties. Every PSUfalls within the boundary of a state. ThePSUs are then grouped into strata on thebasis of independent information, thatis, information obtained from the decen-nial census or other sources. The strataare constructed so that they are ashomogeneous as possible with respectto social and economic characteristicsthat are considered important by ACSdata users. A pair of PSUs were selectedfrom each stratum. The probability ofselection for each PSU in the stratum is

proportional to its estimated 1996population. In the second stage of sam-pling, a sample of housing units withinthe sample PSUs is drawn. Ultimate sam-pling units (USUs) are housing units. TheUSUs sampled in the second stage con-sist of housing units which are system-atically drawn from sorted lists ofaddresses of housing units from theMAF.

Data Collection and Imputation Proce-dures: The American Community Survey(ACS) is conducted every month on inde-pendent samples. Each housing unit inthe independent monthly samples ismailed a prenotice letter announcing theselection of the address to participate, asurvey questionnaire package, and areminder postcard. These sample unitsreceive a second (replacement) ques-tionnaire package if the initial question-naire has not been returned by ascheduled date. In the mail-out/mail-back sites, sample units for which aquestionnaire is not returned in the mailand for which a telephone number isavailable are defined as the telephonenonresponse follow-up universe. Inter-viewers attempt to contact and inter-view these mail nonresponse cases.Sample units from all sites that are stillunresponsive two months after the mail-ing of the survey questionnaires anddirectly after the completion of the tele-phone followup operation are sub-sampled at a rate of 1 in 3. The selectednonresponse units are assigned to FieldRepresentatives, who visit the units,verify their existence or declare themnonexistent, determine their occupancystatus, and conduct interviews. Afterdata collection is completed, anyremaining incomplete or inconsistentinformation was imputed during thefinal automated edit of the collecteddata.

Estimates of Sampling Error: The data inthe ACS products are estimates of theactual figures that would have beenobtained by interviewing the entirepopulation using the same methodol-ogy. The estimates from the chosensample also differ from other samples ofhousing units and persons within thosehousing units.

Other (nonsampling) Errors: NonsamplingError — In addition to sampling error,

Appendix III 949

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 10: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

data users should realize that othertypes of errors may be introduced dur-ing any of the various complex opera-tions used to collect and process surveydata. An important goal of the ACS is tominimize the amount of nonsamplingerror introduced through nonresponsefor sample housing units. One way ofaccomplishing this is by following up onmail nonrespondents.

Sources of Additional Material: U.S. Cen-sus Bureau, American Community Sur-vey Web site available on the Internet,<http://www.census.gov/acs/www/index.html>. U.S. Census Bureau,American Community Survey, Accuracyof the Data documents available on theInternet, <http://www.census.gov/acs/www/UseData/Accuracy/Accuracy1.htm>.

American Housing SurveyUniverse, Frequency, and Types of Data:

Conducted nationally in the fall in oddnumbered years to obtain data on theapproximately 121 million occupied orvacant housing units in the UnitedStates (group quarters are excluded).Data include characteristics of occupiedhousing units, vacant units, new hous-ing and mobile home units, financialcharacteristics, recent mover house-holds, housing and neighborhood qual-ity indicators, and energy character-istics.

Type of Data Collection Operation: Thenational sample was a multistage prob-ability sample with about 61,000 unitseligible for interview in 2003. Sampleunits, selected within 394 PSUs, weresurveyed over a 4-month period.

Data Collection and Imputation Proce-dures: For 2003, the survey was con-ducted by personal interviews. Theinterviewers obtained the informationfrom the occupants or, if the unit wasvacant, from informed persons such aslandlords, rental agents, or knowledge-able neighbors.

Estimates of Sampling Error: For thenational sample, illustrations of the S.E.of the estimates are provided in Appen-dix D of the 2003 report. As anexample, the estimated CV is about 0.2percent for the estimated percentage ofowner-occupied units with two persons.

Other (nonsampling) Errors: Responserate was about 92 percent. Nonsamplingerrors may result from incorrect orincomplete responses, errors in codingand recording, and processing errors.For the 2003 national sample, approxi-mately 2.2 percent of the total housinginventory was not adequately repre-sented by the AHS sample.

Sources of Additional Material: U.S. Cen-sus Bureau, Current Housing Reports,Series H-150 and H-170, American Hous-ing Survey. <http://www.census.gov/hhes/www/ahs.html>.

Annual Survey of Manufac-tures

Universe, Frequency, and Types of Data:The Annual Survey of Manufactures(ASM) is conducted annually, except foryears ending in 2 and 7 for all manufac-turing establishments having one ormore paid employees. The purpose ofthe ASM is to provide key intercensalmeasures of manufacturing activity,products, and location for the publicand private sectors. The ASM providesstatistics on employment, payroll,worker hours, payroll supplements, costof materials, value added by manufac-turing, capital expenditures, inventories,and energy consumption. It also pro-vides estimates of value of shipmentsfor 1,800 classes of manufactured prod-ucts.

Type of Data Collection Operation: TheASM includes approximately 57,000establishments selected from the censusuniverse of 366,000 manufacturingestablishments. Some 27,000 largeestablishments are selected with cer-tainty, and some 30,000 other establish-ments are selected with probabilityproportional to a composite measure ofestablishment size. The survey isupdated from two sources: Internal Rev-enue Service administrative records areused to include new single-unit manu-facturers and the Company OrganizationSurvey identifies new establishments ofmultiunit forms.

Data Collection and Imputation Proce-dures: Survey is conducted by mail withphone and mail follow-ups of nonre-spondents. Imputation (for all nonre-sponse items) is based on previous yearreports, or for new establishments insurvey, on industry averages.

950 Appendix III

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 11: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

Estimates of Sampling Error: Estimatedstandard errors for number of employ-ees, new expenditures, and for valueadded totals are given in annual publica-tions. For U.S. level industry statistics,most estimated standard errors are 2percent or less, but vary considerablyfor detailed characteristics.

Other (nonsampling) Errors: Responserate is about 85 percent. Nonsamplingerrors include those due to collection,reporting, and transcription errors,many of which are corrected throughcomputer and clerical checks.

Sources of Additional Material: U.S. Cen-sus Bureau, Annual Survey of Manufac-tures, and Technical Paper 24.<http://www.census.gov/econ/www/mancen.html>

Annual Surveys of State andLocal Government

Universe, Frequency, and Types of Data:Sample survey conducted annually toobtain data on revenue, expenditure,debt, and employment of state and localgovernments. Universe is all govern-mental units in the United States (about87,500).

Type of Data Collection Operation: Samplesurvey includes all state governments,county governments with 100,000+population, municipalities with 75,000+population, townships with 50,000+population, all independent school dis-tricts with 10,000+ enrollment in March2002, all school districts providing col-lege level (postsecondary) education,and other governments meeting certaincriteria; probability sample for remain-ing units.

Data Collection and Imputation Proce-dures: Field and office compilation ofdata from official records and reportsfor states and large local governments;central collection of local governmentalfinancial data through cooperativeagreements with a number of state gov-ernments; mail canvass of other unitswith mail and telephone follow-ups ofnonrespondents. Data for nonresponsesare imputed from previous year data orobtained from secondary sources, ifavailable.

Estimates of Sampling Error: State andlocal government totals are generallysubject to sampling variability of lessthan 3 percent.

Other (nonsampling) Errors: Nonresponserate is less than 10 percent for localgovernments. Other possible errors mayresult from undetected inaccuracies inclassification, response, and processing.

Sources of Additional Material: Publica-tions: <http://www.census.gov/prod/www/abs/govern.html>:U.S. CensusBureau, Public Employment in 1992, GE92, No. 1, Governmental Finances in1991−1992, GF 92, No. 5, and Census ofGovernments, 1997 and 2002, variousreports. Web site references: Census ofGovernments <http://www.census.gov/govs/www/cog2002.html>, <http://www.census.gov/govs/www/cog.html>.Employment—state and local site:<http://www.census.gov/govs/www/apes.html>. Finance—state and localsite: <http://www.census.gov/govs/www/estimate.html>.

Census of PopulationUniverse, Frequency, and Types of Data:

Complete count of U.S. population con-ducted every 10 years since 1790. Dataobtained on number and characteristicsof people in the U.S.

Type of Data Collection Operation: In1980, 1990, and 2000, complete censusfor some items: age, date of birth, sex,race, and relationship to householder. In1980, approximately 19 percent of thehousing units were included in thesample; in 1990 and 2000, approxi-mately 17 percent.

Data Collection and Imputation Proce-dures: In 1980, 1990, and 2000, mailquestionnaires were used extensivelywith personal interviews in the remain-der. Extensive telephone and personalfollow-up for nonrespondents was donein the censuses. Imputations were madefor missing characteristics.

Estimates of Sampling Error: Samplingerrors for data are estimated for allitems collected by sample and vary bycharacteristic and geographic area. Thecoefficients of variation (CVs) fornational and state estimates are gener-ally very small.

Appendix III 951

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 12: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

Other (nonsampling) Errors: Since 1950,evaluation programs have been con-ducted to provide information on themagnitude of some sources of nonsam-pling errors such as response bias andundercoverage in each census. Resultsfrom the evaluation program for the1990 census indicated that the esti-mated net undercoverage amounted toabout 1.5 percent of the total residentpopulation. For Census 2000, the evalu-ation program indicates a net overcountof 0.5 percent of the resident popula-tion.

Sources of Additional Material: U.S. Cen-sus Bureau, The Coverage of Populationin the 1980 Census, PHC80-E4; ContentReinterview Study: Accuracy of Data forSelected Population and Housing Char-acteristics as Measured by Reinterview,PHC80-E2; 1980 Census of Population,Vol. 1, (PC80-1), Appendixes B, C, andD. 1990 Census of Population and Hous-ing, Content Reinterview Survey: Accu-racy of Data for Selected Population andHousing Characteristics as measured byReinterview, CPH-E-1; 1990 Census ofPopulation and Housing, Effectiveness ofQuality Assurance, CPH-E-2; Programs toImprove Coverage in the 1990 Census,CPH-E-3. For Census 2000, see<http://www.census.gov/pred/www>.

County Business Patterns

Universe, Frequency, and Types of Data:County Business Patterns is an annualtabulation of basic data items extractedfrom the Business Register, a file of allknown single- and multi-location compa-nies maintained and updated by theCensus Bureau. Data include number ofestablishments, number of employees,first quarter and annual payrolls, andnumber of establishments by employ-ment size class. Data are excluded forself-employed persons, domestic serviceworkers, railroad employees, agricul-tural production workers, and most gov-ernment employees.

Type of Data Collection Operation: Theannual Company Organization Surveyprovides individual establishment datafor multi-location companies. Data forsingle establishment companies areobtained from various Census Bureauprograms, such as the Annual Survey of

Manufactures and Current Business Sur-veys, as well as from administrativerecords of the Internal Revenue Serviceand the Social Security Administration.

Estimates of Sampling Error: Not appli-cable.

Other (nonsampling) Error: The data aresubject to non-sampling errors, such asindustry classification errors, as well aserrors of response, keying, and non-reporting.

Sources of Additional Materials: U.S. Cen-sus Bureau, General Explanation ofCounty Business Patterns.

Current Population Survey(CPS)

Universe, Frequency, and Types of Data:Nationwide monthly sample survey ofcivilian noninstitutionalized population,15 years old or over, to obtain data onemployment, unemployment, and anumber of other characteristics.

Type of Data Collection Operation: Multi-stage probability sample of about50,000 households in 754 PSUs in 1996expanded to about 60,000 householdsin July 2001. Over-sampling in somestates and the largest MSAs to improvereliability for those areas of employmentdata on annual average basis. A con-tinual sample rotation system is used.Households are in sample 4 months, outfor 8 months, and in for 4 more. Month-to-month overlap is 75 percent; year-to-year overlap is 50 percent.

Data Collection and Imputation Proce-dures: For first and fifth months that ahousehold is in sample, personal inter-views; other months, approximately 85percent of the data collected by phone.Imputation is done for both item andtotal nonresponse. Adjustment for totalnonresponse is done by a predefinedcluster of units, by MSA size and resi-dence; for item nonresponse imputationvaries by subject matter.

Estimates of Sampling Error: EstimatedCVs on national annual averages forlabor force, total employment, andnonagricultural employment, 0.2 per-cent; for total unemployment and agri-cultural employment, 1.0 percent to 2.5percent. The estimated CVs for familyincome and poverty rate for all persons

952 Appendix III

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 13: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

in 1986 are 0.5 percent and 1.5 percent,respectively. CVs for subnational areas,such as states, would be larger andwould vary by area.

Other (nonsampling) Errors: Estimates ofresponse bias on unemployment are notavailable, but estimates of unemploy-ment are usually 5 percent to 9 percentlower than estimates from reinterviews.Six to 7.0 percent of sample householdsunavailable for interviews.

Sources of Additional Material: U.S. Cen-sus Bureau and Bureau of Labor Statis-tics, Current Population Survey; Designand Methodology, (Tech. Paper 63),available on Internet <http://www.census.gov/prod/2002pubs/tp63rv.pdf>and Bureau of Labor Statistics, Employ-ment and Earnings, monthly, Explana-tory Notes and Estimates of Error,Household Data and BLS Handbook ofMethods, Chapter 1, available on theInternet <http://www.bls.gov/opub/hom/homch1a.htm>.

Foreign Trade—Import Statis-tics

Universe, Frequency, and Types of Data:The import entry documents collectedby U.S. Bureau of Customs and BorderProtection are processed each month toobtain data on the movement of mer-chandise imported into the UnitedStates. Data obtained include value,quantity, and shipping weight by com-modity, country of origin, district ofentry, and mode of transportation.

Type of Data Collection Operation: Importentry documents, either paper or elec-tronic, are required to be filed for theimportation of goods into the UnitedStates valued over $2,000, or for articleswhich must be reported on formalentries. U.S. Bureau of Customs and Bor-der Protection officials collect and trans-mit statistical copies of the documentsto the Census Bureau on a flow basis fordata compilation. Estimates for ship-ments valued under $2,001 and notreported on formal entries are based onestimated established percentages forindividual country totals.

Data Collection and Imputation Proce-dures: Statistical copies of import entrydocuments, received on a daily basisfrom ports of entry throughout the

country, are subjected to a monthly pro-cessing cycle. They are fully processedto the extent they reflect items valued at$2,001 and over or items which must bereported on formal entries.

Estimates of Sampling Error: Not appli-cable.

Other (nonsampling) Errors: The goodsdata are a complete enumeration ofdocuments collected by the U.S. Bureauof Customs and Border Protection andare not subject to sampling errors; butthey are subject to several types of non-sampling errors. Quality assurance pro-cedures are performed at every stage ofcollection, processing and tabulation;however the data are still subject to sev-eral types of nonsampling errors. Themost significant of these include report-ing errors, undocumented shipments,timeliness, data capture errors, anderrors in the estimation of low-valuedtransactions.

Sources of Additional Material: U.S. Cen-sus Bureau, FT 900 U.S. InternationalTrade in Goods and Services, FT 925(discounted after 1996) U.S. Merchan-dise Trade, FT 895 U.S. Trade withPuerto Rico and U.S. Possessions, FT920U.S. Merchandise Trade: selected high-lights, and Information Section onGoods and Services at<http://www.census.gov/ft900>.

Foreign Trade—Export Statis-tics

Universe, Frequency, and Types of Data:The export declarations collected byU.S. Bureau of Customs and Border Pro-tection are processed each month toobtain data on the movement of U.S.merchandise exports to foreign coun-tries. Data obtained include value, quan-tity, and shipping weight of exports bycommodity, country of destination, dis-trict of exportation, and mode of trans-portation.

Type of Data Collection Operation: Ship-per’s Export Declarations (paper andelectronic) are generally required to befiled for the exportation of merchandisevalued over $2,500. U.S. Bureau of Cus-toms and Border Protection officials col-lect and transmit the documents to theCensus Bureau on a flow basis for data

Appendix III 953

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 14: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

compilation. Data for shipments valuedunder $2,501 are estimated, based onestablished percentages of individualcountry totals.

Data Collection and Imputation Proce-dures: Statistical copies of Shipper’sExport Declarations are received on adaily basis from ports throughout thecountry and subject to a monthly pro-cessing cycle. They are fully processedto the extent they reflect items valuedover $2,500. Estimates for shipmentsvalued at $2,500 or less are made,based on established percentages ofindividual country totals.

Estimates of Sampling Error: Not appli-cable.

Other (nonsampling) Errors: The goodsdata are a complete enumeration ofdocuments collected by the U.S. Bureauof Customs and Border Protection andare not subject to sampling errors; butthey are subject to several types of non-sampling errors. Quality assurance pro-cedures are performed at every stage ofcollection, processing and tabulation;however the data are still subject to sev-eral types of nonsampling errors. Themost significant of these include report-ing errors, undocumented shipments,timeliness, data capture errors, anderrors in the estimation of low-valuedtransactions.

Sources of Additional Material: U.S. Cen-sus Bureau, FT 900 U.S. InternationalTrade in Goods and Services, FT 925(discounted after 1996) U.S. Merchan-dise Trade, FT 895 U.S. Trade withPuerto Rico and U.S. Possessions, FT920 U.S. Merchandise trade: selectedhighlights, and Information Section onGoods and Services at<http://census.gov/ft900>.

Monthly Retail Trade andFood Service Survey

Universe, Frequency, and Types of Data:Provides monthly estimates of retail andfood service sales by kind of businessand end-of-month inventories of retailstores.

Type of Data Collection Operation: Prob-ability sample of all firms from a listframe. The list frame is the CensusBureau’s Business Register updated

quarterly for recent birth Employer Iden-tification (EI) Numbers issued by theInternal Revenue Service and assigned akind-of-business code by the SocialSecurity Administration. The largestfirms are included monthly; a sample ofothers is included every month also.

Data Collection and Imputation Proce-dures: Data are collected by mail ques-tionnaire with telephone follow-ups andfax reminders for nonrespondents.Imputation is made for each nonre-sponse item and each item failing editchecks.

Estimates of Sampling Error: For the 2004monthly surveys, CVs are about 0.5 per-cent for estimated total retail sales and1.2 percent for estimated total retailinventories. Sampling errors are shownin monthly publications.

Other (nonsampling) Errors: Imputationrates are about 20 percent for monthlyretail and food service sales, and 28 per-cent for monthly retail inventories.

Sources of Additional Material: U.S. Cen-sus Bureau, Current Business Reports,Annual Benchmark Report for RetailTrade and Food Services.

Monthly Survey of Construc-tion

Universe, Frequency, and Types of Data:Survey conducted monthly of newly con-structed housing units (excludingmobile homes). Data are collected onthe start, completion, and sale of hous-ing. (Annual figures are aggregates ofmonthly estimates.)

Type of Data Collection Operation: Forpermit issuing places probability sampleof 850 housing units obtained from19,000 permit issuing places. For non-permit places,multistage probabilitysample of new housing units selected in169 PSUs. In those areas, all roads arecanvassed in selected enumeration dis-tricts.

Data Collection and Imputation Proce-dures: Data are obtained by telephoneinquiry and field visit.

Estimates of Sampling Error: EstimatedCV of 3 percent to 4 percent for esti-mates of national totals, but may be

954 Appendix III

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 15: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

higher than 20 percent for estimatedtotals of more detailed characteristics,such as housing units in multiunit struc-tures.

Other (nonsampling) Errors: Responserate is over 90 percent for most items.Nonsampling errors are attributed todefinitional problems, differences ininterpretation of questions, incorrectreporting, inability to obtain informationabout all cases in the sample, and pro-cessing errors.

Sources of Additional Material: All dataare available on the Internet at<http://www.census.gov/const/www/newsresconstindex.html>. Furtherdocumentation of the survey is alsoavailable at that site.

Nonemployer StatisticsUniverse, Frequency, and Types of Data:

Nonemployer statistics are an annualtabulation of economic data by industryfor active businesses without paidemployees that are subject to federalincome tax. Data showing the numberof establishments and receipts by indus-try are available for the U.S., states,counties, and metropolitan areas. Mosttypes of businesses covered by the Cen-sus Bureau’s economic statistics pro-grams are included in the nonemployerstatistics. Tax-exempt and agricultural-production businesses are excludedfrom nonemployer statistics.

Type of Data Collection Operation: Theuniverse of nonemployer establishmentsis created annually as a by-product ofthe Census Bureau’s Business Registerprocessing for employer establishments.If a business is active but without paidemployees, then it becomes part of thepotential nonemployer universe. Indus-try classification and receipts are avail-able for each potential nonemployerbusiness. These data are obtained pri-marily from the annual business incometax returns of the Internal Revenue Serv-ice (IRS). The potential nonemployer uni-verse undergoes a series of complexprocessing, editing, and analyticalreview procedures at the Census Bureauto distinguish nonemployers fromemployers, and to correct and completedata items used in creating the datatables.

Estimates of Sampling Error: Not appli-cable.

Other (nonsampling) Errors: The data aresubject to nonsampling errors, such aserrors of self-classification by industryon tax forms, as well as errors ofresponse, keying, nonreporting, andcoverage.

Sources of Additional Material: U. S. Cen-sus Bureau, Nonemployer Statistics:2002 (Introduction; Coverage and Meth-odology). See also <http://www.census.gov/epcd/nonemployer/view/cov&meth.htm>

Service Annual SurveyUniverse, Frequency, and Types of Data:

The U.S. Census Bureau conducts theService Annual Survey to providenational estimates of revenues,expenses, and e-commerce revenues fortaxable and tax-exempt firms classifiedin selected service industries. Estimatesare summarized by industry classifica-tion based on the 1997 North AmericanIndustry Classification System (NAICS).Industries covered by the Service AnnualSurvey include all or part of the follow-ing NAICS sectors: Transportation andWarehousing (NAICS 48−49); Informa-tion (NAICS 51); Finance and Insurance(NAICS 52); Real Estate and Rental andLeasing (NAICS 53); Professional, Scien-tific, and Technical Services (NAICS 54);Administrative and Support and WasteManagement and Remediation Services(NAICS 56); Health Care and Social Assis-tance (NAICS 62); Arts, Entertainment,and Recreation (NAICS 71); and OtherServices, except Public Administration(NAICS 81). Data items collected includetotal revenue, revenue from e-commercetransactions; and for selected industries,revenue from detailed service products,total expenses, and expenses by majortype, revenue from exported services,and inventories. Questionnaires aremailed in January and request annualdata for the prior year. Estimates arepublished approximately 12 monthsafter the initial survey mailing.

Type of Data Collection Operation: TheService Annual Survey estimates aredeveloped using data from a probabilitysample and administrative records. Serv-ice Annual Survey questionnaires aremailed to a probability sample that isperiodically reselected from a universe

Appendix III 955

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 16: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

of firms located in the United States andhaving paid employees. The sampleincludes firms of all sizes and coversboth taxable firms and firms exemptfrom Federal income taxes. Updates tothe sample are made on a quarterlybasis to account for new businesses.Firms without paid employees, or non-employers, are included in the estimatesthrough imputation and/or administra-tive records data provided by other Fed-eral agencies. Links to additionalinformation about confidentiality protec-tion, sampling error, nonsampling error,sample design, definitions, and copiesof the questionnaires may be found onthe Internet at <http://www.census.gov/econ/www/servmenu.html>.

Estimates of Sampling Error: Coefficientsof variation for the 2003 Service AnnualSurvey estimates range from 0.6% to2.2% for total revenue estimates com-puted at the NAICS sector (2-digit NAICScode) level. Sampling errors for moredetailed industries are shown in the cor-responding publications. The full 2003Service Annual Survey results, includingcoefficients of variations, can be foundat <http://www.census.gov/econ/www/servmenu.html>. Links to addi-tional information regarding samplingerror may be found at: <http://www.census.gov/svsd/www/cv.html>.

Other (Nonsampling) Errors: Data areimputed for unit nonresponse, item non-response, and for reported data thatfails edits. The percent of imputed datafor total revenue for the 2003 ServiceAnnual Survey is approximately 14%.

Sources of Additional Material: U.S. Cen-sus Bureau, Current Business Reports,Service Annual Survey, Census BureauWeb site: <http://www.census.gov/econ/www/servmenu.html>.

U.S. DEPARTMENT OF EDUCATION

National Center for EducationStatistics Higher EducationGeneral Information Survey(HEGIS), Degrees and OtherFormal Awards Conferred.Beginning 1986, IntegratedPostsecondary EducationData Survey (IPEDS), Comple-tions

Universe, Frequency, and Types of Data:Annual survey of all institutions andbranches listed in the Education Direc-tory, Colleges and Universities to obtaindata on earned degrees and other for-mal awards, conferred by field of study,level of degree, sex, and by racial/ethniccharacteristics (every other year prior to1989, then annually).

Type of Data Collection Operation: Com-plete census.

Data Collection and Imputation Proce-dures: Data are collected through a Web-based survey in the fall of every year.Missing data are imputed by using dataof similar institutions.

Estimates of Sampling Error: Not appli-cable.

Other (nonsampling) Errors: For 2002−03,approximately 100.0 percent responserate for degree-granting institutions.

Sources of Additional Material: U.S.Department of Education, National Cen-ter for Education Statistics, Postsecond-ary Institutions in the United States: Fall2003 and Degrees and Other AwardsConferred: 2002−03. <http://www.nces.ed.gov/ipeds/>.

National Household EducationSurveys Program

Universe, Frequency, and Types of Data:The National Household Education Sur-veys Program (NHES) is a system of tele-phone surveys of the noninstitution-alized civilian population of the UnitedStates. Surveys in NHES have varyinguniverses of interest depending on theparticular survey. Specific topics cov-ered by each survey are at the NHESWeb site <http://nces.ed.gov/nhes>. Alist of the surveys fielded as part ofNHES, each universe, and the years theywere fielded is provided below. 1. AdultEducation Interviews were conductedwith a representative sample of civilian,noninstitutionalized persons age 16 andolder who were not enrolled in grade 12or below (1991, 1995, 1999, 2001,2003). 2. Before-and After-School Pro-grams and Activities Interviews wereconducted with parents of a representa-tive sample of students in grades Kthrough 8 (1999, 2001). 3. CivicInvolvement Interviews were conductedwith a representative sample of parents,

956 Appendix III

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 17: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

youth, and adults(1996, 1999). 4. EarlyChildhood Program Participation Inter-views were conducted with parents of arepresentative sample of children frombirth through grade 3, with the specificage groups varying by survey year(1991, 1995, 1999, 2001). 5. Householdand Library Use Interviews were con-ducted with a representative sample ofU.S. households (1996). 6. Parent andFamily Involvement in Education Inter-views were conducted with parents of arepresentative sample of children age 3through grade 12 or in grades K through12 depending on the survey year (1996,1999, 2003). 7. School Readiness Inter-views were conducted with parents of arepresentative sample of 3- to 7-year-oldchildren (1993, 1999). 8. School Safetyand Discipline Interviews were con-ducted with a representative sample ofstudents in grades 6−12, their parents,and the parents of a representativesample of students in grades 3−5(1993).

Type of Data Collection Operation: NHESuses telephone interviews to collectdata.

Data Collection and Imputation Proce-dures: Telephone numbers are selectedusing random digit dialing techniques.Approximately 45,000 to 64,000 house-holds are contacted in order to identifypersons eligible for the surveys. Dataare collected using computer-assistedtelephone interviewing (CATI) proce-dures. Missing data are imputed usinghot-deck imputation procedures.

Estimates of Sampling Error: Unweightedsample sizes range between 2,500 and21,000. The average root design effectsof the surveys in NHES range from 1.1to 4.5.

Other (nonsampling) Errors: Because ofunit nonresponse and because thesamples are drawn from householdswith telephone instead of all house-holds, nonresponse and/or coveragebias may exist for some estimates. How-ever, both sources of potential bias areadjusted for in the weighting process.Analyses of both potential sources ofbias in the NHES collections have beenstudied and no significant bias has beendetected.

Sources of Additional Material: Please seethe NHES Web site at <http://nces.ed.gov/nhes>

U.S. FEDERAL BUREAU OF INVESTI-GATION

Uniform Crime Reporting(UCR) Program

Universe, Frequency, and Types of Data:Monthly reports on the number of crimi-nal offenses that become known to lawenforcement agencies. Data are col-lected on crimes cleared by arrest; byage, sex, and race of arrestees and forvictims and offenders for homicides, onfatal and nonfatal assaults against lawenforcement officers, and on hatecrimes reported.

Type of Data Collection Operation: Crimestatistics are based on reports of crimedata submitted either directly to the FBIby contributing law enforcement agen-cies or through cooperating state UCRprograms.

Data Collection and Imputation Proce-dures: States with UCR programs collectdata directly from individual lawenforcement agencies and forwardreports, prepared in accordance withUCR standards, to FBI. Accuracy andconsistency edits are performed by FBI.

Estimates of Sampling Error: Not appli-cable.

Other (nonsampling) Errors: Coverage of93 percent of the population (95 percentin MSAs, 85 percent in ‘‘cities outside ofmetropolitan areas,’’ and 83 percent innonmetropolitan counties) by UCR Pro-gram, through varying number of agen-cies reporting.

Sources of Additional Material: U.S. Fed-eral Bureau of Investigation, Crime inthe United States, annual, Hate CrimeStatistics, annual, Law EnforcementOfficers Killed & Assaulted, annual,<http://www.fbi.gov/ucr.htm>.

U.S. INTERNAL REVENUE SERVICE

Corporation Income TaxReturns

Universe, Frequency, and Types of Data:Annual study of unaudited corporationincome tax returns, Forms 1120,

Appendix III 957

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 18: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

1120-A, 1120-F, 1120-L, 1120-PC, 1120-REIT, 1120-RIC, and 1120S, filed by cor-porations or businesses legally definedas corporations. Data provided on vari-ous financial characteristics by industryand size of total assets, and businessreceipts.

Type of Data Collection Operation: Strati-fied probability sample of approximately147,000 returns for Tax Year 2002, allo-cated to sample classes which are basedon type of return, size of total assetssize of net income or deficit, andselected business activity. Samplingrates for sample classes varied from .25percent to 100 percent.

Data Collection and Imputation Proce-dures: Computer selection of sample oftax return records. Data adjusted duringediting for incorrect, missing, or incon-sistent entries to ensure consistencywith other entries on return and to com-ply with statistical definitions.

Estimates of Sampling Error: EstimatedCVs for Tax Year 2002: Returns withassets over $250 million are self-representing. For other returns groupedby assets, CVs ranged from 0.01 percentto 2.16 percent; for amount of netincome CV is 0.19 percent.

Other (nonsampling) Errors: Nonsamplingerrors include coverage errors, process-ing errors, and response errors.

Sources of Additional Material: U.S. Inter-nal Revenue Service, Statistics ofIncome, Corporation Income TaxReturns, annual.

Partnership Income TaxReturns

Universe, Frequency, and Types of Data:Annual study of unaudited income taxreturns of partnerships, Form 1065.Data provided on various financial char-acteristics by industry.

Type of Data Collection Operation: Strati-fied probability sample of approximately34,800 partnership returns from a popu-lation of 2.4 million filed during calen-dar year 2003. The sample is classifiedbased on combinations of grossreceipts, net income or loss, and totalassets, and on industry. Sampling ratesvary from 0.12 percent to 100 percent.

Data Collection and Imputation Proce-dures: Computer selection of sample oftax return records. Data are adjustedduring editing for incorrect, missing, orinconsistent entries to ensure consis-tency with other entries on return. Datanot available due to regulations are notimputed.

Estimates of Sampling Error: EstimatedCVs for tax year 2002 (latest available):For number of partnerships, 0.29 per-cent; business receipts, 0.34 percent;net income, 0.80 percent; net loss, 1.72percent.

Other (nonsampling) Errors: Processingerrors and errors arising from the use oftolerance checks for the data.

Sources of Additional Material: U.S. Inter-nal Revenue Service, Statistics ofIncome, Partnership Returns and Statis-tics of Income Bulletin, Vol. 24, No. 2(fall 2004).

Sole Proprietorship IncomeTax Returns

Universe, Frequency, and Types of Data:Annual study of unaudited income taxreturns of nonfarm sole proprietorships,form 1040 with business schedules.Data provided on various financial char-acteristics by industry.

Type of Data Collection Operation: Strati-fied probability sample of approximately50,000 sole proprietorships for tax year2002. The sample is classified based onpresence or absence of certain businessschedules; the larger of total income orloss; and size of business plus farmreceipts. Sampling rates vary from 0.05percent to 100 percent.

Data Collection and Imputation Proce-dures: Computer selection of sample oftax return records. Data adjusted duringediting for incorrect, missing, or incon-sistent entries to ensure consistencywith other entries on return.

Estimates of Sampling Error: EstimatedCVs for tax year 2002 are available. Forsole proprietorships, business receipts,0.69 percent; depreciation 1.41 percent.

Other (nonsampling) Errors: Processingerrors and errors arising from the use oftolerance checks for the data.

958 Appendix III

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 19: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

Sources of Additional Material: U.S. Inter-nal Revenue Service, Statistics ofIncome, Sole Proprietorship Returns (foryears 1980 through 1983) and Statisticsof Income Bulletin, Vol. 24, No. 1 (sum-mer 2004, as well as bulletins for earlieryears).

Individual Income TaxReturns

Universe, Frequency, and Types of Data:Annual study of unaudited individualincome tax returns, forms 1040, 1040A,and 1040EZ, filed by U.S. citizens andresidents. Data provided on variousfinancial characteristics by size ofadjusted gross income, marital status,and by taxable and nontaxable returns.Data by state, based on 100 percent file,also include returns from 1040NR, filedby nonresident aliens plus certain self-employment tax returns.

Type of Data Collection Operation: Annual2002 stratified probability sample ofapproximately 176,000 returns brokeninto sample strata based on the larger oftotal income or total loss amounts aswell as the size of business plus farmreceipts. Sampling rates for samplestrata varied from 0.05 percent to 100percent.

Data Collection and Imputation Proce-dures: Computer selection of sample oftax return records. Data adjusted duringediting for incorrect, missing, or incon-sistent entries to ensure consistencywith other entries on return.

Estimates of Sampling Error: EstimatedCVs for tax year 2002: Adjusted grossincome less deficit 0.12 percent; salariesand wages 0.21 percent; and taxexempt interest received 1.78 percent.(State data not subject to samplingerror.)

Other (nonsampling) Errors: Processingerrors and errors arising from the use oftolerance checks for the data.

Sources of Additional Material: U.S. Inter-nal Revenue Service, Statistics ofIncome, Individual Income Tax Returns,annual.

U.S. NATIONAL CENTER FORHEALTH STATISTICS (NCHS)

National Health Interview Sur-vey (NHIS)

Universe, Frequency, and Types of Data:Continuous data collection covering thecivilian noninstitutionalized populationto obtain information on demographiccharacteristics, conditions, injuries,impairments, use of health services,health behaviors, and other health top-ics.

Type of Data Collection Operation: Multi-stage probability sample of 49,000households (in 198 PSUs) from 1985 to1994; 36−40,000 households (358design PSUs) from 1995 on.

Data Collection and Imputation Proce-dures: Some missing data items (e.g.,race, ethnicity) are imputed using a hotdeck imputation value. Unit nonre-sponse is compensated for by an adjust-ment to the survey weights.

Estimates of Sampling Error: Estimates ofStandard Error (SE): For 2003 medicallyattended injury episodes rates in thepast 12 months by falling for: females29.72 (1.94), and males 26.15 (1.82) per1,000 population; for 2003 injury epi-sodes rates during the past 12 monthsinside the home−20.72 (1.18) per 1,000population.

Other (nonsampling) Errors: The responserate was 93.8 percent in 1996; in 2003,the total household response rate was89.2 percent, with the final familyresponse rate of 87.9 percent, and thefinal sample adult response rate of 74.2percent. (Note: The NHIS sample rede-sign was conducted in 1995, and theNHIS questionnaire was redesigned in1997.)

Sources of Additional Material: NationalCenter for Health Statistics, SummaryHealth Statistics for the U.S. Population:National Health Interview Survey, 2003,Vital and Health Statistics, Series 10#224; National Center for Health Statis-tics, Summary Health Statistics for U.S.Children: National Health Interview Sur-vey, 2003, Vital and Health Statistics,Series 10 #223; National Center forHealth Statistics, Summary Health Statis-tics for U.S. Adults: National HealthInterview Survey, 2003, Vital and HealthStatistics, Series 10 #225; U.S. NationalCenter for Health Statistics, Design and

Appendix III 959

U.S. Census Bureau, Statistical Abstract of the United States: 2006

Page 20: Appendix III Limitations of the Data · Appendix III Limitations of the Data Introduction The data presented in this Statistical Abstract came from many sources. The sources include

Estimation for the National Health Inter-view Survey, 1995−2004, Vital andHealth Statistics, Series 2 #130.

National Vital Statistics Sys-tem

Universe, Frequency, and Types of Data:Annual data on births and deaths in theUnited States.

Type of Data Collection Operation: Mortal-ity data based on complete file of deathrecords, except 1972, based on 50 per-cent sample. Natality statistics1951−71, based on 50 percent sampleof birth certificates, except a 20 percentto 50 percent sample in 1967, receivedby NCHS. Beginning 1972, data fromsome states received through Vital Sta-tistics Cooperative Program (VSCP) andcomplete file used; data from otherstates based on 50 percent sample.Beginning 1986, all reporting areas par-ticipated in the VSCP.

Data Collection and Imputation Proce-dures: Reports based on records fromregistration offices of all states, Districtof Columbia, New York City, PuertoRico, Virgin Islands, Guam, AmericanSamoa, and Northern Marianas.

Estimates of Sampling Error: For recentyears, there is no sampling for thesefiles; the files are based on 100 percentof events registered.

Other (nonsampling) Errors: Data onbirths and deaths believed to be at least99 percent complete.

Sources of Additional Material: U.S.National Center for Health Statistics,Vital Statistics of the United States, Vol. Iand Vol. II, annual, and National VitalStatistics Reports. NCHS Web site at<http://www.cdc.gov/nchs/nvss.htm>.

National Highway Traffic SafetyAdministration (NHTSA)

Fatality Analysis ReportingSystem (FARS)

Universe, Frequency, and Types of Data:FARS is a census of all fatal motorvehicle traffic crashes that occurthroughout the United States includingthe District of Columbia and Puerto Ricoon roadways customarily open to thepublic. The crash must be reported to

the state/jurisdiction and at least onedirectly related fatality must occurwithin thirty days of the crash.

Type of Data Collection Operation: One ormore analysts, in each state, extractdata from the official documents andenter the data into a standardized elec-tronic database.

Data Collection and Imputation Proce-dures: Detailed data describing the char-acteristics of the fatal crash, the vehiclesand persons involved are obtained frompolice crash reports, driver and vehicleregistration records, autopsy reports,highway department, etc. Computerizededit checks monitor that accuracy andcompleteness of the data. The FARSincorporates a sophisticated mathemati-cal multiple imputation procedure todevelop a probability distribution ofmissing blood alcohol concentration(BAC) levels in the database for drivers,pedestrians and cyclists.

Estimates of Sampling Error: Since this iscensus data, there are no samplingerrors.

Other (nonsampling) Errors: Fatal motorvehicle traffic crash data are more than97 percent complete. However, thesedata are highly dependent on the accu-racy of the police accident reports.Errors or omissions within police acci-dent reports may not be detected.

Sources of Additional Material: The FARSCoding and Validation Manual, ANSID16.1 Manual on Classification of MotorVehicle Traffic Accidents (Sixth Edition).

960 Appendix III

U.S. Census Bureau, Statistical Abstract of the United States: 2006