Skip Main Navigation Skip To Navigation Content

Coverage Measurement

You are here: Census.govCoverage Measurement › Definitions
Skip top of page navigation

Definitions

Accuracy and Coverage Evaluation (A.C.E.)
– The coverage measurement program for Census 2000.

Address canvassing
– A decennial census field operation to ensure the currency and completeness of the Master Address File and to create the basis of the decennial census address file.

Alternate address
– A respondent-provided address obtained during the CCM Person Interview or Person Follow-up that indicates where, other than the sample address, someone in the household may have been counted on Census Day.

Alternate address matching
– Part of the CCM matching operation that supplements the nationwide computer match. Generally, the search area is the block cluster that contains the alternate address (and the surrounding blocks).

Birth Cohorts
– A group who have shared a particular experience during a particular time span. Demographic Analysis has followed births cohorts since 1935.

Block cluster
– A single collection block or a group of contiguous collection blocks. Used as the primary sampling and work unit for the CCM.

Capture-recapture model
– The capture-recapture model is used to estimate the unknown size of a population; the simplest capture-recapture model is the two-sample model. In this model, the first sample provides the individuals for marking or tagging and these are returned to the population, while the second sample provides the recaptures. Under certain assumptions, using the numbers of individuals captured in both samples (the recaptures) and the numbers captured in just one sample, it is possible to estimate the number not captured in either sample, thus providing an estimate of the total population size.

Census Coverage Measurement (CCM)
– The coverage measurement program that will be used to evaluate coverage of the 2010 Census. Also refers to the post-enumeration survey and other parts of the 2010 Census coverage measurement program.

Census Day
– The reference date for collection of information for a census. For the decennial census, this has been April 1 of the decade year (the year ending in zero) since the 1930 Census.

Census Day residence status
– A determination of the residence status of a person on Census Day. In particular, for a person on the CCM Person Interview roster at the sample address, the result may be resident, nonresident, or unclassified.

Census Identification Number (CID)
– A number associated with each record in the Master Address File (MAF). It is also called the Master Address File Identification Number (MAFID).

Census-only unit
– A nonmatched (from the CCM initial housing unit operations) census unit that is in a CCM sample block.

Characteristic imputation
– Imputation of census or CCM characteristics. The population characteristics include household relationship, sex, race, Hispanic origin, age and date of birth. The housing characteristic is tenure.

Cohorts
– A group who has shared a particular experience during a particular time span. Demographic Analysis has followed births cohorts since 1935.

Collection block
– The smallest geographic area within which census enumeration operations are conducted, usually bounded by visible features.

Components of census coverage
– Includes housing unit or person omissions, erroneous enumerations (including duplicates), correct enumerations, and whole-person census imputations, or some subcategories of these components.

Contamination error
– An error which can result when the census processes carried out in the CCM sample areas are different from those in the remainder of the country in ways that may affect the census counts for those areas.

Conflicting household
– A situation in which the CCM housing unit matches the census housing unit and both contain whole households of nonmatched people.

Correct enumeration
– A census enumeration for an actual person or housing unit that is appropriate, complete, and in the correct location for the relevant tabulation.

Correlation bias
– The tendency towards underestimation by the dual-system estimator if persons enumerated in the census are more likely than those missed in the census to have been enumerated in the coverage measurement survey.

Count imputation
– A stage of census processing that includes imputation for status, occupancy, or household size.

Coverage error
– An error that results from 1) the failure to include in a census or survey all eligible persons or housing units, or 2) the inclusion of some persons or housing units erroneously. Examples of coverage errors include omissions, duplicates, and erroneous enumerations.

See net coverage error and components of census coverage.

Coverage Follow-up (CFU)
– A decennial census operation that provides a follow-up of selected households that are likely to have coverage errors in the person roster. The universe of cases also includes large households from forms that were mailed back.

Data-defined enumeration
– A census enumeration for which two or more non-imputed characteristics have been collected.

Demographic Analysis
– A technique used to develop an understanding of the age, sex, and racial composition of a population and how it has changed over time through the basic demographic processes of birth, death, and migration. The term Demographic Analysis (usually abbreviated as DA) also refers to a specific set of techniques for developing national population estimates by age, sex, and race from administrative records to be used to assess the quality of the decennial census.

Demographic Balancing (or bookkeeping) equation
– The demographic balancing equation is as follows: P = (B - D) + (Mi - Mo). The components of this equation include Population (P), Births (B), Deaths (D), and In (Mi) and Out (Mo) Migration. The population at any time is equal to births minus deaths, plus the amount of in-migration minus the amount of out-migration.

Demographic Benchmark Analysis
– A technique using a variety of demographic benchmarks and analytic indicators to draw inferences about census coverage at the subnational level. These benchmarks can include independent housing unit estimates, administrative data on births (for inferences about the population under age 10), Medicare enrollment data (for inferences about the population aged 65 and over), and sex ratios (for inferences about the difference in coverage of males and females). Demographic Benchmark Analysis can provide useful broad indicators of coverage patterns for specific demographic groups.

Differential undercount rate
– The difference between the net undercount rate for a particular demographic or geographic domain and the net undercount rate either for another domain or for the nation.

Discrepant
– A census or coverage measurement survey record that does not refer to a real person. Also called “fictitious.”

Domain for estimation
– A collection of individuals defined by various characteristics, usually geographic or demographic.

Dual-system estimation (DSE)
– A statistical method that uses the capture-recapture technique to estimate the true size of a population. A sample of census enumerations (the CCM E sample) yields an estimate of the proportion of the population correctly enumerated in the census, while an independent sample (the CCM P sample) yields an estimate of the proportion of the population missed by the census. These two components are used to produce the DSE estimate of the size of the population.

Duplicate
– A person or housing unit that is enumerated more than once in the census or in the coverage measurement survey.

E sample
– A sample of data-defined census enumerations. For CCM, the E sample is composed of housing units and persons in the same sample block clusters as the P sample.

Enumeration status
– The classification of a data-defined census person or a census housing unit into one of the following categories: correct, erroneous, duplicate, or in the wrong location.

Erroneous enumeration
– A census person or housing unit enumeration that should not have been counted for any of several reasons, such as, that the person or housing unit 1) is a duplicate of a correct enumeration; 2) is inappropriate (e.g., the person died before Census Day); or 3) is enumerated in the wrong location for the relevant tabulation.

Fictitious
– See “discrepant.”

Final Housing Unit Clerical Matching
– A CCM operation in which clusters containing units flagged during final housing unit (FHU) computer processing are reviewed by clerical matching staff (using a computer-assisted clerical matching system, FHUMaRCS) in order to assign match, duplicate, enumeration, or unit status to the flagged units. Units with unresolved status are sent to final housing unit follow-up. Clerical matching includes before-follow-up and after-follow-up phases.

Final Housing Unit Computer Processing
– A CCM operation in which information from previous stages of person and housing unit matching (both computer and clerical), along with final census housing unit data, is used to identify P- and E-sample housing unit records that require further information in order to determine their match, duplicate, enumeration, or unit status.

Final Housing Unit Follow-up
– A CCM field operation in which field interviewers collect data from units needing further information to resolve match, duplicate, enumeration, or unit status.

Final Housing Unit Operations
– A set of CCM operations designed to identify geocoding errors and housing unit duplicates; update pre-existing files; and resolve match, enumeration, or unit status for all P- and E-sample housing units not resolved in previous CCM operations. The final housing unit operations include final housing unit computer processing, final housing unit before follow-up clerical matching, final housing unit follow-up, and final housing unit after follow-up clerical matching.

Geocode (geographic code)
– A code or set of codes used to identify a specific geographic entity. For example, the information needed to geocode a CCM respondent-provided address is the state code, county code, and block number.

Geocoding
– The assignment of an address, structure, key geographic location, or business name to a location that is identified by one or more geographic codes. For living quarters, geocoding usually requires identification of a specific census block.

Geocoding error
– The result of assigning an incorrect geocode to an entity. For CCM, this refers to assigning a P- or E-sample housing unit to the incorrect census collection block.

Gross coverage error
– A measure of coverage error that adds together omissions and erroneous enumerations. The 2010 CCM is not planning to produce estimates of gross coverage error, but will estimate these components separately.

Group quarters (GQ)
– Living quarters that are normally owned or managed by an entity or organization providing housing or services for the residents. People living in group quarters are generally not related to each other. Group quarters include institutions such as prisons and nursing homes, and other places such as college dormitories, monasteries, and convents.

Homogeneity assumption
– In the capture-recapture methodology, the homogeneity assumption implies that the probability of being captured in a particular enumeration is the same for each individual in a population. Since this assumption rarely holds, even approximately, across an entire population, the estimate of the population size is often calculated by using post-stratification or modeling approaches like logistic regression.

Household size imputation
– Imputation of a census population count for a housing unit known to be occupied, but with an unknown number of individuals residing in the unit.

Housing unit
– A house, townhouse, mobile home or trailer, apartment, group of rooms, or single room that is occupied as a separate living quarters or, if vacant, is intended for occupancy as a separate living quarters. If occupied by someone who has no other usual residence, a camper, tent, boat, van, motel room, or any other similar unit is classified as a housing unit.

Housing unit status
– See Unit status

Imputation
– A statistical technique for assigning reasonable values for missing, bad, or inconsistent data.

Independence assumption
– In the capture-recapture methodology, the independence assumption implies that the probability of an individual being captured in both enumerations equals the product of the probabilities of being captured in either enumeration separately. In other words, capture status in one enumeration does not affect the probability of capture in the other enumeration.

Independent Listing
– A CCM field operation, conducted separately from the census, in order to compile a list of all housing units and potential housing units in each CCM sample block cluster.

Initial Housing Unit Clerical Matching
– A set of CCM clerical operations that reviews the results of computer matching and determines match, duplicate, and unit status for each CCM and census unit within the CCM block clusters. The matching occurs in two phases: before follow-up and after follow-up.

Initial Housing Unit Computer Matching
– A CCM operation in which keyed data from the Independent Listing are matched against a preliminary census list of housing units and group quarters within each sample block cluster and one ring of surrounding blocks. Duplicate addresses within the Independent Listing and within the preliminary census list are also identified.

Initial Housing Unit Follow-up
– A CCM field operation that collects additional information about CCM or census units in order to resolve differences between the Independent Listing and the preliminary census file identified in the initial housing unit computer matching or before follow-up clerical matching operations. If necessary, it also collects information to resolve match, duplicate, or unit status.

Initial Housing Unit Operations
– A collection of CCM operations designed to produce the list of housing units eligible to receive the CCM Person Interview. The operations include initial housing unit computer matching, initial housing unit before follow-up clerical matching, initial housing unit follow-up, and initial housing unit after follow-up clerical matching.

In-mover
– A person who is an Interview Day resident at the sample address and a Census Day resident of a different address.

Insufficient information for DSE processing
– A census or CCM person record that does not have a complete name and at least two non-imputed demographic characteristics.

Insufficient information for follow-up
– A census or CCM person record that does not have a complete, valid name.

Integrated coverage measurement (ICM)
– A coverage measurement program proposed, but never implemented, for the 2000 Census. A major objective of the program was to produce a “one-number census,” which would have included an adjustment for the estimated census coverage error. Instead, the Accuracy and Coverage Evaluation survey (see A.C.E.) was implemented as the coverage measurement program for the 2000 Census.

Interview Day
– In the CCM, the day the coverage measurement Person Interview is conducted at a particular sample address.

In the census, the day the census enumeration is conducted at a particular address.

Interview Day residence status
– A determination of the residence status of a person on Interview Day. In particular, for a person on the CCM Person Interview roster at the sample address, the result may be resident, nonresident, or unclassified.

Link
– A pair of records identified as referring or possibly referring to the same person or housing unit.

Living quarters
– Any place where people live, stay, or could live. Living quarters are classified as housing units or group quarters.

Local Update of Census Addresses (LUCA)
– A decennial census program that provides an opportunity for local and tribal governments to review and update address and geographic information to improve the currency and completeness of the Master Address File.

Map spot
– The set of coordinates (latitude and longitude) assigned to the location of a structure.

In a paper data collection environment, it is represented by a dot drawn on a census block map by a lister or enumerator that shows the location of a structure containing one or more living quarters.

In an automated data collection environment, it is depicted on the map of the block shown on the screen of the data collection device.

Map spot number (MSN)
– A number assigned to a housing unit or structure, usually associated with a map spot, in order to facilitate field work. Map spots are not assigned to military housing units or structures, but for CCM these units are assigned map spot numbers to facilitate identification.

Master Address File (MAF)
– A database conceptually containing the address or location description of every building (residential or non-residential) known to the Census Bureau, along with geographic information.

Master Address File Identification Number (MAFID)
– A number associated with each record in the Master Address File. It is also called the Census Identification Number (CID).

Matching
– A comparison of records, either clerically or by computer, for an individual person or housing unit from the census and from the coverage measurement survey.

Matching error
– One of three types of error that can arise during matching. The types are
  1. False matches - treating as matched two records that do not correspond to the same person or housing unit;
  2. False nonmatches - a failure to treat as matched two records that do correspond to the same person or housing unit;
  3. Mismatches - cases in which a report in one source that corresponds to a record in the other source is correctly treated as matched but is paired with the wrong record in the other source.
Matching, Review and Coding System (MaRCS)
– A computer-assisted system developed to support clerical matching of persons and housing units. PerMaRCS refers to the system used for person matching; HUMaRCS and FHUMaRCS refer to the systems used for initial and final housing unit matching.

Match status
– For any pair of records under consideration, the different types of match status are match, nonmatch, or possible match.

Mover
– A person whose address on Interview Day differs from his or her address on Census Day.

Mover status
– The determination of a person’s status on CCM Interview Day. The status can be non-mover, in-mover, P-sample out-mover, non-P-sample out-mover, or unclassified mover.

Nationwide computer matching
– This refers to the computerized operation to identify links between persons rostered in the CCM Person Interview and census persons anywhere within the U.S. or within Puerto Rico. The CCM nationwide computer matching is also used to identify persons duplicated in the census through an E-sample-to-census match.

Net coverage error
– The difference between the estimated population (or subpopulation) total obtained through dual- system estimation or demographic analysis and the census count for that population (or subpopulation). A positive net error indicates a census undercount, while a negative net error indicates a census overcount.

Non-interview adjustment
– A statistical technique used to account for persons or housing units for which information should have been collected, but for some reason was not. The CCM performs this adjustment by modifying sample weights.

Non-mover
– A person who is an Interview Day resident and a Census Day resident of the same address.

Non-P-sample out-mover
– An out-mover who should not be included in the P sample, because on Interview Day, he or she resides in another residence eligible to be in the P sample.

Non-response bias
– A bias in survey estimates resulting from not obtaining completed interviews for the entire sample. The size of the bias depends on how much the nonrespondents differ from the respondents in terms of characteristics being measured by the survey.

Non-response Follow-up (NRFU)
– A decennial census field operation that attempts to visit and enumerate all housing units for which a census questionnaire has not been returned.

Occupancy imputation
– Imputation of “occupied” or “vacant” for census housing unit records with unknown occupancy status.

Omission
– A person or housing unit that should have been a census enumeration but was not.

Out of scope for CCM
– Persons who, on Census Day, either lived in a group quarters, were experiencing homelessness, or lived in another country. It also includes persons who were born after or died before Census Day and persons or housing units in remote Alaska.

Out-mover
– A person who is not an Interview Day resident at a particular sample address, but was a Census Day resident of that address.

Overcount
– Occurs when the census enumerates a larger number of persons or housing units in a population than indicated by an independent estimate of the size of that population.

P sample
– Refers to either the housing units in a geographically representative sample, or the persons that reside in those housing units, listed or rostered independently from the census by a post-enumeration survey.

P-sample out-mover
– An out-mover who should be included in the P sample because, on Interview Day, he or she resides in a residence not eligible to be in the P sample.

Person Follow-up (PFU)
– A CCM field operation designed to collect additional information about persons rostered during the PI or persons included in the final census file for which match, duplicate, or enumeration status has not been determined.

Person Interview (PI)
– A CCM field operation designed to collect basic demographic and residence information about all persons associated with an address as of Interview Day and persons who may have moved from that address between Census Day and Interview Day.

Person Clerical Matching
– A set of CCM clerical operations that reviews the results of computer matching and determines match, duplicate, and enumeration status for each CCM and census person within the CCM block clusters. The matching occurs in two phases: before follow-up and after follow-up.

Person Computer Matching
– A set of CCM operations that uses information collected in the CCM Person Interview (PI) and final census information to identify 1) duplicates and possible duplicates of persons in the PI, 2) duplicates and possible duplicates of persons in the census, and 3) matches and possible matches between persons in the PI and in the census.

PI duplicates and possible duplicates are identified by matching the PI to the PI only within the sample block clusters. Census duplicates and possible duplicates are identified by matching the CCM E sample to census enumerations within the sample block cluster search area and throughout the nation. PI-to- census matches and possible matches are identified by matching the PI to the census within the sample block cluster search area and throughout the nation.

Population Composition
– The description of the population defined by characteristics such as age, sex, and race.

Population Pyramid
– A graphical illustration of the composition of the population by age and sex. It normally takes the shape of a pyramid with two back-to-back graphs, one showing males and one showing females. The population is plotted on the X-axis and age on the Y-axis.

Population Size
– The number of individuals in a population (usually denoted P).

Possible match
– A type of link between records for an individual person or housing unit from the CCM and from the census. There is some degree of agreement between the records, but not enough to conclude with certainty that the records refer to the same person or housing unit.

Post-enumeration program (PEP)
– The term used for the coverage measurement program for the 1980 Census. It included both P and E samples. The source of the P sample was the Current Population Survey.

Post-enumeration survey (PES)
– A national survey that is operationally independent of the census, taken shortly after the census has concluded its data collection, for purposes of coverage evaluation. Also, refers to the coverage measurement program for the 1990 Census.

Post-stratification
– The use of covariates to define what are believed to be homogeneous subgroups of the population for which separate dual-system estimation computations are conducted.

Procedure A (PES-A)
– A procedure used in a post-enumeration survey to account for movers since Census Day. This procedure reconstructs the households as they existed on Census Day and matches to the local search area.

Procedure B (PES-B)
– A procedure used in a post-enumeration survey to account for movers since Census Day. This procedure identifies all current residents living or staying at the sample address on Interview Day and matches to the Census Day address.

Procedure B+ (PES-B+)
– A procedure used in a post-enumeration survey to account for movers since Census Day. It includes all the features of Procedure B. In addition, the respondent is asked about out-movers who may currently be living at a residence not eligible to be in the P sample.

Procedure C (PES-C)
– A procedure used in a post-enumeration survey to account for movers since Census Day. Procedure C combines features of Procedure A and Procedure B, but matches to the local search area.

Proxy interview
– An interview in which the respondent is not a member of the household being enumerated. The respondent might be a neighbor or some other knowledgeable person.

Recall bias
– The bias introduced if a respondent provides an inaccurate answer to a question due to his or her inability to remember the true answer.

Reference person
– The housing unit resident compared to whom other household members’ relationships are stated. This is usually the person in whose name the housing unit is owned or rented.

Reinterview
– A quality control operation in which a sample of respondents to the original interview are contacted a second time. During the reinterview, some or all of the original questions may be asked in order 1) to determine if the respondent had been contacted for an initial interview, or 2) to compare responses from the two interviews in order to measure response error.

Remote Alaska
– A type of census enumeration area that includes hard-to-access communities and fishing villages in Alaska. Persons and housing units in these areas are not covered by the 2010 CCM.

Residence status
– Identifies each person as either a resident or non-resident of a housing unit on Census Day or Interview Day.

Respondent
– The person who provides answers to an interview. For the CCM, this person is either a resident of the housing unit being enumerated or a proxy.

Search area
– Defines the geographic area that will be searched to call a record a match/nonmatch, a duplicate, or a correct/erroneous enumeration.

Sensitivity Analysis
– Strategies for investigating the robustness of the estimates to systematic changes to the input data. The outcomes reflect uncertainty in the estimates.

Sex Ratios
– The ratio of males to females in a population (usually defined as the number of males per 100 females or Pm/Pf * 100).

Status imputation
– In the census, refers to imputing a status (occupied, vacant, delete) for address records with conflicting or insufficient information about whether the address represents a valid, non-duplicated housing unit.

In the CCM, refers to imputing a match, correct enumeration, or residence probability for unresolved cases.

Surrounding block
– A census collection block that is geographically contiguous to at least one block in a CCM block cluster or that is contained completely within the contiguous blocks.

Synthetic estimation
– A statistical method for carrying estimates down to subpopulations.

Triple-system estimation
– A statistical method that uses the capture-recapture technique to estimate the true size of a population. This method is similar to dual-system estimation, but it features a second “recapture,” for which administrative records are often used.

Undercount
– Occurs when the census enumerates a smaller number of persons or housing units in a population than indicated by an independent estimate of the size of that population.

Unit status (Housing unit status)
– The determination of a housing unit’s status for census or CCM purposes. For CCM, the types of unit status are occupied, vacant and intended for occupancy, under construction, future construction, unfit for habitation, boarded up, empty trailer/lot site, structure used for storage of household goods, or structure used for another purpose.

Unresolved
– Indicates the inability to assign a final match status, enumeration status, duplicate status, or residence status (persons only) to a person or housing unit record.

Usual residence
– The place where a person lives or sleeps most of the time.
Source: U.S. Census Bureau | Decennial Statistical Studies Division | Last Revised: March 26, 2012