The following terms were selected from the UN
Glossary of Classification Terms approved by the Expert Group on
International Economic and Social Classifications of the United Nations.
A | B
| C | D | E | F
| G | H | I | J | K
| L | M | N
O | P
| Q | R | S | T | U
| V | W | X | Y | Z | Bibliography
A
Aggregation
Aggregation is the combination of related categories, usually within a
common branch of a hierarchy, to provide information at a broader level to
that at which detailed observations are taken.
B
Boundary
Represents the limit of a known or recognizable quantity, area or
scope. Each classification has its own boundary, as do its constituent
categories, such as activities, commodities, occupations etc. Whilst it is
possible for the boundaries of individual classifications to overlap,
there should be no overlap within individual classifications.
Building blocks (elementary items)
Are the most elementary items of a statistical classification, i.e. the
most detailed code for a variable. They may be used alone or in
combination to describe a category in one or more classifications, or to
compare classifications. A prime example is the Harmonized Commodity
Description and Coding System (HCDCS or HS), the categories of which are
used not only for the construction of country specific tariff and trade
classifications, but also as the building blocks of the Standard
International Trade Classification (SITC Rev. 3) and the goods component
of the Central Product Classification (CPC). The SITC and the CPC regroup
individual HS categories to meet differing statistical needs. Another
example is the General Industrial Classification of Economic Activities
within the European Community (NACE) which can be combined to reconstruct
higher levels of ISIC.
C
Category
Is the generic term for items at any level within a classification,
typically tabulation categories, sections, subsections, divisions,
subdivisions, groups, subgroups, classes and subclasses. Classification
categories are usually identified by codes (alphabetical or numerical)
which provide both a unique identifier for each category and denote their
place within the hierarchy. They contain elements which are subsets of the
classification to which they belong, such as activities, products, types
of occupations, types of education, etc.
Refer also to Item.
Class
Is a title/name used in classifications to depict a particular level
within a hierarchy (e.g. Section, Division, Group, Class). It usually
refers to the one of the lower levels of a classification, often the
lowest (e.g. in ISIC Rev. 3 the lowest level - 4 digit - is referred to as
the class, while in the CPC the class level is the second lowest level).
Its use is not mandatory.
Refer also to Division, Level.
Classification
Is a set of discrete, exhaustive and mutually exclusive observations
which can be assigned to one or more variables to be measured in the
collation and/or presentation of data. The terms 'classification' and
'nomenclature' are often used interchangeably, despite the definition of a
'nomenclature' being narrower than that of a 'classification'.
The structure of a classification can be either hierarchical or flat.
Hierarchical classifications range from the broadest level (e.g. division)
to the detailed level (e.g. class). Flat classifications (e.g. sex
classification) are not hierarchical.
The characteristics of a good classification are as follows:
- the categories are exhaustive and mutually exclusive (i.e. each
member of a population can only be allocated to one category without
duplication or omission);
- the classification is comparable to other related (national or
international) standard classifications.
- the categories are stable i.e. they are not changed too frequently,
or without proper review, justification and documentation;
- the categories are well described with a title in a standard format
and backed up by explanatory notes, coding indexes, coders and
correspondence tables to related classifications (including earlier
versions of the same classification);
- the categories are well balanced within the limits set by the
principles for the classification (i.e. not too many or too few
categories). This is usually established by applying significance criteria
(e.g. size limits on variables such as employment, turnover,
etc.)
- the categories reflect realities of the field (e.g. the society or
economy) to which they relate (e.g. in an industry classification, the
categories should reflect the total picture of industrial activities of
the country); and
- the classification is backed up by availability of instructions,
manuals, coding indexes, handbooks and training.
Refer also to Nomenclature.
Classification structure
Refers to how the categories of a classification are arranged, grouped
and sub-divided.
The categories of a classification can be arranged in either a
hierarchical or flat structure. In flat classifications the categories are
arranged at a single level. Hierarchical classifications have several
levels corresponding to different degrees of resolution (detail) in the
measurement (specification) of the variable being observed.
Classification unit
Is the basic unit to be classified in the classification (e.g. in an
activity classification this would be the establishment or enterprise, in
an occupational classification it will be the job).
Refer also to Observation
Unit.
Coding index
A detailed, comprehensive list of entries reflecting the information
required as the basis for consistent coding.
The code list (of descriptions and codes) is ordered either
alphabetically or numerically and could be stored electronically or in
hard copies. It acts as a link between responses and the classification,
enabling responses to be coded accurately and quickly to the appropriate
category of the classification. It may be necessary to compile different
coding indexes for coding different observations to the same
classification because of the differences in the information provided. The
content and structure of a coding index may also depend on the coding
methodology being used (i.e. whether coding is done manually, with
computer assistance or automatically).
Coding structure (coding system)
Refers to the systematic numbering/lettering of all the categories in a
classification. The use of standardized conventions when creating coding
structures helps in using and comparing classifications.
The following numbering systems could be used:
- a strict decimal numbering system - i.e. each category at a
particular level in a classification system would have the same type and
number of numeric characters;
- a non-strict decimal numbering system - i.e. at a particular level
each category in a classification system would not have the same number of
numeric characters. This would have been the case in ISCO-88 if a '0' had
not been used to indicate that a particular minor group has not been
further sub-divided into unit groups . In a mixed character system, e.g.
ISIC Rev. 3, where the tabulation categories are identified by letters,
the divisions, groups and classes are identified by two, three, and four
digit numbers respectively.
Concepts
Are abstract summaries, general notions, knowledge etc. of a whole set
of behaviors, attitudes or characteristics which are seen as having
something in common.
Concepts are used to assist in presenting/conveying precise meaning,
categorizing, interpreting, structuring and making sense of
classifications.
Concordance or correspondence Correlation
Refer to Correspondence table and
Concordance.
Content (of a classification category)
Refers to what is included within the boundary of a classification
category.
For example, in an industry classification, the content of a detailed
classification category (e.g. class) would include a descriptor or title,
a summary stating its scope, a list of inclusions and exclusions and a
list of primary activities.
Controlled vocabulary
Refers to the vocabulary to be used for specific classifications which
have specific meaning as given by the author or agreed by experts.
'Commodity', 'product' and 'service' would be examples of terms in
controlled vocabulary used in industry and product classifications.
Convention
A general rule, method, or practice established by consensus, usage, or
a formal agreement between parties. A convention can apply to both the
adoption of particular classifications and to the ways they are
interpreted and used.
Formal international Conventions are often signed to establish
organizations such as the Customs Co-operative Council (CCC). The
functions and purposes of the organization are included under the terms of
the Convention. They may also be designed to commit governments to certain
policies or legislation upon ratification, as do e.g. the ILO Labor
Conventions.
Correspondence table
Is a tool for the linking of classifications. A correspondence table
systematically explains where, and to what extent, the categories in one
classification may be found in other classifications, or in earlier
versions of the same classification. Methodologically, correspondence
tables (also referred to as tables) describe the way in which the value
sets of classifications are related, by describing how the units
classified to the groups defined for a classification would be classified
in other classifications.
Tables are important for the development and
harmonization of
international classifications. There are many different circumstances
under which one may want to establish relations between classifications,
and many forms which these relations may take.
Tables can be precise, depending on convention in order to describe the
type of link between tables (e.g. historical, hierarchical or whether they
overlap).
Source: United Nations Statistics Division, Fourth Meeting of the
Expert Group on International Economic and Social Classifications, New
York, 2-4 November (1998): A Statement of Best Practices.
Coverage
Specifies the population from which observations for a particular topic
can be drawn.
An understanding of coverage is required to facilitate the comparison
of data. Coverage issues are often explained through the use of tables
showing linkages (e.g. part or full correspondence); and can also be used
to explain the ratio of coverage.
The rules and conventions of coverage are largely determined by concept
definitions, scope rules, information requirements and, in the case of
statistical collections and classifications, collection and counting units
and the collection methodology.
Refer also to Scope.
Coverage ratio
Measures the extent to which observations designated as primary to a
particular category are undertaken by units primarily involved with the
observations related to that category. In industry statistics, the
coverage ratio is the output of goods and services characteristic of a
particular industry in proportion to the total output of the same goods
and services by the economy as a whole).
Cross reference
Is the linking, tracing or comparing of concepts/categories in one
classification or between classifications.
This could be done by specifying inclusions/exclusions, footnotes or
descriptors in an annotation. Cross referencing draws users' attention to
related concepts/categories, inclusions/exclusions etc. in the same or
other related classifications.
Refer also to Table.
Custodian of a classification
Refers to an institution or statistical area which has responsibility
for development, maintenance, implementation, promulgation and
interpretation of classifications. Collaboration among custodians is
essential for harmonization of classifications.
Classifications are often constructed by, or on behalf of, those
responsible for policy implementation. In such cases, the administrative
agency will normally be the custodian, sometimes in cooperation with the
statistical agency. For example, Customs agencies are often the custodians
of tariff classifications, even though such classifications are also used
for statistical purposes.
D
Definition
Is a statement of the precise meaning of something. In classifications
this refers to the explanation of the concepts encompassed in category
description and often includes specific examples of what is and is not
included in particular categories.
Refer also to Explanatory
note.
Derived classifications
Are based upon reference classifications. Derived classifications may
be prepared either by adopting the reference classification structure and
categories, providing additional detail beyond that provided by the
reference classification, or they may be prepared through rearrangement or
aggregation of items from one or more reference classifications. Derived
classifications are often tailored for use at the national or
multi-national level (e.g. NACE).
Refer also to Related
classifications.
Source: United Nations Statistics Division, Fourth Meeting of the
Expert Group on International Economic and Social Classifications, New
York 2-4 November (1998): Preamble: International Family of Economic and
Social Classifications
Description/descriptor
Is normally a one line statement/heading/index entry of a category in a
classification, designed to convey its content.
Disaggregation
Disaggregation is the breakdown of observations, usually within a
common branch of a hierarchy, to a more detailed level to that at which
detailed observations are taken. With standard hierarchical
classifications, statistics for related categories can be grouped or
collated (aggregated) to provide a broader picture, or categories can be
split (disaggregated) when finer details are required and made possible by
the codes given to the primary observations.
Division
Is a title/name used in classifications to depict a particular level
within a hierarchy (e.g. Section, Division, Group, Class). It usually
refers to one of the upper levels of a classification (e.g. in ISIC Rev. 3
the highest level - 1 digit - is referred to as the division, while in the
CPC the division level is the second highest level). Its use is not
mandatory.
Refer also to Class, Level.
E
Economic entity
Refers to a legal or social entity, or a group of entities, that
engage in economic activities and transactions in its/their own right,
such as corporations, non-profit institutions or government units. An
economic entity has legal, administrative, or fiduciary arrangements,
organizational structures or other parties having the capacity to
efficiently allocate resources in order to achieve objectives. Economic
entities are often used as a specific classification unit or a statistical
unit.
Source: SNA93 4.5.
Enterprise unit
An institutional unit or the smallest combination of institutional
units that encloses and directly or indirectly controls all necessary
functions to carry out its production activities.
Source: ISIC, Rev 3.
F
Family (of classifications)
The family of international economic and social classifications is
comprised of those classifications that have been internationally approved
as guidelines by the United Nations Statistical Commission or other
competent inter-governmental board on such matters as economics,
demographics, labor, health, education, social welfare, geography,
environment and tourism.
Source: United Nations Statistics Division, Fourth Meeting of the
Expert Group on International Economic and Social Classifications, New
York, 2-4 November (1998): Preamble: International Family of Economic and
Social Classifications.
Field
Generally refers to the area or sphere of operation, observation,
activity etc. In classifications, field could describe the scope of
individual classifications or their constituent groups, categories or
items at a given level.
Refer also to Scope.
Format(s)
Format(s) refers to the style or manner of arrangement or procedure
used for the presentation of the content and structure of a
classification, and may also refer to hard copy or electronic storage,
software used, and text font, size and attributes.
Framework
A framework is a multi dimensional classification system that seeks to
bring in a range of elements. A framework could include a combination of
classifications, code lists and/or data items modules, and generally
metadata.
Source: ABS, Standard Procedure for Creation of Alternate Industry
Views and Frameworks, 1999.
The term framework can also be used to describe the skeleton of
classification from which a detailed classification is developed. Such a
framework encompasses the concepts to be embedded in a classification
(e.g. product and activity) and provides the structure for the
classification.
At a broader level, the term framework may be used to describe a family
of related classifications, such as those produced by the UN.
G
Grouping/de-grouping
In a hierarchical or tree structure classification, categories are
grouped ranging from broad to detailed levels for each set. The categories
within each set can be grouped (aggregated) or de-grouped
(disaggregated).
For example, a multi level hierarchical classification would be
structured such that the sum of the detail of each level equates to the
level above. In this way, observations can be taken at the level of detail
of interest of particular purposes. Observations at the lower levels can
be summed to provide observations at more aggregated levels (grouped) and,
with appropriate manipulations, observations at higher levels can be
inferred at lower levels (de-grouped).
Guidelines
Refer to the directions or principles used in the development/building,
maintenance and application of classifications. Guidelines are not
necessarily mandatory, but are provided as an aid to interpretation and
use of classifications.
Refer also to Rules.
H
Harmonization
Classification harmonization involves the alignment, wherever possible,
of the underlying concepts and definitions of both similar and disparate
classifications to produce classifications which can related to the
maximum extent possible within the constraints of the requirements of
individual classifications.
Harmonization is the process of combining or comparing data for
purposes of analysis, either through the use of similar standard
definitions and classifications, or through a complex set of explanations
on how to achieve comparisons across standards and classifications
In the harmonization of classifications, building blocks for common
groupings and regroupings of items from different structures of the
classifications are identified. The process is facilitated by reducing or
eliminating minor differences among the classifications
Harmonization of classifications requires continuous co-ordination and
exchange of information between the custodians of the relevant
classifications on a regular basis. Without such exchange, different
interpretations of similar concepts and categories will occur.
In the harmonization process the classifications could be described as
reference, derived or related classifications.
Source: United Nations Statistics Division, Fourth Meeting of the
Expert Group on International Economic and Social Classifications, New
York, 2-4 November (1998): A Statement of Best Practices.
Heading
Generally this refers to the title or caption of a document or section
of a document. In the context of a classification it is the main
descriptor of a category, also referred to as Name or Title (e.g. the
heading under Division 01 in ISIC Rev. 3 is Agriculture, Hunting and
Related Service Activities).
Refer also to Name (of items)
and Title (of a category in a
classification).
Hierarchy
Refers to the classification structure where a classification is
arranged in levels of detail from the broadest to the most detailed level.
Each level of the classification is defined in terms of the categories at
the next lower level of the classification.
Homogeneity (homogeneous)
One of the characteristics of a good classification is reasonably high
homogeneity for its categories. Homogeneity is the measure of the degree
to which categories consist of components with similar characteristics and
is achieved by systematic grouping and stratifying members of the
population being classified.
Homogeneity ratios are defined on a mathematical basis to minimize the
variance within a classification.
I
Index
A listing, usually alphabetical, providing pointers to the location
within classifications of the observations contained therein. This may be
achieved by references to page numbers, paragraph numbers or
classification codes. Indexes often contain terminology not expressly used
within classifications (synonyms) and may contain cross-references to
related observations.
Refer also to Coding index.
Item
Refers to an article or a unit included in enumeration. In
classifications, the term item generally applies to a classification
category.
Refer also to Category.
K
Keyword
Is part of the controlled vocabulary often used to assist in
identification or retrieve information about a classification (e.g.
keywords used within a software search facility). The term 'Keywords'
reflects this function of the words.
L
Legal entities
Are entities created for purposes of production, mainly corporations
and non-profit institutions (NPIs), or government units, including social
security funds. They are capable of owning goods and assets, incurring
liabilities and engaging in economic activities and transactions with
other units in their own right.
Source: SNA93, 1.13.
Level
Level denotes the position within the hierarchy of a category or a
group of categories.
Refer also to Class, Division.
Life cycle
Refers to the creation, changes and death of a given classification. A
classification can be revised due to a number of factors e.g. changes in
industries, changes in international standard classifications, etc. Such
changes may include the aggregation of disaggregation of items, changes in
terminology, additions and/or deletions etc. These changes will result in
either a revised version of the existing classification (where changes are
essentially in the detail), or a replacement version (where the changes
are substantial, involving structural changes, etc).
The life span of a classification is dependent upon a number of
factors, most importantly the rate of change of the observations it
describes and time series (stability) requirements.
Limit
Refer to Boundary and Scope
(universe).
Linkages
Refers to mapping or linking one classification to another. That is
each individual group in one classification should be linked with the most
appropriate corresponding group(s) in the other. This allows for better
management of classifications in a coordinated way, and for the transfer
from using one classification to using the other.
The first step when establishing linkages should always be to give to
the most detailed groups of one classification the code of the most
detailed appropriate group in the other. This then allows, when needed,
the groups of one classification to be subsequently aggregated to most of
the relevant aggregated groups of the other.
Source: United Nations Statistics Division, Fourth Meeting of the
Expert Group on International Economic and Social Classifications, New
York, 2-4 November (1998): A Statement of Best Practices.
Local unit
Is defined as an enterprise, or part of an enterprise, which engages in
productive activity at or from one location. The definition has only one
dimension in that it does not refer to the kind of activity that is
carried out. Location may be interpreted according to the purpose-
narrowly, such as specific address, or more broadly, such as within
province, state, country, etc. Local units are also used as Statistical
Units.
Source: SNA93, 5.20.
M
Maintenance
Refers to an institution or statistical area which has or has been
given the responsibility for maintaining and/or updating or revising the
classification.
Source: United Nations Statistics Division, Fourth Meeting of the
Expert Group on International Economic and Social Classifications, New
York, 2-4 November (1998): Preamble: International Family of Economic and
Social Classifications.
Refer also to Custodian.
N
Name (of items)
Refers to the descriptions of classification categories, subcategories
or their elements e.g. primary activities.
Refer also to Title (of a category in a classification)
and Heading.
Network
Applies to a chain of interconnected persons, things, operations etc.
In classifications, networking could result in reference, derived or
related classifications.
Exchange of information and knowledge across classifications would be
facilitated and implemented if national classifications could be presented
as part of the web sites of statistical offices and an international cyber
platform on the Internet is used for all the major international
classification debates.
Source: United Nations Statistics Division, Fourth Meeting of the
Expert Group on International Economic and Social Classifications, New
York, 2-4 November (1998): A Statement of Best Practices.
Nomenclature
Systematic naming of things or a system of names or terms for
things.
In classifications, nomenclature involves a systematic naming of
categories or items.
The terms "nomenclature" and "classification" are often used
interchangeably, despite the definition of a "classification" being
broader than that of a "nomenclature". A nomenclature is essentially a
convention for describing observations, whereas a classification
structures and codifies the observations as well.
Refer also to Classification.
Normalized heading and codes
This refers to standardization of headings/titles and their codes in
the classifications. For example, the titles and use of n.e.c. (not
elsewhere classified) categories and the codes given to them, as well as
the code ending in zero should normally be standardized, e.g. by using the
title of the aggregate group when naming its n.e.c. group, and by using 9
as the last digit for n.e.c. categories and 0 to be equivalent with coding
to the higher level, e.g. because the information needed for a more
detailed code is not available.
Not elsewhere classified (n.e.c.), not elsewhere included (n.e.i.)
or not elsewhere specified (n.e.s.) residual category
Applies to a subset of a category (e.g. class, group, etc.) which
represents those members of the category that do not belong to any of the
other, separately identified, categories. (Note that it should not be used
as a 'dump' code for observations for which there is insufficient
information to assign a detailed code. Such units should be given a code
ending in zero to indicate the appropriate aggregate group.) The
significance of the observations (e.g. income, employment etc.) for this
category should be relatively low compared to those of the other
categories in the same more aggregated group of the hierarchical set.
O
Overlapping
Is defined as to partly cover, cover or extend beyond two or more
classification concepts - i.e. not to be completely separate (e.g. "the
group xy in classification N is partly 'overlapping' with category yx in
classification N, rev. 1").
P
Partial correlation/correspondence
Occurs where a category of one classification can be coded to two or
more categories of other classifications.
Refer also to Correspondence Table and
Concordance.
Partition
Refer to Subset.
Population
Is the total membership or population or 'universe' of a defined class
of people, objects, or events.
There are two types of population viz. target population and survey
population. A target population is the population outlined in the survey
objects about which information is to be sought and a survey population is
the population from which information can be obtained in the survey. The
target population is also known as the scope of the survey and the survey
population is also known as the coverage of the survey. For administrative
records the corresponding populations are: the 'target population' as
defined by the relevant legislation and regulations, and the actual
'client population'.
Preamble
An introductory statement to a classification, outlining the intent,
purpose, scope and coverage of a classification.
Primary unit
Is a unit which always can take one and only one value for the variable
for which the classification represents a value set. (e.g. the primary
unit to be classified by 'occupation' is the 'job'. To classify a 'person'
by 'occupation' one first needs to establish a link between the person and
an appropriate 'job').
Source: United Nations Statistics Division, Fourth Meeting of the
Expert Group on International Economic and Social Classifications, New
York, 2-4 November (1998): A Statement of Best Practices
Process
A systematic series of actions directed to some end. It is a usually a
set of continuous actions or operations undertaken in a defined manner.
Process, when applied to classification development and maintenance, is
the means by which the concepts and methodology supporting and underlying
classifications are incorporated within classifications, thereby promoting
classification best practice.
R
Reference classifications
Are those economic and social classifications that are a product of
international agreements approved by the United Nations Statistical
Commission or another competent inter-government board, such as that of
the International Labor Organization (ILO), the International Monetary
Fund (IMF), the United Nations Educational, Scientific and Cultural Organization
(UNESCO), World Health Organization (WHO), or the World
Customs Organization (WCO) depending upon the subject matter area. Thus
reference classifications have achieved broad acceptance and official
agreement and are approved and recommended as guidelines for the
preparation of classifications. They may be used as models for the
development or revision of other classifications, both with respect to the
structure and with respect to the character and definition of the
categories.
Source: United Nations Statistics Division, Fourth Meeting of the
Expert Group on International Economic and Social Classifications, Fourth
Meeting of the Expert Group on International and Social Classifications,
New York, 2-4 November (1998): Preamble: International Family of Economic
and Social Classifications.
Related categories
Are those categories which have some form of elementary relationship.
Such related categories can be meaningfully aggregated to give a broad
picture or disaggregated when finer details are required. Related
categories often have commonality in their codes, due to their common
starting point (e.g. an international standard), although relationships
can be applied between classifications with different structures and
coding systems provided the concepts embedded within the observations
under consideration are consistent.
Related classifications
Are classifications which encompass the same or similar observations
within different structures and/or to different levels of detail. They
often occur as part of a family of classifications, sometimes with a
common starting point, such as an international standard
classification.
Refer also to Derived classification.
Revised classification
Refers to a classification which replaces the previous classification.
A change in a classification does not necessary result in a change of the
name of the classification but can be distinguished by use of a version
number (e.g. ISIC Rev. 2 was replaced by ISIC Rev. 3).
A revised classification will normally represent a rethinking of
conceptual basis, similarity criteria and/or scope, and should be
distinguished from an up-dated classification.
Rules
Are statements, decisions, judgments, or precedents which provide
operational guidelines on the implementation, use, updating and revision
of classifications.
S
Scope (universe)
The scope of a classification is the coverage or sphere of what is to
be observed. It is the total membership or population of a defined set of
people, objects, or events.
Refer also to Coverage and
Population.
Segmentation
Relates to the splitting/separation of topics. Once the scope/coverage
of a classification has been defined (e.g. age, language, industry),
breakdowns (e.g. beginning from the top) are made, based on similarity
criteria, to form groups of elementary blocks or topics and different
categories or levels.
Refer also to Aggregation, Disaggregation
and Coverage.
Similarity criteria
Refers to the criteria used to define categories in hierarchical
classifications (e.g. the grouping of elementary building blocks).
In ISCO-88 the main similarity criteria are the skill level and skill specialization
needed to carry out the tasks and duties of the jobs. Skill
level is the main criterion to delineate the most aggregate categories,
while skill specialization is used to delineate the more detailed
categories within the aggregate categories.
Source: ISCO-88.
Specialization ratio
Aids in the assessment of the homogeneity of categories within a
classification. Specialization ratios measure the extent to which
observations contained within a category are representative of the
population of those observations as a whole (e.g. in industry statistics,
the specialization ratio is the output by an industry of goods and
services characteristic to that industry in proportion to its total
output).
Source: ISIC Rev. 3, page 35, paragraph 155.
Refer also to Coverage Ratio,
Homogeneity.
Standard classifications
Are those that follow prescribed rules and are generally recommended
and accepted. They aim to ensure that information is classified
consistently regardless of the collection, source, point of time etc.
Statistical classification
Refers to a classification constructed for the collection and
presentation of numerical facts systematically collected (i.e.
statistics). The usefulness of a statistical classification is enhanced if
based on or representing a standard classification.
Structure (tree)
Provides the means for identifying relationships, usually hierarchical,
between categories. A hierarchical classification is based on a tree
structure where each set of its detailed categories are subsets of
categories at the level about the one in which they contained.
Refer also to Hierarchy.
Structural Links
Are correspondence links where opportunities for (direct)
correspondence between the categories of different classifications are
difficult or not possible to establish, owing to significant structural
differences in the defined value sets that do not allow for common
correspondence at a similar hierarchical level in the structure. In some
circumstances, an approximate or truncated correspondence may be made by
aggregating subclasses of one classification to different structural
levels of the other classification.
Source: United Nations Statistics Division, Fourth Meeting of the
Expert Group on International Economic and Social Classifications, New
York, 2-4 November (1998): A Statement of Best Practices.
Subdivision
Refers to a breakdown of a division in a hierarchical classification.
Two or more subdivisions form a division.
Subset
Refers to a secondary or subordinate set whose elements belong to a
larger given set. In classifications a group of categories which are
defined for a given purpose and can be defined at a given level. In a
hierarchical classification, for example, a subset could be those detailed
categories below the broad category of the set.
Refer also to Partition.
T
Table
Statistically this refers to a list of numbers systematically arranged
in columns and/or rows. In classification this usually refers to a
correspondence/concordance table or a code list.
Term
Refers to the specific meaning of a word used to define/express a
concept. The definitions of terms for a particular classification should
be precise, accompanied by explanatory definitions which will make them
readily understood by non-experts, and agreed upon by the classification
experts and main users of the resulting statistics.
Terminology (classification)
Refers to the system of terms commonly used or adapted for use in a
classification. Wording or terminology, which may have broader meanings
within the wider community, may have specific meaning within the context
of given classifications. For example, 'industry' and 'homogeneity ratios'
have unique definitions in the context of industry classifications.
Title (of a category in a classification)
Refers to a word or a limited set of words chosen to reflect the
content of the category and to differentiate between categories.
In a hierarchical classification the titles of aggregate groups should
always be designed to include all the component subgroups, and nothing
else. The title of a 'not elsewhere classified' group should reproduce the
title of the broader group to which it belongs with n.e.c. added.
Refer also to Heading and
Title (of a category in a classification).
Topic
Refers to the specific meaning of a word used to define/express a
concept. The definitions of terms for a particular classification should
be precise, accompanied by explanatory definitions which will make them
readily understood by non-experts, and agreed upon by the classification
experts and main users of the resulting statistics.
U
Units (classified)
Refer to entities, respondents to a survey or things used for purpose
of calculation or measurement. Their statistics are collected, tabulated
and published. They include, among others, businesses, government
institutions, individual organizations, institutions, persons, groups,
geographical areas and events. They form the population from which data
can be collected or upon which observations can be made.
V
Variable
Is a characteristic of a unit being observed that may assume more than
one of a set of values to which a numerical measure or a category from a
classification can be assigned (e.g. income, age, weight, etc. and
'occupation', 'industry', 'disease' etc).
Validity period
Refers to the time period that any document, classification, etc. could
be applicable or used. In classification, usually there is an overlapping
time period where the old classification could still be used before being
superseded by the revised edition.
Bibliography
A Statement of Best Practices, United Nations, New York (1998):
Fourth Meeting of the Expert Group on International Economic and Social
Classifications, 2-4 November (1998).
Balance of Payments Manual Fifth Edition (BPM5), International
Monetary Fund (1993).
International Standard Industrial Classification of All Economic
Activities (ISIC), Rev. 3, Adopted by the twenty-fifth session of the
United Nations Statistical Commission, New York, 6-15 February,
1989.
International Standard Classification of Occupation, ISCO-88,
International Labour Office, Geneva (1990).
Preamble: International Family of Economic and Social
Classifications. United Nations Statistics Division, Fourth Meeting of
the Expert Group on International Economic and Social Classifications, New
York, 2-4 November (1998).
System of National Accounts, Inter-Secretariat Working Group on
National Accounts (1993).
System of National Accounts Glossary, OECD
(1999)
|
