Generic Statistical Information Model (GSIM)

GSIM is a reference framework of internationally agreed

definitions, attributes and relationships that describe the
pieces of information that are used in the production of
official statistics (information objects).

This framework enables generic descriptions of the

definition, management and use of data and metadata
throughout the statistical production process.
 BUSINESS The„„GSIM covers the whole statistical process. It includes information objects used in
the Specify Needs phase (such as Statistical Need and Business Case) right through to
information objects used in the Evaluation phase (such as Evaluation and
Assessment).
PRODUCTION
may change
Statistical applies
Statistical program design has Process Step Process has
Program Rule
Need Design Method
Design

ma
y

has
init
iat
e
„A Statistical Program is the overarching, ongoing activity that an The GSIM was designed to enable an explicit separation between the design and
organization undertakes to produce statistics. A Statistical Program has execution of statistical processes. The Production Group assumes that each process step
„The model identifies different types of activities will be designed during a design phase. During the production phase, Process Steps are
Statistical an associated set of Statistical Program Designs that identify the

has
that represent the major steps in the statistical “executed”„(performed) in accordance with their design. For each Process Step that is
Program methodology used for the Statistical Program.
production process . These types have been executed, there will be a Process Step Execution Record.
specifically identified in the model but other types
is c
could be defined. om
pri
sed
of

Acquisition, Process Step

Production, has subtype Statistical includes has
Process Step Execution
Dissemination Activity
Activity Record
The execution of a Process Step will supply Process
Inputs and result in Process Outputs. In many cases,

records
these Process Inputs and Outputs are the conceptual and
„The model supports collecting data via surveys or registers. A Data Channel
uses

structural information objects that are described in the

identifies the Instrument used to collect data and the Data Resource from which
GSIM Concepts and Structures groups.
the data is collected. An Instrument is the description of the tool that will be used
to collect data (for example a survey instrument). Process Input,
Process
Data Channel Output
The GSIM supports both static and dynamic dissemination outputs. This includes
Products (for example publications, press releases, etc) and Representations
(which may contain data, structural metadata or conceptual metadata).
rom
ects f
bj
m eo p
cts fro lud grou
in clu de obje c
y in this
ma y oup ma
this gr

STRUCTURES
The GSIM model includes rich detail on
classifications. Classification management is

CONCEPTS
uses

important to statistical agencies who have Classification

complicated representations of concepts
within their statistical information systems

The GSIM makes a distinction between unit and dimensional data. There are a
number of data structure components defined by the model. These include
identifier, measure and attribute components.
Instance uses Represented uses
Code List

represents
Variable Variable

has
Data The model differentiates between the objects used to
Data
Structure conceptually describe information, and those that are
Structure
s
ibe

Components

re
representational. For example, a Variable does not contain any

pr
uses
scr

es
information on how the resulting value may be represented. This
de

en
ts
information is put into the Represented Variable. This enables the
Variable definition to be reused
has

includes observation of
Data contains comprises measures defines groups Concept
Data Set Data Point Unit Population Variable Concept
Resource System
Variable relationship