Isabel Morgado isabel.morgadoine.pt

1

Documenting Variables

• Isabel Morgado (isabel.morgado_at_ine.pt)
• Monica Isfan (monica.isfan_at_ine.pt)
• National Statistical Institute Portugal

24-26. April. 2006

2
Documenting Variables

• General Guidelines
• Why a Variables System?
• Defining Variables
• Management
• Problems and Solutions
• (Naming Convention)
• Benefits
• References

3
General Guidelines

• ISO/ IEC 11179
• IMBD (Integrated Meta Data Base)
• (Statistics Canada)

back

4
Why a Variables System?

• Central Variables Repository
• Statistical Survey Design
• Automatic Questionnaire Generations
• Statistical Dissemination
• Facilitate Standardization
• Identify Duplicates
• Facilitate Data Sharing

5
Why a Variables System?


6
Why a Variables System? (Implications)

• Information on variables must be consistently
  structured
• Variables must be meaningfully named (to be
  listed in web pages, questionnaires, etc.)
• Variables must be effectively searchable
  (consistent named)
• Etc.

back

7
Defining Variables (ISO/ IEC 11179)

Property
Object Class
Conceptual Variable
Representation Class
Value Domain
Variable

8
Defining Variables (ISO/ IEC 11179)

Property
A characteristic or attribute common to all
members of an object class (ISO/ IEC
11179). Example Age group, Geography.
back

9
Defining Variables (ISO/ IEC 11179)

Object Class
A set of ideas, abstractions, or things in the
real world that can be identified with explicit
boundaries and meaning whose properties and
behaviour follow the same rule (ISO/ IEC 11179).

• Fundamental and Derivate Statistical Units
• Example Establishment, Person, Groom, etc.
• Macro Statistical Units or Populations
• Example Employees, Urban population, etc.

back

10
Defining Variables (ISO/ IEC 11179)

Conceptual Variable
A concept that can be represented in the form of
a data element, described independently of any
particular representation (ISO/ IEC
11179). Example Age group of person
back

11
Defining Variables (ISO/ IEC 11179)

Representation Class
The representation describes how the data are
represented (ISO/ IEC 11179). Example Name,
Type, Quantity, Ratio, etc.
back

12
Defining Variables (ISO/ IEC 11179)

Value Domain
Set of permissible values and their associated
meanings (ISO/ IEC 11179). Example Age group
table 0,...,120 etc.
back

13
Defining Variables (ISO/ IEC 11179)

Variable
A unit of data for which the definition,
identification, representation and permissible
values are specified by means of a set of
attributes (ISO/ IEC 11179).

14
Defining Variables (ISO/ IEC 11179)

Object Class
Property
Person
Age group
Conceptual Variable
Age group_Person
Representation Class
Value Domain
Code
Age group Table
Variable
Age group_Person_Code_(Age group Table)
Formal Name of Variable
back

15
Management

• Variables System
• Decentralized Management
• Survey managers use/ propose variables
• Centralized Coordination
• Metadata Unit accepts/ rejects the proposals

back

16
Problems and Solutions

• Production System
• Survey
• Methodological Document
• Protocols with Public Organizations
• (Administrative Sources)

• Dissemination System
• Official Statistics Portal
• Dissemination Data Base
• All Indicators registered previously in the
  Variables System

Problems

17
Problems and Solutions

• Formal name not user friendly
• Formal name very long
• Variables must supply both production systems and
  dissemination systems
• Variables effectively searchable

Solutions

18
Problems and Solutions

• External Name
• General Rule Property (Qualifier term)
  Object Class
• Abbreviate name
• General Rule Property (Qualifier term)

Qualifier term
A word or words which help define and
differentiate a name within the database

19
Problems and Solutions

• Rules (examples)
• Representation Class Count / Estimation
  Qualifier Count/ Estimation
• Example
• Resident Population (Estimation)
• Representation Class Quantity Qualifier
  Symbol of Unit of Measure
• Example
• Average income () of person (External Name)
• Average income ()(Abbreviate Name)

20
Problems and Solutions
21
Problems and Solutions

back
22
Problems and Solutions
back
go
23
Problems and Solutions
back
24
Problems and Solutions

back
25
Problems and Solutions

Official Statistics Portal

• Step 1 Search
• Step 2 List of aggregate variables (Name of
  conceptual variable)
• Example
• Unemployed population
• Unemployment Rate
• Resident population

Name of Conceptual Variable

26
Problems and Solutions

Official Statistics Portal

• Step 3 List of aggregate variables by dimensions,
  periodicity and source
• Example
• Unemployed population (Estimation) by age group
  and sex
• Unemployed population (Estimation) by age group
  and economic
• 
  
  activity
• Step 4 Select

External Name
Abbreviate name
back

27
Benefits

• Increased chances of sharing data and metadata
  with other agencies
• Improved understandability of data and survey
• Single point of reference for data harmonization
• Central reference for survey re-engineering and
  re-design
• Reduce ongoing production costs
• Reduce statistical burdens

back

28
References

• Bo Sundgren, The aß?t-model A Theory of
  Multidimensional Structure of Statistics, 2-4
  April 2001.
• Bo Sundgren, Objects and their Classifications,
  Relations, and Life Histories as Reflected by
  Official Statistics, 2004.
• Bo Sundgren, The Contents of a Statistical System
  as a Whole, 2004.
• Bruce E. Bargmeyer, Daniel W. Gillman, Metadata
  Standards and Metadata Registries An Overview.
• K. Mechanda, Recommendation for Accessing
  Definitions on the Internet.
• Isabel Morgado, Metadata and Survey Documentation
  Portuguese NSI Experience, 2004.
• ISO/ IEC 11179, Information Technology
  Specification and Standardization of Data
  Element, 1999.
• P. Johanis, B. Brooks, T. Dunstan, J. S.
  Lévesque, Statistics Canadas Implementation of
  the Data Element Model, 2003, Santa Fe, USA.

29
(No Transcript)
30
http//www.isi2007.com.pt isi2007_at_ine.pt
31
back