EPUNet Training Course 2005 Day 2 Tutors: Olaf J

1
EPUNet Training Course 2005Day 2Tutors Olaf
Jürgens and Christian SchmittBerlin, April 11th
to April 15th 2005
2
Overview

• Cross-Sectional Matching
• Transcription Routines (Raw Data into
  Statistics-Package-Format)
• Exploring Data-Sets

EPUNet 2005 Training Course
3
ECHP Overview - Data Sets

• Contents of the ECHP UDB
• Personal File
• Household File
• Register File
• Relational File
• Above One file per wave
• Country File
• Link File

4
ECHP Overview - Data Sets

• Contents of the ECHP UDB
• Personal File
• All person related information in the ECHP
• Only adult respondents
• Similar variable structure across countries
• Similar variable structure across waves

5
ECHP Overview - Data Sets

• Contents of the ECHP UDB
• Household File
• Unit of analysis Household
• General information that is applicable for all
  household members
• Similar variable structure across countries
• Similar variable structure across waves

6
ECHP Overview - Data Sets

• Contents of the ECHP UDB
• Register File
• Unit of analysis persons
• General information
• Regardless of age or participation
• Similar variable structure across countries
• Similar variable structure across waves

7
ECHP Overview - Data Sets

• Contents of the ECHP UDB
• Relational File
• Unit of analysis persons (caution repeated
  observations of persons!)
• Display of the relation-matrix of persons within
  a household
• Kinship, parent-child relationships, sibblings,
  etc.
• Similar variable structure across countries
• Similar variable structure across waves

8
ECHP Overview - Data Sets

• Contents of the ECHP UDB
• Country File
• Unit of analysis country
• Display of general country specific information
  (PPP, Exchange rates, Population)
• Single file
• One set of variables per wave

9
ECHP Overview - Data Sets

• Contents of the ECHP UDB
• Link File
• Heart of the ECHP
• Unit of analysis persons
• Regardless of age or participation
• General information for cross-sectional and
  longitudinal matches
• Household membership in a given wave
• Sampling information
• Weighting information
• Single file

10
Cross-Sectional Matching Procedures

• Cross Sectional Matches
• Person-level Matching
• Household-level Matching
• Relational Matching
• Central identifiers for all of the above
• country
• pid/hid
• Base for all matches ECHP Link File

EPUNet 2005 Training Course
11
Cross-Sectional Matching Procedures

• Matching Logical order
• First country variable (country)
• Second household identifier (HID)
• Third personal identifier (PID)
• Always use this logical order!
• sort by country hid pid
• (hid may be left out if no household based
  information is included in the data generation)
• Base for all matches ECHP Link File

EPUNet 2005 Training Course
12
Cross-Sectional Matching Procedures

• Person-level matching I - Same Individual
• country
• pid
• Examples of matches
• Matching information of one Person across files
• Using personal information from the Register File
  and the Personal File
• Adding information from the Personal File to the
  Link File

13
Cross-Sectional Matching Procedures

• Person-level matching - Across Individuals
• Examples of matches
• Matching information of a child to the mother -
  Unit of analysis Mother additional child
  related information
• Matching information of a husband to his wife -
  Unit of analysis Wife additional information of
  the husbands income, e.g.
• Central information stored within the
  Relation-File

14
Cross-Sectional Matching Procedures

• Identifiers within Files
• Unit of analysis Basic info / File
  structure

Personal File (Register File) Country PID (HID)
Person Level Information
Link File Country HIDwaveN PID Linking across
waves
Country File Country General country specific
info
Relationship File Country PID (HID) Linking
across individuals
Household File Country HID Household Level
Information
15
Lab Session Day 2

• Transcription Routines
• Transformation from PDB to UDB
• Exploring Data Sets
• Cross-sectional matching procedures

16
Transcription Routines (Raw Data into
Statistics-Package-Format)

• From PDB to UDB
• Raw ECHP data comes in comma separated ASCII
  format.
• Raw ECHP data comes without any labels!
• Transcription Routines for SPSS
• In SPSS open syntax file

EPUNet 2005 Training Course
17
Transcription Routines (Raw Data into
Statistics-Package-Format)

• From PDB to UDB
• Raw ECHP ASCII format without
• Transcription Routines for SPSS
• In SPSS open syntax file
• UDB_readin.SPS
• UDB_label.SPS
• Adjust pathnames to fit your file structure
• Run

EPUNet 2005 Training Course
18
Transcription Routines (Raw Data into
Statistics-Package-Format)

• From PDB to UDB - Result
• ECHP UDB Files in SPSS .sav - format
• Link File (1 file ulink)
• Personal File (pfilen wave 1 to 8)
• Household File (hfilen wave 1 to 8)
• Register File (rfilen wave 1 to 8)
• Relationship File (relatn wave 1 to 8)
• Country File (1 file ctryvars)

EPUNet 2005 Training Course
19
Exploring Data Sets - See Doc-Pan 166

• Personal File
• Open Pfile
• Get file X\path1\a_w8p.sav.
• Descriptives variables PE001.
• For self defined employment status
• Continue with a_w7p.sav, a_w8h.sav,
  ulink.sav, etc.

EPUNet 2005 Training Course
20
Exploring Data Sets

• Personal File - contents
• Demographic information
• Employment and activity
• Calendar of activities
• Income
• Educational attainment
• Current education and training
• Health/Care
• Migration
• Satisfaction

EPUNet 2005 Training Course
21
Exploring Data Sets

• Household File - contents
• Demographic information
• Household income
• Household related benefits
• Accommodation and housing situation
• Durables
• Persons in household.

EPUNet 2005 Training Course
22
Exploring Data Sets

• Register File - contents
• Panel specific information (personal identifier
  PID, household identifier, weights, etc.
• Demographic information (age, sex, etc.)

EPUNet 2005 Training Course
23
Exploring Data Sets

• Relationship File - contents
• Always lists two persons per case!
• Central relation between person one and person
  two (pid1 relation pid2)

EPUNet 2005 Training Course
24
Exploring Data Sets

• Country File - contents
• One record for each country/panel
• One block of variables for each wave
• RATE Exchange rates in Euro
• PPP Purchasing power parities
• POPTOT Total population in private
• households
• POP16P Number of persons aged 16 living
• in private households
• POPHHD Number of private households

EPUNet 2005 Training Course
25
Exploring Data Sets

• Link File - contents
• General structural information for linking
  households and individuals within and across
  waves
• All panel household members (regardless of age or
  panel participation)
• Basic demographic information
• Cross sectional and longitudinal weights
• Sample status
• Wave specific household identifiers

EPUNet 2005 Training Course
26
Exploring Data Sets

• General structure
• Identical naming of variables across waves
• First letter of variable describes file (P for
  Personal File)
• Second letter of variable describes module (PM
  for Personal File, module migration)
• Subsequent numbers describe exact information
• (PM001 for Personal File, module migration 001
  for migration trajectory)

EPUNet 2005 Training Course
27
Exploring Data Sets

• Getting information
• Central tool Codebook (Doc-Pan 166) containing
• List and description of all ECHP UDB variables
  and
• information on availability and comparability
  of variables
• across countries and
• across waves

EPUNet 2005 Training Course
28
Files to use

• Personal File
• Household File
• Register File
• Relationship File
• Country File
• Link File
• UDB_readin.sps
• UDB_label.sps