TwinLife Documentation TwinLife Documentation TwinLife Documentation
  • Overview and getting started
    • Table of content
    • Getting started
    • Data access & documentation
  • 1. About TwinLife
    • 1.1 Basic Concept
    • 1.2 Study Design and Sample Structure
    • 1.3 Where and how to get the Data
  • 2. Documentation of the study
    • 2.1 Data Documentation Website and ShortGuide
    • 2.2 Documentation within the Data Sets
    • 2.3 paneldata.org
    • 2.4 Codebooks
    • 2.5 Technical Report Series, Methodology Reports, and Working Paper Series
  • 3. Data Structure
    • 3.1 Data Formats and Data Files
    • 3.2 Person Types
    • 3.3 System of Variable Names
    • 3.4 ID Variables, Wave and Data Collection Identifiers
    • 3.5 Missing Types and their Meanings
    • 3.6 Delivered Para Data
    • 3.7 Weights
    • 3.8 Pecularities of Data
    • 3.9 How to match the Data Files
    • 3.10 Matching information from the parent-about-child questionnaire to the child's data set
  • 4. Check Routines
    • 4.1 Check routines
    • 4.2 Data Adjustment
  • 5. Generated Variables and Scales
    • 5.1 Generated Variables
    • 5.2 Generated Scales
  • 6. Publications and Citation
    • 6.1 Publications and Literature Database
    • 6.2 Citation
  • 7. Useful Links
  • 8. Workshops
  • Terms and Privacy
  • Downloads
  • Start
  • Overview and getting started
    • Table of content
    • Getting started
    • Data access & documentation
  • 1. About TwinLife
    • 1.1 Basic Concept
    • 1.2 Study Design and Sample Structure
    • 1.3 Where and how to get the Data
  • 2. Documentation of the study
    • 2.1 Data Documentation Website and ShortGuide
    • 2.2 Documentation within the Data Sets
    • 2.3 paneldata.org
    • 2.4 Codebooks
    • 2.5 Technical Report Series, Methodology Reports, and Working Paper Series
  • 3. Data Structure
    • 3.1 Data Formats and Data Files
    • 3.2 Person Types
    • 3.3 System of Variable Names
    • 3.4 ID Variables, Wave and Data Collection Identifiers
    • 3.5 Missing Types and their Meanings
    • 3.6 Delivered Para Data
    • 3.7 Weights
    • 3.8 Pecularities of Data
    • 3.9 How to match the Data Files
    • 3.10 Matching information from the parent-about-child questionnaire to the child's data set
  • 4. Check Routines
    • 4.1 Check routines
    • 4.2 Data Adjustment
  • 5. Generated Variables and Scales
    • 5.1 Generated Variables
    • 5.2 Generated Scales
  • 6. Publications and Citation
    • 6.1 Publications and Literature Database
    • 6.2 Citation
  • 7. Useful Links
  • 8. Workshops

3.3 System of Variable Names

  • Print
  • Email
  • Generally, the variable name is a composite of the variable stem (three letters), the item block (two digits), and the item number (two digits). Furthermore, depending on the person- or family-format of the data sets, person code letters in variable names indicate that proxy information is given from and/or about a certain family member (see chapter 3.2 for person codes).

    Examples:

    1.    What information is in the variable name edu0100 in the ‘person format’ (‘long’ dataset)?

    → edu is the variable stem (type of education - self report), 0100 is ‘school attendance’
    within the facet ‘type of education’.

     

    2.    What information is in the variable name pas0100m in the ‘person format’ (‘long’ dataset)?
     
    → pas is the variable stem (parental style - child report), the four digits 0100 are the
    item number within the construct. In most cases , for s cale variables the first two
    digits are the item block within the construct (01, parental style) and the last two
    digits are the item number (00, shows affection); the variable extension m is the
    person code (statement about the mother). Thus, this item as sesses the child’s rating
    of parental style, more precisely on how much affection the parent, here the mother,
    shows.
     
    3.    What further information is in the variable names in the ‘family format’ (‘wide’ dataset), e.g. pas0100m_s1?
     
    → _s is the person code of the person who provides the information (sibling), _1 is the
    data collection code (first data collection ). Thus, this item assesses the child’s rating of
    parental style, more precisely on how much affection the parent (in this case mother)
    shows - rated by the sibling in the first data collection.
     
    In the family format, two particularities have to be mentioned:
    First, information on the households get the code _1 to _x (x= maximum number of households within families) instead of a person code.
    • For example, ptyp_hq_1_1 provides information on the type of respondent for the household questionnaire in data collection one for the first household which belongs to the family.
    • The variable pih_$_$ shows the household number of a person (e. g. pih_t_1 is the household number of the first-born twin in data collection 1).
    Second, information given by a parent about their children is not filed under the type of person of the interviewed parent. Instead, the letter suffix reflects for which person information was given.
    • For example, the variable name eca0300t_1 shows that the statement was made for the first-born twin, but provides no information about the parent who filled in the questionnaire.
    • Instead, the variable ptyp_cp_1 indicates which parent completed the questionnaire about their children in data collection 1.
     
    Variable labels contain a short description of the variable content and information about filter conditions (e.g., which person types and which age groups have answered the question). Generated variables are marked with ‘(gen)’; labels of variables with differing age filters over time include the minimum age of respondents across data collections.
    • Prev
    • Next
    • Terms and Privacy
    • Downloads
    • Start