Codes and Frequencies
Comparability — Index
This variable is largely comparable across countries, though there are some surveys with limited information about household composition and different age bands (such as the number of children under age 12 or 15). There are adjustments and corrections made when possible. Users should consult documentation below for more information.
Comparability — Canada [top]
In 2005 and 2010, while creating the variable NCHILD instead of using the original variable CHH0014C which is supposed to show the number of children under 15 in the household but fails to do so because there are many "none" responses in this variable and the age of the youngest child of the respondent and in the household is less than 14 (this must be a bug on the dataset), we use 4 different variables to create NCHILD: CHH0014C (Number of children aged from 0 to 14 years living in the respondent's household, this variable is capped at 3, 3 represents 3 or more children), CHRINHSDC (number of respondent's children living in the household (any age or marital status. This variable is capped at 4, 4 represents 4 or more children), AGECHRYC (age of respondent's youngest single child living in the household), and AGEHSDYC (age of youngest household member living in the respondent's household). Therefore, NCHILD=3 represents 3 or more children and NCHILD=4 represents 4 and more children.
In 2015, the number of children are those that are younger than 14 years old. The code of "3" indicates three or more children.
Comparability — Finland [top]
In 1979, the number of children in the household are the number of own, not total, children. All respondents under the age of 18 left this question blank. Therefore, the number of household children is missing for all respondents under the age of 18. However, it should be noted that these households have at least one child - the respondent.
Comparability — France [top]
For 1985, using information about the age and relationship to respondent/reference person for all household members it was possible to create the NCHILD and AGEKID variables, and also a variable which denotes whether a parent of the respondent is living in the household, useful for FAMSTAT. There was information about number and age of children in the data but 'child' could be someone 18yrs or older.
In 2009, Cut-off point at 8 or more children (code 8) – as in the original data.
Comparability — Netherlands [top]
Due to limitations in the source data, this variable only includes children under 16.
Comparability — South Africa [top]
In 2000, the original variable (CHILD18H, number of children under 18 household) was used to compute NCHILD; however, this variable had an overwhelming number of missing cases. Those respondents who were 18+ and had missing values for CHILD18H were coded as NCHILD=0; however, those respondents 18 with missing values for CHILD18H (for a majority, this was the case) were coded as NCHILD=1. Note that it is not possible to determine whether these respondents are the only children 18 in the household; therefore, NCHILD=1 could be inflated.
Comparability — United Kingdom [top]
In 1995, NCHILD includes the number of children under 16 plus 1 if the diarist is between 16 and 18.
- All persons.
- Austria: 1992
- Bulgaria: 2001
- Canada: 2005, 2010, 2015
- Finland: 1979, 2009
- France: 1985, 2009
- Hungary: 1999, 2009
- Israel: 1991
- Italy: 2002, 2008
- Netherlands: 1975, 1980, 1985, 1990, 1995, 2000, 2005
- South Africa: 2000, 2010
- South Korea: 2004, 2009
- Spain: 2002, 2009
- United Kingdom: 1974, 1983, 1987, 1995, 2000, 2005, 2014
- United States: 1965, 1975, 1985, 1993, 1995, 1998, 2003-2018