Comparability — Austria [top]

In the Austria 1992, there is no variable per se for the age of the youngest child. All household members aged 10+ were asked to keep diaries, and MTUS was able to compute the age of the youngest diarist. They also knew the number of children aged lt 15 in the household. If there are children aged lt 15 and no child diarists, these children must be at least aged 5 or older. In these cases, AGEKID is set to 2 and AGEKID2 to 9. We then turn to the diary codes. If the household includes a child aged lt 10 and one or more diarists in the household report looking after an infant, then the youngest child in the household most likely is aged lt 5. Also, if the childcare time is above 250 for the household for the day, the likelihood is high that the youngest child is very young. In such cases, AGEKID is set to 1 and AGEKID2 to 3.

Comparability — Canada [top]

In 2005, AGEKID2 was created using the original variable AGEHSDYC (corresponding age category under 18). For the age codes between 18 and 25, we relied on the age of the children of the respondents, so the number likely underrepresents because there could be other children in the household other than the respondent's children. For ages 25-60, the age was replaced by the age of the youngest member in the household if that age is 25 or older and were all coded as 25. Ages 60+ are coded as 60.

Comparability — Finland [top]

In 1979, there is no relationship to the household reference person variable. There is a variable which includes a category for an unmarried person still living with one or both parents. Some of the people coded in this category are coded as adult children. Nevertheless, users should note that some adult respondents who live with a parent cannot be identified as such for this survey.

Comparability — Italy [top]

In 2002 and 2008 Italian samples, AGEKID2 is top coded at 60.

Comparability — South Korea [top]

There were three households with pregnant women. These cases were coded as 1 and youngest child age as 0.

Also it was not possible to distinguish between children and son/daughter in-law in case of married children in the household. For AGEKID2, the possible son/daughter-in-law cases were coded as -8 when aged over 18. (84 households)

Comparability — United Kingdom [top]

In 1974, this variable is not directly available for all cases. In cases where we knew a child in a particular age range was present, we coded AGEKID2 to the midpoint of the range. Some values of 3 represent the age range 2-4; some values of 8 represent the age range of 5-11; some values of 14 represent the range 12-16.

In 1983, there are variables which record the number of children in the household of certain ages. The midpoints of these age categories are used to create AGEKID2 in several cases.

Comparability — United States [top]

In 1992-94, the age of the youngest co-resident child is only available if there was a household member under 18.

For 1998-2001, 1998-99 FISCT did not collect the age of the youngest child. 1999-2001 NSP is a sample of parents with at least one child under 18. The NSP did not collect household grid information, and it is not possible to distinguish where parents of young children also live with their parents.