Hi, how do I determine if an ATUS respondent is a parent versus other adult caretaker (e.g. grandparent)? I’m analyzing parents’ time spent on child care activities, and want to make sure to exclude respondents who are reporting child care but are not parents.
Thanks much, SG
To clarify: I want to exclude ATUS respondents who are not parents of the children for whom the child care activity variables record their time spent on those activities.
You can use the who variables to identify time spent with own children. These variables are attached to episode records and provide information on who the activity was done with. LINENOW is a variable that reports the line number in the household of the person with whom an activity was done. This variable is only available for activities performed with one or more persons within the ATUS respondent household. RELATEW is a variable that reports the relationship to the respondent of the individual with whom the activity was performed. For example, an activity performed by a respondent with their own household child would have a value of RELATEW=0202. An activity performed by a respondent with their own child who does not live in their household would have a value of RELATEW=0300.
Also see the family interrelationship variables. These variables can be used to determine the relationships between most household members. MOMLOC and POPLOC report the line number within the household of each individual’s mother and father, respectively (if they live in the same household). NCHILD reports the number of own children residing with each individual. RELATE reports each individual’s relationship to the ATUS respondent.
The time use variable builder tool from IPUMS ATUS allows you to create a custom time use variable measuring the amount of time a person spent on a certain activity. Using this tool, you can require that your variable includes only time a respondent spent with specified people, in specified locations, during specified times of day, and/or while performing specified secondary activities. For example, could create a variable that measures the amount of time each respondent spent with their own child. You can access this tool by clicking the yellow button labeled “create time use variable” in the data extract system.
Thanks very much. If I simply excluded respondents for whom ‘KIDUND18’ = 0, does that effectively do the same thing, i.e. eliminate respondents who are not parents of the children for whom the child care activity variables record their time spent on those activities? Thanks again, SG
No, excluding respondents without children under the age of 18 in the household from your analysis will not accomplish your goal. An ATUS respondent with an own child under age 18 in the household (KIDUND18=1) may spend time with other children who are not their own or with their own child who does not reside in their household. These other children could be related or unrelated to the respondent, such as nieces and nephews, grandchildren, children of roommates, friends, or romantic partners, children the respondent babysits, etc.
To determine the nature of the relationship between the ATUS respondent and the child with whom they spent time, you should use the family interrelationship variables, the who variables, and/or create a custom time use variable, as described in my previous post.
Thank you again Isabel, and to make sure I understand, does this apply to already existing variables like BLS_CAREHH_KID? (We’re not analyzing time spent with children generally, just time spent on child care activities.)
That is, (i) it’s possible BLS_CAREHH_KID reports the time spent on child care activity by an ATUS respondent with an own child under age 18 in the household, but it’s that respondent’s time spent on child care of another child (not their own, e.g. niece) in their household?
And (ii) it’s possible BLS_CAREHH_KID does not report the time spent on child care of own child by an ATUS respondent with an own child under age 18 in the household, because that own child is not in the household?
Thanks again, SG
The time use variable BLS_CAREHH_KID reports the amount of time spent on caring for and helping household children. An ATUS respondent can spend time on caring for a household child that is not their own, regardless of whether they have their own child, and regardless of whether their own child lives in their household. The variable BLS_CAREHH_KID does not distinguish between the time spent caring for the respondent’s own child or another household child.
You can recreate a custom, edited version of this variable using the time use variable builder tool I described in an earlier post. Clicking on any of the existing time use variables, such as BLS_CAREHH_KID, will show you information on which activities are included. This table shows how the BLS published table major categories correspond to IPUMS ATUS activity codes.
Below are the time use activities that are included in BLS_CAREHH_KID:
Your custom variable can include the same time use activities but specify that the respondent was with their own household child during the activity.
