All about Time Use Variables
The creation of time use variables is a central component of the ATUS Data Extract Builder. Time use variables are combinations of activities and/or filters that represent the amount of time respondents spend in those activities. A time use variable may be more than, for example, a measure of time spent playing baseball. The ATUS data also allow users to specify, in addition to activities, what we call filters. The filters available at this time include, times of day, secondary activity (childcare, eating, drinking), locations, and with whom. The possible components that may be combined to create a time use variable are discussed in further detail.
All of the activities respondents report doing over a 24-hour period from 4:00 a.m. to 4:00 a.m. on an assigned diary day are recorded as part of the American Time Use Survey. Activities are coded using a six-digit scheme in which the first two digits represent the broadest level of detail, the second set of two-digit codes represents an intermediate level, and the final two-digits represent the detailed code within the two broader categories. The scheme represents over 400 activity groupings at the most detailed level. The major activity categories include:
- Personal care
- Household activities
- Caring for and helping household members
- Caring for and helping non-household members
- Working and work-related activities
- Consumer purchases
- Professional and personal care services
- Household services
- Government services and civic obligations
- Eating and drinking
- Socializing, relaxing and leisure
- Sports, exercise and recreation
- Religious and spiritual activities
- Volunteer activities
- Telephone calls
We represent these distinctions hierarchically in an activity coding tree. The tree expands to show intermediate and detailed subcategories, and clicking on any given code takes you to more detailed information about the code, including rules and examples. This additional information is also available inside the extract system.
Users concerned with comparability over time may want to refer to this table, which shows the activity categories that change over time. The majority of code changes occur between 2003 and 2004. Users building time use variables and selecting from the most detailed codes may want to consider combining categories as suggested to avoid comparability issues. Comparability is only an issue at the most detailed level of the activity coding structure.
Users analyzing particular types of travel will want to be aware of how travel activities are coded. Travel activities are generally coded to match the activity that follows the travel. For example, if the respondent reports driving and then working (050101), the driving episode will be coded "travel related to working"-180501. The last four digits of the travel-related activity code match the first four digits of the following activity's code.
There are two exceptions. If the activity that follows the travel episode takes place at the respondent's home, then the travel is coded to match the previous activity. Similarly, if the activity that follows the travel episode takes place in an unspecified location, the travel is coded to match the previous activity.
Also, if the respondent uses several modes of transportation in one trip, for example, walking followed by a bus ride followed by more walking, all three travel activities will be coded to match the activity that follows the final travel activity.
This coding scheme is important to keep in mind for analyses of particular sorts of travel (e.g., commuting time). A researcher interseted in commuting time may want to take assign new codes in certain circumstances. For example, individuals often run errands on the way to or from work. If a respondent stops at the grocery store on the way home from work, both the leg of the trip from work to the store and the leg from the store to the respondent's home will be coded as travel related to shopping. In this instance, travel related to work is under-represented, and it may be more appropriate to assign new activity codes to travel episodes to better suit the research goal.
Filters are essentially restrictions that users may request when defining their time use variables. For example, a user might be interested in Research or Homework (activity code 060300) done at the respondent's home (location filter) between the hours of 6 p.m. and 9 p.m. (time of day filter). In this case, activities are selected, and location and time filters are applied. To take another example, users might want to differentiate between time spent doing homework at home in the evening while engaging in secondary child care or not engaging in secondary child care.
Time of Day
ATUS respondents report when they began and ended each activity they engage in over the 24-hour diary day. You may specify any combination of times between 4:00am and 4:00am. Please note, however, that you may not specify times that cross the 4:00am threshold. For example, 3:00am to 6:00am is not a valid selection. This specification requires constructing two separate time use variables (3:00am to 4:00am and 4:00am to 6:00am) which you can then sum in your statistics package. Users should not, however, that 3:00am to 4:00am occurs at the end of the reporting period while 4:00am to 6:00am occurs at the beginning of the reporting period. This is due to the design of the study in which respondents report activities from 4:00am of the previous day to 4:00am of the reporting day.
ATUS respondents report only whether they were engaged in secondary childcare or secondary eating and drinking activities while they were also doing primary activities. These data are collected after respondents have reported their main activities for the 24-hour diary data.
Extract system selections
The following are the secondary activity selections available through our website for use in the construction of time use variables. Time use variables may include either filters for secondary child care or eating and drinking as a secondary activity, but not both in a given time use variable.
- Duration of entire activity
- Time spent during activity on secondary child care of all children (SCC_ALL_LN)
- Time spent during activity on secondary child care of household and own, non-household children (SCC_HHNHHOWN_LN)
- Time spent during activity on secondary child care of non-own, non-household children (SCC_NOWNNHH_LN)
- Time spent during activity on secondary eldercare (SEC_ALL_LN)
- Time spent during activity on secondary eating and drinking (SED_ALL_LN)
Names in parentheses above represent the variables that serve as filters. These are variable through the data extract builder with the hierarchical data selection. Additional secondary child care variables are also available on the ACTIVITY record: SCC_HH_LN and SCC_OWN_LN.
User note: Application of secondary activity filters during the time use variable creation process assumes "any" rather than "only" in the inclusion process. For example, users who select "Time spent during activity on secondary child care of household and own, non-household children" will be given a time use variable that includes all time spent in the care of household and own, non-household children even if the respondent was also caring for non-own, non-household children.
Secondary child care is care for children under age 13 that is done while the respondent is doing something else as a primary activity. Information on secondary child care is available for both household and non-household children. Secondary child care data are collected as part of a series of summary questions after the main diary has been completed. For respondents with a child under age 13 on the household roster, the interviewer asks the respondent during which times or activities households spend doing secondary care of household children is restricted to time between when the first household child under 13 woke up and the last household child under 13 went to bed. All respondents are asked whether a non-own, non-household child under 13 was in their care at any time during the diary day.
Not all primary activities are eligible to be combined with secondary child care. Secondary child care is only recorded during times when the respondent is awake, and, for household children, secondary child care can only take place when at least one household child is awake. Further, secondary child care cannot take place if the respondent's primary activity is also a child care activity.
The availability of secondary child care information changed between 2003 and 2004. The information on secondary child care is more limited for 2003 than for 2004 forward. Only SCC_HHNHHOWN_LN and SCC_NOWNNHH_LN are available for 2003. Users who want to use secondary child care filters should be aware of this limitation when considering comparability of estimates over time.
Secondary child care is assumed to last for the entire duration of the primary activity during which it occurs, with one exception. If the first household child wakes up or the last household child goes to bed during the primary activity, then the secondary child care for household children time is assumed to begin at the wake time or end at the bed time, rather than at the beginning or end of the respondent's primary activity.
Secondary eldercare is care for elderly person(s) that is done while the respondent is engaged in another activity as his or her primary activity. Elderly persons are defined as adult individuals who need assistance in performing different activities, because of conditions related to aging. Only respondents who report to have provided care or assistance (excluding financial assistance or assistance as part of the job) for an elderly person at least once in the three months prior to the interview date, are included in the universe of this variable. The care may have been provided for one or more household or non-household members.
Secondary eldercare is assumed to last for the entire duration of the primary activity during which it occurs. However, secondary eldercare may not be performed while the respondent is engaging in personal care activities (01XXXX) or personal care services (0805XX).
Information on secondary eldercare is available starting in 2011. Time spent in secondary eldercare is available in the variable SEC_ALL_LN, which is available on the ACTIVITY record.
Eating and Drinking
Secondary eating and drinking data were collected from 2006 through 2008 as part of the Eating and Health Module.
Secondary eating and drinking is eating or drinking that occurs while the respondent is doing something else as the primary activity. Like secondary child care, these data are collected at the conclusion of the main ATUS interview. During the module, the interviewer asks the respondent if there were any times during the day when the respondent was eating while doing something else, and for how long during that activity the respondents were eating. Those questions are repeated for drinking, although drinking plain water is specifically excluded. These data indicate whether respondents were engaged in the secondary activity but not when. Users should note that beginning in October 2006, respondents were able to report engaging in secondary eating and/or drinking "all day."
Secondary eating and drinking can take place during any primary activity except sleeping, eating and drinking as a primary activity, and eating and drinking as part of one's job. Secondary eating and drinking can occur during sleeplessness
There are three secondary eating and drinking variables. Unlike secondary child care which is assumed to last the duration of the primary activity except when the first household child wakes or the last household child goes to bed, secondary eating and drinking may occur for less than the duration of the primary activity. SED_DRINK_LN reports the number of minutes during the primary activity the respondent was also drinking something other than plain water, SED_EAT_LN reports the number of minutes the respondent spent eating, and SED_ALL_LN reports the number of minutes the respondent spent eating and drinking.
The data do not specify when during the primary activity secondary eating/drinking occurred or the amount of overlap between time spent in secondary eating and time spent in secondary drinking. ATUS-X follows the ERS guidelines for calculating overlap between secondary eating and drinking and the timing of these activities as follows.
- If the secondary eating and drinking option is selected and either secondary eating or secondary drinking is equal to the duration of the primary activity, then the ATUS-X returns the entire duration as the amount of time in secondary eating and drinking.
- If neither secondary activity is as long as the primary activity, the minimum amount of time during which one or the other could have occurred is the larger of the two times; the maximum amount of time during which one or the other could have occurred is either the sum of the two times or the total length of the activity, whichever is less. ATUS-X takes the average of these minimum and maximum times to be the duration of secondary eating or drinking during the activity. For example, if a respondent reports 10 minutes of secondary eating and 20 minutes of secondary drinking during a primary activity that lasts more than 30 minutes, the ATUS-X will return a total secondary eating and drinking time of 25 minutes for that activity. This is the average of the minimum time (20 minutes) and the maximum time (10 minutes plus 20 minutes).
- In cases where the time spent in secondary eating and/or drinking is not for the duration of the primary activity, we assume that the secondary eating and/or drinking occurred in the middle of the primary activity. For example, if a respondent reported secondary eating for 10 minutes during a primary activity that lasted from 1:00 p.m. to 3:00 p.m., the ATUS-X defines the secondary eating time as 1:55 p.m. to 2:05 p.m. This is only a concern in cases where a time filter is also applied in the time use variable.
Researchers who use the secondary eating and drinking variables should be sure to use an alternative weighting variable, EHWT, rather than the standard estimation weighting variable WT06. The reason a different weight is needed is that the Eating and Health Module has a slightly lower response rate than the main ATUS interview.
ATUS respondents report where the activities they engage in during their diary days occur. Exceptions are Sleeping (0101xx), Grooming (0102xx), Personal activities (0104xx), "respondent refused to answer" (500105), and "respondent didn't remember" (500106). In these cases, the location is recorded as blank.
The locations where activities occur are grouped into two broad categories: places and modes of transportation, as well as a residual "missing" category. In October of 2004, three locations were added: bank, gym/health club, and post office. The list of possible locations to use as filters is as follows:
Place Respondent's home or yard Respondent's workplace Someone else's home Restaurant or bar Place of worship Grocery store Other store/mall School Outdoors away from home Library Bank Gym/health club Post office Other place Unspecified place
Mode of Transportation Car, truck, or motorcycle (driver) Car, truck, or motorcycle (passenger) Walking Bus Subway/train Bicycle Boat/ferry Taxi/limousine service Airplane Other mode of transportation Unspecified mode of transportation Missing Don't know Blank Outdoors away from home Library Bank Gym/health club Post office Other place Unspecified place Outdoors away from home Library Bank Gym/health club Post office Other place Unspecified place
ATUS respondents report who was with them during the activities they engaged in over the 24-hour diary day. Respondents may report more than one person who was in the room with them or who accompanied them during each activity. The list of possible "with whom" responses to use as filters is as follows:
Household Persons Spouse Unmarried partner Own household child Grandchild Parent Brother/Sister One related person Foster child Housemate/roommate Roomer/boarder Other non-relative
Non-Household Persons Own non-household child <18 Parents (not living in household) Other non-household family members <18 Other non-household family members 18+ Friends Co-workers/colleagues/clients (non-work activities only) Boss or manager (work activities only, 2010+) People whom I supervise (work activities only, 2010+) Co-workers (work activities only, 2010+) Customers (work activities only, 2010+) Neighbors/acquaintances Other non-household children <18 Other non-household adults 18+
In the extract builder, users may also specify the age and sex of household children who did or did not accompany the ATUS repsondent during the activity. Note that using this feature also allows for handling the own household children not specified in the age and sex criteria. For example, the ATUS-X allows researchers to specify activities done with own male household children when own female household children were not present.
Quality of Life
Quality of life data were collected in 2010 as part of the Well-Being Module.
For each of three randomly selected activities reported by each respondent, quality of life data were collected. The quality of life data consist of seven questions: five on affect (SCHAPPY, SCPAIN, SCSAD, SCSTRESS, and SCTIRED), one about the meaningfulness of the activity (MEANING), and one about whether the respondent was interacting with anyone during the activity (INTERACT). The order of the affect questions was randomly assigned and is also available (OHAPPY, OPAIN, OSAD, OSTRESS, and OTIRED).
Quality of life data were not collected for all activities. The three randomly selected activities for which quality of life data were collected were required to last five or more minutes and could not include the following ACTIVITY codes: sleeping (0101xx), grooming (0102xx), personal activities (0104xx), refusal (500105), don't know (500106).
At this time the quality of life data are only available to ATUS-X users by choosing a hierarchical data extract. Because quality of life information is not available for all activities, this information cannot be used to create time use variables in the same way as other activity-level data are used. Users who want to analyze these data should consult the Well-Being Module Data Dictionary.