Microdata: Time Use

The Time Use Survey (TUS) is designed to provide a snapshot of time spent per day on activities such as unpaid work, employment and education, free time, personal care, and caring activities undertaken at the same time.

The 2024 TUS can provide insights into:

• the average time spent per day on activities 
• the proportion of people who participated in activities 
• the time of day activities were undertaken 
• household and personal characteristics of persons undertaking activities 
• division of time use within couples. 

• DataLab - approved users can access detailed microdata for in-depth and interactive data analysis using a range of statistical software packages. This product is available for TUS for 1997, 2006 and 2024. For more information, including prerequisites for access, refer to the DataLab User Guide.
• MicrodataDownload – approved users can download and analyse basic microdata (unit record data) in their own environment. This is available for TUS for 1992, 1997 and 2006. For more information refer to MicrodataDownload. 
• TableBuilder – no TableBuilder products are available for TUS. 

The previous survey 2020-21 TUS did not release microdata.

Before you apply for access, read Responsible use of ABS microdata. 

For all data items refer to the Data Item List in Data Downloads.

Data items include: 

• Time spent per day on activities such as unpaid work, employment and education, free time and personal care (for a full list of activities refer to the Activity Classification in the Data Item List in Data Downloads) 
• Time spent per day on supervisory care of a child aged 14 years and under 
• Time spent per day on supervisory care of an adult (due to disability, illness, problems related to old age) 
• Time spent per day working from home 
• Workday and non-workday 
• Weekday and weekend day 
• How often feels rushed or pressed for time 
• Person items – age, sex, gender, couple status, parent status, grandparent status, disability status/type, labour force status, occupation, industry, education status/attainment, income 
• Partner items – age, sex, gender, labour force status, education status/attainment, income of partner
• Household items – household composition, number of children in household, age of youngest child, household income, geography, Socio-Economic Indexes for Areas (SEIFA).

The 2024 TUS microdata is provided across two levels: 

• Day level – contains information about whether the day was a weekday or weekend, or a workday or non-workday, time spent per day on activities, time spent per day on supervisory care, time spent per day working from home, and demographic and socioeconomic characteristics of persons and their household
• Time Interval level – contains information on the time of day, the activity participated in, and whether supervisory care and working from home were participated in at that time.

The following table shows the hierarchical file structure and the relationship between the two levels. 

Every record on each level has a unique identifier which can be used when merging items from the Day level with the Time Interval level.

For the 2024 TUS, every record on the Day level is uniquely identified by a combination of identifiers, the Household (ABSHIDD), Person (ABSPID) and Day (ABSDID) identifiers. The household identifier assists with linking records from the same household. 

These identifiers are repeated on each record on the Time Interval level and are used when merging data with the day level. 

Each record on the Time Interval level has a Time Interval identifier (ABSIID). Combining the Household, Person, Day and Time Interval identifiers will uniquely identify each record on the Time Interval level. 

For details on identifiers refer to the Data Item List in Data Downloads. 

To analyse data within a couple, combining the household identifier (ABSHIDD) with the Couple identifier (ABSCID) will uniquely identify couples in a household where both members of the couple returned at least one diary day. Each person in a couple who returned a diary will have the same ABSCID value. Where a person is in a couple relationship, but the partner did not return at least one diary day, there will be no record on the file for the partner, and the person will have a value of ABSCID=0.

The Partner linking key (PRTNRLNK) assists with linking a person’s record to their partner’s record. This enables analysis of how time was spent by each person within a couple on the same day.

These identifiers are on each record of the Day and Time Interval levels. For details on these identifiers refer to the Data Item List in Data Downloads . 

For 2024 TUS, each record on the day level has a Day Weight (FINWGT), which indicates how many persons are represented by the day record. The weights produce estimates that are designed to represent the demographic spread of the entire in-scope population and correct for bias in survey selection and response. The weights are also designed to account for the type of day (weekday or weekend day) and whether a respondent completed one or two diary days. For more information refer to Estimation methods in the Methodology.

There is no weight associated with the Time Interval level because the records are repeated for each day.

The TUS is a sample survey, so to produce estimates you must use FINWGT in your calculations. Once weights have been applied, the resulting estimates represent the number of persons in the in-scope population.

When analysing data from the Time Interval level, care must be taken when applying the weights. It is recommended to transform the statistic of interest from the time interval level to the day level to ensure weights are applied correctly.

In a ‘Time spent per day’ data item, a value of 0 minutes indicates that there was no participation in that activity. Examples on how to calculate 'average time spent per day' on an activity: 

• mean of ACT3102 will provide the average time spent per day on housework of the total population 

• mean of ACT3102 where ACT3102>0 will provide the average time spent per day of people who participated in housework. 

In a ‘Time spent per day on activity while doing supervisory care’ data item, a value of 0 minutes indicates that there was no participation in supervisory care while doing that activity. The following examples demonstrate calculations for the 'average time spent per day on an activity while doing supervisory care': 

• mean of CHI3102 will provide the average time spent per day the whole population spent on housework while doing supervisory care of child

• mean of CHI3102 where ACT3102>0 will provide ‘of all persons that did housework, the average time spent doing housework while supervising a child’ 

• mean of CHI3102 where ACT3102>0 and CHI3102 >0 will provide ‘of all persons that did housework while supervising a child, the average time spent per day doing so’. 

The 2024 TUS Activity Classification provides hierarchical grouping of activities for analysis and output. Refer to the Activity Classification in the Data Item List in Data Downloads.

Several questions in the survey allow respondents to select more than one response category. Each response category is treated as a separate data item. These data items share the same SAS name but are suffixed with a letter A, B, C and so on.

For example, the multi-response data item ‘Disability type’ has seven response categories. In the data item list, the SAS name is labelled as DISTYPE(A-G). There are seven data items named DISTYPEA, DISTYPEB, DISTYPEC, DISTYPED, DISTYPEE, DISTYPEF, DISTYPEG. Each data item has a code corresponding to the response category and a null response code of 0.

Note that the sum of the individual multi-response categories will be greater than the applicable population for the data item as respondents can select more than one response. 

Some continuous data items are allocated special codes for certain responses (e.g. 9999 = 'Not applicable'). These are specified in the Data Item List in Data Downloads. These special values should be excluded when calculating means. 

Sampling error should be considered when interpreting estimates as this gives an indication of accuracy. Sampling error is the expected difference between survey estimates and the true value that would result from surveying the whole population. Sampling error can be estimated using measures of data variance including standard error (SE), relative standard error (RSE) and margin of error (MoE). Only estimates with RSEs below 25% are considered reliable for most purposes. Further information on interpreting these measures refer to the Accuracy section of Methodology. 

These measures of sampling error can be estimated using the replicate weights. These are different to the weight used to produce the estimate. TUS uses 60 replicate groups for the day weight labelled RPWT01 to RPWT60.

The TUS sample was designed to support national estimates. Sub-national estimates may be subject to high relative standard errors.

How Australians use their time methodology, 2024 

The methodology contains information about the 2024 TUS on: 

• Improvements implemented for the 2024 Time Use Survey
• Data collection
• Data quality 
• Processing the data 
• Key concepts 
• Accuracy 
• Glossary.

How Australians use their time, 2024

03/07/2026 - Datalab product for 2024 Time Use Survey released.