Interpretation and use of physical activity data

Physical activity and sleep data measured by accelerometers, and physical activity and sleep data that people report for themselves, require careful consideration when comparing the two.

• Accelerometers record all movement over a set period, while self-report questionnaires depend on what people remember and choose to report.
• Self-reported physical activity data is based on survey questions about activities such as walking for exercise, recreation or sport or to get to places, moderate or vigorous physical activity (including at work) and strength or toning activities.

Self-report questionnaires play an important role in collecting data about people’s behaviours and risk factors, which are not obtainable through directly measured methods using accelerometers.

The questionnaires used in the IHMHS have undergone rigorous testing and validation to ensure the highest standard of data collection. However, it is important to consider factors that may impact reliability and accuracy. These can include:

• what people remember (for example, they may not remember to include time spent running upstairs or what time they went to bed)
• their ability to estimate total time per day spent doing an activity at each level of intensity across a 24-hour period (for example, moderate physical activity may occur for 30 minutes in the morning, 10 minutes at lunch time, 45 minutes in the afternoon)
• their assessment of the intensity of their physical activity (moderate or vigorous)
• some people providing responses they felt were expected, rather than those that accurately reflect their own situation
• difficulties for parents or guardians being able to estimate the amount of physical activity their child had done, particularly for school-aged children or in shared parenting arrangements.

Directly measured physical activity data using accelerometers provide valuable insights into physical activity by how people move each day. Accelerometers present limitations which may affect comparability with self-reported data. These can include:

• the type of accelerometer device (i.e., devices worn on the hip or wrist) impacts how movement is detected. For example wrist worn devices may not accurately capture physical activity done on a stationary exercise bike or treadmill or during resistance training and weightbearing activities such as weightlifting
• accelerometer measured activity is independent of a person’s perceived behaviour. For example inactivity as an accelerometric classification may not directly align with a person’s self-reported sedentary behaviour
• sleep and naps are difficult to accurately measure from movement alone, as some people may move a lot during sleep, which may be misclassified as other activity types
• compliance and adherence to wearing the accelerometer affects how accurate the results are for a given person.

Self-reported and directly measured data using accelerometers should be treated as complementary information about physical activity, sedentary behaviour and sleep. Users should consider the advantages and limitations of each type of data collection in conjunction with current subject literature when comparing estimates from self-reported and directly measured data using accelerometers.

While each of the four health surveys in the IHMHS collected physical activity, sedentary behaviour and sleep data, the types of questions were tailored to each survey. This is due to survey specific priorities, consistency and timeseries with previous survey iterations and target populations.

For a full list of data items collected across the IHMHS, please see the complete data item list in the Data Downloads section. This includes a concordance of comparable data items for all topics by survey collection.

The NNPAS 2023 and NATSINPAS 2023 provide self-reported data on physical activity, sedentary behaviour and sleep for children and young people 2–17 years. Major updates to sedentary behaviour topics were implemented in NNPAS 2023 and NATSINPAS 2023. Substantial updates have been made to physical activity items in NATSINPAS 2023 which may affect comparability with previous surveys.

See Content Changes and the Data Item List in the NNPAS 2023 Methodology and Content Changes and the Data Item List in the NATSINPAS 2023 Methodology for further information. 

The NHS 2022 and NATSIHS 2022-23 provide self-reported physical activity data for young people aged 15–17 years. Estimates published from these surveys, however, are for all persons aged 15 years and over.

Published estimates of physical activity and sedentary behaviour from accelerometers is available in NNPAS 2023 and NATSINPAS 2023. Processed and raw accelerometer data are available in the DataLab.

The NHS 2022 provides self-reported data on physical activity for adults.

The NATSIHS 2022-23 provides physical activity estimates for Aboriginal and Torres Strait Islander people in non-remote and remote areas. The NATSINPAS 2023 collected a range of comparable items to the NATSIHS 2022-23 with additional physical activity information and topics specific to sedentary behaviour. Substantial updates were made to NATSINPAS 2023 between 2012-13 and 2023 to align physical activity questions with NHS and NATSIHS which has resulted in limited time series data. 

Estimates of self-reported physical activity were collected for NATSINPAS 2023 and were not collected for adults in NNPAS 2023. The ABS prioritised publishing commentary and data cubes for directly measured physical activity using accelerometers as self-reported estimates are already available in NHS 2022 and NATSIHS 2022-23. 

Estimates of physical activity and sedentary behaviour from accelerometers are available in NNPAS 2023 and NATSINPAS 2023. Processed and raw accelerometer data is available in the DataLab.

NNPAS 2023 and NATSINPAS 2023 are the only sources of self-reported and directly measured sleep data. Self-reported sleep duration from NNPAS 2023 and NATSINPAS 2023 maintain timeseries with previous iterations.

Directly measured sleep data from accelerometers is available in NNPAS 2023 and NATSINPAS 2023. Processed and raw accelerometer data is available in the DataLab.