ABSTRACT
Wearable, objective PA measurement tools provide researchers and end-users with information about the frequency, intensity, and time spent in PA, but generally do not capture PA context data. Accelerometry is the most popular objective method of PA measurement in free-living and discrete settings. Accelerometers are valid and reliable tools to measure PA in a range of populations, but they do not measure the physiological cost of PA, and the output relies on researchers to make data processing decisions related to wear time and thresholds for PA intensities. Consumer-focused wearable PA devices with multiple integrated sensors are readily available but with the exception of step counts, their accuracy is less than that of research-grade monitors. An estimation of the physiological cost of PA (e.g., EE) is important to understanding energy balance and its contribution to health. Laboratory-based physiological measures of PA EE are highly accurate, but may not be practically feasible to use in many studies due to the high cost and required specialised expertise in the technique and data analyses. Self-report tools are the most widely used measures of PA in youth and adult populations. They are mostly administered in paper-based format but can also be completed electronically on computers or tablet devices. The validity of self-report tools to accurately classify time spent in specific PA intensities such as MVPA is generally weak, and participant responses are prone to social desirability bias and inaccurate recall.
