Measurement of Physical Activity
JAMES F. SALLIS
MARION F. ZABINSKI
It is necessary to assess physical activity to understand the etiology, prevention, and treatment of obesity. This chapter highlights some of the best available self-report and objective measures.
Self-reports are self- or interviewer-administered recalls, activity logs, and proxy reports. They are widely used because of ease of administration, low cost, and adaptability to fit a variety of research questions. Self-reports can assess all dimensions of physical activity: intensity, frequency, duration, and type. However, they cannot be used with young children or persons with limited cognitive abilities. Absolute estimates of activity duration seem to be biased, particularly with subjects’ overreporting of vigorous physical activity.
In general, interviews seem to have better validity than self-administered questionnaires. Importantly, the age range of the population under investigation will influence type of activity and possibly also response to phraseology. Questionnaires with the best psychometric properties are described for children and adolescents, adults, and older adults.
There are seventeen self-report measures with some evidence of validity in youth aged 10–18 years; however, reliability and validity are strongly related to age. The Seven-Day Physical Activity Recall interview takes 15–20 minutes and assesses the past 7 days. It has strong test–retest reliability (r = .77) and adequate validity (r = .53) compared to heart rate monitoring. The Self-Administered Physical Activity Checklist takes approximately 35 minutes for group administration and assesses activity during the previous day be-