There are some limitations to this study. Only broad categories of
physical environments are assessed in the ATUS. Consequently, all
other plausible environments where physical activity can occur (e.g.,
indoor golf driving ranges, ice rinks, indoor swimming pools, climbing
walls) are included in the “other” category, making this group quite
large and heterogeneous. Second, there may be some confusion and
ambiguity experienced by the respondent with regards to how to
report certain environments. For example, outdoor exercise bouts
reported during working might have been reported as occurring
“outdoors” instead of at “work,” which might have lead to an
underestimation of sports and exercise in the work setting. Also, the
current study limited its description of social and physical environments
to those of recreational sports and exercise. Activities
performed for the purpose of transportation, work, or household
chores were not included in the analyses. Lastly, recall errors and
biases are still possible with a 24-hour recall. However, other methods
of assessing exercise environments such as real-time data capture (via
electronic diaries) and direct observation are not feasible in a national
sample of this size due to cost and practical limitations.