Combining accelerometer, GPS, and GIS data will allow for more sensitive and accurate measures of behavior in their context, both in time and in space. Having these measures will make it possible to more accurately determine which environments people are exposed to, for how long, and during which behavior. These combined measures will also aid in evaluation of interventions, as well as allow testing of specific hypotheses, such as the Activity Stat hypothesis, which posits that increased activity in one domain will be compensated for by decreased activity in another domain(14). However, it is important to stress that this type of improved measure, used cross-sectionally, does not increase the possibility to detect causal relations; they will show where and when behavior takes place, not why. As such, these types of measures cannot directly be used to predict behavior in a given environment. Intervention studies, however, with GPS assessing changes in PA locations after local environment improvements will provide better evidence of causality(20). However, they will allow a wealth of new research questions focused on better matching of behavior and environment, resulting in more specific assessments of PA as related to locations, times, and behaviors.