There is growing interest in understanding environmental impact on physical activity (PA) behaviors. Traditional measures of PA, however, focus only on the activities themselves and provide little information on the interaction between environment and PA. Zhu (2003) proposed a concept/measure of physical activity space (PAS), which recognizes that PA is a spatial human behavior and its spatial characteristics must be measured so that it can be better understood. The purpose of this study was to develop a walking space measure and evaluate its psychometric quality. The walking space measure is an integration of two measurement devices: AMP331 activity monitor system and GeoLogger GPS data logger. The former is an accurate device that can record a person's steps and the time when steps are taken and the latter is versatile device for collecting detailed spatial and temporal travel data. A total of 31 health adults (16 males and 15 females; M±SD: Age = 39.35±14.56 yr., Height = 174.77±10.54 cm, Weight = 79.74±17.07 kg, & BMI = 26.13±5.58) participated in the study and they wore both devices, from when they got up to when they retired, for 21 days. In addition, to check the accuracy of location recording and travel distance, they were asked to walk with a research assistant (RA), who served as a “shadow observer,” for about a one-hour “validity” walk in the study. During the walk, the RA also wore a set of devices, recorded the major building or street intersections and recorded travel distance using a measuring wheel. Preliminary data analyses found that, on average, this group of participants took 7,628 (±4,399; ranged from 3,642 to 13,679) steps a day. During the “validity” walk, they walked on average 4,654±2,168 steps and 3,211.63±1,025.69 Meters, which is consistent with those recorded by the measuring wheel (3,338.29±833.03 Meters) and by the RA's AMP 331 (3,363.77±1,027.86 Meters; ICC = .94; Repeated ANOVA F = 1.39, p > .05). Some inconsistencies, however, were found between GeoLogger's recorded locations and locations recorded by the RA (e.g., mean absolute error between locations ranged from 5.35 Meters to 39.23 Meters). An effort is being made to understand the sources of the error. In conclusion, preliminary results of the study indicate that the walking space of an individual can be accurately measured by the walking space measure created and the large location errors should be investigated. Keyword(s): measurement/evaluation, research