Self-report continues to be the most widely used method of physical activity assessment, yet the need to improve the validity and reliability of these measures remains a research priority (Wood, 2000). Recently, the International Physical Activity Questionnaire (IPAQ, Sjöström, Bull, & Craig, 2002) was developed as a comprehensive assessment of physical activity in a variety of contexts to provide comparable measures of activity both cross-culturally and across research studies. The purpose of this study was to examine the test-retest reliability of the past 7-day long form IPAQ, which has been recommended for use in research settings. The measure’s validity and reliability as a population surveillance tool has been reported; however, its usefulness in research contexts is unknown (Craig et al., in press). The IPAQ is particularly appealing as it assesses physical activity in a variety of contexts, potentially providing a more accurate and reliable measure of activity. Individuals report frequency and duration of vigorous, moderate, and walking activities that occur on the job, in transportation, at home, and during leisure time, over the past 7 days. Participants were 151 university student volunteers (M age = 24.15, SD = 5.01) who completed the IPAQ on two separate occasions, 5 days apart. At time two, participants referred to the same 7-day period assessed at time one. Intraclass correlation coefficients from one-way ANOVAs were calculated for vigorous, moderate, and walking activity in each context, and for total activity. Coefficients for activity in transportation were low for moderate activity (R=.51) and moderate for walking (R=.73). For home activity, coefficients ranged from low for moderate outdoor activity (R=.38) and vigorous activity (R=.64 ) to high for moderate indoor activity (R=.82). For leisure time activity, reliabilities were low for moderate activity (R=.49) and walking (R=59) and moderate for vigorous activity (R=.72). For those with jobs (N=121), reliabilities for job-related vigorous (R=.73) and walking (R=.78) behavior were moderate, while reliability for job-related moderate activity was low (R=.60). Reliability for total weekly activity was moderate (R=.79). These findings suggest that, similar to other self-report measures, reliable assessment of moderate intensity activity is problematic. Coefficients for vigorous activity, except at home, were consistent with other self-report measures (Sallis & Saelens, 2000). Reliabilities for transportation and job-related walking, and moderate indoor home activity suggest the measure’s potential usefulness in these activity contexts. Future studies should however, focus on improving IPAQ reliabilities if its full benefits are to be realized.Keyword(s): exercise/fitness, measurement/evaluation, physical activity