Instruments to assess physical activity are needed for (inter)national surveillance systems and comparison. Male and female adults were recruited from diverse sociocultural, educational and economic backgrounds in 9 countries (total n = 2657). GPAQ and the International Physical Activity Questionnaire (IPAQ) were administered on at least 2 occasions. Eight countries assessed criterion validity using an objective measure (pedometer or accelerometer) over 7 days. Reliability coefficients were of moderate to substantial strength (Kappa 0.67 to 0.73; Spearman's rho 0.67 to 0.81). Results on concurrent validity between IPAQ and GPAQ also showed a moderate to strong positive relationship (range 0.45 to 0.65). Results on criterion validity were in the poor-fair (range 0.06 to 0.35). There were some observed differences between sex, education, BMI and urban/rural and between countries. Overall GPAQ provides reproducible data and showed a moderate-strong positive correlation with IPAQ, a previously validated and accepted measure of physical activity. Validation of GPAQ produced poor results although the magnitude was similar to the range reported in other studies. Overall, these results indicate that GPAQ is a suitable and acceptable instrument for monitoring physical activity in population health surveillance systems, although further replication of this work in other countries is warranted.