Objectives: To assess the validity and reproducibility of a new computer-based physical activity questionnaire (cPAQ) to estimate the habitual physical activity levels in Malaysian adolescents. Design: Cross-sectional study. Methods: About 425 adolescent boys and girls participated in the study. A heart rate monitor and a 7-days physical activity logbook were used as objective and subjective method criteria, respectively. Results: The mean estimated weekly total physical activity levels measured by the cPAQ were 16.5±17.5h and 13.1 ± 16.9h for boys and girls, respectively. Analyses of the gender-specific Spearman’s correlation coefficient between cPAQ and both objective and subjective method criteria showed that r values ranged from 0.54 to 0.76 (p < 0.001). Method comparisons assessed by the Bland–Altman plot analysis found that the total physical activity levels, calculated from the cPAQ, fell within the mean ± 2 standard deviation. Reproducibility of the cPAQ was re-examined after two-weeks, showing that intra-class correlations for most PA intensity levels in boys and girls were high with a mean r value of 0.72. The effectiveness of the cPAQ showed a significantly shorter assessment time compared to the manual paper-administered questionnaire in adolescents (p < 0.001) and the majority of participants (93%) accepted the overall appearance of this new cPAQ. Conclusions: The cPAQ provides a valid and reliable method for assessing the habitual physical activity levels for the past year in Malaysian adolescents. It was also proven to be highly effective, based on duration time and its overall appearance for the adolescent participants.