To compare physical activity (PA) reported through pedometer registrations (step counts) with PA reported in four different questionnaires; to compare step count thresholds (7500, 10 000 and 12 500 steps/d) with the PA guideline of 30 min of moderate to vigorous PA (MVPA) per day.

A sample of 310 healthy adults, mean age 38·7 (sd 11·9) years, volunteered to participate. Forty-seven per cent was male and 93 % of the sample was employed.

PA was assessed by interview (Minnesota Leisure Time Physical Activity Questionnaire (MLTPAQ)), three self-administered questionnaires (long version and short version of the International Physical Activity Questionnaire (long-form IPAQ, short-form IPAQ), Baecke questionnaire) and seven consecutive days of pedometer registration.

Step counts correlated positively with questionnaire-based PA. The strongest correlations were found between step counts and total PA reported in the long-form IPAQ (rs = 0·37), moderate PA reported in the short-form IPAQ (rs = 0·33), total and moderate PA reported in the MLTPAQ (rs = 0·32), and the total and leisure-time PA indices (excluding sport) reported in the Baecke questionnaire (rs = 0·44). According to step counts, 22·6 % of the participants were somewhat active, 18·7 % active and 39·4 % highly active. As assessed by the long-form IPAQ, short-form IPAQ and MLTPAQ, the guideline of 30 min MVPA/d was reached by respectively 85·4 %, 84·8 % and 68·0 % of participants.

Pedometer-based data offer adequate information to discriminate between levels of PA. Caution is needed when comparing active samples based on different PA recommendations.