Background: Adaptation of physical activity self-report questionnaires is sometimes required to reflect the activity
behaviours of diverse populations. The processes used to modify self-report questionnaires though are typically
underreported. This two-phased study used a formative approach to investigate the validity and reliability of the
Physical Activity Questionnaire for Adolescents (PAQ-A) in English youth. Phase one examined test content and
response process validity and subsequently informed a modified version of the PAQ-A. Phase two assessed the
validity and reliability of the modified PAQ-A.
Methods: In phase one, focus groups (n = 5) were conducted with adolescents (n = 20) to investigate test content
and response processes of the original PAQ-A. Based on evidence gathered in phase one, a modified version of
the questionnaire was administered to participants (n = 169, 14.5 ± 1.7 years) in phase two. Internal consistency
and test-retest reliability were assessed using Cronbach’s alpha and intra-class correlations, respectively. Spearman
correlations were used to assess associations between modified PAQ-A scores and accelerometer-derived physical
activity, self-reported fitness and physical activity self-efficacy.
Results: Phase one revealed that the original PAQ-A was unrepresentative for English youth and that item
comprehension varied. Contextual and population/cultural-specific modifications were made to the PAQ-A for use
in the subsequent phase. In phase two, modified PAQ-A scores had acceptable internal consistency (? = 0.72) and
test-retest reliability (ICC = 0.78). Modified PAQ-A scores were significantly associated with objectively assessed
moderate-to-vigorous physical activity (r = 0.39), total physical activity (r = 0.42), self-reported fitness (r = 0.35), and
physical activity self-efficacy (r = 0.32) (p ? 0.01).
Conclusions: The modified PAQ-A had acceptable internal consistency and test-retest reliability. Modified PAQ-A
scores displayed weak-to-moderate correlations with objectively measured physical activity, self-reported fitness,
and self-efficacy providing evidence of satisfactory criterion and construct validity, respectively. Further testing
with more diverse English samples is recommended to provide a more complete assessment of the tool.