International Physical Activity Questionnaire – Long Form

Adults:

(Craig, 2003; n = 2213 [long form]; 12 countries) 

• Excellent test-retest reliability for overall score (ICC = 0.81) 

• Excellent test-retest reliability for physical activity (ICC = 0.84-1.00) 

• Adequate test-retest reliability for sitting (ICC = 0.70) 

(Rosenberg, 2008; n = 289; mean age 35.9 years) 

• Excellent test-retest reliability for sitting items for both sexes combined (ICC = 0.81) 

  • Excellent for sitting items for men (ICC = 0.83) 

  • Adequate for sitting items for women (ICC = 0.77) 

(Alomari, 1993; n = 111; mean age 20.4 (1.3) years; College-enrolled) 

• Excellent test-retest reliability for walking items per week (ICC = 0.97) 

• Excellent test-retest reliability for moderate physical activity per week (ICC = 0.96) 

• Excellent test-retest reliability for vigorous physical activity per week (ICC = 0.97) 

• Excellent test-retest reliability for overall physical activity per week (ICC = 0.96) 

• Excellent test-retest reliability for sitting per week (ICC = 0.97)

Adults: (Hagströmer, 2006; n = 46, 19-62 years old, mean age 40.7 ± 10.3 years) 

• Excellent concurrent validity: correlations of time spent in vigorous physical activity compared to accelerometer monitoring (ρ = 0.71) 

• Adequate concurrent validity: correlation of total time spent in physical activity, in combined vigorous and moderate physical activity compared to accelerometer monitoring (ρ = 0.55, ρ = 0.36) 

• Poor concurrent validity: correlation of time spent in moderate physical activity and total time spent sitting compared to accelerometer monitoring (ρ = 0.20, ρ = 0.17 respectively) 

• Excellent concurrent validity: correlations of time spent in sitting and physical activity at work compared to log book monitoring (ρ = 0.75, ρ = 0.64 respectively) 

• Adequate concurrent validity: correlations of time spent in physical activity at home and during leisure time compared to log book monitoring (ρ = 0.47, ρ = 0.58 respectively) 

(Rosenberg, 2008; n = 289; mean age 35.9 years) 

• Adequate correlation of sitting time and accelerometer recorded measurements for both sexes and for women only (ρ = 0.34, ρ = 0.43 respectively) 

• Poor correlation of sitting time for men only (ρ = 0.24) 

  (Kwak, 2012; n = 441; mean age 49.4 years – 18-67 years old) 

   

• Adequate concurrent validity: correlation with time spent on occupational physical activity for both sexes, for men, and for women compared with accelerometer (r = 0.46, r = 0.44, r = 0.49 respectively) 

• Adequate concurrent validity: correlation with time spent on occupational physical activity for normal weight and overweight workers compared with accelerometer (r = 0.44, r = 0.55 respectively) 

• Poor concurrent validity: correlation with time spent on occupational physical activity for obese workers compared with accelerometer (r = 0.27) 

  (Craig, 2003; n = 744 [long form vs. accelerometer]; 18-65 years old, 12 countries)

   

• Adequate concurrent validity for all physical activity and total sitting time compared to accelerometer (ρ = 0.33) 

  (Chastin, 2014; n = 69; 18-65 years old)

   

• Poor concurrent validity: correlation weekday and weekend sitting time compared with accelerometer (r = 0.17, r = 0.01 respectively) 

  (Alomari, 1993) 

   

• Poor concurrent validity of vigorous physical activity compared to: 

  • Percent body fat (r = -0.2; p=0.02)) 

  • Muscle mass (r = 0.3; p<.01) 

  • Maximal handgrip strength (r = 0.3; p<.01) 

  • Six minute walk distance (r = 0.3; p<.01) 

• Poor concurrent validity of total physical activity compared to: 

  • Percent body fat (r = -0.11; p = 0.26)) 

  • Muscle mass (r = 0.2; p<.01) 

  • Maximal handgrip strength (r = 0.2; p<.01) 

  • Six minute walk distance (r = 0.3; p<.01)

Adults: 

(Rosenberg, 2008; n = 289; mean age 35.9 years) 

• • Poor agreement between time spent sitting and IPAQ categories of activity (kappa = 0.014, P = 0.72; x²(4) = 2.52, P = 0.64) 

(Kim, 2013; meta-analysis) 

• • Adequate convergent validity (corrected mean effect size = ESp) with other instruments for walking, total moderate physical activity, vigorous physical activity, and total physical activity (ESp = 0.32, 0.45, 0.49, and 0.39 respectively) 

  • Poor convergent validity with other instruments for moderate physical activity (ESp = 0.27) 

  • Interviewer administered IPAQ had higher correlated mean effect size compared to self-administered IPAQ 

(Hagströmer , 2006; n = 46, mean age 40.7 years – 19-62 years old) 

• • Poor construct validity between physical activity time and moderate-intensity physical activity time compared to aerobic fitness (ρ=0.21, P < 0.05). 

  • Poor construct validity between total amount of physical activity, moderate physical activity, and vigorous physical activity compared to Body Mass Index (ρ= 0.25, ρ=0.27, and ρ=0.17 respectively; P<0.01) 

  • Poor construct validity between time spent in moderate or vigorous physical activity and aerobic fitness or percent body fat (ρ = 0.21, ρ = 0.14 respectively; P < 0.05)

Not assessed. However, because frequency, duration, and intensity of physical activity and sedentary behavior are assessed, the IPAQ is believed to have a high level of content validity (Craig, 2003)

Not assessed. However, excellent to adequate ICC values are consistent with a reliable face validity to assess levels of physical activity