Limits...
Reliability and Validity of the Self- and Interviewer-Administered Versions of the Global Physical Activity Questionnaire (GPAQ).

Chu AH, Ng SH, Koh D, Müller-Riemenschneider F - PLoS ONE (2015)

Bottom Line: Participants had similar baseline characteristics in both administration groups and no significant difference was found between the two formats in terms of validity (correlations between the GPAQ and accelerometer).Our findings showed comparability between both self- and interviewer-administration modes of the GPAQ.It is advisable to incorporate accelerometers in future studies, particularly when measuring different intensities of physical activity.

View Article: PubMed Central - PubMed

Affiliation: Saw Swee Hock School of Public Health, National University of Singapore, Singapore, Singapore.

ABSTRACT

Objective: The Global Physical Activity Questionnaire (GPAQ) was originally designed to be interviewer-administered by the World Health Organization in assessing physical activity. The main aim of this study was to compare the psychometric properties of a self-administered GPAQ with the original interviewer-administered approach. Additionally, this study explored whether using different accelerometry-based physical activity bout definitions might affect the questionnaire's validity.

Methods: A total of 110 participants were recruited and randomly allocated to an interviewer- (n = 56) or a self-administered (n = 54) group for test-retest reliability, of which 108 participants who met the wear time criteria were included in the validity study. Reliability was assessed by administration of questionnaires twice with a one-week interval. Criterion validity was assessed by comparing against seven-day accelerometer measures. Two definitions for accelerometry-data scoring were employed: (1) total-min of activity, and (2) 10-min bout.

Results: Participants had similar baseline characteristics in both administration groups and no significant difference was found between the two formats in terms of validity (correlations between the GPAQ and accelerometer). For validity, the GPAQ demonstrated fair-to-moderate correlations for moderate-to-vigorous physical activity (MVPA) for self-administration (rs = 0.30) and interviewer-administration (rs = 0.46). Findings were similar when considering 10-min activity bouts in the accelerometer analysis for MVPA (rs = 0.29 vs. 0.42 for self vs. interviewer). Within each mode of administration, the strongest correlations were observed for vigorous-intensity activity. However, Bland-Altman plots illustrated bias toward overestimation for higher levels of MVPA, vigorous- and moderate-intensity activities, and underestimation for lower levels of these measures. Reliability for MVPA revealed moderate correlations (rs = 0.61 vs. 0.63 for self vs. interviewer).

Conclusions: Our findings showed comparability between both self- and interviewer-administration modes of the GPAQ. The GPAQ in general but especially the self-administered version may offer a relatively inexpensive method for measuring physical activity of various types and at different domains. However, there may be bias in the GPAQ measurements depending on the overall physical activity. It is advisable to incorporate accelerometers in future studies, particularly when measuring different intensities of physical activity.

No MeSH data available.


Flowchart of participants’ recruitment.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4556683&req=5

pone.0136944.g001: Flowchart of participants’ recruitment.

Mentions: A total of 110 participants were included in the test-retest reliability study, of which 108 participants who met the wear time criteria were included in the validity study. A flow chart of study participants’ recruitment is shown in Fig 1.


Reliability and Validity of the Self- and Interviewer-Administered Versions of the Global Physical Activity Questionnaire (GPAQ).

Chu AH, Ng SH, Koh D, Müller-Riemenschneider F - PLoS ONE (2015)

Flowchart of participants’ recruitment.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4556683&req=5

pone.0136944.g001: Flowchart of participants’ recruitment.
Mentions: A total of 110 participants were included in the test-retest reliability study, of which 108 participants who met the wear time criteria were included in the validity study. A flow chart of study participants’ recruitment is shown in Fig 1.

Bottom Line: Participants had similar baseline characteristics in both administration groups and no significant difference was found between the two formats in terms of validity (correlations between the GPAQ and accelerometer).Our findings showed comparability between both self- and interviewer-administration modes of the GPAQ.It is advisable to incorporate accelerometers in future studies, particularly when measuring different intensities of physical activity.

View Article: PubMed Central - PubMed

Affiliation: Saw Swee Hock School of Public Health, National University of Singapore, Singapore, Singapore.

ABSTRACT

Objective: The Global Physical Activity Questionnaire (GPAQ) was originally designed to be interviewer-administered by the World Health Organization in assessing physical activity. The main aim of this study was to compare the psychometric properties of a self-administered GPAQ with the original interviewer-administered approach. Additionally, this study explored whether using different accelerometry-based physical activity bout definitions might affect the questionnaire's validity.

Methods: A total of 110 participants were recruited and randomly allocated to an interviewer- (n = 56) or a self-administered (n = 54) group for test-retest reliability, of which 108 participants who met the wear time criteria were included in the validity study. Reliability was assessed by administration of questionnaires twice with a one-week interval. Criterion validity was assessed by comparing against seven-day accelerometer measures. Two definitions for accelerometry-data scoring were employed: (1) total-min of activity, and (2) 10-min bout.

Results: Participants had similar baseline characteristics in both administration groups and no significant difference was found between the two formats in terms of validity (correlations between the GPAQ and accelerometer). For validity, the GPAQ demonstrated fair-to-moderate correlations for moderate-to-vigorous physical activity (MVPA) for self-administration (rs = 0.30) and interviewer-administration (rs = 0.46). Findings were similar when considering 10-min activity bouts in the accelerometer analysis for MVPA (rs = 0.29 vs. 0.42 for self vs. interviewer). Within each mode of administration, the strongest correlations were observed for vigorous-intensity activity. However, Bland-Altman plots illustrated bias toward overestimation for higher levels of MVPA, vigorous- and moderate-intensity activities, and underestimation for lower levels of these measures. Reliability for MVPA revealed moderate correlations (rs = 0.61 vs. 0.63 for self vs. interviewer).

Conclusions: Our findings showed comparability between both self- and interviewer-administration modes of the GPAQ. The GPAQ in general but especially the self-administered version may offer a relatively inexpensive method for measuring physical activity of various types and at different domains. However, there may be bias in the GPAQ measurements depending on the overall physical activity. It is advisable to incorporate accelerometers in future studies, particularly when measuring different intensities of physical activity.

No MeSH data available.