This article was downloaded by: [University of Guelph] On: 05 January 2015, At: 01:49 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Sports Sciences Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/rjsp20
TM
Reliability and validity of a GPS-enabled iPhone “app” to measure physical activity a
a
a
Amanda Clare Benson , Lyndell Bruce & Brett Ashley Gordon a
RMIT University, Discipline of Exercise Sciences, School of Medical Sciences, Australia Published online: 02 Jan 2015.
Click for updates To cite this article: Amanda Clare Benson, Lyndell Bruce & Brett Ashley Gordon (2015): Reliability and validity of a GPSTM
enabled iPhone
“app” to measure physical activity, Journal of Sports Sciences, DOI: 10.1080/02640414.2014.994659
To link to this article: http://dx.doi.org/10.1080/02640414.2014.994659
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Sports Sciences, 2014 http://dx.doi.org/10.1080/02640414.2014.994659
Reliability and validity of a GPS-enabled iPhoneTM “app” to measure physical activity
AMANDA CLARE BENSON, LYNDELL BRUCE & BRETT ASHLEY GORDON RMIT University, Discipline of Exercise Sciences, School of Medical Sciences, Australia
Downloaded by [University of Guelph] at 01:49 05 January 2015
(Accepted 1 December 2014)
Abstract This study assessed the validity and reliability of an iPhoneTM “app” and two sport-specific global positioning system (GPS) units to monitor distance, intensity and contextual physical activity. Forty (23 female, 17 male) 18–55-year-olds completed two trials of six laps around a 400-m athletics track wearing GPSports ProTM and WiSpiTM units (5 and 1 Hz) and an iPhoneTM with a Motion X GPSTM “app” that used the inbuilt iPhoneTM location services application programming interface to obtain its sampling rate (which is likely to be ≤1 Hz). Overall, the statistical agreement, assessed using t-tests and Bland–Altman plots, indicated an underestimation of the known track distance (2.400 km) and average speed by the Motion X GPSTM “app” and GPSports ProTM while the GPSports WiSpiTM device overestimated these outcomes. There was a ≤3% variation between trials for distance and average speed when measured by any of the GPS devices. Thus, the smartphone “app” trialled could be considered as an accessible alternative to provide high-quality contextualised data to enable ubiquitous monitoring and modification of programmes to ensure appropriate intensity and type of physical activity is prescribed and more importantly adhered to. Keywords: global positioning system, smartphone, aerobic exercise, physical activity, reliability and validity
Introduction Global positioning system (GPS) use in health research is becoming more prevalent. However, research to date has primarily used commercially available units and watches. Researchers and practitioners are always seeking ways to optimise population-based physical activity assessment as well as novel physical activity interventions to improve adherence and enhance health outcomes (Haskell et al., 2007; Nelson et al., 2007; Wilmot et al., 2012; Zinman, Ruderman, Campaigne, Devlin, & Schneider, 2004). As an objective measure of physical activity, GPS is more accurate than self-report (Badland, Duncan, Oliver, Duncan, & Mavoa, 2010; Duncan & Mummery, 2007; Stopher, Fitzgerald, & Xu, 2007), is less time consuming and impractical than direct observation (Maddison & Ni Mhurchu, 2009) and unlike pedometers and accelerometers provides contextual information of where activities occur (Krenn, Titze, Oja, Jones, & Ogilvie, 2011). Contextual information provides valuable insight when developing interventions to target specific cohorts (Chen, Janz, Zhu, & Brychta, 2012) and to enable the monitoring of adherence to prescribed physical activity.
GPS has been used for both navigation and by sporting teams to measure and monitor player performance for some time. Within a sporting context, the reliability and validity of various GPS units across a wide range of sports (Brewer, Dawson, Heasman, Stewart, & Cormack, 2010; Coutts & Duffield, 2010; Gray, Jenkins, Andrews, Taaffe, & Glover, 2010; Jennings, Cormack, Coutts, Boyd, & Aughey, 2010) has been conducted. While the sporting arena has been examined, there are limited data available on GPS units used to measure or monitor community-based physical activity (Maddison & Ni Mhurchu, 2009), with the majority of such studies investigating active transport and the role of the environment in physical activity (Krenn et al., 2011; Maddison & Ni Mhurchu, 2009). To our knowledge, no studies have measured physical activity using a smartphone and associated GPS applications (“apps”). While GPS units are popular among professional and semi-professional sporting teams, they do possess a number of limitations. The majority of commercially available units are generally expensive to purchase for individual use and therefore unlikely to be feasible for large-scale population-based research or community-
Correspondence: Amanda Clare Benson, RMIT University, Discipline of Exercise Sciences, School of Medical Sciences, PO Box 71, Melbourne 3083, Australia. E-mail: [email protected] © 2014 Taylor & Francis
Downloaded by [University of Guelph] at 01:49 05 January 2015
2
A. C. Benson et al.
based interventions. A single sport-specific unit can cost $2000–3000 AUD for the latest models, while a GPS-based watch can cost $250–500 AUD. As a result, if an individual wants to measure or monitor their personal physical activity level, they need to outlay a considerable amount to do so. Individuals may purchase or own heart rate-based watches with built-in GPS units; however, these units also have a single purpose and need to be purchased for a specific task (measuring physical activity). In contrast, smartphones are often already owned by the individual, and serve multiple purposes and therefore the additional outlay for an “app” is small. Currently, approximately half (46%) of US adults own a smartphone (Pew Internet Project, 2012). Similarly, Australian data show that approximately 59% of the population own a smartphone (Sensis e-Business Report, 2012). Physical activity has important health benefits; however, improving adherence to the recommended guidelines is challenging (Department of Health and Human Services, 1996; Haskell et al., 2007; Physical Activity Guidelines Advisory Committee, 2008; World Health Organisation, 2002, 2004). Finding ways to monitor physical activity on a large scale and within the participant’s own community are problematic (Marcus et al., 2000). Thus, the advent of mobile technology (smartphones and their associated “apps”), already embedded within an individual’s lifestyle, could potentially be a more accessible (cost-effective) method with less participant burden (having to carry or wear multiple devices) that could be used for population-based physical activity monitoring and interventions. The aim of the current study was to assess the validity and reliability of a common iPhoneTM “app” for the purpose of monitoring physical activity distance, duration, intensity and context and two commercially available sport-specific GPS devices (GPSports ProTM and GPSports WiSpiTM) which sample at different rates. This study will provide information on the ability of the Motion X GPSTM “app” to measure and monitor physical activity; to determine the usefulness for monitoring adherence to aerobic-based physical activity ubiquitously within the community.
Methods Study participants Forty apparently healthy participants aged 18–55 years were recruited for the study in Melbourne, Australia. Participants received information about the purpose and procedures of the study and provided written informed consent before participating. Ethical approval was received from the University Human Research Ethics Committee. Participants completed
a health risk-screening questionnaire to ensure no musculoskeletal injuries or medical conditions contraindicated for walking or running were present. Prior to beginning the first trial, height and body mass were measured using standardised procedures (Gore & Australian Sports Commission, 2000).
Experimental protocol This study evaluated the validity and reliability of an inexpensive GPS-enabled iPhoneTM “app” (Motion X GPSTM, Fullpower Technologies, Inc., Santa Cruz, CA, USA) that used the inbuilt iPhoneTM location services application programming interface (API) to obtain its sampling rate (which is likely to be ≤1 Hz) and two sport-specific GPS units that sampled at two different rates (5 and 1 Hz) (GPSports, ACT, Australia) during a walking and jogging protocol around a 400 m all-weather running track. Participants wore two GPS units and one iPhone™ with the GPS “app” while completing two trials of walking and jogging around the 400 m track for a total of six laps [2.400 km; two laps walking, three laps (100 m walk, 100 m light jog, 100 m walk, 100 m fast jog); one lap walking] to try and replicate the types and transitions between intensities that typically occur during aerobic-based physical activity. During each trial, participant wore the GPSports ProTM GPS unit on the middle of their back in a standard bib, a GPSports WiSpiTM GPS unit in an armband on their upper-right arm and the iPhoneTM using the Motion X GPSTM “app” in an armband on their upper-left arm. All units were turned on at the start and off at the end of each trial at the same location on the 400 m track. Data were cleaned to remove any data recorded prior to and after the end of each trial by finding the zero speed point for each unit and confirming it with the manually recorded start and end times for each trial. This was necessary as units require time for initialisation prior to beginning movement to ensure that the units obtain a satellite fix. Two trials were deemed, by two independent reviewers, to have outliers in the GPSports WiSpiTM trial 1 data and Motion X GPSTM trial 2 data and demonstrated a failure of the technology in recording and therefore were excluded from analysis. The Motion X GPSTM “app” was selected as it was inexpensive, to maximise generalisability, had no membership requirements and it enabled data to be emailed in a usable format for analysis. Ending a trial using the Motion X GPSTM “app” creates a “track” that was emailed after each trial. The GPSportsTM units were downloaded in a standard docking station and data analysed using the
iPhoneTM app to measure physical activity GPSportsTM software (GPSports Team AMS Release R1 2011.8; Canberra, ACT, Australia).
Statistical analysis All data were analysed using IBM SPSS Statistics 19 for Mac (SPSS, Chicago, IL, USA) with the exception of Bland–Altman analysis performed with GraphPad Prism 5.01 for Windows. Data are presented as mean ± s or as specified, with significance set at P ≤ 0.05.
Downloaded by [University of Guelph] at 01:49 05 January 2015
Validity Differences between GPS-measured distances and actual trial distances (2.400 km) were examined using paired t-tests. In addition, paired t-tests examined differences between calculated average speed [displacement (actual distance of 2.400 km)/time taken to complete the distance] used as a criterion and GPS device-measured average speed. Coefficient of variation (CV) and Bland–Altman bias and 95% limits of agreement were calculated for distance and average speed.
Reliability Intra-class correlation was used to assess the statistical correlation between trials for GPSports ProTM, Motion X GPSTM “app” and GPSports WiSpiTM derived results. Paired samples t-tests were performed to examine differences in distance travelled between trials (trial 1 versus trial 2) for each of the three GPS units. To further analyse the accuracy (bias) and assess for systematic errors, statistical agreement between trials for each of the GPS units for distance and average speed using a Bland– Altman plot were performed with GraphPad Prism 5.01 for Windows.
3
Results Forty participants (23 female and 17 male; age 24.93 ± 6.73 years; height 170.68 ± 7.12 cm; body mass 69.06 ± 9.04 kg) completed two trials using all three GPS-measuring tools simultaneously, with validity and reliability analysis conducted on data from 38 participants. All three devices recorded accurate contextual information about the location of the activity (data not shown). Validity When compared with the known track distance of 2.400 km, paired t-tests showed that the GPSports ProTM and the Motion X GPSTM “app” significantly underestimated the distance covered, while the GPSports WiSpiTM overestimated the distance (Table I). When comparing GPS-measured average speed with the calculated criterion for average speed, GPSports ProTM and Motion X GPSTM “app” both underestimated average speed, while the GPSports WiSpiTM overestimated the average speed (Table I). When trials 1 and 2 were combined, the bias ± s of bias between the actual distance of 2.400 km and GPSports ProTM, GPSports WiSpiTM and Motion X GPSTM “app” were −0.020 ± 0.014 km, 0.063 ± 0.018 km and −0.065 ± 0.094 km, respectively (Table I). When comparing the criterion for average speed [displacement (actual distance of 2.400 km)/time taken to complete the distance], the bias ± s of bias were −0.060 ± 0.090 km/h, 0.172 ± 0.091 km/h and −0.186 ± 0.275 km/h for GPSports ProTM, GPSports WiSpiTM and Motion X GPSTM “app”, respectively (Table I; Figure 1(A)–(F)). Reliability There were significant intra-class correlations for both distance and average speed between trials 1 = 0.97, and 2 for GPSports ProTM ( = 0.79;
∝
∝
Table I. Validity of distance and average speed for each GPS unit compared with the criterion value.
Measurement tool Criterion value GPSports ProTM GPSports WiSpiTM Motion X GPSTM “app”
Distance mean (s) (km)
Bias ± 95% LOA (km)
s CV difference (%)
P
Average speed mean (s) (km/h)
Bias ± 95% LOA (km/h)
s CV difference (%)
P
2.400 (0.000) – 2.380 (0.013) −0.020 ± 0.026 2.463 (0.017) 0.063 ± 0.035
– 0.01 0.02
– 6.783 (0.671) – 0.56
Journal of Sports Sciences Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/rjsp20
TM
Reliability and validity of a GPS-enabled iPhone “app” to measure physical activity a
a
a
Amanda Clare Benson , Lyndell Bruce & Brett Ashley Gordon a
RMIT University, Discipline of Exercise Sciences, School of Medical Sciences, Australia Published online: 02 Jan 2015.
Click for updates To cite this article: Amanda Clare Benson, Lyndell Bruce & Brett Ashley Gordon (2015): Reliability and validity of a GPSTM
enabled iPhone
“app” to measure physical activity, Journal of Sports Sciences, DOI: 10.1080/02640414.2014.994659
To link to this article: http://dx.doi.org/10.1080/02640414.2014.994659
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Sports Sciences, 2014 http://dx.doi.org/10.1080/02640414.2014.994659
Reliability and validity of a GPS-enabled iPhoneTM “app” to measure physical activity
AMANDA CLARE BENSON, LYNDELL BRUCE & BRETT ASHLEY GORDON RMIT University, Discipline of Exercise Sciences, School of Medical Sciences, Australia
Downloaded by [University of Guelph] at 01:49 05 January 2015
(Accepted 1 December 2014)
Abstract This study assessed the validity and reliability of an iPhoneTM “app” and two sport-specific global positioning system (GPS) units to monitor distance, intensity and contextual physical activity. Forty (23 female, 17 male) 18–55-year-olds completed two trials of six laps around a 400-m athletics track wearing GPSports ProTM and WiSpiTM units (5 and 1 Hz) and an iPhoneTM with a Motion X GPSTM “app” that used the inbuilt iPhoneTM location services application programming interface to obtain its sampling rate (which is likely to be ≤1 Hz). Overall, the statistical agreement, assessed using t-tests and Bland–Altman plots, indicated an underestimation of the known track distance (2.400 km) and average speed by the Motion X GPSTM “app” and GPSports ProTM while the GPSports WiSpiTM device overestimated these outcomes. There was a ≤3% variation between trials for distance and average speed when measured by any of the GPS devices. Thus, the smartphone “app” trialled could be considered as an accessible alternative to provide high-quality contextualised data to enable ubiquitous monitoring and modification of programmes to ensure appropriate intensity and type of physical activity is prescribed and more importantly adhered to. Keywords: global positioning system, smartphone, aerobic exercise, physical activity, reliability and validity
Introduction Global positioning system (GPS) use in health research is becoming more prevalent. However, research to date has primarily used commercially available units and watches. Researchers and practitioners are always seeking ways to optimise population-based physical activity assessment as well as novel physical activity interventions to improve adherence and enhance health outcomes (Haskell et al., 2007; Nelson et al., 2007; Wilmot et al., 2012; Zinman, Ruderman, Campaigne, Devlin, & Schneider, 2004). As an objective measure of physical activity, GPS is more accurate than self-report (Badland, Duncan, Oliver, Duncan, & Mavoa, 2010; Duncan & Mummery, 2007; Stopher, Fitzgerald, & Xu, 2007), is less time consuming and impractical than direct observation (Maddison & Ni Mhurchu, 2009) and unlike pedometers and accelerometers provides contextual information of where activities occur (Krenn, Titze, Oja, Jones, & Ogilvie, 2011). Contextual information provides valuable insight when developing interventions to target specific cohorts (Chen, Janz, Zhu, & Brychta, 2012) and to enable the monitoring of adherence to prescribed physical activity.
GPS has been used for both navigation and by sporting teams to measure and monitor player performance for some time. Within a sporting context, the reliability and validity of various GPS units across a wide range of sports (Brewer, Dawson, Heasman, Stewart, & Cormack, 2010; Coutts & Duffield, 2010; Gray, Jenkins, Andrews, Taaffe, & Glover, 2010; Jennings, Cormack, Coutts, Boyd, & Aughey, 2010) has been conducted. While the sporting arena has been examined, there are limited data available on GPS units used to measure or monitor community-based physical activity (Maddison & Ni Mhurchu, 2009), with the majority of such studies investigating active transport and the role of the environment in physical activity (Krenn et al., 2011; Maddison & Ni Mhurchu, 2009). To our knowledge, no studies have measured physical activity using a smartphone and associated GPS applications (“apps”). While GPS units are popular among professional and semi-professional sporting teams, they do possess a number of limitations. The majority of commercially available units are generally expensive to purchase for individual use and therefore unlikely to be feasible for large-scale population-based research or community-
Correspondence: Amanda Clare Benson, RMIT University, Discipline of Exercise Sciences, School of Medical Sciences, PO Box 71, Melbourne 3083, Australia. E-mail: [email protected] © 2014 Taylor & Francis
Downloaded by [University of Guelph] at 01:49 05 January 2015
2
A. C. Benson et al.
based interventions. A single sport-specific unit can cost $2000–3000 AUD for the latest models, while a GPS-based watch can cost $250–500 AUD. As a result, if an individual wants to measure or monitor their personal physical activity level, they need to outlay a considerable amount to do so. Individuals may purchase or own heart rate-based watches with built-in GPS units; however, these units also have a single purpose and need to be purchased for a specific task (measuring physical activity). In contrast, smartphones are often already owned by the individual, and serve multiple purposes and therefore the additional outlay for an “app” is small. Currently, approximately half (46%) of US adults own a smartphone (Pew Internet Project, 2012). Similarly, Australian data show that approximately 59% of the population own a smartphone (Sensis e-Business Report, 2012). Physical activity has important health benefits; however, improving adherence to the recommended guidelines is challenging (Department of Health and Human Services, 1996; Haskell et al., 2007; Physical Activity Guidelines Advisory Committee, 2008; World Health Organisation, 2002, 2004). Finding ways to monitor physical activity on a large scale and within the participant’s own community are problematic (Marcus et al., 2000). Thus, the advent of mobile technology (smartphones and their associated “apps”), already embedded within an individual’s lifestyle, could potentially be a more accessible (cost-effective) method with less participant burden (having to carry or wear multiple devices) that could be used for population-based physical activity monitoring and interventions. The aim of the current study was to assess the validity and reliability of a common iPhoneTM “app” for the purpose of monitoring physical activity distance, duration, intensity and context and two commercially available sport-specific GPS devices (GPSports ProTM and GPSports WiSpiTM) which sample at different rates. This study will provide information on the ability of the Motion X GPSTM “app” to measure and monitor physical activity; to determine the usefulness for monitoring adherence to aerobic-based physical activity ubiquitously within the community.
Methods Study participants Forty apparently healthy participants aged 18–55 years were recruited for the study in Melbourne, Australia. Participants received information about the purpose and procedures of the study and provided written informed consent before participating. Ethical approval was received from the University Human Research Ethics Committee. Participants completed
a health risk-screening questionnaire to ensure no musculoskeletal injuries or medical conditions contraindicated for walking or running were present. Prior to beginning the first trial, height and body mass were measured using standardised procedures (Gore & Australian Sports Commission, 2000).
Experimental protocol This study evaluated the validity and reliability of an inexpensive GPS-enabled iPhoneTM “app” (Motion X GPSTM, Fullpower Technologies, Inc., Santa Cruz, CA, USA) that used the inbuilt iPhoneTM location services application programming interface (API) to obtain its sampling rate (which is likely to be ≤1 Hz) and two sport-specific GPS units that sampled at two different rates (5 and 1 Hz) (GPSports, ACT, Australia) during a walking and jogging protocol around a 400 m all-weather running track. Participants wore two GPS units and one iPhone™ with the GPS “app” while completing two trials of walking and jogging around the 400 m track for a total of six laps [2.400 km; two laps walking, three laps (100 m walk, 100 m light jog, 100 m walk, 100 m fast jog); one lap walking] to try and replicate the types and transitions between intensities that typically occur during aerobic-based physical activity. During each trial, participant wore the GPSports ProTM GPS unit on the middle of their back in a standard bib, a GPSports WiSpiTM GPS unit in an armband on their upper-right arm and the iPhoneTM using the Motion X GPSTM “app” in an armband on their upper-left arm. All units were turned on at the start and off at the end of each trial at the same location on the 400 m track. Data were cleaned to remove any data recorded prior to and after the end of each trial by finding the zero speed point for each unit and confirming it with the manually recorded start and end times for each trial. This was necessary as units require time for initialisation prior to beginning movement to ensure that the units obtain a satellite fix. Two trials were deemed, by two independent reviewers, to have outliers in the GPSports WiSpiTM trial 1 data and Motion X GPSTM trial 2 data and demonstrated a failure of the technology in recording and therefore were excluded from analysis. The Motion X GPSTM “app” was selected as it was inexpensive, to maximise generalisability, had no membership requirements and it enabled data to be emailed in a usable format for analysis. Ending a trial using the Motion X GPSTM “app” creates a “track” that was emailed after each trial. The GPSportsTM units were downloaded in a standard docking station and data analysed using the
iPhoneTM app to measure physical activity GPSportsTM software (GPSports Team AMS Release R1 2011.8; Canberra, ACT, Australia).
Statistical analysis All data were analysed using IBM SPSS Statistics 19 for Mac (SPSS, Chicago, IL, USA) with the exception of Bland–Altman analysis performed with GraphPad Prism 5.01 for Windows. Data are presented as mean ± s or as specified, with significance set at P ≤ 0.05.
Downloaded by [University of Guelph] at 01:49 05 January 2015
Validity Differences between GPS-measured distances and actual trial distances (2.400 km) were examined using paired t-tests. In addition, paired t-tests examined differences between calculated average speed [displacement (actual distance of 2.400 km)/time taken to complete the distance] used as a criterion and GPS device-measured average speed. Coefficient of variation (CV) and Bland–Altman bias and 95% limits of agreement were calculated for distance and average speed.
Reliability Intra-class correlation was used to assess the statistical correlation between trials for GPSports ProTM, Motion X GPSTM “app” and GPSports WiSpiTM derived results. Paired samples t-tests were performed to examine differences in distance travelled between trials (trial 1 versus trial 2) for each of the three GPS units. To further analyse the accuracy (bias) and assess for systematic errors, statistical agreement between trials for each of the GPS units for distance and average speed using a Bland– Altman plot were performed with GraphPad Prism 5.01 for Windows.
3
Results Forty participants (23 female and 17 male; age 24.93 ± 6.73 years; height 170.68 ± 7.12 cm; body mass 69.06 ± 9.04 kg) completed two trials using all three GPS-measuring tools simultaneously, with validity and reliability analysis conducted on data from 38 participants. All three devices recorded accurate contextual information about the location of the activity (data not shown). Validity When compared with the known track distance of 2.400 km, paired t-tests showed that the GPSports ProTM and the Motion X GPSTM “app” significantly underestimated the distance covered, while the GPSports WiSpiTM overestimated the distance (Table I). When comparing GPS-measured average speed with the calculated criterion for average speed, GPSports ProTM and Motion X GPSTM “app” both underestimated average speed, while the GPSports WiSpiTM overestimated the average speed (Table I). When trials 1 and 2 were combined, the bias ± s of bias between the actual distance of 2.400 km and GPSports ProTM, GPSports WiSpiTM and Motion X GPSTM “app” were −0.020 ± 0.014 km, 0.063 ± 0.018 km and −0.065 ± 0.094 km, respectively (Table I). When comparing the criterion for average speed [displacement (actual distance of 2.400 km)/time taken to complete the distance], the bias ± s of bias were −0.060 ± 0.090 km/h, 0.172 ± 0.091 km/h and −0.186 ± 0.275 km/h for GPSports ProTM, GPSports WiSpiTM and Motion X GPSTM “app”, respectively (Table I; Figure 1(A)–(F)). Reliability There were significant intra-class correlations for both distance and average speed between trials 1 = 0.97, and 2 for GPSports ProTM ( = 0.79;
∝
∝
Table I. Validity of distance and average speed for each GPS unit compared with the criterion value.
Measurement tool Criterion value GPSports ProTM GPSports WiSpiTM Motion X GPSTM “app”
Distance mean (s) (km)
Bias ± 95% LOA (km)
s CV difference (%)
P
Average speed mean (s) (km/h)
Bias ± 95% LOA (km/h)
s CV difference (%)
P
2.400 (0.000) – 2.380 (0.013) −0.020 ± 0.026 2.463 (0.017) 0.063 ± 0.035
– 0.01 0.02
– 6.783 (0.671) – 0.56
