Purpose:Current Direct Observation(DO)methods are burdensome to the observer and relevant limitations can be identified on the outcome information.This study aims to characterize DO from the observers’point of view a...Purpose:Current Direct Observation(DO)methods are burdensome to the observer and relevant limitations can be identified on the outcome information.This study aims to characterize DO from the observers’point of view and to analyze the feasibility and usefulness of visual information in the form of video.Method:data collection was done with ten adult males,observed while performing different physical activities in an indoor court.Sessions were video recorded by two cameras.Three observers performed systematic observation,using SOPLAY,with different sampling rates.Inter observer’s agreement and with reference data was analyzed by Cohen’s Kappa statistic.Results:SOPLAY highest agreement between observers was 0.611 on vigorous category and walking category had the lowest 0.188.It doubles the time needed to annotate the video with pauses,but half of player’s behavior is preserved.Conclusion(s):using video to support DO permits to collect more accurate data and a more detailed behavior categorization is warranted.Developments in computer vision are expected to reduce the human efforts in DO methods.展开更多
基金This work was funded by the Portuguese Foundation for Science and Technology(FCT),grant SFRH/BPD/71332/2010,Pest-OE/SAU/UI0617/2011 and UIDB/50014/2020The work was also developed in the context of Project QREN 33910 ARENA,a R&D project funded by ERDF through ON2 as part of the NSRF,and managed by IAPMEI-Agência para a Competitividade e Inovação,I.P..
文摘Purpose:Current Direct Observation(DO)methods are burdensome to the observer and relevant limitations can be identified on the outcome information.This study aims to characterize DO from the observers’point of view and to analyze the feasibility and usefulness of visual information in the form of video.Method:data collection was done with ten adult males,observed while performing different physical activities in an indoor court.Sessions were video recorded by two cameras.Three observers performed systematic observation,using SOPLAY,with different sampling rates.Inter observer’s agreement and with reference data was analyzed by Cohen’s Kappa statistic.Results:SOPLAY highest agreement between observers was 0.611 on vigorous category and walking category had the lowest 0.188.It doubles the time needed to annotate the video with pauses,but half of player’s behavior is preserved.Conclusion(s):using video to support DO permits to collect more accurate data and a more detailed behavior categorization is warranted.Developments in computer vision are expected to reduce the human efforts in DO methods.
