The Ethical Collection of Athlete Data in Sport: A Realist Process

Citation

Kirkland A & Neupert E (2024) The Ethical Collection of Athlete Data in Sport: A Realist Process. In: 6th International Coaching Conference, University of Northumbria, Newcastle-Upon-Tyne, 14.06.2024-16.06.2024. Cluster in Coaching Research. https://cricsportcoaching.com/product/6th-international-coaching-conference-2024-registration-late/#program

Abstract
Background Recent growth in the collection data in sport, including geospatial; biomechanical; physiological; athlete management and medical, has been rapid. Whilst there a potential benefits, a report (Australian Academy of Science, 2022) raised ethical concerns in relation to this growth. Challenges highlighted were the: amount of data being collected; intimacy of the data; limited and often tenuous benefits; a lack of ethical oversight of handling and use of the data. Similarly, Williams and Manley (2014) raised concerns of a surveillance culture in sport, arguing that social influences of data collection can be detrimental to athlete wellbeing. Neupert et al. (2022) also reported that data collection methods can be implemented without clear-rationale, sound evidence, misinterpreted, or used as an athlete sanctioning tool. Notwithstanding, valid athlete data monitoring and evaluation can play an important role in coach decision-making in sport (Collins et al., 2015). We therefore suggest that the use of data in sport should be based on sound ethical principles, in which benefits, harms and risks are considered a-priori. Therefore, the purpose of this study was to design a framework for coaches and multi-disciplinary team (MDT) practitioners as a decision-making process to support ethical collection of athlete data. Methods and Model Description The underlying assumption of the framework is that data collection in sport represents a complex intervention (Craig et al., 2008) because decisions made surrounding its use are made within complex, open and dynamic social systems. Therefore, we used a Realist Evaluation Logic Model (Pawson and Tilley, 1997) as shown in Figure 1 to outline context-mechanism-outcome relationships on the pathway to effective data use. Data Inputs into the System The data collected within a system must be proportionate to the capacity to process it (Ashby, 1958). Capacity is complex and dependent on the capabilities (knowledge and skills), opportunities (including social agency) and motivations of system actors (Michie et al., 2005). System coherency is intrinsically linked to capacity, reflecting ability of actors to aggregate the data and work together towards impactful system outputs. Is the data likely to result in effective intervention(s)? We used the Medical Research Council process evaluation guidelines of complex interventions (Moore et al., 2015) to inform on model design. They suggest valid and reliable methods which work in context are key to intervention effectiveness. Consider complexity of decision-making and how the data may be used. Needs analyses processes are required to explore potential benefits, harms and risks within the sport specific context. Considerations for implementation are based on the model of Ingham (2016). Less tangible factors reflect social and philosophical dimensions of the model which merit consideration. Power relationships are illustrated by Nyberg’s (1981) model. What are the system outputs? The goal of the system to use data to inform on effective decision-making. Clarity of the benefits, potential harms and risks of collecting data should emerge through using the model a-priori to data collection. Keywords: Data, Complex systems, Ethics

Keywords
Data; Complex systems; Ethics