In recent years, scrutiny on sport-science research has intensified from both internal and external sources.1,2 Several debates have arisen concerning methodological and theoretical issues, such as magnitude-based inferences (MBI)3 and the acute chronic workload ratio (ACWR).4 For example, MBI is a statistical method developed specifically for use in sport science research to address the perceived inability of inferential statistics (null-hypothesis testing) to represent uncertainty and the prevalence of p-hacking.3,5,6 MBI has been applied in hundreds of peer-reviewed sport science journal articles. However, MBI has been criticized as statistically and mathematically flawed.7 A further recent high-profile methodological debate is the ACWR.8 Proponents of the ACWR claim that injury may be predicted by calculating the ratio between acute and chronic training loads.3 Similar to MBI, the ACWR has been criticized as a statistically flawed method, despite its use in numerous peer-reviewed publications.1,8 These high-profile cases, along with other longstanding issues in sport science, including reductionism versus holism,9,10 monodisciplinary research,11 research-practice gaps,12,13 and a lack of accepted concepts,14 highlight that multiple issues are negatively impacting the discipline.
Scientific debate is healthy, as it ensures that attempts are made at self-correcting; however, efforts to understand the complex issues within sport science research have mostly focused on isolated issues, such as methodological or statistical flaws,2 philosophical research approaches,15 and the research practice gap.12,13 In recent times, systems thinking has emerged as a useful approach in sport science research.10,16–18 From a systems thinking perspective, it is important to understand the array of systemic influences that contribute to issues. Understanding complex issues can only be achieved by taking the overall system as the unit of analysis.19 In this article, we argue that, as sport science research represents a complex system,20 it is useful to consider the broader system, its interrelated components, and system behavior when attempting to understand how and why issues emerge, as well as how they can be resolved.
Some of the world’s most complex problems have been investigated and better understood using systems thinking methods.21,22 One such method, causal loop diagrams (CLDs), has been used to better understand and respond to complex crises, including COVID-19,23 obesity,22 climate change,24 food security,25 terrorism,26 and transport-related trauma,27 to name a few. These analyses have guided policy makers to develop and apply systemic interventions that will enact change on the goals of the system and, subsequently, the systems’ behavior.
All systems are composed of interacting networks of reinforcing (positive) and balancing (negative) feedback loops that influence system behavior and equilibrium.28 Reinforcing (or positive feedback) loops are actions that afford change in one direction to produce even more change in the same direction, whereas balancing (or negative feedback) loops work to keep the system in a state of equilibrium. Balancing feedback loops resist change in one direction by producing change in the opposite direction.28,29 CLDs provide a method to represent these dynamic interrelations via visual representation, which assists in communicating the complexity of a given system.29,30 CLDs are composed of variables connected by arrows, which depict the causal influences between the variables (Figure 1).28
—Example causal loop diagram showing the reinforcing loop (R) of research funding on publications and publications on research funding. + indicates a positive influence on the variables.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
—Example causal loop diagram showing the reinforcing loop (R) of research funding on publications and publications on research funding. + indicates a positive influence on the variables.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
—Example causal loop diagram showing the reinforcing loop (R) of research funding on publications and publications on research funding. + indicates a positive influence on the variables.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
Systems thinking methods are gaining traction in sports research, including performance analysis, injury, coaching, and sport science generally,9,16,31–34 and such approaches may be suited to understanding the pertinent issues associated with sport science research. In this article, we argue that CLD provides a useful method to help understand the complex set of interacting variables, which influence the quality and impact of sport science research. The aim of this study was to develop a CLD to model the interacting network of variables within the sport science research “system” to understand the dynamics of the system, which influence key disciplinary issues (Figure 1).
Methods
Study Design
This qualitative study was designed to develop a CLD depicting the positive and negative feedback loops influencing the dynamic behavior of the sport science research system. Data for the current study were sourced from relevant peer-reviewed literature and subject-matter experts (SMEs). The CLD was created by the study authors using a group model-building process,28,35 using Vensim software (Ventana Systems Inc, Harvard, MA). The boundary for the analysis included the general academic system through to sport science research processes and to sporting practice, for example, sport science research activities that are undertaken by academics working for an academic institution. Other factors outside of this boundary that are not considered include governments, the economy, academic institutions, sporting associations, and professional clubs and academies, as well as the history of sport science. These are, however, represented within the CLD to provide context of their influence on the system.
Procedure
There are 2 common methods of CLD development.35 The first involves researchers constructing models using data derived from sources such as the peer-reviewed literature and organizational documentation. The second involves engaging relevant SMEs to assist in model building via workshops, surveys, or Delphi studies.23 The involvement of SMEs is critical, with research demonstrating the need to incorporate multiple stakeholder perspectives and, ultimately, enhance model validity.22 In the present study, both processes were utilized. First, peer-reviewed literature was sourced to identify issues with academia in general and, specifically, within sport science, which were subsequently included in the CLD. Second, a group model-building process involving the authors was undertaken. The authors were considered to be SMEs for the sport science academic system, with extensive experience in academic and sport science research,36,37 systems modeling in sport,17,34 and systems thinking and systems modeling applications generally.38–40 Combined, the authors have published more than 300 peer-reviewed journal articles and 21 books across several disciplines, including 2 books on methods applications in sport.41,42
Model-building workshops were conducted in person and via video conferencing. Each of the CLD variables, feedback loops, and link polarity (+ or −) were reviewed and refined, and any disagreements were discussed until consensus was achieved. Variables for the CLD were described as nouns or noun phrases where possible, which allows a determination of the direction of a variable with feedback; for example, variables can fluctuate up or down based on the type of feedback. Upon completion of the final CLD, reinforcing loops (R) were identified to understand where a certain behavior was driving more of the same behavior.43 Finally, causal trees were used to demonstrate the systemic influence of key variables within the CLD. Causal trees demonstrate, through direct and indirect links, the systemic influence of variables across the system on individual variables. In the current analysis, the causal trees were presented, showing the primary and secondary influencing variables.
Results
The sport science research system CLD is presented (Figure 2). In total, 48 interacting variables were identified, many of which are contributing to issues with sport science research. Definitions for each of the CLD variables are provided as Supplementary Material (available online). Within the CLD, 3 distinct segments were identified (left to right), the sport science research system rules and structures (left/green), the processes and behaviors of the system (middle/orange), and the outcomes/performance of the system (right/blue) (Figure 2). Eleven reinforcing loops (R) were identified that are potentially enabling recurring systemic issues (Table 1). Causal trees are presented to show the systemic influences on, and influence of, methodological issues (Figure 3), as well as the variables interacting to create a research-practice gap (Figure 4). The methodological issues in sport science research are influenced by multiple interacting variables from the general academic system and the sport science system processes, which are then propagated onto other processes and overall system performance (Figure 3). In the CLD, the research-practice gap was influenced by 5 primary variables: accuracy/validity/reliability, conclusions not supported by data, ecological validity, perception of utility of sport science research, and translation to practice (Figure 4).
—Causal loop diagram of the sport-science research system. Shaded circles indicate factors outside of the study boundary. R indicates reinforcing loop. The sport-science system structure is represented on the left of the diagram, the processes and behaviors are represented in the middle of the diagram, and the system outcomes and performance are represented on the right of the diagram.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
—Causal loop diagram of the sport-science research system. Shaded circles indicate factors outside of the study boundary. R indicates reinforcing loop. The sport-science system structure is represented on the left of the diagram, the processes and behaviors are represented in the middle of the diagram, and the system outcomes and performance are represented on the right of the diagram.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
—Causal loop diagram of the sport-science research system. Shaded circles indicate factors outside of the study boundary. R indicates reinforcing loop. The sport-science system structure is represented on the left of the diagram, the processes and behaviors are represented in the middle of the diagram, and the system outcomes and performance are represented on the right of the diagram.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
Reinforcing Loop Descriptions
| Reinforcing loop | Title |
|---|---|
| R1 | Academic survival loop |
| R2 | Journal quality loop |
| R3 | Publishing loop |
| R4 | Workload loop |
| R5 | Funding loop |
| R6 | Reductionism loop |
| R7 | Social media loop |
| R8 | Blowhard loop |
| R9 | Blackbox loop |
| R10 | Conceptual model loop |
| R11 | Bias loop |
—Causal tree showing the systemic influences on and because of methodological issues. Variables in parentheses indicate that the variable is represented elsewhere within the causal tree.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
—Causal tree showing the systemic influences on and because of methodological issues. Variables in parentheses indicate that the variable is represented elsewhere within the causal tree.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
—Causal tree showing the systemic influences on and because of methodological issues. Variables in parentheses indicate that the variable is represented elsewhere within the causal tree.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
—Causal tree for systemic influences on the research–practice gap. Variables in parentheses indicate that the variable is represented elsewhere in the causal tree.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
—Causal tree for systemic influences on the research–practice gap. Variables in parentheses indicate that the variable is represented elsewhere in the causal tree.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
—Causal tree for systemic influences on the research–practice gap. Variables in parentheses indicate that the variable is represented elsewhere in the causal tree.
Citation: International Journal of Sports Physiology and Performance 16, 10; 10.1123/ijspp.2020-0934
Discussion
All models are wrong, but some are useful.44
The impetus for the CLD was to attempt to identify and understand the systemic factors influencing the quality and impact of sport science research. It is acknowledged that there is an ongoing body of ethical and high-quality sport science research being conducted across the world, evidenced by a retraction rate of 3.44 per 10,000 publications between 2010 and 2018.45 However, for sport science to have greater impact, an understanding of the systemic factors that interact to create issues is required, followed by interventions that will enact positive change on the entire system. Due to the complexity of the model, it is not possible to discuss each of its components; rather, we discuss key behaviors of the entire system through well-known dynamics of complex systems.21,29,43,46
First, the CLD demonstrates the inherent complexity of the sport science research system. Multiple interrelated variables were identified, along with multiple reinforcing and balancing feedback loops. This suggests that, as a complex system, there will be no quick fix to many of the issues identified here and being discussed in the sport science literature. For example, appropriate corrective actions are often not a quick fix, involve intervention on multiple variables, and require a delay to bring the current state of the system toward the intended goal.47 Moreover, it is clear that many of the interventions suggested in the literature, which focus on issues in isolation, will do little to improve the impact of sport science research.
The CLD demonstrates how the general academic system influences and creates many of the issues impacting sport science research. In particular, according to the CLD, 3 key sets of variables within the general academic system have a negative influence on scientific rigor and associated variables throughout the system. The first set of variables identified are associated with peer-reviewed academic journals, including journal revenue generation, journal quality, and peer reviewing. The second set of variables identified are associated with job security and promotion, pressure to publish, fixation on bibliometrics, and mechanisms around securing research funding. The third set of variables identified are associated with academic workload, quality of supervision, and higher-degree research (HDR) education and training. According to the CLD, these 3 sets of variables have a strong and negative influence on sports science research. As such, it can be inferred that the academic system, in which the sport science research system is embedded, is driving many of the issues that reduce the quality and impact of sport science research. Interestingly, these variables from the general academic system could apply to any scientific discipline and are not limited to sport science. This finding indicates a potential problem with the overall academic system, which may negatively impact research in general.
Multiple reinforcing loops (R) that potentially create or enable recurring systemic issues were identified. Many of these loops were identified in the general academic system segment of the CLD. There is a need to keep these reinforcing loops in check, as a system with unchecked reinforcing loops will ultimately destroy itself.43 According to the CLD, the academic survival loop (R1), which is one of the main institutional structures, is creating multiple emergent factors within the system, such as increased academic workloads and decreased scientific rigor. The academic survival loop is also creating a known system archetype, called “success to the successful,”29 that is widespread throughout academia. This suggests that success is as dependent on the system structure as it is on talent and quality of research.30 For example, if we consider 2 fictitious academics of equal ability, one having job security and the other not having job security and both having the academic pressures indicated in the CLD, the pressures on the academic without job security (or ongoing funding) are heightened. From this viewpoint, it is easy then to see how the academic survival system structure will produce unintended consequences, such as less rigorous science in order to get quick academic outputs (eg, publications) and enhanced bibliometrics to support funding applications. This issue of quantity over quality in terms of research projects and outputs is well known in many areas.48,49 The CLD here demonstrates how these behaviors negatively impact the entire system, yet they are an inherent part of its structure. Interventions to reduce the issues within the sport science research should aim to introduce balancing loops to slow or impede self-reinforcing system behaviors.
Sensitive dependence on initial starting conditions is known to be a key driver of behavior within complex systems.20,50 That is, decisions and actions made at the onset of a system (even many years prior) influence current behavior.50 This was revealed in the present analysis, with the history of sport science shown to be driving multiple behaviors within the system. In particular, the CLD demonstrates that many of the values or approaches adopted during the early years of sport science research are continuing to drive behavior in a manner that may be detrimental to outcomes. These include the value placed on quantifiable data,51 reductionism, and linear thinking.9 A positive for sport science research is that we are witnessing a greater proportion of research applying qualitative or mixed methods, studies applying systems thinking and complexity science, and the emergence of numerous multidisciplinary journals. The CLD suggests that further such shifts in theoretical and methodological approaches are critical requirements for future sports science research.
Methodological issues are one of the most debated issues affecting the quality of sport science research.1,5,7,14 This is supported by the CLD, which shows multiple factors influencing and contributing to methodological issues, as well as the reverberations of these issues throughout the system. Although numerous articles have indicated that this is where interventions are required,2,5 the CLD indicates that such solutions may have low leverage for changing systemic behavior.47 The blackbox (R9), conceptual model (R10), and bias (R11) reinforcing loops indicate that methodological issues will likely continue if they remain unchecked. The CLD indicates that systemic structures will not be changed by targeting interventions directly at the methodological issues themselves, as other variables will continue to influence research. For example, simply arguing that sport science needs more collaboration with statistical experts or that sample sizes need to be bigger is not enough and could even increase confusion and create further methodological issues. Rather, interventions should be targeted at the variables influencing the methodological issues. These include variables around the quality of HDR supervision and the education and supervision of PhD students and early career researchers, as well as variables in the CLD influencing scientific rigor. Desirable proceeding links to methodological issues in the CLD could include a change to the variables influencing academic workload in order to improve the quality of HDR supervision and the quality of education/training.
Emanating from the general academic system of the CLD are the system processes and behaviors undertaken within sport science research. These include many well-documented issues, including statistical and equipment black boxism,5 conclusions not supported by data,52 and overstated practical implications.14 Most pertinent are the numerous variables interacting to create the research practice gap. As demonstrated in the CLD, the research practice gap is created and facilitated by various interacting variables from across the entire system. Without sports practice, there is no sport science; therefore, research aimed at improving practice should be the goal of the entire system. Changing the goal or purpose of the system is one of the most powerful ways to enact systemic change,43 as the system structures, functions and processes will conform to the goal.47 However, the goal of the current sport science research system appears to be survival in academia for many researchers, which encourages vastly different behaviors around the number of publications, bibliometrics, and obtaining grant funding. Indeed, simply changing the goal of the sports science research system is likely not possible, given its current structure and the influence of academic system-related loops. A key finding from this study is the need to revisit and clarify the goals of sport science research, as well as communicating them clearly and often to researchers at all levels.
A final consideration is that interventions to improve sport science research, should be targeted at systemic change. There are several models and frameworks that have attempted to improve the impact of sport science research in practice12,53; however, they have not been aimed at the broader system. The Applied Research Model for Sports Sciences12 provides a framework that aims to improve the uptake of sport science research in practice. The Applied Research Model for Sports Sciences is comprehensive and highly applicable; however, it was not developed to enact change on the entire system, particularly the variables identified within in the present CLD. Despite the Applied Research Model for Sports Sciences being published 12 years ago, there remains a substantial research practice gap in sport science.13 This highlights how recent calls for targeting interventions specifically at methodological issues may be too narrow a focus to enact systemic change. As such, interventions aimed at systemic change are required to enhance the reputation of sport science as a scientific discipline markedly impacting practice. This will require alternative thinking. Some noteworthy interventions have already commenced; for example, the introduction of preregistered reports could have substantial change on methodological issues.54 In addition, an open research framework would promote transparent, open, and reproducible science and, subsequently, more credible research.55 Furthermore, industry placements in sport science degree programs and master’s degree programs will emphasize the importance of applied sport science research.
There appears to be less innovation when it comes to changing the general academic system, which is potentially driving many of the identified issues. Academic metrics and track records were designed to improve the system,49 yet the CLD suggests they are potentially damaging its discipline subsystems. Instead, measures of research’s impact on sports practice may be a more appropriate metric for sport science research. Such measures are especially important if the goal of the system was changed to focus on improving practice; however, it is acknowledged that, without a similar change to the goals of the broader academic system, the sports science research system would quickly collapse if academic metrics were abandoned. It is our contention that positive systemic change to sport science research will only be made by interventions that influence the entire system behavior, which includes the academic system generally. While such change is difficult to enact, the CLD presented in this article suggests that it is the only way in which sport science research can achieve its primary aim of optimizing sports practice.
Limitations
A limitation of the current study is the small sample of SMEs to build the CLD. However, the SMEs have extensive experience within the academic system, sport science, and systems modeling in multiple domains across Australia, the United Kingdom, and Europe. An extension of the current study could be to conduct a Delphi study with global sport science researchers to review, refine, and validate the CLD. A future research direction could be to develop a CLD of the optimal sport science research system to identify differences between that and the current model and appropriate interventions.
Practical Implications
The variables within the CLD should be considered when designing, conducting, and disseminating sport science research. While it is unlikely that the academic system will change anytime soon, the CLD highlights multiple variables where alternative interventions could be implemented to improve sports science research. As stated in other literature,13 removing the research practice gap will require collaboration between researchers and practitioners. Stakeholders could develop CLDs using group model building to develop and design programs of research to better understand complex issues in sport. Such a process would enable the alignment of mental models, the achievement of consensus, and group involvement around decision making between parties.
Conclusions
This study involved the development of a CLD to depict the causal structure of current issues in sports science research. It concluded that the general academic system negatively influences multiple variables associated with scientific research in general, including sport science. Furthermore, the early traditions of sport science may be contributing to current issues, such as reductionism, linear thinking, and causality. Improving methodological issues has long been a focus within sport science research; however, the current analysis suggests that multiple variables influencing methodological issues may be a better place to intervene to enact systemic change. Finally, the emergent outcomes of the interactions from the variables included in the CLD are creating a poorly performing system, as indicated by a lack of ecological validity, translation to practice, and, ultimately, a research-practice gap.
References
- 1.↑
Impellizzeri FM, Tenan MS, Kempton T, Novak A, Coutts AJ. Acute:chronic workload ratio: conceptual issues and fundamental pitfalls. Int J Sports Physiol Perform. 2020;15(6):907–913. doi:10.1123/ijspp.2019-0864
- 2.↑
Sainani KL, Borg DN, Caldwell AR, et al. Call to increase statistical collaboration in sports science, sport and exercise medicine and sports physiotherapy. Br J Sports Med. 2021;55(2):118–122. PubMed ID: 32816788 doi:10.1136/bjsports-2020-102607
- 3.↑
Batterham AM, Hopkins WG. Making meaningful inferences about magnitudes. Int J Sports Physiol Perform. 2006;1(1):50–57. doi:10.1123/ijspp.1.1.50
- 4.↑
Hulin BT, Gabbett TJ, Blnach P, Chapman P, Bailey D, Orchard JW. Spikes in acute workload are associated with increased injury risk in elite cricket fast bowlers. Br J Sports Med. 2014;48(8):708–712. PubMed ID: 23962877 doi:10.1136/bjsports-2013-092524
- 5.↑
Lohse K, Sainani K, Taylor JA, Butson ML, Knight E, Vickers A. Systematic review of the use of “magnitude-based inference” in sports science and medicine. PLoS One. 2020;15(6):e0235318. PubMed ID: 32589653 doi:10.1371/journal.pone.0235318
- 6.↑
Bernards JR, Sato K, Haff GG, Bazyler CD. Current research and statistical practices in sport science and a need for change. Sports. 2017;5(4):87. doi:10.3390/sports5040087
- 7.↑
Sainani KL, Lohse KR, Jones PR, Vickers A. Magnitude-based inference is not Bayesian and is not a valid method of inference. Scand J Med Sci Sports. 2019;29(9):1428–1436. PubMed ID: 31149752 doi:10.1111/sms.13491
- 8.↑
Maupin D, Schram B, Canetti E, Orr R. The relationship between acute:chronic workload ratios and injury risk in sports: a systematic review. Open Access J Sports Med. 2020;11:51–75. PubMed ID: 32158285 doi:10.2147/OAJSM.S231405
- 9.↑
McLean S, Read GJM, Hulme A, et al. Beyond the tip of the iceberg: using systems archetypes to understand common and recurring issues in sports coaching. Front Sports Act Living. 2019;1:49. PubMed ID: 33344972 doi:10.3389/fspor.2019.00049
- 10.↑
Salmon PM, McLean S. Complexity in the beautiful game: implications for football research and practice. Sci Med Football. 2020;4(2):162–167. doi:10.1080/24733938.2019.1699247
- 11.↑
Piggott B, Müller S, Chivers P, Papaluca C, Hoyne G. Is sports science answering the call for interdisciplinary research? A systematic review. Eur J Sport Sci. 2019;19(3):267–286. PubMed ID: 30198825 doi:10.1080/17461391.2018.1508506
- 12.↑
Bishop D. An applied research model for the sport sciences. Sports Med. 2008;38(3):253–263. PubMed ID: 18278985 doi:10.2165/00007256-200838030-00005
- 13.↑
Fullagar HH, McCall A, Impellizzeri FM, Favero T, Coutts A. The translation of sport science research to the field: a current opinion and overview on the perceptions of practitioners, researchers and coaches. Sports Med. 2019;49(12):1817–1824. PubMed ID: 31214978 doi:10.1007/s40279-019-01139-0
- 14.↑
Impellizzeri FM, Ward P, Coutts AJ, Bornn L, McCall A. Training load and injury: part 2—questionable research practices hijack the truth and Mislead well-intentioned clinicians. J Orthop Sports Phys Ther. 2020;50(10):577–584. doi:10.2519/jospt.2020.9211
- 15.↑
Marino FE. Is it time to retire the ‘central governor’? A philosophical and evolutionary perspective. Sports Med. 2010;40(3):265–268. PubMed ID: 20229620 doi:10.2165/11533680-000000000-00000
- 16.↑
Duarte R, Araújo D, Folgado H, Esteves P, Marques P, Davids K. Capturing complex, non-linear team behaviours during competitive football performance. J Syst Sci Complex. 2013;26(1):62–72. doi:10.1007/s11424-013-2290-3
- 17.↑
McLean S, Salmon PM, Gorman AD, Read GJ, Solomon C. What’s in a game? A systems approach to enhancing performance analysis in football. PLoS One. 2017;12(2):e0172565. PubMed ID: 28212392 doi:10.1371/journal.pone.0172565
- 18.↑
Hulme A, McLean S, Read GJM, Dallat C, Bedford A, Salmon PM. Sports organisations as complex systems: using cognitive work analysis to identify the factors influencing performance in an elite Netball organisation. Front Sports Act Living. 2019;1:56. PubMed ID: 33344979 doi:10.3389/fspor.2019.00056
- 20.↑
Cilliers P. ‘Complexity and postmodernism. Understanding complex systems’ Reply to David Spurrett. S Afr J Philos. 1999;18(2):275–278. doi:10.1080/02580136.1999.10878188
- 21.↑
Forrester JW. Industrial dynamics. A major breakthrough for decision makers. Harv Bus Rev. 1958;36(4):37–66
- 22.↑
Allender S, Owen B, Kuhlberg J, et al. A community based systems diagram of obesity causes. PLoS One. 2015;10(7):e0129683. PubMed ID: 26153893 doi:10.1371/journal.pone.0129683
- 23.↑
Sahin O, Salim H, Suprun E, et al. Developing a preliminary causal loop diagram for understanding the wicked complexity of the COVID-19 pandemic. Systems. 2020;8(2):20. doi:10.3390/systems8020020
- 24.↑
Sterman JD. Communicating climate change risks in a skeptical world. Clim Change. 2011;108(4):811. doi:10.1007/s10584-011-0189-3
- 25.↑
Purwanto A, Sušnik J, Suryadi F, de Fraiture C. Using group model building to develop a causal loop mapping of the water-energy-food security nexus in Karawang Regency, Indonesia. J Clean Prod. 2019;240:118170. doi:10.1016/j.jclepro.2019.118170
- 26.↑
Schoenenberger L, Schenker-Wicki A, Beck M. Analysing terrorism from a systems thinking perspective. Perspect Terror. 2014;8(1):16–36.
- 27.↑
Shepherd S. A review of system dynamics models applied in transportation. Transp B Transp Dyn. 2014;2(2):83–105.
- 28.↑
Sterman JD. Business dynamics: systems thinking and modeling for a complex world. Working Paper Series, ESD Internal Symposium; 2002.
- 29.↑
Senge PM. The Fifth Discipline. The Art and Practice of the Learning Organization. New York, NY: Doubleday/Currency; 1990.
- 30.↑
Kim DH. Systems Archetypes I: Diagnosing Systemic Issues and Designing High-Leverage Interventions. Waltham, MA: Pegasus Communications; 1993.
- 31.↑
Hulme A, Mclean S, Salmon PM, Thompson J, Lane BR, Nielsen RO. Computational methods to model complex systems in sports injury research: agent-based modelling (ABM) and systems dynamics (SD) modeling. Br J Sports Med. 2019;53(24):1507–1510. PubMed ID: 30448782 doi:10.1136/bjsports-2018-100098
- 32.
Bittencourt N, Meeuwisse W, Mendonça L, Nettel-Aguirre A, Ocarino J, Fonseca S. Complex systems approach for sports injuries: moving from risk factor identification to injury pattern recognition—narrative review and new concept. Br J Sports Med. 2016;50(21):1309–1314. PubMed ID: 27445362 doi:10.1136/bjsports-2015-095850
- 33.↑
Mooney M, Charlton PC, Soltanzadeh S, Drew MK. Who ‘Owns’ the Injury or Illness? Who ‘Owns’ Performance? Applying Systems Thinking to Integrate Health and Performance in Elite Sport. BMJ Publishing Group Ltd and British Association of Sport and Exercise Medicine; 2017.
- 34.↑
McLean S, Hulme A, Mooney M, Read GJM, Bedford A, Salmon PM. A systems approach to performance analysis in women’s netball: using work domain analysis to model elite netball performance. Front Psychol. 2019;10:201. PubMed ID: 30787902 doi:10.3389/fpsyg.2019.00201
- 35.↑
Bérard C. Group model building using system dynamics: an analysis of methodological frameworks. Electron J Bus ResMethods. 2010;8(1):35.
- 36.↑
McLean S, Kerhervé H, Lovell GP, Gorman AD, Solomon C. The effect of recovery duration on vastus lateralis oxygenation, heart rate, perceived exertion and time motion descriptors during small sided football games. PLoS One. 2016;11(2):e0150201. PubMed ID: 26919064 doi:10.1371/journal.pone.0150201
- 37.↑
Kerhervé HA, Samozino P, Descombe F, et al. Calf compression sleeves change biomechanics but not performance and physiological responses in trail running. Front Psychol. 2017;8:247. doi:10.3389/fphys.2017.00247
- 38.↑
Salmon P, Read GJ. Many model thinking in systems ergonomics: a case study in road safety. Ergonomics. 2019;62(5):612–628. PubMed ID: 30460881 doi:10.1080/00140139.2018.1550214
- 39.↑
McLean S, Finch CF, Goode N, Clacy A, Coventon LJ, Salmon PM. Applying a systems thinking lens to injury causation in the outdoors: evidence collected during 3 years of the understanding and preventing led outdoor accidents data system. Inj Prev. 2021;27(1):48–54.
- 40.↑
Stevens NJ, Salmon PM, Walker GH, Stanton NA. Human Factors in Land Use Planning and Urban Design: Methods, Practical Guidance, and Applications. Boca Raton, FL: CRC Press; 2018.
- 41.↑
Salmon P, Stanton NA, Gibbon A, Jenkins D, Walker GH. Human Factors Methods and Sports Science: A Practical Guide. Boca Raton, FL: CRC Press; 2009.
- 42.↑
Salmon PM, McLean S, Dallat C, Mansfield N, Solomon C, Hulme A. Human Factors and Ergonomics in Sport: Applications and Future Directions. Boca Raton, FL: CRC Press; 2020.
- 44.↑
Box GE. Science and statistics. J Am Stat Assoc. 1976;71(356):791–799. doi:10.1080/01621459.1976.10480949
- 45.↑
Kardeş S, Levack W, Özkuk K, Aydın EA, Karabulut SS. Retractions in rehabilitation and sport sciences journals: a systematic review. Arch Phys Med Rehabil. 2020;101(11):1980–1990. PubMed ID: 32402503 doi:10.1016/j.apmr.2020.03.010
- 46.↑
Sterman JD. Learning from evidence in a complex world. Am J Public Health. 2006;96(3):505–514. PubMed ID: 16449579 doi:10.2105/AJPH.2005.066043
- 47.↑
Meadows DH. Leverage points: places to intervene in a system. 1999. Accessed October 15, 2020. https://donellameadows.org/archives/leverage-points-places-to-intervene-in-a-system/
- 48.↑
Ioannidis JP, Baas J, Klavans R, Boyack KW. A standardized citation metrics author database annotated for scientific field. PLoS Biol. 2019;17(8):e3000384. PubMed ID: 31404057 doi:10.1371/journal.pbio.3000384
- 49.↑
Hicks D, Wouters P, Waltman L, De Rijcke S, Rafols I. Bibliometrics: the Leiden Manifesto for research metrics. Nature. 2015;520(7548):429–431. PubMed ID: 25903611 doi:10.1038/520429a
- 50.↑
Dekker S. Drift into Failure: From Hunting Broken Parts to Understanding Complex Systems. Burlington, VT: Ashgate; 2011.
- 51.↑
Williams SJ, Kendall LR. A profile of sports science research (1983–2003). J Sci Med Sport. 2007;10(4):193–200. PubMed ID: 17000134 doi:10.1016/j.jsams.2006.07.016
- 52.↑
Halperin I, Vigotsky AD, Foster C, Pyne DB. Strengthening the practice of exercise and sport-science research. Int J Sports Physiol Perform. 2018;13(2):127–134. PubMed ID: 28787228 doi:10.1123/ijspp.2017-0322
- 53.↑
Finch C. A new framework for research leading to sports injury prevention. J Sci Med Sport. 2006;9(1–2):3–9. PubMed ID: 16616614 doi:10.1016/j.jsams.2006.02.009
- 54.↑
Impellizzeri FM, McCall A, Meyer T. Registered reports coming soon: our contribution to better science in football research. Sci Med Football. 2019;3(2):87–88. doi:10.1080/24733938.2019.1603659
- 55.↑
Nosek BA, Alter G, Banks GC, et al. Promoting an open research culture. Science. 2015;348(6242):1422–1425. PubMed ID: 26113702 doi:10.1126/science.aab2374
