University of Groningen The importance of tactical skills ... · Visscher, C. (2010) Journal of Sport Sciences, 28, 521-528 Chapter 4 Tactical skills of world-class youth soccer teams

University of Groningen

The importance of tactical skills in talent developmentKannekens, Rianne

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2010

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Kannekens, R. (2010). The importance of tactical skills in talent development. s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 22-11-2020

https://www.rug.nl/research/portal/en/publications/the-importance-of-tactical-skills-in-talent-development(dbea6b54-c066-41da-b426-06d7b19def85).html

https://www.rug.nl/research/portal/en/publications/the-importance-of-tactical-skills-in-talent-development(dbea6b54-c066-41da-b426-06d7b19def85).html

The importance of tactical skills

in talent development

The research presented in this thesis was conducted under the auspices of the Center forHuman Movement Sciences, part of the University Medical Center Groningen, Universityof Groningen, the Netherlands. The studies described in this thesis were supported by a grantfrom NOC*NSF.

Paranimfen: Barbara HuijgenAnnelies Kannekens

Cover design:Annemieke van der Hoek, Rotterdam, www.annemiekevanderhoek.nl Produced by: Fontijn & Partners, Smilde, www.fontijn.nlISBN: 978-90-367-4385-3

© 2010 R. Kannekens, Groningen, The NetherlandsAll right reserved. No part of this publication may be reproduced or transmitted in any formor by any means, electronic or mechanical, including photocopy, recording, or any informationstorage or retrieval system, without permission in writing from the author.

RIJKSUNIVERSITEIT GRONINGEN

The importance of tactical skills in talent development

Proefschrift

ter verkrijging van het doctoraat in deMedische Wetenschappen

aan de Rijksuniversiteit Groningenop gezag van de

Rector Magnificus, dr. F. Zwarts,in het openbaar te verdedigen op

woensdag 2 juni 2010om 13.15 uur

doorRianne Kannekens

geboren op 30 november 1981te Zwolle

Promotor: Prof. dr. C. Visscher

Copromotor: Dr. M.T. Elferink-Gemser

Beoordelingscommissie: Prof. dr. R.J. BoskerProf. dr. G.J.P. SavelsberghProf. dr. E.J.A. Scherder

ISBN: 9789036743853

Content

Chapter 1 General Introduction 1

Chapter 2 Self-assessed tactical skills of elite youth soccer players with 7different academic achievement Kannekens, R., Elferink-Gemser, M.T., Visscher, C. Submitted to European Journal of Sport Science

Chapter 3 Knowing what to do and doing it: Difference in self-assessed 19tactical skills of regional, sub-elite, and elite youth field hockey playersElferink-Gemser, M.T., Kannekens, R., Lyons, J., Tromp, E.J.Y., Visscher, C. (2010)Journal of Sport Sciences, 28, 521-528

Chapter 4 Tactical skills of world-class youth soccer teams 37Kannekens, R., Elferink-Gemser, M.T., Visscher, C. (2009)Journal of Sport Sciences, 27, 807-812

Chapter 5 Self-assessed tactical skills in elite youth soccer players: A 51longitudinal studyKannekens, R., Elferink-Gemser, M.T., Post, W.J., Visscher, C. (2009)Perceptual and Motor Skills, 109, 459-472

Chapter 6 Positioning and deciding: The key to professional soccer 69Kannekens, R., Elferink-Gemser, M.T., Visscher, C. Scandinavian Journal of Medicine & Science in Sports (in press)

Chapter 7 General Discussion and Conclusion 85

Summary 95Samenvatting 101Dankwoord 107Curriculum Vitae 111List of Publications 115

Chapter 1

General Introduction

1.1 General Introduction

In sports, talent identification and talent development programmes have gained popularity in recent decades. Talent identification programmes are, by definition, directed atrecognizing the potential of an athlete to excel in a particular sport, while talent developmentprogrammes are aimed at providing the most optimal learning environment to help thesepromising youth athletes realize their potential (Russell, 1989; Williams & Reilly, 2000). Especially in invasion sports such as soccer, (association football) considerable resources areallocated to help identify and develop youth talent to thus enable athletes to reach the top intheir sport (Abbott & Collins, 2004; Reilly et al., 2000). Nevertheless, only few of these promising athletes ultimately succeed. Although up till now there is no uniformly acceptedtheoretical framework to guide current talent programmes, consensus has been reached inthat such a model should take into account both the dynamic and the multidimensional natureof talent (Vaeyens et al., 2008). A more in-depth understanding of which performance characteristics of talented athletes are most closely associated with future success, implies astep closer to finding out what it takes for aspiring youth athletes to acquire the necessaryskills and expertise to make it to the top in their sport.

In invasion games players constantly have to deal with a complex and rapidly changingenvironment (Janelle & Hillman, 2003; Starkes, 1987). In order to be successful, they have toperform the right action at the right moment and punctually adapt to new configurations ofplay and the circulation of the ball (Elferink-Gemser et al., 2004b; Grehaigne & Godbout,1995). It is therefore crucial for talented players to acquire well-developed tactical skills. Highproficiency in tactical skills is a requisite for expert performance in virtually all achievementdomains and thus, besides in sports, also in the academic domain (Janelle & Hillman, 2003).In sports being proficient in tactical skills not only involves having the theoretical knowledgeto determine which strategy is most appropriate in a given situation, but also the practical capability to execute the strategy successfully within the constraints of the required movements(see e.g., McPherson, 1994; Starkes, 1993). This raises an interesting question as to the potential association between the two achievement domains: Are talented youth athletes thatperform at a higher educational level also the ones that have the better tactical skills in theirfield of sport?

There are two main components to tactical skills: ‘knowing what to do’ or declarativeknowledge and ‘doing it’ or procedural knowledge. Declarative knowledge comprises theknowledge of the rules and goals of the game (French & Thomas, 1987; Williams & Davids,1995) and procedural knowledge the selection and execution of an appropriate action withinthe context of the game (McPherson, 1994). If athletes wish to reach the top in their sport,they need to develop and constantly improve their performance over time (see Figure 1.1).The studies reported in the current thesis focus exclusively on the aspect of tactical skillswithin the context of team sports. In the investigations it is, of course, always kept in mind

2

Chapter 1

that to perform well in sports the multidimensionality of performance characteristics thatcover no less than four different domains, i.e., the physiological, technical, tactical, and psychological domains, in combination with a player’s anthropometric characteristics plays afundamental role (Elferink-Gemser et al., 2004a).

Figure 1.1. Tactical skills and their role in the performance development of youth athletes.

A team player’s performance is always the result of the interplay between the relevant taskcharacteristics, in the studies of the current thesis the demands of the highly structured andanalytical nature of soccer (and field hockey) with their different field positions, the player’spersonal characteristics, among which tactical skills (i.e. the level of declarative and proceduralknowledge) and various environmental characteristics (e.g. the talent development programme).

In conclusion, given the relevance and status of (team) sports in today’s society and themoney involved in elite sports, it has become increasingly important to know what makes anathlete a top athlete. Therefore, it is important to gain insight into the construct of tacticalskills and explore which aspects explain and predict future success best and how these skillsdevelop during the adolescent years. The research presented in this thesis was designed tocontribute to this knowledge.


Time

Person

Tactical skillsDeclarative KnowledgeProcedural Knowledge

Task

Development

Performance

Environment

3

1.2 Outline of the thesis

The main purpose of this thesis is to gain more insight into the talent developmentprocess by examining the importance of tactical skills in the performance and success of talented youth team players over time. In Chapter 2, the constructs of declarative and procedural knowledge are addressed by studying the tactical skills of elite youth soccer playersrelative to their different academic performance levels. Chapter 3 presents a study that compares the tactical skills in regional, sub-elite and elite youth field hockey players. To expand the knowledge of what makes a top player and to ultimately foster talent, the study inChapter 4 looks for relationships between the tactical skills of promising players of two national youth soccer teams from two countries and the level at which they compete. Chapter5 features a longitudinal study that focuses on the change in the tactical skills among eliteyouth soccer players over time in relation to their field positions. To help establish which factors predict which talented players are most likely to make it to the top one needs to followthem into adulthood. The study in Chapter 6 accordingly looked for factors that are crucialto professional soccer. In this longitudinal study the tactical skills of elite youth soccer playerswere assessed during their adolescence and related these outcomes to their adult performancelevels. Chapter 7 comprises the general discussion in which the findings of the various studiesare summarized and put into context and their implications for trainers, coaches, scouts, players, parents and other interested parties discussed.

Chapter 1

4

References

• Abbott, A. & Collins, D. (2004). Eliminating the dichotomy between theory and practicein talent identification and development: considering the role of psychology. Journal ofSports Sciences, 22, 395-408.

• Elferink-Gemser, M. T., Visscher, C., Lemmink, K. A. P. M., & Mulder, T. W. (2004a).Relation between multidimensional performance characteristics and level of performancein talented youth field hockey players. Journal of Sports Sciences, 22, 1053-1063.

• Elferink-Gemser, M. T., Visscher, C., Richart, H., & Lemmink, K. A. P. M. (2004b). Development of the tactical skills inventory for sports. Perceptual and Motor Skills, 99, 883-895.

• French, K. E. & Thomas, J. R. (1987). The Relation of Knowledge Development to Children's Basketball Performance. Journal of Sport Psychology, 9, 15-32.

• Grehaigne, J. F. & Godbout, P. (1995). Tactical Knowledge in Team Sports from A Constructivist and Cognitivist Perspective. Quest, 47, 490-505.

• Janelle, C. M. & Hillman, C. H. (2003). Expert performance in sport: Current perspectivesand critical issues. In J.L. Starkes & K. A. Ericsson (Eds.), Expert performance in sports: Advances in research on sport expertise (pp. 49-83). Champaign, IL: Human Kinetics.

• McPherson, S. L. (1994). The Development of Sport Expertise - Mapping the TacticalDomain. Quest, 46, 223-240.

• Reilly, T., Williams, A. M., Nevill, A., & Franks, A. (2000). A multidisciplinary approachto talent identification in soccer. Journal of Sports Sciences, 18, 695-702.

• Russell, K. (1989). Athletic talent: From detection to perfection. Science Periodical on Researchand Technology in Sport, 9, 1-6.

• Starkes, J. L. (1987). Skill in Field Hockey - the Nature of the Cognitive Advantage. Journalof Sport Psychology, 9, 146-160.

• Starkes, J. L. (1993). Motor experts: Opening thoughts. In J.L. Starkes & F. Allard (Eds.),Cognitive issues in motor expertise (pp. 3-16). Amsterdam: Elsevier.

• Vaeyens, R., Lenoir, M., Williams, A. M., & Philippaerts, R. M. (2008). Talent identification and development programmes in sport - Current models and future directions. Sports Medicine, 38, 703-714.

• Williams, A. M. & Reilly, T. (2000). Talent identification and development in soccer.Journal of Sports Sciences, 18, 657-667.

• Williams, M. & Davids, K. (1995). Declarative Knowledge in Sport - A By-Product of Experience Or A Characteristic of Expertise. Journal of Sport & Exercise Psychology, 17, 259-275.


5

Chapter 2

Self-assessed tactical skills of elite youth soccer

players with different academic achievement

Kannekens, R., Elferink-Gemser, M.T., Visscher, C.

Submitted to European Journal of Sport Science

AcknowledgementsThe authors would like to thank all players and staff of the Dutch youth soccer

teams for their participation in this study. In addition, we are grateful to the students ofthe Center for Human Movement Sciences for their help in collecting the data.

Abstract

This study investigates the tactical skills of 93 elite youth soccer players (mean age17.3±0.6) attending either pre-vocational or pre-university schools. All athletes completedthe declarative and procedural knowledge scales of the Tactical Skills Inventory for Sport(TACSIS). Multivariate analyses of covariance including educational level with field positionand repeating class in school as covariates, revealed differences between elite youth soccerplayers attending pre-vocational and pre-university educational level on the total constructof tactical skills (P<.05). Players attending pre-university schools may benefit from the cognitive skills they acquire in their academic environment. In contrast, it seems that playersattending pre-vocational schools have an advantage on the scale Acting in changing situations.It is recommended to investigate which players will become professional versus who will stayat the amateur level and whether this is related to their tactical skills and academic achievements.

Keywords: talent development, performance level, expert athletes, youth, education

Chapter 2

8

2.1 Introduction

Academic achievement is a good predictor of sport knowledge and physical educationknowledge (Dexter, 1999). Studies which focused on associations between sport performanceand academic achievement suggest that young, elite athletes tend not only to perform well insports but also in an academic setting (Brettschneider, 1999; Durand-Bush & Salmela, 2002;Jonker, Elferink-Gemser & Visscher, 2009; Umbach, Palmer, Kuh & Hannah, 2006; Watt &Moore, 2001). They have higher academic standards and graduation rates in comparison totheir peers performing at lower competitive levels (Watt & Moore, 2001). These studies suggest that academic achievement is related to sport knowledge.

In the Netherlands a child’s educational level and hence type of school is determinedbased on their academic achievement. Dutch teenagers either attend the pre-vocational orthe pre-university level, with pre-vocational schools offering a mix of theoretically and practically oriented training (c.f. International Standard Classification of Education (ISCED)levels 4 and 5). Given the schools’ more practical orientation, students are specifically enabledto acquire the skills they need for their future work careers. Before the students start withtheir work careers, they will attend a follow-up study (vocational level) in the same directionas their pre-vocational study. At the pre-university level students are prepared for a universitystudy (ISCED level 6; UNESCO, 1997), acquiring general knowledge for use in their futurestudies (Ministerie van OCW, 2008). If a student has a low Grade Point Average in eitherschool type, he or she has to repeat class. Hence, repeating class is also an indication of academic achievement.

In order to be successful in invasion games such as soccer, elite players need well-developed tactical skills besides well-developed physiological and technical skills (Elferink-Gemser, Visscher, Lemmink & Mulder, 2007; Helsen & Starkes, 1999; McPherson & Kernodle, 2003; Nougier & Rossi, 1999; Starkes, 1987). Hence, tactical skills are fundamentalfor high-level soccer performance (Kannekens, Elferink-Gemser & Visscher, 2009b). Academic achievement and tactical skills both rely on cognitive competencies. The cognitivecompetencies of tactical skills are typically categorized as declarative and procedural knowledge (McPherson, 1994). Declarative knowledge is taken to denote knowledge of therules and goals of the game (French & Thomas, 1987; McPherson, 1994; Williams & Davids,1995), in other words ‘knowing what to do,’ whereas procedural knowledge describes the selection of an appropriate action within the context of the game, i.e., ‘doing it’ (McPherson,1994).

Studies on tactical skill show that athletes performing at higher performance levels aresuperior to players showing lower performance levels on many tactical aspects (e.g., Grehaigne, Godbout & Bouthier, 2001; Kannekens et al., 2009b; Starkes, 1987; Williams,Davids, Burwitz & Williams, 1993). More specifically, expert or elite athletes are more accurate at picking up task-relevant information and better able to integrate this information

Tactical skills of soccer players attending different educational levels

9

with earlier experiences, which enables them to choose the best option available (e.g., French& Thomas, 1987; Kannekens et al., 2009b; Williams et al., 1993). They, moreover, have amore complex hierarchically organized knowledge base stored in their memory and are ableto cluster items into larger and more meaningful units (Williams et al., 1993). In addition toperformance related differences in tactical skills, players of different field positions (e.g., defenders, midfielders, and attackers) distinguish themselves on the basis of their tactical skillswith defenders scoring best on scales for defensive situations and attackers scoring best onscales for offensive situations (Kannekens, Elferink-Gemser, Post & Visscher, 2009a).

In sum, although in elite youth athletes a positive relationship has been established between sport performance and tactical skills and between sport performance and academicachievement, it has not yet been investigated whether elite youth soccer players attending different educational levels score differently on tactical skills. To advance our knowledge aboutwhat makes a top athlete and to cultivate talent, the present study investigates declarative andprocedural tactical knowledge of elite youth soccer players attending different levels of education.

2.2 Methods

Participants and procedureWe invited 93 soccer players in the 17-18 year age bracket (mean age = 17.3, SD = 0.6)

to take part in a cross-sectional study. All played with national (Dutch) premier league clubsand participated in their club’s talent development programme. They competed at the highestlevel and belonged to the top 0.5% of all soccer players in their age group (Royal NetherlandsFootball Association; KNVB, 2008).

Permission for the study was obtained from both the clubs and the trainers and the athletes were individually informed about the study and its procedures, following which theyprovided their informed consent. All procedures were in accordance with the standards of theethics committee of the University Medical Center Groningen and the University of Groningen.

The data were collected at the end of the competitive season and considered based onthe type of education the athletes attended, with players receiving pre-vocational and vocational level (n = 45; further mentioned as pre-vocational level) and players studying at thepre-university level (n = 48). Table 2.1 presents general characteristics concerning age, sportpractice, field position and repeating class in school per type of education.

Chapter 2

10

Table 2.1. General characteristics (means; standard deviations) of the elite youth soccer players specified per

type of education

Educational levelPre-vocational Pre-university

n=45 n=48

Age (years) 17.25 (0.61) 17.38 (0.56)Accumulated organized soccer experience (years) 10.36 (2.11) 11.41 (1.67)

Soccer practice (hours/week) 8.38 (1.48) 8.84 (2.38)Non-specific soccer practice (hours/week) 2.08 (2.38) 2.45 (2.44)Soccer matches (per week) 1.06 (0.19) 1.04 (0.18)Field position (n)

Defender 22 15Midfielder 12 21Attacker 11 12

Repeating class in school (n) 14 10

Tactical skills knowledgeThe 22-item Tactical Skills Inventory for Sports (TACSIS; Elferink-Gemser, Visscher,

Richart & Lemmink, 2004) with separate scales for declarative and procedural knowledge wasused to assess the extent and nature of the athletes’ tactical skills. This questionnaire is developed especially for players of invasion games such as soccer (Elferink-Gemser et al.,2004). The scales Knowing about ball actions (e.g., “I know exactly when to pass the ball to ateammate or when not to”) and Knowing about others (e.g., “I know quickly how the opponentis playing”) contain four and five items respectively that were related to declarative knowledge.The scales Positioning and deciding (e.g., “My getting open and choosing positions is”) and Acting in changing situations (e.g., “My interception of the opponents is”) consisted of nine andfour items respectively that gauged procedural knowledge. When rating their sport performance, the respondents were instructed to compare themselves with top players in thesame age category, scoring the items on a 6-point Likert scale ranging from ‘Very poor’ to‘Excellent’ or from ‘Almost never’ to ‘Always’.

In previous research, the TACSIS was shown to have good psychometric characteristicswith internal consistency coefficients (Cronbach’s alphas) of all four subscales ranging from0.72 to 0.89 (Elferink-Gemser et al., 2004). The Intraclass Correlation Coefficients (ICC) forrepeated measures was 0.76 for Knowing about others and 0.88 and 0.82 for Positioning and deciding, and Acting in changing situations, respectively. The scale Knowing about ball actions hadan ICC of 0.60 (Elferink-Gemser et al., 2004). For detailed information about the development of the TACSIS, see Elferink-Gemser et al. (2004). In the current study, the


11

internal consistency of the four TACSIS scales indicates good internal consistency (Nunally,1978) with Cronbach’s alpha of 0.84 for Knowing about ball actions, 0.79 for Knowing aboutothers, 0.90 for Positioning and deciding, and 0.79 for Acting in changing situations.

Data analyses For each subgroup the players’ mean scores and standard deviations were calculated for

the four subscales of the TACSIS and multivariate analysis of covariance (MANCOVA) wasconducted (factor of educational level) to examine group (educational level) differences. Fieldposition and repeating class in school were included as covariates. Standardized mean scoresor effect sizes (d) were also computed; we followed Cohen’s (988) suggestion in that we classified effect sizes around 0.20 as small, around 0.50 as moderate, and around 0.80 as large.Statistical significance was set at P < 0.05.

2.3 Results

Table 2.2 presents the means, standard deviations and effect sizes of the declarative andprocedural knowledge scales of the TACSIS. The effect sizes for the TACSIS subscales Knowing about ball actions and Positioning and deciding were small (d =0.14 and d = 0.21) i.e., thetalented soccer players attending pre-university schools scored higher than their talentedcounterparts attending pre-vocational schools. The subscales Knowing about others and Acting in changing situations had moderate effect sizes (d = 0.39 and d = 0.41).On the subscaleKnowing about others the players attending pre-university schools outscored the players attending pre-vocational schools whereas on the subscale Acting in changing situations the players attending pre-vocational schools outscored their counterparts attending pre-universityschools. The MANCOVA revealed a significant main effect for the differences between pre-vocational and pre-university level on the TACSIS (F1,88=3.15, P<0.05). The scores onthe individual TACSIS subscales did not reveal any significant differences (P>0.05). The covariate repeating class in school had no influence (P>0.05), while field position was significant (P<0.05) (See Table 2.3).

Chapter 2

12

Table 2.2. Means (SDs) and effect sizes (d) for scores on the TACSIS subscales for the elite youth soccer player

per type of education.

Educational levelPre-vocational Pre-university d

n=45 n=48

Declarative knowledgeKnowing about ball actions 4.21 (0.71) 4.31 (0.69) 0.14Knowing about others 3.84 (0.65) 4.09 (0.62) 0.39

Procedural knowledgePositioning and deciding 3.75 (0.64) 3.88 (0.59) 0.21Acting in changing situations 4.21 (0.69) 3.92 (0.73) 0.41

d around 0.20 is small; d around 0.50 is moderate; d around 0.80 is large (Cohen, 1988)

Table 2.3. Results of MANCOVA for type of education with repeating class in school and field position as

covariate

Wilks’ Lambda F-value Hypothesis df Error df P-value

Type of education 0.871 3.147 4 85 0.018Field position 0.884 2.782 4 85 0.032Repeating class in school 0.980 0.436 4 85 0.782

2.4 Discussion

Although evidence of a positive association between tactical skills and sport performanceand between sport performance and academic achievement has been reported previously, theseearlier studies did not compare the tactical skills of elite youth athletes at different educationallevels even though tactical skills and education both rely on cognitive skills. The current studyis one of the first to directly compare the young players’ sport-related cognitive competenciesto their academic achievements with the aim to investigate the differences on declarative andprocedural tactical knowledge of elite youth soccer players attending different levels of education.

The results of this study suggest that there are differences between elite youth soccerplayers attending pre-vocational versus those attending pre-university educational levels onthe total construct of tactical skills of elite youth soccer players. The results showed that pre-university players outscored the pre-vocational players on the TACSIS subscales Knowingabout ball actions, Knowing about others, and Positioning and deciding. On the TACSIS subscale


13

Acting in changing situations the pre-vocational players outscored the pre-university players.Only the subscales Knowing about others and Acting in changing situations had moderate effectsizes (0.39; 0.41); the other subscales showed small differences between players of differenteducational levels.

The subscale Knowing about others consists of questions concerning declarative knowledge.Declarative knowledge is taken to denote the factual information of a skill domain (Alexander& Judy, 1988; Anderson, 1982), which, for invasion sports, means the rules and goals of thegame. Players attending pre-university schools outscored players attending pre-vocationalschool on declarative knowledge, this may be explained by the objective of pre-universityschools. Pre-university schools prepares teenagers for an academic career with the knowledgebeing taught focusing on continuing education and general knowledge (Ministerie van OCW,2008). They learn to gain knowledge and to use this in their future studies. Pre-universitystudents learn to store knowledge pertaining to cognitive domains, and, given that the athletesthat attended pre-university schools had higher scores on declarative tactical skills knowledge,the cognitive skills they acquire during their academic studies may help them or are consistentwith the skills to develop their declarative knowledge in sport situations.

The TACSIS subscale Acting in changing situations consists of questions related to procedural knowledge. The quintessence of procedural knowledge is the ability to correctlyinterpret a specific situation, to make decisions in a split second, and to subsequently executethe right action (French & Thomas, 1987; McPherson & Thomas, 1989; McPherson, 1994;Thomas & Thomas, 1994). This part of our knowledge system is thus all about ‘doing it,’ inother words about athletes having to recognize the situation on the field and choosing theappropriate response. It seems that players attending pre-vocational schools have a slight advantage on this scale. The objective of pre-vocational schools is to help teenage childrenattain general skills and prepare them for their future working lives with a focus on skills required in practically oriented professions (Ministerie van OCW, 2008). Children attendingpre-vocational schools learn skills, which they have to apply in other situations. This is inagreement with the content of the questions of the subscale Acting in changing situations.

Nevertheless, to develop high levels of tactical skills most likely practice in sport-specificsituations is important (Mountakis, 2009). The participants of this study belonged to the top0.5% of all soccer players in their age group and have accumulated over 10 years of organizedsoccer training. This confirmed that, regardless of their level of education, all athletes in oursample were experts in their sports, competing at the highest level in their age category. Hencedifferences within this homogenous group will be small. This is in consistent with researcherswho have indicated that there is a positive relationship between practice and expert performance (Baker & Horton, 2004) and with research that found that decision-making hasto be learned through practice within the context of the game (Turner & Martinek, 1999).

Even though all the participants of this study were elite players, they still have to moveup from youth competitions through adult competition and in the subsequent years it becomes

Chapter 2

14

apparent whether they actually reach the status of professional soccer players or not. Abouthalf of them ultimately reach the top (Roescher, Elferink-Gemser, Huijgen & Visscher, inpress). Since the pre-vocational players had in general a lower score than the pre-universityplayers on the subscales Knowing about ball actions, Knowing about others and Positioning and deciding but scored higher on the subscale Acting in changing situations it is recommended toinvestigate which players will become professional versus who will stay at the amateur leveland whether this is related to their tactical skills and academic achievements.


15

References

• Alexander, P. A., & Judy, J. E. (1988). The Interaction of Domain-Specific and StrategicKnowledge in Academic-Performance. Review of Educational Research, 58, 375-404.

• Anderson, J. R. (1982). Acquisition of Cognitive Skill. Psychological Review, 89, 369-406.• Baker, J., & Horton, S. (2004). A review of primary and secondary influences on sport

expertise. High Ability Studies, 15, 211-228.• Brettschneider, W. D. (1999). Risk and opportunities: Adolescents in top-level sport

growing up with the pressures of school and training. European Physical Education review, 5,121-133.

• Cohen, J. (1988). Statistical power analysis for the behavioral sciences.Hillsdale, NJ: LawrenceErlbaum Associates.

• Dexter, T. (1999). Relationships between sport knowledge, sport performance and academicability: Empirical evidence from GCSE Physical Education. Journal of Sports Sciences, 17,283-295.

• Durand-Bush, N., & Salmela, J. H. (2002). The development and maintenance of expertathletic performance: Perceptions of World and Olympic champions. Journal of AppliedSport Psychology, 14, 154-171.

• Elferink-Gemser, M. T., Visscher, C., Lemmink, K. A. P. M., & Mulder, T. (2007). Multidimensional performance characteristics and standard of performance in talentedyouth field hockey players: A longitudinal study. Journal of Sports Sciences, 25, 481-489.

• Elferink-Gemser, M. T., Visscher, C., Richart, H., & Lemmink, K. A. P. M. (2004). Development of the tactical skills inventory for sports. Perceptual and Motor Skills, 99, 883-895.

• French, K. E., & Thomas, J. R. (1987). The Relation of Knowledge Development to Children's Basketball Performance. Journal of Sport Psychology, 9, 15-32.

• Grehaigne, J. F., Godbout, P., & Bouthier, D. (2001). The teaching and learning of decisionmaking in team sports. Quest, 53, 59-76.

• Helsen, W. F., & Starkes, J. L. (1999). A Multidimensional Approach to Skilled Perceptionand Performance in Sport. Applied Cognitive Psychology, 13, 1-27.

• Jonker, L., Elferink-Gemser, M. T., & Visscher, C. (2009). Talented athletes and academicachievements: a comparison over 14 years. High Ability Studies, 20, 55-64.

• Kannekens, R., Elferink-Gemser, M. T., Post, W. J., & Visscher, C. (2009a). Self-assessedtactical skills in elite youth soccer players: A longitudinal study. Perceptual and Motor Skills,109, 459-472.

• Kannekens, R., Elferink-Gemser, M. T., & Visscher, C. (2009b). Tactical skills of world-class youth soccer teams. Journal of Sports Sciences, 27, 807-812.


Chapter 2

16

• McPherson, S. L., & Kernodle, M. W. (2003). Tactics, the Neglected Attribute of Expertise. Problem Representations and Performance Skills in Tennis. In J. L. Starkes &K. A. Ericsson (Eds.), Expert Performance in Sports. Advances in Research on Sport Expertise(pp. 137-167). Champaign (IL): Human Kinetics.

• McPherson, S. L., & Thomas, J. R. (1989). Relation of Knowledge and Performance inBoys' Tennis - Age and Expertise. Journal of Experimental Child Psychology, 48,190-211.

• Ministerie van Onderwijs, Cultuur en Wetenschappen (2008). Voortgezet onderwijs 2008-2009 (2008) [secondary education 2008-2009] Retrieved March 13, 2009,http://www.minocw.nl/vo/235/Over-voortgezet-onderwijs.html

• Mountakis, C. (2009). Differences between physical education and top-level sport. EuropeanPhysical Education review, 7, 92-106.

• Nougier, V., & Rossi, B. (1999). The Development of Expertise in the Orienting of Attention. International Journal of Sport Psychology, 30, 246-260.

• Nunally, J. C. (1978). Psychometric theory. New York: McGraw-Hill.• Roescher, C., Elferink-Gemser, M. T., Huijgen, B. C. H., & Visscher, C. (in press). Soccer

endurance development in professionals. International Journal of Sports Medicine.• Starkes, J. L. (1987). Skill in Field Hockey - the Nature of the Cognitive Advantage. Journal

of Sport Psychology, 9, 146-160.• Thomas, K. T., & Thomas, J. R. (1994). Developing Expertise in Sport - the Relation of

Knowledge and Performance. International Journal of Sport Psychology, 25, 295-312.• Turner, A. P., & Martinek, T. J. (1999). An investigation into teaching games for

understanding: Effects on skill, knowledge, and game play. Research Quarterly for Exerciseand Sport, 70, 286-296.

• Umbach, P. D., Palmer, M. M., Kuh, G. D., & Hannah, S. J. (2006). Intercollegiate athletesand effective educational practices: Winning combination or losing effort? Research inHigher Education, 47, 709-733.

• UNESCO (1997). International Standard Classification of Education: ISCED (1997). Retrieved March 13, 2009, http://www.uis.unesco.org/ev.php?ID=3813_201&ID2=DO_TOPIC

• Watt, S. K., & Moore, J. L. (2001). Who are student athletes? New Directions For StudentServices, 7-18.

• Williams, M., & Davids, K. (1995). Declarative Knowledge in Sport - A By-Product of Experience Or A Characteristic of Expertise. Journal of Sport & Exercise Psychology, 17, 259-275.

• Williams, M., Davids, K., Burwitz, L., & Williams, J. (1993). Cognitive Knowledge andSoccer Performance. Perceptual and Motor Skills, 76, 579-593.


17

Chapter 3

Knowing what to do and doing it: Difference in

selfassessed tactical skills of regional, sub-elite,

and elite youth field hockey players

Elferink-Gemser, M.T., Kannekens, R. , Lyons, J., Tromp, E.J.Y.,

Visscher, C. (2010)

Journal of Sport Sciences, 28, 521-528

AcknowledgementsThis study has been supported by a grant of the Dutch National Olympic CommitteeNOC*NSF. The authors thank all players, trainers and staff of the field hockey clubs

HC ’s-Hertogenbosch, HC Rotterdam, and GHBS for their cooperation.

Abstract

To determine whether youth athletes with an ‘average’ (regional), ‘high’ (sub-elite), and‘very high’ (elite) level of performance differ with respect to their self-assessed tactical skills,191 youth field hockey players (mean age 15.5 years, s=1.6) completed the Tactical Skills Inventory for Sports (TACSIS) with scales for declarative (‘knowing what to do’) and procedural (‘doing it’) knowledge. Multivariate analyses of covariance with age as covariateshowed that elite and sub-elite players outscored regional players on all tactical skills (P<0.05)whereas elite players had better scores than sub-elite players on Positioning and deciding(P<0.05) only. The sex of the athletes had no influence on the scores (P>0.05). With increasinglevel of performance, scores on declarative and procedural knowledge were higher. Close toexpert performance level, declarative knowledge no longer differentiated between elite andsub-elite players (P>0.05), in contrast to an aspect of procedural knowledge (i.e. positioningand deciding), where elite players outscored sub-elite players (P<0.05). These results may haveimplications for the development of talented athletes.

Keywords: sports, talent development, performance level, positioning and deciding, questionnaire

Chapter 3

20

3.1 Introduction

To be successful in team sports such as field hockey, where players have to deal with acomplex and rapidly changing environment, it is crucial to perform the right action at theright moment- that is, to have well-developed tactical skills (Elferink-Gemser, Visscher, Lemmink & Mulder, 2004a). Players constantly need to adapt to new playing configurationsand to the circulation of the ball (Gréhaigne & Godbout, 1995). Playing well means choosingthe right action at the right moment and performing that course of action efficiently and consistently throughout the game (Baker, Côté & Abernethy, 2003; Gréhaigne, Godbout &Bouthier, 2001). Tactical skills can be defined as knowledge regarding in-game adaptationsand decision-making activity on the field of play. Such knowledge can be distinguished fromstrategy, which refers to choices discussed in advance with the trainer in order for the team toorganize itself (Bjurwill, 1993; Gréhaigne, Godbout & Bouthier, 1999; Mouchet, 2005). Insports, not only the decision concerning the appropriate action in the context of the game situation is important, but also the ability to execute the complex sport skill. This is the distinction between declarative (‘knowing what to do’) and procedural (‘doing it’) knowledge(e.g., Anderson, 1982; McPherson & Kernodle, 2003; Thomas & Thomas, 1994; Turner &Martinek, 1999). Declarative knowledge includes knowledge of the rules and goals of thegame (French & Thomas, 1987; Williams & Davids, 1995), whereas procedural knowledgeinvolves the selection of an appropriate action within the context of the game (McPherson,1994).

Elite athletes distinguish themselves from novices by their enhanced declarative and procedural knowledge (e.g., French & Thomas, 1987; Williams, Davids, Burwitz & Williams,1993). The relationship between tactical skills and performance level has been comparedmostly among elite athletes and novices (e.g., Enns & Richards, 1997; Ericsson & Kintsch,1995; French & McPherson, 1999; McPherson, 1994; Williams & Davids, 1995). Novicescan be defined as ‘beginning athletes with a low level of skill performance that characterizesthe athlete at the beginning of the learning continuum’, whereas the expertise of elite athletesrefers to ‘a high level of skill performance that characterizes a person at the extreme oppositeend of the learning continuum from the beginner’ (Magill, 2003). Although valuable, differences in tactical skills between athletes of opposite ends of the learning continuum donot offer insight in possible differences between athletes in the middle and athletes towardsthe expert end of the learning continuum. Especially in the field of talent identification andtalent development, information on what distinguishes players with an ‘average’, ‘high’ and‘very high’ performance level (i.e. more towards the expert end of the learning continuum) isessential. This information could help national Olympic committees and sporting bodies toidentify talented athletes. In addition, it might prove valuable in developing talented athletes’potential towards expertise more efficiently. Therefore, the aim of this study is to examinewhether youth athletes with ‘average’ (regional), ‘high’ (sub-elite), and ‘very high’ (elite)

Knowing what to do and doing it

21

performance levels differ with respect to their self-assessed tactical skills (i.e. declarative andprocedural knowledge). In line with recent research on tactical skills of world-class youth soccer teams (Kannekens, Elferink-Gemser & Visscher, 2009), it is hypothesized that eliteyouth players outscore sub-elite and regional youth players on tactical skills.

3.2 Methods

ParticipantsParticipants were 191 youth field hockey players (mean age 15.5, s=1.6) of whom 94 were

boys (mean age 15.7, s=1.7) and 97 girls (mean age 15.3, s=1.5). At the time of this study, therewere 172,477 competitive field hockey players (n = 73,931 adult players and n = 98,546 youthplayers) in the country. Of all the Dutch youth players, most (n = 97,010) were regional playerswho played in a regional competition. In our study, talented players were defined as youthplayers in the 12-19 age categories who perform better than their peers during training andcompetition, and have the potential to reach the elite level (Elferink-Gemser, 2005; Helsen,Hodges, Van Winckel, & Starkes, 2000; Howe, Davidson, & Sloboda, 1998). At the time ofthe current study, 1536 Dutch youth players (1.6%) were considered “talented players” bytrainers and coaches, were part of a talent development program of a club of national prestige,and competed at the highest national level for their age. Based on their level of performance,they were further subdivided into elite and sub-elite youth players. Elite players also playedin a district- or national youth team of the Dutch Field Hockey Association. In the Netherlands, there were 492 elite youth players, 0.5% of all Dutch youth field hockey playersat that time.

All players had at least three years of field hockey experience (i.e. more than 250 trainingsessions and over 125 games) and were competing in regional (n=59; mean age 15.3, s=1.6),sub-elite (n=86; mean age 15.6, s=1.8), or elite (n=46; mean age 15.4, s=1.3) competition. Table3.1 presents the characteristics of the participants.

ProceduresAll players and parents were informed of the procedures used in the study and provided

informed consent to participate. The field hockey clubs and trainers gave their permissionfor this study, as all procedures were in accordance with the standards of the ethics committeeof the University of Groningen. Data were collected at the end of the competitive field hockeyseason.

Chapter 3

22

Table 3.1. Characteristics of youth field hockey players divided by performance level (mean±s)

Regional Sub-elite Elite(n = 59) (n = 86) (n = 46)

Age (years) 15.3±1.6 15.6±1.8 15.4±1.3Field hockey experience (years) 6.2±2.1* 8.2±2.1 8.5±1.6Field hockey practice (hrs/week) 2.6±1.0* 4.2±1.4+ 4.8±1.8Non-field hockey practice (hrs/week) 2.4±3.2 2.6±2.8 3.6±3.7Games (per week) 1.0±0.2 1.0±0.2 1.1±0.3

* Significantly lower scores than sub-elite and elite youth players (P<0.05)+ Significantly lower scores than elite youth players (P<0.05)

Tactical characteristicsThe tactical skills Knowing about ball actions, Knowing about others, Positioning and deciding,

and Acting in changing situations were measured with the Tactical Skills Inventory for Sports(TACSIS; Elferink-Gemser, Visscher, Richart & Lemmink, 2004). This self-assessment questionnaire was based on the theoretical elements of tactical skills according to the framework of McPherson (1994) with one continuum that moves from response selection toresponse execution and another that moves from knowledge (knowing what to do) to action(doing it).

The Tactical Skills Inventory for Sports consists of twenty-two items that have to be answered on a 6-point Likert scale regarding sport performance, ranging from 1 (very poor)to 6 (excellent), or from 1 (almost never) to 6 (always), while the player has to compare himselfor herself with top players in the same age category (see Table 3.2). The scales Knowing aboutball actions and Knowing about others contain questions more related to declarative knowledgewhereas the scales Positioning and deciding and Acting in changing situations contain questionsmore related to procedural knowledge. In addition, questions regarding Knowing about ballactions and Positioning and deciding mostly concern situations in which the team possesses theball. Questions in Knowing about others and Acting in changing situations, on the other hand,mostly concern situations in which the opposing team possesses the ball. By combining bothways of categorizing elements of tactical skills, (i.e. declarative versus procedural knowledgeand attack versus defense), the four factors in the Tactical Skills Inventory for Sports cover allthese four aspects of tactical skills.

In previous research, the Tactical Skills Inventory for Sports was shown to be a reliablepsychometric instrument with internal consistency coefficients of all four scales ranging fromCronbach’s alpha 0.72 to 0.89 (Elferink-Gemser et al., 2004b). Positioning and deciding, Knowingabout others, and Acting in changing situations had Intraclass Correlation Coefficients of repeatedmeasures varying of between 0.76 and 0.88. The scale Knowing about ball actions had an Intraclass Correlation Coefficient of 0.60 (Elferink-Gemser et al., 2004b).


23

Table 3.2. Items of the Tactical Skills Inventory for Sports (TACSIS; Elferink-Gemser et al., 2004b)

Knowing about ball actions1

1 I know exactly when to pass the ball to a teammate or when not to2 If we receive the ball (getting ball possession), I know exactly what to do3 While executing an action in a match, I know exactly what to do subsequently4 If I possess the ball, I know exactly whom I have to pass toKnowing about others2

1 My judgment of the opponent’s play is2 I know quickly how the opponent is playing3 Although I do not see my opponents, I know where they are going4 Without seeing my teammates, I know where they are going5 If an opponent receives the ball, I know exactly what he is going to doPositioning and Deciding3

1 Decisions I make during matches about proceeding actions are generally2 I know how to get open during a match3 My positioning during a match is generally4 My overview (in ball possession or in team’s ball possession) is5 My anticipation (thinking about proceeding actions) is6 I am good at taking the right decisions at the right moments7 In the opinion of my trainer, my understanding of the game is8 My getting open and choosing position is9 In the opinion of my trainer, my positioning isActing in changing situations4

1 My interception of the opponent’s ball is2 My interception of the ball is3 If our team loses the ball during a match, I quickly switch to my task as defender4 I quickly react to changes, as from not possessing the ball to ball possession

Note: items had to be answered on a 6-point scale, ranging from 1 = very poor to 6 = excellent or from1 = almost never to 6 = always, while comparing oneself with top players in the same age category.1 scale for declarative knowledge in attacking situations2 scale for declarative knowledge in defensive situations3 scale for procedural knowledge in attacking situations4 scale for procedural knowledge in defensive situations

Training characteristicsIn the general questions of the questionnaire, players were asked how many years they

had played competitive field hockey. Players also recorded in the hours of field hockey trainingsessions they attended per week. In addition, players were asked how many field hockey gamesthey played per week and how many hours per week they trained in other sports, or by themselves, throughout the year. Time spent in physical education at school, which was onaverage 2.5 hours per week, was excluded. The variables representing training characteristics

Chapter 3

24

for the present study were field hockey experience (years), field hockey practice (hours perweek), non-field hockey practice (hours per week), and number of field hockey games perweek.

Data analyses Mean scores and standard deviations were calculated for each subgroup on each scale of

the Tactical Skills Inventory for Sports. To interpret the scores, effect sizes (d) between elite,sub-elite, and regional players were calculated and related to behaviour (Andersen, McCullagh& Wilson, 2007). Cohen (1988) suggested that effect sizes around 0.20 are small, around 0.50are moderate, and around 0.80 are large. In addition, a multivariate analysis of covariance(MANCOVA) was carried out (factors: performance level and gender) to determine the effectof performance level and gender on the dependent variables, the four scales of the TacticalSkills Inventory for Sports. Where appropriate, univariate analyses of covariance (ANCOVA)on each dependent variable were conducted as follow-up tests to the MANCOVA to determine the nature and source of differences found.

Finally, to determine whether the self-assessed tactical skills could predict group membership of the players (i.e. whether they belonged to the elite, sub-elite, or regionalgroup), a stepwise discriminant function analysis was used with the tactical skills as independent variables and performance level as the dependent variable. An alpha of 0.05 wasadopted for all tests of significance.

3.3 Results

Table 3.3 presents results of the MANCOVA and Table 3.4 presents means and standarddeviations of regional, sub-elite and elite youth players on the four scales of the Tactical SkillsInventory for Sports. In addition, effect sizes between the three performance groups are presented.

Table 3.3. Results of MANCOVA for performance level, gender, and performance level X gender with age

as covariate

Wilks’ Lambda F-value Hypothesis df Error df P-value

Age 0.952 2.299 4 181 0.061Performance level 0.812 4.957 8 362 <0.001Gender 0.979 0.967 4 181 0.427Performance level X gender 0.969 0.712 8 362 0.681

Note: df=degrees of freedin


25

Effect sizes between regional and sub-elite players were moderate on scales for declarativeknowledge (0.42<d<0.49) and moderate to large on scales for procedural knowledge and sumof scales (0.57<d<0.69). Effect sizes between regional and elite players were moderate to largeon all scales for tactical skills (0.59<d<1.20). Effect sizes between elite and sub-elite playerswere small on all scales (d<0.33) with the exception of Positioning and deciding which was moderate (d=0.49).

Multivariate analyses showed a significant main effect for the level of performance(P<0.05). Gender had no influence on the scores (P>0.05) (see Table 3.3). Univariate analysesof covariance showed that elite players outscored regional players on all tactical skills as measured with the Tactical Skills Inventory for Sports; Knowing about ball actions (F2,184=15.1;P<0.05), Knowing about others (F2,184=11.3; P<0.05), Positioning and deciding (F2,184=36.9;P<0.05), Acting in changing situations (F2,184=8.3; P<0.05), and sum of scales (F2,184=27.2;P<0.05). In addition, sub-elite players outscored regional players on all tactical skills as measured with the Tactical Skills Inventory for Sports; i.e., Knowing about ball actions(F2,184=8.2; P<0.05), Knowing about others (F2,184=6.5; P<0.05), Positioning and deciding(F2,184=16.8; P<0.05), Acting in changing situations (F2,184=10.6; P<0.05), and sum of scales(F2,184=16.4; P<0.05).

Table 3.4. Tactical skills of youth field hockey players measured with the TACSIS based on performance level

(mean±s)

Regional (R)Sub-elite (S) Elite (E) d d dn = 59 n = 86 n = 46 R vs S R vs E S vs E

Scales for declarative knowledgeKnowing about ball actions 3.85±0.76* 4.21±0.71 4.38±0.60 0.49 0.77 0.26Knowing about others 3.46±0.71* 3.73±0.57 3.91±0.64 0.42 0.67 0.30

Scales for procedural knowledgePositioning and deciding 3.50±0.57* 3.89±0.56+ 4.15±0.51 0.69 1.2 0.49Acting in changing situations 3.88±0.64* 4.24±0.62 4.26±0.65 0.57 0.59 0.03

Sum of scales 3.67±0.53* 4.02±0.48 4.17±0.43 0.69 1.03 0.33

Note: d around 0.20 is small; d around 0.50 is moderate; d around 0.80 is large (Cohen, 1988) * Significantly lower scores than sub-elite and elite youth players (P<0.05)+ Significantly lower scores than elite youth players (P<0.05)

Chapter 3

26

Univariate analyses of covariance showed no differences between elite and sub-elite players on the scales Knowing about ball actions (F2,184=2.1; P>0.05), Knowing about others(F2,184=2.7; P>0.05), Acting in changing situations (F2,184=0.04; P>0.05), and sum of scales(F2,184=3.5; P>0.05). On the tactical skill Positioning and deciding elite players scored higherthan sub-elite players (F2,184=6.8; P<0.05).

The results of the stepwise discriminant analysis are presented in Table 3.5. The modelpredicted that the scale Positioning and deciding would successfully discriminate between theelite, sub-elite, and regional youth players.

Table 3.5. Summary of the stepwise discriminant analysis: Variables entered/removed. a,b,c,d

Wilks’ Lambda

Exact F

Step Entered Statistic df1 df2 df3 Statistic df1 df2 P-value

1 Positioning and deciding 0.833 1 2 188 18.800 2 188 < 0.001

Note: At each step, the variable that minimises the overall Wilks’ lambda is entered.a. Maximum number of steps is 10.b. Minimum partial F to enter is 3.84.c. Maximum partial F to remove is 2.71.d. F level, tolerance, or VIN insufficient for further computation.

The average squared canonical correlation was 0.408. Based on scores on Positioning anddeciding, estimation of the percent variance accounted for 41%. When group membership ispredicted from Positioning and deciding, 48.2% of the original grouped players are classifiedcorrectly (see Table 3.6). The other self-assessed tactical skills provided no additional information when discriminating further between the three groups of players.

Table 3.6. Classification results of the stepwise discriminant analysis (n and %)a

Predicted group membership

Performance level Regional Sub-elite Elite Total

Original Regional 37 (62.7%) 13 (22%) 9 (15.3%) 59 (100%)Sub-elite 31 (36%) 27 (31.4%) 28 (32.6%) 86 (100%)

Elite 7 (15.2%) 11 (23.9%) 28 (60.9%) 46 (100%)

a 48.2% of original grouped cases correctly classified


27

3.4 Discussion

The aim of this study was to determine whether youth athletes with ‘average’ (regional),‘high’ (sub-elite), and ‘very high’ (elite) performance levels differ with respect to their self-assessed tactical skills. All athletes in this study had a substantial amount of competitive fieldhockey experience that included at least 250 field hockey training sessions in addition to competing in at least 125 field hockey games. Where previous studies were based on differences between experts and novices (i.e. between athletes on opposite sides of the learningcontinuum), this study focused on athletes in the middle to the expert end of this continuum.Similar to studies showing elite athletes outscoring novices on declarative and proceduralknowledge bases (e.g. Allard, Deakin, Parker & Rodgers, 1993; Thomas, French &Humphries, 1986; Williams et al., 1993), the current study shows superior declarative andprocedural knowledge of both elite and sub-elite players in contrast to regional players. Eliteand sub-elite players, who form the top 1.6% players of all Dutch youth field hockey players,not only know what to do better than regional players during a field hockey game but are alsobetter at performing the right action at the right moment. The minimum score on the TacticalSkills Inventory for Sports is 1 and the maximum score is 6. On average, the difference between regional and talented players on the scales was 0.5 of a point, indicating that thelatter had 10% higher scores than the regional players. Moderate to large effect sizes representmeaningful differences between the two groups. For example, the talented players seem toknow better how to get open during a match, to have a better understanding of the game, andto make better decisions during matches about proceeding actions. The different scores onthe Tactical Skills Inventory for Sports between regional and talented players represent thedifference between ‘sometimes’ or ‘mostly’ making the right decision at the right moment, ormaking ‘fair’ in contrast to ‘good’ decisions.

An explanation for these results could be found in deliberate practice theory (Ericsson,1996; Ericsson, Krampe & Tesch-Römer, 1993). According to this theory, expert performanceis the result of many hours of deliberate practice and it is this accumulated practice that underlies the main difference between athletes of different performance levels (Ericsson, 2002,2003). In the present study, talented field hockey players (i.e. elite and sub-elite players) ofabout 15 years of age had already accumulated significantly more years of field hockey experience than regional players (see Table 3.1), which appears to be related to better developed tactical skills. On average, regional players started playing field hockey at 9 yearsof age, whereas the talented players started on average at the age of 7 years. Also, the qualityof practice seems important. Since all talented (elite and sub-elite but not regional) playersmembers of a talent development program of a prestigious club, they were guided by verygood trainers and coaches. On the other hand, genetics might play a significant role in explaining variation in phenotypes related to sports performance (Button & Abbott, 2007;Klissouras, Geladas & Koskolou, 2007), with talented athletes having a more favourable

Chapter 3

28

genetic make up for a high performance level in field hockey than regional players. Intuitively,it is thought that they possess certain personality characteristics that facilitate learning, training, and competition (Williams & Reilly, 2000). Therefore, the differences between regional and talented players have presumably developed over time.

An important task for those involved in talent identification and talent development isto assess the future potential of the athletes (Button & Abbott, 2007) and a possible implicationof these results is that training of tactical skills could be an integral part of a talent developmentprogram. Nevertheless, the results should be interpreted with caution. The discriminant function analysis showed that, based on tactical skills, 48% of the players were classified correctly. Over 60% of the regional and elite players were classified correctly but only 31%of sub-elite players. In addition, results of the canonical correlation accounted for 41% of thevariance. This percentage is considered highly relevant since only one domain of performancecharacteristics is taken into account in this study- tactical skills. Other performance characteristics that were not incorporated in the analysis, susch as biological and behaviouralcharacteristics, could explain some additional variance. Previous research has shown that inaddition to tactical skills, other performance characteristics are needed to become an elitefield hockey player, such as outstanding physiological, technical, and psychological skills(Elferink-Gemser et al., 2004a; Elferink-Gemser, Visscher, Lemmink, & Mulder, 2007). Thereis a long history in psychology of using multidimensional test batteries to predict either expertperformance or the probability of attaining expert levels of performance in tasks that involvea combination of perceptual and motor skills, for example in aircraft pilots (e.g., Damos, 1996;Fleishman, 1966; Hunter & Burke, 1994; Starkes, 1987). Not surprisingly, the underlyingcomponents of expertise in sports have been found to be not only tactical in nature, but alsophysical and include sport-specific technical aspects, as well as motivation (Helsen & Starkes,1999; Reilly, Williams, Nevill, & Franks, 2000; Starkes, 1987).

Moving along the learning continuum towards the expert end, we made a comparisonbetween elite and sub-elite youth players. In contrast to the differences between elite athletesand novices, or between regional players and talented athletes, elite players did not outscoresub-elite players on all tactical skills. Sub-elite youth players showed equally high levels ofdeclarative knowledge as elite youth players, meaning they knew very well what to do on thefield. However, differences in Positioning and deciding were observed, one of the scales for procedural knowledge, with the elite players outscoring the sub-elite ones. The discriminantfunction analysis also revealed this scale to be the most important scale in classifying playersaccording to their level of performance. Sport is unique in that tactical skills do not only involve the ability to determine what action is most appropriate in a given situation, but alsowhether this action can be successfully executed within the constraints of the required movement. These constraints are cognitive, physiological as well as technical, and limit theactions available to the athlete (Janelle & Hillman, 2003). So, although athletes are free incirculating between plan (strategy) and tactical adaptation, this freedom in circulation is


29

margined by personal characteristics and the game situation circumstances (Mouchet, 2005).This means that the ability to make a certain tactical decision may not always equate with theability to carry out or execute the decision. In line with this, the current study shows that talented (elite and sub-elite) players appear to have well-developed knowledge about procedures, the rules of the game, team mates and opponents, and ball actions. However,when it comes to the interpretation of a specific situation, to decide in a split second with opponents restricting time and space, to be at the right place at the right moment and to perform the right action at the right moment, elite players are better than sub-elite players.Questions for the scale Positioning and deciding, the one scale distinguishing between elite andsub-elite youth players, mostly concern situations in which the team possesses the ball (i.e.questions in this scale are more related to the attack than to defense). When a team has possession of the ball and is attacking, time seems more constrained than in defense. Playershave less time to make a decision and perform the action if the ball is in their optic array fora relatively short period of time, as is the case when attacking. Elite players appears to be ableto deal with these time constraints better than sub-elite players. Again, some caution in interpretation of these results is warranted, since the discriminant function analysis revealedthat only a of sub-elite players were correctly classified, in contrast to over 60% of all eliteplayers. It would appear that performance characteristics other than tactical skills determinethe performance of sub-elite players. Results from a longitudinal study on multidimensionalperformance characteristics (i.e. anthropometric, physiological, technical, tactical, and psychological characteristics) showed that elite youth players scored better than sub-eliteyouth players on technical and tactical variables. Future elite players already seemed to excelin tactical skills by the age of 14 already (Elferink-Gemser et al., 2007).

An important issue concerns the reasons for the differences found. What are the underlying mechanisms resulting in expert positioning and deciding in field hockey? Eliteplayers did not accumulate more years of field hockey experience than sub-elite players (seeTable 3.1). In addition, both elite and sub-elite players train predominantly with their clubteam, so that training sessions are similar for both elite and sub-elite players because they arepart of the same team competing in the highest national league and comprise the best 1.6%Dutch field hockey players of their age. An explanation for the difference observed could bethat elite players benefit more from this training than sub-elite players-(Williams & Reilly,2000). Another explanation relates to the extra field hockey-specific training elite players havewith a district or national team (see Table 3.1). About 30 minutes of field-hockey specifictraining per week at the expert level is not only significantly more in relation to sub-elite players, it also seems highly relevant in terms of positioning and deciding. In district or nationalteams, players have to train and compete at an even higher level of performance than in theirclub team. The consequence of training with players at the highest performance level is havingless time for picking up relevant cues, making a decision, and performing the appropriate action. Compared with sub-elite players, elite players have more experience in these highly

Chapter 3

30

demanding field hockey situations and it is speculated that they therefore learn better how todeal with the time constraints needed in positioning and deciding. Further research is neededto assess the relationship between learning and constructs facilitating learning, such as selfregulation with tactical skills of talented athletes.

Since no golden standard exists to measure tactical skills, ways to assess tactical skills include: in a laboratory setting with pictures or film slides of game situations (e.g., Allard etal., 1993), the expert opinion of coaches, and self-assessment (e.g., Elferink-Gemser et al.,2004b). Unfortunately, each form of assessment has drawbacks, especially in terms of theirvalidity. For example, laboratory studies lack real-world settings in which a player has to dealwith pressure of opponents, time constraints and stress (McPherson & Kernodle, 2003). Incontrast, the expert opinion of coaches and self-assessment of the players have problems ofsubjectivity, although they can take into account real-world settings. In this study, tacticalskills were measured with the Tactical Skills Inventory for Sports; a practical and reliablemeasure of tactical skills in invasive game players (Elferink-Gemser et al., 2004b). Althoughthis questionnaire was constructed with help of expert trainers and embedded in theory, itcannot be ascertained that it measures the whole concept of tactical skills. In addition, self-report measures are susceptible to a person’s self-confidence and, since confidence is associated with elite performance in various sports, this might have affected the results (e.g.,Mahoney, Gabriel and Perkins, 1987; Woodman and Hardy, 2003).

Caution should be taken when generalizing the results to other populations. The presentsample consisted of competitive youth field hockey players from the Netherlands and Dutchis the original language in which the Tactical Skills Inventory for Sports was constructed. Todate, the English version of the inventory has yet to be applied and it should not be assumedthat similar results would be obtained. Future research could be directed to populations ofcompetitive sports athletes in other categories of formal games and in other countries. Moreover, it would be valuable to study the tactical skills from the Tactical Skills Inventoryfor Sports with other measuring instruments than the self-reported inventory.

In conclusion, the relation between tactical skills (i.e. declarative and procedural knowledge) as shown in novice-expert studies also holds when the performance level are regional and sub-elite level. However, when performance level reaches the expert level, declarative knowledge is no longer able to differentiate between elite and sub-elite players incontrast to an aspect of procedural knowledge (i.e. positioning and deciding).


31

References

• Allard, F., Deakin, J., Parker, S. & Rodgers, W. (1993). Declarative knowledge in skilledmotor performance: byproduct or constituent? In Cognitive issues in motor expertise (editedby J. L. Starkes & F. Allard), pp. 95-107. Amsterdam: North Holland Elsevier science publishers.

• Andersen, M. B., McCullagh, P. & Wilson, G. J. (2007). But what do the numbers reallytell us? Arbitrary metrics and effect size reporting in sport psychology research. Journal ofSport & Exercise Psychology, 29, 664-672.

• Anderson, J. R. (1982). Acquisition of cognitive skill. Psychological Review, 89, 369-406.• Baker, J., Côté, J. & Abernethy, B. (2003). Learning from the Experts: Practice activities

of expert decision makers in sport. Research Quarterly for Exercise and Sport, 74, 342–347.• Bjurwill, C. (1993). Read and react: the football formula. Perceptual and Motor Skills, 76,

1383-1386.• Button, C. & Abbott, A. (2007). Nature-nurture and sport performance. International

Journal of Sport Psychology, 38, 83-88.• Cohen, J. (1988). Statistical power analysis for the behavioural sciences.Hillsdale, NJ: Lawrence

Erlbaum Associates.• Damos, D. L. (1996). Pilot selection batteries: shortcomings and perspectives. International

Journal of Aviation Psychology, 6, 199-209.• Elferink-Gemser, M.T. (2005). Today’s talented youth field hockey players, the stars of tomorrow?

A study on talent development in field hockey. Wageningen: Ponsen & Looijen bv.• Elferink-Gemser, M. T., Visscher, C., Lemmink, K. A. P. M. & Mulder, Th. (2004).

Relation between multidimensional performance characteristics and level of performancein talented youth field hockey players. Journal of Sports Sciences, 22, 1053-1063.

• Elferink-Gemser, M. T., Visscher, C., Lemmink, K. A. P. M. & Mulder, Th. (2007).• Multidimensional performance characteristics and performance level in talented youth field

hockey players: A longitudinal study. Journal of Sports Sciences, 25, 481-489.• Elferink-Gemser, M. T., Visscher, C., Richart, H. & Lemmink, K. A. P. M. (2004).

Development of the Tactical Skills Inventory for Sports. Perceptual and Motor Skills, 99,883-895.

• Enns, J. & Richards, J. (1997). Visual attentional orienting in developing hockey players.Journal of Experimental Child Psychology, 64, 255-275.

• Ericsson, K. A. (1996). The acquisition of expert performance: An introduction to some ofthe issues. In The road to excellence: The acquisition of expert performance in the arts and sciences,sports and games (edited by K. A. Ericsson), pp. 1-50. Mahwah, NJ: Lawrence Erlbaum Associates.

Chapter 3

32

• Ericsson, K. A. (2002). Attaining excellence through deliberate practice: Insights from thestudy of expert performance. In The pursuit of excellence in education (edited by M. Ferrari),pp. 21-55. Hillsdale, NJ: Erlbaum.

• Ericsson, K. A. (2003). Development of elite performance and deliberate practice. In Expertperformance in sports: Advances in research on sport expertise (edited by J. L. Starkes & K. A.Ericsson), pp. 49-87. Champaign, IL: Human Kinetics.

• Ericsson, K. A. & Kintsch, W. (1995). Long Term Working Memory. Psychological Review,102, 211-245.

• Ericsson, K. A., Krampe, R. T. & Tesch-Römer, C. (1993). The role of deliberate practicein the acquisition of expert performance. Psychological Review, 100, 363-406.

• Fleishman, B. A. (1966). Human abilities and the acquisition of skill. In B. A. Bilodeau(Ed.), Acquisition of skill (pp. 147-167). New York: Academic Press.

• French, K. E. & McPherson, S. L. (1999). Adaptations in response selection processes usedduring sport competition with increasing age and expertise. International Journal of SportPsychology, 30,173-193.

• French, K. E. & Thomas, J. R. (1987). The relation of knowledge development to children’s basketball performance, Journal of Sport Psychology, 9, 15-32.

• Gréhaigne, J. F. & Godbout, P. (1995). Tactical knowledge in team sports from a constructivist and cognitivist perspective. Quest, 47, 490 - 505.

• Gréhaigne, J. F., Godbout, P. & Bouthier, D. (1999). The foundations of tactics and strategyin team sports. Journal of Teaching in Physical Education, 18, 159-174.

• Gréhaigne, J. F., Godbout, P. & Bouthier, D. (2001). The teaching and learning of decisionmaking in team sports. Quest, 53 (1).

• Helsen, W. F. & Starkes, J. L. (1999). A multidimensional approach to skilled perceptionand performance in sport. Applied Cognitive Psychology, 13, 1-27.

• Helsen, W. F., Hodges, N. J., Van Winckel, J., & Starkes, J. L. (2000). The roles of talent,physical precocity and practice in the development of soccer expertise. Journal of Sports Sciences, 18, 727-736.

• Howe, M. J. A., Davidson, J. W., & Sloboda, J. A. (1998). Innate talents: Reality or myth.Behavioral and Brain Sciences, 21, 399-442.

• Hunter, D. R., & Burke, E. F. (1994). Predicting aircraft pilot-training success: A metaanalysis of published research. International Journal of Aviation Psychology, 4, 297-313.

• Janelle, C. M. & Hillman, C. H. (2003). Expert performance in sport. Current perspectivesand critical issues. In Expert performance in sports: Advances in research on sport expertise (editedby J. L. Starkes & K. A. Ericsson), pp. 19-47. Champaign, IL: Human Kinetics.

• Kannekens, R., Elferink-Gemser, M..T. & Visscher, C. (2009). Tactical skills of worldclassyouth soccer teams. Journal of Sports Sciences, 27, 807-812.

• Klissouras, V., Geladas, N. & Koskolou, M. (2007). Nature prevails over nurture.International Journal of Sport Psychology, 38, 35-67.


33

• Magill, R. A. (2003). Motor learning and control, concepts and applications (7th ed.). New York:McGraw Hill companies.

• Mahoney, M. J., Gabriel, T. J. & Perkins, T. S. (1987). Psychological skills and exceptionalathletic performance. The Sport Psychologist, 1, 181-199.

• McPherson, S. L. (1994). The development of sport expertise: mapping the tactical domain.Quest, 46, 223-240.

• McPherson, S. L. & Kernodle, M. W. (2003). Tactics, the neglected attribute of expertise.Problem representations and performance skills in tennis, In Expert performance in sports.Advances in Research on Sport Expertise (edited by J. L. Starkes & K. A. Ericsson), pp 137-167. United Kingdom: Human Kinetics.

• McPherson, S. L. & Thomas, J. R. (1989). Knowledge development and sport skill performance: Directions for motor behaviour research. Journal of Experimental Child Psychology, 48, 190-211.

• Mouchet, A. (2005). Subjectivity in the articulation between strategy and tactics in teamsports: an example in rugby. Italian Journal of Sport Sciences, 12, 24-33.

• Reilly, T., Williams, A. M., Nevill, A. & Franks, A. (2000). A multidisciplinary approach totalent identification in soccer. Journal of Sports Sciences, 18, 695-702.

• Starkes, J. L. (1987). Skill in field hockey: The nature of the cognitive advantage. Journalof Sport Psychology, 9, 146-160.

• Thomas, K. T. & Thomas, J. R. (1994). Developing expertise in sport: the relation ofknowledge and performance. International Journal of Sport Psychology, 25, 295-312.

• Thomas, J. R., French, K. E. & Humphries, C. A. (1986). Knowledge development andsport skill performance: Directions of motor behaviour research. Journal of Sport Psychology,8, 259-272.

• Turner, P. & Martinek, J. (1999) An investigation into teaching games for understanding:effects on skill, knowledge, and game play. Research Quarterly for Exercise and Sport, 70, 286-296.

• Williams, A. M. (2000). Perceptual skill in soccer: Implications for talent identification anddevelopment. Journal of Sports Sciences, 18, 737 – 750.

• Williams, M. & Davids, K. (1995). Declarative knowledge in sport: a by-product of experience or a characteristic of expertise. Journal of Sport and Exercise Psychology, 17, 259-275.

• Williams, A. M. & Reilly, T. (2000). Talent identification and development in soccer. Jour-nal of Sports Sciences, 18, 657-667.

• Williams, A. M., Davids, K., Burwitz, L. & Williams, J. G. (1993). Cognitive knowledgeand soccer performance. Perceptual and Motor Skills, 76, 579-593.

• Woodman, T. & Hardy, L. (2003). The relative impact of cognitive anxiety and selfconfidence upon sport performance: a meta-analysis. Journal of Sports Sciences, 21, 443-457.

Chapter 3

34


35

Chapter 4

Tactical skills of world-class youth soccer teams

Kannekens, R., Elferink-Gemser, M.T., Visscher, C. (2009)

Journal of Sport Sciences, 27, 807-812

AcknowledgementsThe authors would like to thank all players, and staff of the national youth soccer teams for their participation in this study. In addition,

we thank Steven Doeven as well as the students of the Center for Human Movement Sciences for their help in collecting the data.

Abstract

In this study, we examined the relationship between tactical skills and competition standard of two youth soccer teams by comparing 18 players (age 18-20 years) from the Dutchand 19 (age 18-23 years) from the Indonesian national youth team. All players completed thedeclarative and procedural knowledge scales of the Tactical Skills Inventory for Sport (TACSIS). Multivariate analyses of variances (MANOVA) and effect sizes were conducted toassess between- and within-team differences. There was a positive relationship between competitive standard and level of tactical skills: the higher-ranked (FIFA World Ranking 2005-2006) Dutch players outscored their Indonesian counterparts on the TACSIS subscales Knowing about ball actions (F1.36= 10.58, P<0.01), Knowing about others (F1.36=28.88, P<0.01),and Positioning and deciding (F1.36=10.10, P<0.01). Multivariate analysis of variance revealedno relationship between tactical skills and playing time (P>0.05) in the Dutch team, whereasin the Indonesian team one procedural knowledge factor (Positioning and deciding) did show apositive association (effect size=0.99). In conclusion tactical skills are fundamental to high-level soccer performance. Ample, expert-led training and match experience at a highcompetitive standard, starting at a young age, and high-quality talent development programmes are suggested as key ingredients for the development of good tactical skills.

Keywords: talent development, competitive level, young athletes, declarative knowledge,procedural knowledge

Chapter 4

38

4.1 Introduction

In team sports, tactical skills refer to the ability of an individual player to perform theright action at the right moment and quickly adapt to new configurations of play and the circulation of the ball (Elferink-Gemser, Visscher, Lemmink, & Mulder, 2004a; Grehaigne& Godbout, 1995). For a player to perform the right action at the right moment, with a successful performance or outcome, a proper understanding of the game is required. Hence,besides well-developed physiological and technical characteristics, elite players also need well-developed tactical skills (French & Thomas, 1987; Helsen & Starkes, 1999; Nougier &Rossi, 1999; Starkes, 1987; Williams, Davids, Burwitz, & Williams, 1993). This applies especially to players of invasion games such as soccer, in which players compete on the samefield of action as their opponents. Because the environment in these games changes constantly,decisions must be made quickly and accurately, requiring good tactical skills.

Tactical skills rely on a range of cognitive competencies, including knowledge of thegame and its goals and actions, knowledge of monitoring skills, and knowledge of actionswithin the context of the game (Thomas, French, & Humphries, 1986). These cognitive skillsare typically categorized as declarative and procedural knowledge (Anderson, 1982; Thomas& Thomas, 1994; Turner & Martinek, 1999). Declarative knowledge is taken to denote knowledge of the rules and goals of the game, and hence refers to ‘knowing what to do’(French & Thomas, 1987; McPherson, 1994; Williams & Davids, 1995). Procedural knowledge describes the selection of an appropriate action within the context of the game, inother words ‘doing it’ (McPherson, 1994). Sport is unique in that tactical skills do not onlyinvolve the ability to determine what decision is most appropriate in a given situation, butalso whether this decision can be successfully executed within the constrains of the requiredmovement. These constraints are physiological as well as technical and limit the tactical options available to the athlete (Janelle & Hillman, 2003).

Research into the relationship between tactical skills and performance or competitionstandard has mostly been restricted to identifying expert-novice differences – that is, comparisons of experienced athletes with a high degree of domain-specific declarative as wellas procedural knowledge and athletes that have a limited amount of domain-specific knowledge (e.g., French et al., 1996; Starkes, 1987: Williams & Davids 1995). These studiesall show that expert athletes outscore novice athletes on many aspects of tactical skills: an expert athlete is better able to select the appropriate response for a situation within the contextof the game’s goal structure, based on less information, and can do so more quickly than anovice athlete (e.g., Thomas et al., 1986; Williams et al., 1993).

Although valuable, these studies do not tell us anything about how good players can bedistinguished from the best, a prerequisite in the field of talent development. Differences inhighly experienced players will never be as apparent as differences between experts andnovices, while within an elite team the best players will have different performance profiles

Tactical skills of youth soccer teams

39

from the ‘less able’ players in their team. Soccer teams commonly consist of about 20 playersbut during matches only 11 players can take the field. The best players play more matchesthan the players trainer/coach judges will contribute less to the team performance. To furtherour knowledge of what constitutes a top player and the ultimately foster talent and help improve training and player selection procedures, we assessed promising soccer players of twonational youth teams by exploring the relationship between their scope of tactical skills andthe level at which they compete successfully.

4.2 Methods

ParticipantsParticipants were 18 players of the Dutch (mean age 19.52 ± 0.67) and 19 player of the

Indonesian (mean age 20.86 ± 1.63) national youth soccer team. All players were highly experienced and considered to be the top players of their age category in their respective countries. The two teams differed substantially in terms of competitive standard, with theDutch team being ranking considerably higher than the Indonesian team (FIFA 2005-2006world ranking, 3 versus 110; http://www.fifa.com). Moreover, the Dutch players had 14.24years (s = 1.30) of accumulated organized soccer experience, whereas the Indonesian had 11.74years (s = 2.47).

Both teams were assessed in the Netherlands during their preparation for two formalsoccer events (see Procedures). For each team, the players were allocated to two subgroupsbased on the number of playing minutes each player played at the relevant tournament dividedby the number of games each team had played, with the Dutch team having played five andthe Indonesian team having played three games. Allocation to subgroups was determined bythe median of playing time per game; group 1 for the Dutch team comprised players with22.90 and group 2 with >22.90 playing minutes per game; group 1 for the Indonesian teamcomprised players with � 51.00 and group 2 with > 51.00 playing minutes per game.

ProceduresAll players were informed about the procedures used in the study, following which they

provided their informed consent. The national soccer federations, trainers, and coaches ofthe teams gave their permission for the study. All procedures were in accordance with thestandards of the ethics committee of the University of Groningen. The data of both teamswere collected in the Netherlands: for the Dutch soccer players during their preparation forthe Under-20 World Championship (2005) and for the Indonesian players during their 6-month preparation –conducted under the supervision of Dutch trainers and coaches- forthe qualification tournament for the Asian Games (2006).

Chapter 4

40

The TACSISTo assess the players’ tactical skills, we used the Tactical Skills Inventory for Sports

(TACSIS; Elferink-Gemser, Visscher, Richart & Lemmink, 2004b), which has four subscales:Knowing about ball action, Knowing about others, Positioning and deciding, and Acting in changingsituations. The Dutch team completed the original Dutch inventory, while the Indonesianplayers completed the Bahasa Indonesian version, which had been translated by an educatednative speaker from Indonesia and an expert on soccer. The original TACSIS comprised 34questions. Principal component analysis with the four factors fixed, followed by varimax rotation, yielded a structure that accounted for 50% of the response variance. Twenty-twoitems met the criterion of having a factor loading greater than or equal to 0.55 (Elferink-Gemser et al., 2004b). The Knowing about ball actions scale contains four items related to declarative knowledge of attacking situations, an example of which reads: ‘I know exactly whento pass the ball to a teammate or when not to’. The five items of Knowing about others addressdeclarative knowledge of defensive situations (e.g., ‘I know quickly how the opponent is playing’). Positioning and deciding has nine items on procedural knowledge of attack situations(e.g. ‘My getting open and choosing positions is’), while the four items of Acting in changingsituations gauge procedural knowledge of defensive situations (‘My interception of the opponent’s ball is’). When rating his soccer performance, the respondent was instructed tocompare himself with top players in the same age category, scoring the items on a 6-pointLikert scale ranging from ‘very poor’ to ‘excellent’ or from ‘almost never’ to ‘always’. By testing the four key elements of tactical skills (i.e. declarative versus procedural and attackingversus defensive situations), the TACSIS captures all four aspects of tactical skills.

In previous research the TACSIS was shown to have good psychometric characteristicswith internal consistency coefficients (Cronbach’s alphas) of all four subscales ranging from0.72 to 0.89 (Elferink-Gemser et al., 2004b). The Intra-class correlation coefficients (ICC)for repeated measures was 0.76 for Knowing about others and 0.88 and 0.82 for Positioning anddeciding, and Acting in changing situations, respectively. The scale Knowing about ball actions hadan ICC of 0.60 (Elferink-Gemser et al., 2004b). [For detailed information about the development of the TACSIS, see Elferink-Gemser et al. (2004b).] The TACSIS subscales ofthe Bahasa Indonesia version had good psychometric characteristics with Cronbach’s alphaof 0.73 for Knowing about ball actions, 0.88 for Knowing about others, and 0.87 for Positioning anddeciding. The Acting in changing situations scale had a Cronbach’s alpha of 0.61.

To validate the players’ scores on the TACSIS, three of the teams’ regular trainers wereasked to rate the tactical skills of each Dutch player on a 6-point Likert scale, ranging from‘very poor’ to ‘excellent’. The Spearman’s correlation coefficient between the trainers’ ratingsand the players’ TACSIS sum score was high (rho=0.79, P<0.01). The trainers of the Indonesian team were unable to provide a valid judgment of the players’ tactical skills becausethey only worked with them during the 6 months of preparation for the Asian Games qualification tournament.


41

Data analysesThe teams’ mean scores and standard deviations were calculated for each subscale of the

TACSIS and multivariate analyses of variance (MANOVA) were conducted (factor of team)to examine group (team) differences. Standardized mean scores or effect sizes (d) were alsocomputed; we followed Cohen’s (1988) suggestion in that we classified effect sizes around0.20 as small, around 0.50 as moderate and around 0.80 as large. Next, we analyzed the dataof each team separately. As explained earlier, for each team the players were divided into twogroups based on their playing time during the games of the tournaments. We then comparedthe scores on the TACSIS of these within-team subgroups using a MANOVA (factor of subgroups). Effect sizes were also computed for the two groups of the teams. Statistical significance was set at P < 0.05 for all tests.

4.3 Results

Figure 4.1 presents the means and standard deviations of the four TACSIS subscales forthe Dutch and the Indonesian national youth teams. In table 4.1, effect sizes between theteams and within the teams are presented.

Figure 4.1. Mean scores (s) for the four TACSIS subscales for the two national youth soccer teams

* Significant difference between the Dutch and the Indonesian national team (P<0.05)

Chapter 4

****

**

6

5

4

3

2

1

1.5

2.5

3.5

4.5

5.5

Knowing about ball actions Knowing about others

Declarative knowledge

Positioning and deciding Acting in changing situations

Procedural knowledge

The Netherlands (n=18)Indonesia (n=19)

Scor

es

42

The MANOVA revealed a significant main effect for the differences between the Dutchand Indonesian teams on the TACSIS subscales (F=7.29, P<0.01). The Dutch playersoutscored the Indonesian players on Knowing about ball actions (F1,36=10.58, P<0.01), Knowingabout others (F1,36=28.88, P<0.01), and Positioning and deciding (F1,36=11.53, P<0.01).

Table 4.1. Effect sizes (d) of tactical skills between the teams and within the teams divide by the number of

minutes played per game during international tournaments

TACSIS The Netherlands The Netherlands Indonesiavs. Indonesia most PT vs. less PT most PT vs. less PT

Declarative knowledge subscalesKnowing about ball actions 1.07 0.06 0.06Knowing about others 1.78 0.47 0.07Procedural knowledge subscalesPositioning and deciding 1.12 0.05 0.99Acting in changing situations 0.64 0.10 0.25

Note: A Cohen’s d around 0.20 reflects a small, around 0.50 a moderate and around 0.80 a large effectsize

Figure 4.2 lists the means and standard deviations for the four TACSIS subscales for theDutch and Indonesian players by the number of minutes played per game (PT) during international tournaments. For the Dutch team the MANOVA yielded no significant maineffect (P>0.05). Also for the Indonesian team, MANOVA revealed no significant main effect(P>0.05). However, the players with the most playing time tended to score higher than thosehaving played less on the Positioning and deciding scale (d=0.99, see table 4.1).

4.4 Discussion

Comparing the tactical skills of two national youth soccer teams, we found that the Dutchteam, whose ranking is for superior to that of the Indonesian team according to FIFA’s 2005-2006 world ranking (http://www.fifa.com), outscore the Indonesian team on the TACSIS subscales Knowing about others, Knowing about ball actions, and Positioning and deciding. Althoughsome previous studies found evidence of a positive relationship between tactical skills andcompetitive standard based on expert-novice differences (e.g., McPherson, 1999; Ward &Williams, 2003), the current study is one of the first to confirm such an association within apopulation consisting exclusively of soccer players selected for their national teams.


43

Figure 4.2. TACSIS mean scores (s) for the teams’ soccer players by the number of minutes played per game

(PT) during international tournaments

That the players’ knowledge and application of tactical skills coincided with the standardat which they competed may be explained by the differences in accumulated experience in organized soccer. The Dutch players had spent more time playing organized soccer at ayounger age than their Indonesian counterparts: with a mean age of 21 years, the Indonesianplayers had an average of 12 years of organized soccer experience; in contrast, the Dutch players had a mean age of 20 years and had 14 years of organized soccer experience. Our findings are consistent with the deliberate practice theory, which states that differences between athletes and competitive standard can mainly be attributed to accumulated practice(e.g., Ericsson, Krampe & Teschromer, 1993). Other researchers have also found support forthe importance of practice and competitive experience in the development of highly skill performance (e.g., French & McPherson, 1999; Janelle & Hillman, 2003). The standard ofyouth soccer competition is higher in the Netherlands than it is in Indonesia, creating a morechallenging environment in which not only sound soccer-specific physiological and technicalskills are developed, but also (complex) tactical skills. Advanced physiological and technicalskills will provide the player with more and better tactical options (Janelle & Hillman, 2003).

Recently, empirical research on soccer has been stepped up considerably (see e.g.,Castagna, D’Ottavio, & Abt, 2003; Reilly & Gilbourne, 2003). In the Netherlands many premier league clubs work together with universities to enhance their training methods andtalent development programmes. It is mandatory for Dutch first- and second division soccerclubs to apply for official approval of their talent-development programmes with the RoyalNetherlands Football Association (KNVB, 2008a). This KNVB accreditation guarantees that

Chapter 4

6

5

4

3

2

1

1.5

2.5

3.5

4.5

5.5


Declarative knowledge


Procedural knowledge

Dutch players with most PT (n=9)Dutch players with less PT (n=9)Indonesia players with most PT (n=9)Indonesia with less PT (n=10)

Scor

es

44

the programmes meet the highest standards. One requisite for certification is that the talentdevelopment programmes are delivered by highly skilled coaches and trainers. The trainingcourses for coaches/trainers the KNVB Academy (the Educational Department of the Dutchsoccer federation) are among the world’s best (KNVB, 2008b). As the Academy also offersstate-of-the-art coaching courses for youth team trainers at all levels, this is likely to haveplayed a role in the variance between the Dutch and the Indonesia soccer players. Future research will need to determine whether and in what way differences in the organization andstandard of competition and the quality of coach/trainer educational programmes in differentcountries affect the tactical skills of their elite youth soccer players.

The TACSIS Acting in changing situations scale, which gauges defence situations, failedto show any significant differences between the two national teams. This absence of variationmay be explained by the fact that defensive actions are generally less complicated than attacking ones and that the differences in competitive standard had less of an effect than theyhad in other areas. This is in line with earlier findings of our research group (Elferink-Gemser,Visscher, Lemmink, & Mulder, 2007), in that elite female field hockey players scored higheron tactical skills when they were in possession of the ball than their sub-elite counterparts,but that this difference became non-significant when players were not in possession of theball.

The Indonesian players with the longest playing time per game (i.e. the best players) hada positive association with one procedural knowledge factor. They scored higher than theirteam-mates who had played less long during the tournament on the TACSIS scale Positioningand deciding. Procedural knowledge entails the interpretation of a specific situation and refersto the ability to make a decision in a split second and to be at the right place at the right moment (French & Thomas, 1987; McPherson, 1994; McPherson & Thomas, 1989; Thomas& Thomas, 1994). As all players were selected for the national youth team, the Indonesianplayers must all have been highly skilled and, accordingly, have had similar declarative knowledge in terms of playing time. However, when it came to procedural expertise (i.e. whatthey do with this knowledge), their achievements did make a difference when a trainer/coachhas to decide which players would play and which players would remain in reserve. The playerswith more playing time had more experience in specific game situations and this additionalexperience will have raised their competitive standard (Ward, Hodges, Starkes, & Williams,2007). In contrast, within the Dutch team (ranked third in 2005-2006), no such playing-timedifferences in TACSIS scores were found, possibly because all members were world-class players and any difference between them would have been very small. Moreover, the Dutchplayers had completed almost every stage of their talent development programme and survivedmany selection procedures, as the talent development teams only holds on the best players,making the Dutch team a rather homogenous group. Perhaps the TACSIS is not sensitiveenough to detect small differences in tactical skills when the competing standard of the playerstested is particularly high.


45

In summary, in our study of two national youth soccer teams, we largely confirmed thepositive relationship between level of tactical skills and competitive standard as reported inexpert-novice studies. We hence conclude that tactical skills are indispensable for a successful,high-profile career in soccer. Also within this elite population, the more skilled players outperformed the lesser skilled counterparts on aspects of declarative and procedural knowledge. Based on our findings, we tentatively conclude that core factors for the development of good tactical skills include ample, expert-led training and high-quality talentdevelopment programmes.

Chapter 4

46

References

• Anderson, J. R. (1982). Acquisition of cognitive skill. Psychological Review, 89, 369-406.• Castagna, D., D'Ottavio, S., & Abt, G. (2003). Activity profile of young soccer players

during actual match play. Journal of Strength and Conditioning Research, 17, 775-780.• Cohen, J. (1988). Statistical power analysis for the behavioural sciences. Hillsdale, NJ: Lawrence

Erlbauw Associates.• Elferink-Gemser, M. T., Visscher, C., Lemmink, K. A. P. M., & Mulder, Th. (2004a).

Relation between multidimensional performance characteristics and level of performancein talented youth field hockey players. Journal of Sports Sciences, 22, 1053-1063.

• Elferink-Gemser, M. T., Visscher, C., Richart, H., & Lemmink, K. A. P. M. (2004b). Development of the tactical skills inventory for sports. Perceptual and Motor Skills, 99, 883-895.

• Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., & Mulder, T. (2007). Multidimensional performance characteristics and standard of performance in talentedyouth field hockey players: A longitudinal study. Journal of Sport Sciences, 25, 481-489.

• Ericsson, K. A., Krampe, R. T., & Teschromer, C. (1993). The role of deliberate practicein the acquisition of expert performance. Psychological Review, 100, 363-406.

• French, K. E. & McPherson, S. L. (1999). Adaptations in response selection processes usedduring sport competition with increasing age and expertise. International Journal of SportPsychology, 30, 173-193.

• French, K. E. & Thomas, J. R. (1987). The relation of knowledge development to children's basketball performance. Journal of Sport Psychology, 9, 15-32.

• French, K. E., Nevett, M. E., Spurgeon, J. H., Graham, K. C., Rink, J. E., & McPherson,S. L. (1996). Knowledge representation and problem solution in expert and novice youthbaseball players. Research Quarterly for Exercise and Sport, 67, 386-395.

• Grehaigne, J. F. & Godbout, P. (1995). Tactical knowledge in team sports from A constructivist and cognitivist perspective. Quest, 47, 490-505.

• Helsen, W. F. & Starkes, J. L. (1999). A multidimensional approach to skilled perceptionand performance in sport. Applied Cognitive Psychology, 13, 1-27.

• http://www.fifa.com• Janelle, C. M. & Hillman, C. H. (2003). Expert performance in sport. Current perspectives

and critical issues. In Expert Performance in Sports: Advances in Research on Sport Expertise(edited by J. L. Starkes & K. A. Ericsson), pp. 49-83. Champaign, IL: Human Kinetics.

• KNVB (2008a). Clubs op sterrenjacht. Retrieved November 28, 2008, fromhttp://www.knvb.nl/organisatie/cert._jeugdopl.

• KNVB (2008b). Over KNVB. Retrieved November 28, 2008, from http://www.knvb.nl/organisatie/academie/elearning


47

• McPherson, S. L. (1994). The development of sport expertise - Mapping the tactical domain. Quest, 46, 223-240.

• McPherson, S. L. (1999). Expert-novice differences in performance skills and problem representations of youth and adults during tennis competition. Research Quarterly for Exercise and Sport, 70, 233-251.

• McPherson, S. L. & Thomas, J. R. (1989). Relation of knowledge and performance in boys'tennis - Age and expertise. Journal of Experimental Child Psychology, 48, 190-211.

• Nougier, V. & Rossi, B. (1999). The development of expertise in the orienting of attention.International Journal of Sport Psychology, 30, 246-260.

• Reilly, T. & Gilbourne, D. (2003). Science and football: A review of applied research inthe football codes. Journal of Sports Sciences, 21, 693-705.

• Starkes, J. L. (1987). Skill in field hockey - The nature of the cognitive advantage. Journalof Sport Psychology, 9, 146-160.

• Thomas, K. T. & Thomas, J. R. (1994). Developing expertise in sport - The relation ofknowledge and performance. International Journal of Sport Psychology, 25, 295-312.

• Thomas, J. R., French, K. E., & Humphries, C. A. (1986). Knowledge development andsport skill performance - Directions for motor behavior research. Journal of Sport Psychology,8, 259-272.

• Turner, A. P. & Martinek, T. J. (1999). An investigation into teaching games for understanding: Effects on skill, knowledge, and game play. Research Quarterly for Exerciseand Sport, 70, 286-296.

• Ward, P., Hodges, N. J., Starkes, J. L., & Williams, M. (2007). The road to excellence: Deliberate practice and the development of expertise. High Ability Studies, 18, 119-153.

• Ward, P. & Williams, A. M. (2003). Perceptual and cognitive skill development in soccer:The multidimensional nature of expert performance. Journal of Sport & Exercise Psychology,25, 93-111.

• Williams, M. & Davids, K. (1995). Declarative knowledge in sport - A by-product of experience or a characteristic of expertise. Journal of Sport & Exercise Psychology, 17, 259-275.

• Williams, M., Davids, K., Burwitz, L., & Williams, J. (1993). Cognitive Knowledge andSoccer Performance. Perceptual and Motor Skills, 76, 579-593.

Chapter 4

48


49

Chapter 5

Self-assessed tactical skills in elite youth soccer

players: A longitudinal study

Kannekens, R. , Elferink-Gemser, M.T., Post, W.J., Visscher, C. (2009)

Perceptual and Motor Skills, 109, 459-472

AcknowledgementsThe authors would like to thank all players, trainers, coaches and staff of

the youth soccer teams for their participation in this study. In addition, we thank Steven Doeven as well as the students of the Center for

Human Movement Sciences for their help in collecting the data.

Abstract

Self-assessed tactical skills were investigated among 191 youth soccer players from age14 through 18 playing in different field positions. On a yearly basis, all players completed theTactical Skills Inventory for Sports with scales for attacking and defensive situations and fordeclarative and procedural knowledge. A model to assess whether tactical skills change overtime in each field position was developed using multilevel analysis. The models indicated thatdefenders and midfielders did not improve their tactical skills, whereas attackers increasedtheir tactical skills from age 14 to 18 years. The representing part of tactical skills for defendersis Acting in changing situations, for midfielders Positioning and deciding whereas Knowing aboutball actions was the qualifying factor for attackers. Possible explanations for these differencesin tactical skills among elite youth soccer players are the selection procedures at a youngerage and task-specific experiences.

Keywords: sport, position in the field, attack and defense, task-specific experience, talentdevelopment

Chapter 5

52

5.1 Introduction

Expertise in soccer depends on the accurate and efficient execution of movement patternsand sport-specific cognitive factors (Williams, Davids, Burwitz, & Williams, 1993). To be successful, a soccer player needs the ability to execute the necessary movements, as well asthe ability to perform these movements at the right time, i.e., he needs excellent tactical skills.Tactical skills are based on declarative and procedural knowledge structures (McPherson,1994), so it is not surprising that expert performers possess superior declarative and proceduralknowledge bases as an essential component of performance in sports (Thomas, French,Thomas, & Gallaghers, 1988). Declarative knowledge is taken to denote knowledge of therules and goals of the game, in other words “knowing what to do” (French & Thomas, 1987;McPherson, 1994; Williams & Davids, 1995), whereas procedural knowledge describes theselection of an appropriate action within the context of the game, and hence refers to “doingit” (McPherson, 1994).

Research on tactical skills has so far focused mainly on differences between novices andexpert athletes, showing that expert athletes have better tactical skills than novice athletes,i.e., they have more condition and action concepts that are associated with the goal structureof the game (McPherson & Thomas, 1989). Overall, these studies showed that experts arebetter able to select the appropriate response for a situation within the context of the game’sgoal structure based on less information, and do so more quickly than novice athletes (e.g.,Egan & Schwartz, 1979; Thomas, French, & Humphries, 1986; Williams, et al., 1993). Anexplanation could be that the declarative and procedural knowledge structures of experts aremuch more advanced than those of novice athletes (Anderson, 1982). Unknown, however, ishow declarative and procedural knowledge develops within elite youth soccer players and itsrelevance for talent development.

Before their teenage years, elite youth players are unable to discriminate task-relevantfrom task-irrelevant information (French, Nevett, Spurgeon, Graham, Rink, & McPherson,1996; McPherson & Thomas, 1989; McPherson, 1999). Nevertheless, between the ages of 7and 12 years, problem representations of experts are more highly structured than those ofnovices, although still limited compared to those of adult experts (French, et al., 1996; Nevett& French, 1997). During the development of a successful sports career, action-plan profilesand current-event profiles increase with greater experience, age, and task-specific practice.The ability to synthesize contextual information with expectations stored in memory via acquisition, adaptation and development of domain-specific skills develops integrally (Ericsson& Kintsch, 1995). However, research on the development of tactical skills so far has focusedmainly on comparing groups of experts and novices cross-sectional. In contrast, in the currentstudy, repeated measurements were made for individual players on self-assessed tactical skillsamong elite youth soccer players from age 14 to 18 years.

Tactical skills in elite youth soccer players

53

Since there is no standard measurement of tactical skills, there are different ways to assessthem; for example, in a laboratory setting with picture or film slides of game situations (e.g.,Allard & Burnett, 1985; Allard, Deakin, Parker, & Rodgers, 1993; Chamberlain & Coelho,1993), by using the opinion of expert coaches, or through self-assessment (Elferink-Gemser,Visscher, Richart, & Lemmink, 2004). Unfortunately, besides advantages, each manner of assessment has drawbacks, especially in terms of validity. For example, laboratory studies aredifferent from the real-world settings in which a player has to deal with pressure of opponents,time constraints, and stress (McPherson & Kernodle, 2003). On the other hand, the opinionof expert coaches or players’ self-assessment are subjectivity, although the techniques can takereal-world settings into account. This is confirmed by McPherson (1994), who found that experts not only have more sophisticated declarative and procedural knowledge, they alsohave more sophisticated domain-specific metacognitive strategies compared to novices.

In soccer, efficient organization of the team is essential for optimal development of theabilities of every player, control of the opponents, and successful resolution of a match. Therefore, players are selected for certain positions to perform specific tasks (Gil, Gil, Ruiz,Irazusta, & Irazusta, 2007). Each playing position has different functions and goals during amatch (Williams, et al., 1993). Defensive organization is more structured than offensive formation, with attacking play likely to be much less structured because of its emphasis onflexibility, creativity and improvisation (Hughes, 1990). Therefore, each position has differenttechnical and tactical aspects (Grehaigne, Godbout, & Bouthier, 1999; Taylor, Stephen, &Nic, 2004). The question arises as to whether players in different field positions develop theirtactical skills differently. The aim of this study was to measure the changes over time of self-assessed tactical skills among elite youth soccer players from age 14 to 18 years, playingin different field positions.

5.2 Methods

ParticipantsIn the 2001-2008 period, 256 talented elite youth soccer players (age range = 14-18 years)

were invited to participated in a longitudinal study. They were tested on a yearly basis. Dueto practical problems, testing did not take place in 2004, resulting in seven measurementstimes. If the player changed field position during this period or played fewer than two yearsin the talent development program of the soccer club, he was excluded from the study. Thiswas the case for 65 players, whose data were excluded from this study, resulting in data from191 elite youth soccer players. A total of 410 measurements were taken. The age of the participants was recorded in months at time of measurement to create standardized agegroups, i.e., a 14-years-old player was defined as a player tested within the age range of 13.50to 14.49 years. The distribution of measurements per player per field position and per agecategory is reported in Table 5.1.

Chapter 5

54

All participants played with national (Dutch) premier league clubs and participated intheir club’s talent development program. Players were recruited to the talent developmentprogram at age 12 years. At the start of each season, which lasted on average 37 wk. with apreparation period of 6 wk., the clubs monitored the players’ skills. Throughout the season,the development of the players was evaluated three times. At the third evaluation time in theseason, the coaches and the staff decided if a player was to continue in the talent developmentprogram. The goal of the program is to develop a player to a professional level. All teams inthe program competed at the highest level and players belonged to the top 0.5% of all association soccer players of their age group (Royal Netherlands Football Association; KNVB,2008). The youth had started playing soccer at a mean age of 6.50 years. On average, 14-years-old players had of 6.42 hours soccer practice per week, 15-years-old players 6.97hours per week, 16-years-old players 8.10 hours per week, and 17-years-old players practicedan average of 8.53 hours per week. Eighteen-years-old players had an average of 8.83 hoursof soccer practice per week.

Table 5.1. Number of measurements per player per field position and per age category

Position Age Number of measurements1 2 3 4 5 Total

Defenders 14 8 17 5 2 3 3515 3 11 9 7 4 3416 3 6 13 6 2 3017 2 14 6 14 2 3818 4 6 4 5 2 21

Midfielders 14 3 10 7 3 2 2515 6 13 9 4 3 3516 2 10 8 4 4 2817 4 7 6 4 6 2718 8 2 5 3 4 22

Attackers 14 3 8 3 4 2 2015 6 12 7 4 2 3116 2 9 10 4 1 2617 3 4 5 6 2 2018 9 1 5 2 1 18

Total measurements 66 130 102 72 40 410Number of players 66 65 34 18 8 191


55

ProceduresAll players were informed about the procedures employed in the study, following which

they provided their informed consent. The soccer clubs and trainers gave their permissionfor the study. All procedures were conducted in accordance with the standards of the ethicscommittee of the University Medical Center Groningen and the University of Groningen.

Tactical Skills Inventory for SportsSelf-assessed tactical skills were measured with the Tactical Skills Inventory for Sports

(TACSIS; Elferink-Gemser, et al., 2004). The TACSIS includes questions within the generalareas of Knowing about ball actions, Knowing about others, Positioning and deciding, and Acting inchanging situations; the total consists of 22 items. When rating his own soccer performance,the respondent was instructed to compare himself with top players in the same age category,scoring the items on a 6-point Likert scale ranging from 1: Very poor to 6: Excellent, or from1: Almost never to 6: Always. The scale Knowing about ball actions consists of four items, thescale Knowing about others has five items, the scale Positioning and deciding consists of nine items,and the scale Acting in changing situations has four items. Knowing about ball actions and Knowingabout others contain questions related to declarative knowledge. Questions in the scales Positioning and deciding and Acting in changing situations relate to procedural knowledge. Questions in the scales Positioning and deciding and Knowing about ball actions concern situationsin which the team possesses the ball (i.e., are related to attacking situations). Questions in thescales Knowing about others and Acting in changing situations concern situations in which theopposing team possesses the ball (i.e., are related to defensive situations). The items of theTACSIS are presented in Table 5.2. In previous research the TACSIS was shown to have goodpsychometric characteristics (Elferink-Gemser, et al., 2004). In the current study, the internalconsistency (Cronbach alpha) of Knowing about ball actions, Knowing about others, Positioningand deciding, and Acting in changing situations is respectively .82, .74, .90, and .71.

Data analysisThe multilevel modeling program MlwiN 2.02 (Rasbash et al., 1999) was used to

investigate possible longitudinal changes in self-assessed tactical skills for different positions.Multilevel modeling is a regression model, appropriate for analyzing hierarchically structureddata. It properly accounts for correlations amongst repeated measurements within individualplayers (Peugh & Enders, 2005). The advantage of multilevel modeling is that the number ofmeasurements as well as the temporal spacing of measurements may vary between players,which is the case in this study, where some players participated four or five times whereas others participated only once, twice, or three times (Landau, & Everitt, 2004; Peugh, & Enders, 2005). The analysis gives valid results as long as the missing data are at random. Inthe current study this could be assumed, since the missing observations did not depend ontactical skills. The multilevel procedure created a model of the assessed tactical skills that

Chapter 5

56

included age and field position as independent factor. Random as well as fixed effects wereconsidered. Moreover different interaction terms were considered. By comparing the devianceof the empty model (model without predicting variables) and the final model, the model fitwas evaluated. An alpha of .05 was adopted for all tests of significance.

Table 5.2. Items of the Tactical Skills Inventory for Sports (TACSIS; Elferink-Gemser et al., 2004)

Knowing about ball actions (declarative knowledge: attack)1 I know exactly when to pass the ball to a teammate or when not to2 If we receive the ball (getting ball possession), I know exactly what to do3 While executing an action in a match, I know exactly what to do subsequently4 If I possess the ball, I know exactly whom I have to pass to

Knowing about others (declarative knowledge: defense)1 My judgment of the opponent’s play is2 I know quickly how the opponent is playing3 Although I do not see my opponents, I know where they are going4 Without seeing my teammates, I know where they are going5 If an opponent receives the ball, I know exactly what he is going to do

Positioning and deciding (procedural knowledge: attack)1 Decisions I make during matches about proceeding actions are generally2 I know how to get open during a match3 My positioning during a match is generally4 My overview (in ball possession or in team’s ball possession) is5 My anticipation (thinking about proceeding actions) is6 I am good at taking the right decisions at the right moments7 In the opinion of my trainer, my understanding of the game is8 My getting open and choosing position is9 In the opinion of my trainer, my positioning is

Acting in changing situations (procedural knowledge: defense)1 My interception of the opponent’s ball is2 My interception of the ball is3 If our team loses the ball during a match, I quickly switch to my task as defender4 I quickly react to changes, as from not possessing the ball to ball possession

Note: items had to be answered on a 6-point scale, ranging from 1: Very poor to 6: Excellent or from1 : Almost never to 6 : Always, while comparing oneself with top players in the same age category.


57

5.3 Results

The descriptive statistics of the self-assessed tactical skills of the players are presented inTable 5.3. Based on the descriptives it is shown that the courses are not linear, hence age isincluded as a categorical variable with age 14 as reference. The estimate models of the scalesof the TACSIS for the different field positions are displayed in Table 5.4. For all models, onlythe random intercept was significant.

The model for the TACSIS scale Knowing about ball actions indicated no increasing scoreof the defenders and midfielders from age 14 to 18 years, the attackers’ scored increased. Inaddition, there was a difference in apparent development between defenders/midfielders andattackers: the self-reported development of the attackers was significantly different from theother positions at age 16, 17 and 18 years. There was no difference in scores between the defenders, midfielders and attackers at age 14 years. The model for the TACSIS scale Knowingabout others indicated no increasing score of the defenders and midfielders from age 14 to 18

Chapter 5

Table 5.3. Mean scores, standard deviations, and ranges of self-assessed tactical skills in elite youth soccer players

(N = 191) presented by age and field position

n Knowing about ball actions Knowing about others

M SD Range M SD Range

Defenders 14 35 4.15 (0.76) 2.25-6.00 3.94 (0.59) 3.00-5.40

15 34 4.07 (0.78) 3.00-6.00 3.88 (0.71) 2.20-5.40

16 30 3.96 (0.69) 2.40-5.25 3.91 (0.54) 3.00-5.00

17 38 4.15 (0.70) 2.75-5.50 3.80 (0.56) 2.80-4.80

18 21 4.31 (0.67) 3.25-5.60 4.12 (0.57) 3.20-5.40

Midfielders 14 25 4.33 (0.75) 3.00-6.00 3.89 (0.62) 3.00-5.80

15 35 4.18 (0.64) 2.75-5.40a 3.81 (0.61) 2.80-5.60

16 28 4.12 (0.50) 3.00-5.25 3.79 (0.58) 2.40-4.80

17 27 4.05 (0.72) 2.75-5.25 3.90 (0.64) 2.75-4.80

18 22 4.36 (0.84) 3.00-5.50 4.05 (0.69) 3.20-5.60

Attackers 14 20 3.96 (0.77) 2.00-5.75 3.39 (0.60) 2.40-4.80

15 31 4.24 (0.78) 3.20-6.00 3.59 (0.65) 2.40-5.20a

16 26 4.35 (0.66) 3.25-5.75 3.71 (0.48) 3.20-4.80

17 20 4.42 (0.60) 2.50-5.25 3.91 (0.59) 2.80-5.20

18 18 4.66 (0.63) 3.40-5.75 4.10 (0.50) 3.00-4.80

Note. Knowing about ball actions: declarative knowledge (attack). Knowing about others:

declarative knowledge (defense). Positioning and deciding: procedural knowledge (attack). Acting in

changing situations: procedural knowledge (defense).

a One missing value. b Two missing values. c Three missing values.

58

years. The attackers’ score increased from age 16 years compared to age 14 years. The attackers had a lower score at age 14 years in comparison to the means for players in the othertwo positions. The changes in the attackers’ scores were significantly different from the otherfield positions at age 17 and 18 years. Similarly, the model for the TACSIS scale Positioningand deciding indicated a similar stable pattern for defenders and midfielders. However, the attackers’ mean scores was higher at age 17 and 18 years compared to age 14 years Defenders’,midfielders’ and attackers’ mean scores at age 14 years differed, with midfielders scoring highest, followed by defenders and attackers. The change in attackers’ mean scores were significantly different from those of the defenders/midfielders from age 17 years. The modelfor the TACSIS scale Acting in changing situations indicated an increasing score for the attackersat age 15, 16, 17, and 18 years compared to their score at age 14 years. The defenders had thehighest score, followed by the midfielders and attackers at age 14 years. The changes in theattackers’ mean scores differed significantly from those of the players in other field positionsfrom age 15 years.



M SD Range M SD Range

3.68 (0.41) 3.11-4.64 4.49 (0.58) 3.00-5.50

3.75 (0.55) 2.44-5.33 4.41 (0.70) 3.25-6.00

3.57 (0.50) 2.56-4.78c 4.09 (0.50) 2.75-5.25

3.61 (0.51) 2.22-4.78 4.29 (0.62) 3.00-5.50

3.74 (0.52) 2.67-4.78a 4.54 (0.46) 3.50-5.25

3.96 (0.41) 3.11-4.67a 3.75 (0.69) 2.25-4.75

3.86 (0.41) 3.11-5.00b 3.81 (0.65) 2.50-5.25

3.86 (0.64) 2.67-5.33 4.02 (0.72) 2.50-5.50

3.96 (0.68) 2.78-5.33 3.87 (0.60) 2.75-5.00

3.95 (0.75) 2.44-5.44 3.85 (0.67) 2.50-5.50

3.52 (0.41) 2.89-4.67a 3.46 (0.53) 2.00-4.50

3.50 (0.58) 2.22-4.56 3.79 (0.72) 2.50-5.50

3.57 (0.53) 2.78-4.89 3.72 (0.43) 3.00-4.25

3.84 (0.36) 3.22-4.78 3.75 (0.66) 2.75-5.25

3.81 (0.44) 3.11-5.00a 4.03 (0.64) 3.00-5.25

59

Table 5.4. Multilevel model for scales of the TACSIS based on 410 measurements of elite youth soccer players

aged 14–18 (N = 191)


Parameter Estimate SE Estimate SE

Fixed effectsIntercept 4.114 0.118 3.956 0.101Age 15 -0.045 0.147 -0.037 0.129Age 16 -0.140 0.159 -0.021 0.138Age 17 0.068 0.152 -0.140 0.132Age 18 0.128 0.181 0.126 0.155Midfielder 0.204 0.176 -0.051 0.152Attacker -0.154 0.186 -0.595 0.161Midfielder * age 15 -0.073 0.210 0.010 0.184Midfielder * age 16 -0.035 0.234 -0.097 0.204Midfielder * age 17 -0.298 0.230 0.146 0.200Midfielder * age 18 -0.091 0.255 0.026 0.221Attacker * age 15 0.347 0.221 0.268 0.195Attacker * age 16 0.621 0.239 0.380 0.209Attacker * age 17 0.477 0.248 0.668 0.216Attacker * age 18 0.605 0.274 0.587 0.237

Random effectsIntercept Between-individuals 0.199 0.038 0.127 0.027Residual intercept 0.287 0.027 0.226 0.021

Deviance 812.5 695.9Deviance empty model 838.9 727.6


Fixed effectsIntercept 3.700 0.083 4.497 0.102Age 15 0.079 0.094 -0.021 0.125Age 16 -0.130 0.104 -0.385 0.135Age 17 -0.109 0.100 -0.177 0.130Age 18 -0.019 0.118 -0.023 0.152Midfielder 0.252 0.122 -0.633 0.151Attacker -0.264 0.129 -1.065 0.160Midfielder * age 15 -0.099 0.134 0.030 0.177Midfielder * age 16 0.073 0.152 0.561 0.199Midfielder * age 17 0.150 0.149 0.264 0.195Midfielder * age 18 -0.048 0.166 0.020 0.216Attacker * age 15 0.015 0.140 0.384 0.187Attacker * age 16 0.262 0.155 0.724 0.203Attacker * age 17 0.522 0.163 0.592 0.211Attacker * age 18 0.412 0.184 0.693 0.232

Random effectsIntercept Between-individuals 0.165 0.023 0.168 0.029Residual intercept 0.104 0.010 0.201 0.019

Deviance 494.0 691.6Deviance empty model 532.8 763.2

Chapter 5

60

5.4 Discussion

The aim of this longitudinal study was to investigate the changes from age 14 to 18 yearsof self-assessed tactical skills among elite youth soccer players playing different field positions.To provide players with an external reference, the players were told to compare themselveswith the top players in their age category. Therefore, although the self-reported skills levelwere relatively stable, the players’ actual level of tactical skills did increase with age. This issupported by the fact that technical skills, which are related to tactical skills, are known toimprove in elite players during adolescence (Huijgen, Elferink-Gemser, Post, & Visscher,2009). Tactical skills do not only involve the ability to determine appropriate actions in givensituations, but also whether actions can be successfully executed within the individual’s constraints (McPherson 1994; Starkes, 1993). These constraints are physiological and technical and limit the tactical options of the individual athlete (Janelle & Hillman, 2003). Asimilar concept is found in research about self-perception, in which studies showed that although self-perception stays stable related to peers during adolescence, the individual scoresdiffer with age (Auweele, Bakker, Biddle, Durand, & Seiler, 1999).

Defenders’ and midfielders’ mean scores on self-assessed tactical skills did not increasecompared to the top players of their age category. In contrast, the attackers’ mean score increased on the TACSIS scale Knowing about ball actions from ‘Mostly’ (3.96) to ‘Almost Always’ (4.66) and on the scale Knowing about others from ‘Sometimes’ (3.39) to ‘Mostly’ (4.10).On the TACSIS scales Positioning and deciding and Acting in changing situations, the increasingscore of the attackers were smaller, from ‘Fair’ to ‘Good’. In comparison to defenders andmidfielders, attackers started with lower mean scores on tactical skills when they were 14 yearsof age, but the scores increased during adolescence. An explanation could be the selectionprocedure used by trainers and technical staff for younger age players. In contrast to defendersand midfielders, attackers were probably selected not because of their excellent tactical skillsbut for their physical skills. Attackers must perform the most physical actions at a high intensity level compared to midfielders and defenders (Bloomfield, Polman, & O'Donoghue,2007; Gil, et al., 2007). Trainers and technical staff may select players with the best physicalskills for the attacking positions because they believe that the success of a match depends primarily on this particular group of soccer players (Gil, et al., 2007). On average, around age14 boys reach maximum height (Malina, 2004), and well-developed physical skills are not expected to improve more. So, to be a successful attacker, a player must proceed to developbetter tactical skills. After maturation, it is only the well-developed physical skills that are notexpected to improve more, so to be a successful attacker, a player must proceed to developbetter tactical skills. Future research is recommended to assess the influence of selection procedures and maturation on tactical skills of elite youth athletes.


61

Relations of scores to players’ positionsPositions-specific self-ratings.- Knowing about ball actions taps declarative knowledge in

attacking situations and its content is therefore most closely related to attackers’ skills. An example of a questions in this scale is: “If we receive the ball (getting ball possession), I knowexactly what to do”. This is a specific task for an attacker. The task of an attacker is to try toscore a goal. Time seems more constrained in attacking situations than in defensive situations.Attackers have less time to make a decision and perform the action as the ball is in their visualfield for a relatively short period of time; most of the time the ball is positioned behind theminstead of in front of them. When the team gets in possession of the ball, the attackers mustreact immediately.

Of the groups of players in three positions, midfielders had the highest mean scores atevery age on Positioning and deciding. An example of a self-rating item of this scale is: “myoverview (in ball possession or in team’s ball possession) is”. This is a specific task for midfielders. Midfielders are the connection between the defenders and the attackers; theirtask is to set up the attacking situations and create circumstances as well as possible for theattackers. When the team loses the ball, the midfielders must change their task quickly andprevent the opposing players from reaching their part of the field, thereby creating the bestas possible circumstances for the defenders.

The TACSIS scale Acting in changing situations is related to procedural knowledge in defensive situations and therefore is the scale most representative of the tactical skills neededby defenders. Results showed that defenders’ mean self-ratings on this scale were higher thanthose of midfielders and attackers at all ages. An example item is “my interception of the opponent’s ball is”. This is typically a defensive task, since a defender acts to prevent the opponent from scoring and to take over the ball.

Experience in play.- A possible explanation of the results related to the different field positions could be their task-specific experiences; i.e., that the skills of each player, his teammates and opponents, and his task or position influence any decision a players makesabout upcoming actions. Every position has specific environmental components and tasks onthe field. As a result, attackers have a different environment and tasks than defenders or midfielders. Experience and practice in specific task contexts support the development of tactical skills (Davids & Myers, 1990; French & Thomas, 1987; McMorris, 1999). A commonway to categorize the elements of tactical skills related to the nature of match play in invasiongames is by making a distinction between on-the-ball and off-the-ball situations (Oslin,Mitchell & Griffin, 1998). Tactics related to scoring or attack can be distinguished from tacticsrelated to preventing scoring or defense (Bjurwill, 1993).

In this study, self perceptions of tactical skills were measured with the TACSIS, a practicaland reliable measure of tactical skills in invasive game players (Elferink-Gemser, et al., 2004).Although this questionnaire was constructed with the help of expert trainers and is embeddedin theory (Elferink-Gemser, et al., 2004), one cannot ascertain that it measures the whole

Chapter 5

62

concept of tactical skills. In addition, self-report measures are susceptible to a person’s self-confidence and, since confidence is associated with elite performance in various sports, thismight have affected the results (e.g., Mahoney, Gabriel, & Perkins, 1987; Woodman & Hardy,2003). However, all participants in this study started playing competitive soccer on average atage 7 years, which means they had already accumulated seven years of experience by the timethey were 14 years of age. In addition, they were all part of a developmental program of aprofessional soccer club of national prestige, which means they have been confronted frequently with selection moments. Talent development programs evaluated the developmentof players’ skills at various time throughout the season. When players are confronted withevaluations of their tactical skills by significant others like trainers, coaches, peers, and parentson a regular basis for many years, they ultimately should be good judges of their own abilitiesin comparison to the top in their age category. Therefore, regardless of their confidence, eliteyouth soccer players are thought to have a realistic perspective of their tactical skills. This issupported by research finding higher performance associated with higher scores measuredwith the TACSIS (Kannekens, Elferink-Gemser, & Visscher, 2009). Future investigationscould focus on possible explanations of the differences in tactical skills found among eliteyouth soccer players, e.g., selection procedures at younger ages and task-specific experience.


63

References

• Allard, F. & Burnett, N. (1985). Skill in sport. Canadian Journal of Psychology-Revue Canadienne de Psychologie, 39, 294-312.

• Allard, F., Deakin, J., Parker, S., & Rodgers, W. (1993). Declarative knowledge in skilledmotor performance: Byproduct or constituent? In J.L. Starkes & F. Allard (Eds.), Cognitiveissues in motor expertise (pp. 95-107). Amsterdam: Elsevier Science Publisers B.V.

• Anderson, J. R. (1982). Acquisition of cognitive skill. Psychological Review, 89, 369-406.• Araujo, D., Davids, K., & Hristovski, R. (2006). The ecological dynamics of decision

making in sport. Psychology of Sport and Exercise, 7, 653-676.• Auweele, Y.V., Bakker, F., Biddle, S., Durand, M., & Seiler, R. (1999). Psychology for physical

educators. Champaign, IL: Human Kinetics. • Bjurwill, C., (1993). Read and react: The football formula. Perceptual and Motor Skills, 3,

1383-1386. • Bloomfield, J., Polman, R., & O'Donoghue, P. (2007). Physical demands of different

positions in FA Premier League soccer. Journal of Sports Science and Medicine, 6, 63-70.• Burgess, D. J., Naughton, G., & Norton, K. I. (2006). Profile of movement demands of

national football players in Australia. Journal of Science and Medicine in Sport, 9, 334-341.• Chamberlain, C. J. & Coelho, A. J. (1993). The perceptual side of action: Decision-making

in sport. In J.L. Starkes & F. Allard (Eds.), Cognitive Issues in Motor Expertise (pp. 135-157).Amsterdam: Elsevier Science Publishers B.V.

• Davids, K. & Myers, C. (1990). The role of tacit knowledge in human skill performance.Journal of Human Movement Studies, 19, 273-288.

• Egan, D. E. & Schwartz, B. J. (1979). Chunking in recall of symbolic drawings. Memory &Cognition, 7, 149-158.

• Elferink-Gemser, M. T., Visscher, C., Richart, H., & Lemmink, K. A. P. M. (2004). Development of the tactical skills inventory for sports. Perceptual and Motor Skills, 99, 883-895.

• Ericsson, K. A. & Kintsch, W. (1995). Long-term working-memory. Psychological Review,102, 211-245.

• French, K. E., Nevett, M. E., Spurgeon, J. H., Graham, K. C., Rink, J. E., & McPherson,S. L. (1996). Knowledge representation and problem solution in expert and novice youthbaseball players. Research Quarterly for Exercise and Sport, 67, 386-395.

• French, K. E. & Thomas, J. R. (1987). The relation of knowledge development to children’s basketball performance. Journal of Sport Psychology, 9, 15-32.

• Gil, S. M., Gil, J., Ruiz, F., Irazusta, A., & Irazusta, J. (2007). Physiological and anthropometric characteristics of young soccer players according to their playing position:Relevance for the selection process. Journal of Strength and Conditioning Research, 21, 438-445.

Chapter 5

64

• Grehaigne, J. F., Godbout, P., & Bouthier, D. (1999). The foundations of tactics and strategy in team sports. Journal of Teaching in Physical Education, 18, 159-174.

• Hughes, C. (1990). The winning formula: the Football Association soccer skills and tactics. London: Collins.

• Huijgen, B.C.H., Elferink-Gemser, M.T., Post, W.J., & Visscher, V. (2009) Soccer skill development in professionals. International Journal of Sports Medicine ahead of printing.

• Janelle, C.M., & Hillman C.H., (2003). Expert performance in sport: Current perspectivesand critical issues. In J.L. Starkes & K.A. Ericsson (Eds.), Expert performance in sports: Advances in research on sport expertise (pp. 49-83). Champaign, IL: Human Kinetics.

• Kannekens, R., Elferink-Gemser, M.T., & Visscher, C. (2009). Tactical skills of world-classyouth soccer teams. Journal of Sports Sciences. Epublished ahead of printing

• KNVB (2008). KNVB Jaarverslag 2007/'08. Retrieved November 20, 2008, fromhttp://sportlink.com/manager/dms/index.php?option=com_content&view=article&id=21

• Landau, S. & Everitt B.S. (2004). Analysis of repeated measures II: Linear mixed model. BocaRaton, Fl: Chapman & Hall.

• Mahoney, M. J., Gabriel, T. J., & Perkins, T. S. (1987). Psychological skills and exceptionalathletic performance. The Sport Psychologist, 1, 181-199.

• Malina, R.M., (2004) Growth and Maturation: Basic principles and effects of training. InM.C. Coelho e Silva & R.M. Malina (Eds.), Children and youth in organized sports (pp. 137-161). Coimbra university press.

• McMorris, T. (1999). Cognitive development and the acquisition of decision-making skills.International Journal of Sport Psychology, 30, 151-172.

• McPherson, S. L. (1994). The development of sport expertise - Mapping the tactical domain. Quest, 46, 223-240.

• McPherson, S. L. (1999). Expert-novice differences in performance skills and problem representations of youth and adults during tennis competition. Research Quarterly for Exercise and Sport, 70, 233-251.

• McPherson, S. L. & Kernodle, M. W. (2003). Tactics, the neglected attribute of expertise.Problem representations and performance skills in tennis. In J.L. Starkes & K. A. Ericsson(Eds.), Expert Performance in Sports. Advances in Research on Sport Expertise (pp. 137-167).Champaign (IL): Human Kinetics.

• McPherson, S. L. & Thomas, J. R. (1989). Relation of knowledge and performance in boystennis - Age and expertise. Journal of Experimental Child Psychology, 48, 190-211.

• Nevett, M. E. & French, K. E. (1997). The development of sport-specific planning, rehearsal, and updating of plans during defensive youth baseball game performance. Research Quarterly for Exercise and Sport, 68, 203-214.


65

• Oslin, J. L., Mitchell, S. A. & Griffin, L. L. (1998). The game performance assessment instrument (GPAI): Development and preliminary validation. Journal of Teaching in PhysicalEducation, 17, 231-243.

• Peugh, J. L. & Enders, C. K. (2005). Using the SPSS mixed procedure to fit cross-sectionaland longitudinal multilevel models. Educational and Psychological Measurement, 65, 811-835.

• Rasbash, J., Browne, W., Goldstein, H., Yang, M., Plewis, I., & Draper, D., et al. (1999).A user's guide to MlwiN. London: Institute of Education.

• Starkes, J.L. (1993). Motor experts: Opening thoughts. In J.L. Starkes & F. Allard (Eds.),Cognitive issues in motor expertise (pp. 3-16). Amsterdam: Elsevier.

• Taylor, J. B., Stephen, D. M., & Nic, J. (2004). Behavioural comparisons of positional demands in professional soccer. International Journal of Performance Analysis in Sport, 4, 81-97.

• Temprado, J. J. & Laurent, M. (1999). Perceptual-motor coordination in sport: Currenttrends and controversies. International Journal of Sport Psychology, 30, 417-436.

• Thomas, J. R., French, K. E., & Humphries, C. A. (1986). Knowledge development andsport skill performance - Directions for motor behavior research. Journal of Sport Psychology,8, 259-272.

• Thomas, J. R., French, K. E., Thomas, K. T., & Gallaghers, J. D. (1988). Children's knowledge development and sport performance. In F.L. Smoll, R. A. Magill, & M. J. Ash(Eds.), Children in Sport (3rd ed., pp. 179-202). Champaign (IL): Human Kinetics Books.

• Vereijken, B. & Whiting, H. T. A. (1990). In defense of discovery learning. Canadian Journal of Sport Sciences-Revue Canadienne des Sciences du Sport, 15, 99-106.

• Williams, M. & Davids, K. (1995). Declarative knowledge in sport - A by-product of experience or a characteristic of expertise. Journal of Sport & Exercise Psychology, 17, 259-275.

• Williams, M., Davids, K., Burwitz, L., & Williams, J. (1993). Cognitive knowledge andsoccer performance. Perceptual and Motor Skills, 76, 579-593.

• Woodman, T. & Hardy, L. (2003). The relative impact of cognitive anxiety and self-confidence upon sport performance: a meta-analysis. Journal of Sports Sciences, 21, 443-457.

Chapter 5

66


67

Chapter 6

Positioning and deciding: the key to

professional soccer

Kannekens, R. , Elferink-Gemser, M.T., Visscher, C.

Scandinavian Journal of Medicine & Science in Sports (in press)

AcknowledgementsThe authors would like to thank all players, and staff of the youth soccer teams fortheir participation in this study. In addition, we thank the students of the Center for

Human Movement Sciences for their help in collecting the data. We would also like to thank Jim Lyons for his textual and editorial suggestions.

Abstract

Talent identification and development implicate recognizing youth players who will besuccessful in the future and guiding them to the top. A major determinant of such success istactical skills. To identify possible key factors that help in predicting success over time, thisstudy assesses the tactical skills of 105 elite youth soccer players who participated in a talentdevelopment program at an earlier stage of their sport career (mean age 17,8 ± 0,9). Theseskills were related to their adult performance level, specifically whether they became professionals (n = 52) or amateurs (n = 53). Defenders, midfielders, and attackers completedthe Tactical Skills Inventory for Sports with scales for declarative and procedural knowledgein either attacking or defensive situations. A logistic regression analysis was performed toidentify the tactical skills that contribute to professional performance level in adulthood. Positioning and deciding appeared the tactical skill that best predicts adult performance level(P<0.05). This is especially true for midfielders, with the correct classification of elite youthplayers in the range of 80%. For players scoring high on this skill the odds ratios indicated a6.60 times greater chance that a player became a professional than players scoring low(P<0.05).

Keywords: experts, position in the field, adolescence, tactics

Chapter 6

70

6.1 Introduction

Millions of children play soccer worldwide; however, only a few will reach the top. Themain goal in talent identification and development is to recognize youth players who will besuccessful in the future and help guide them to the top. However, the answer to the intriguingquestion as to which characteristics must be present at an early stage of a player’s sport careerin order for them to become successful in the future is still largely unknown. To date, mostresearch in the field of talent identification and development has focused on comparing successful youth players with their less successful counterparts on one or more performancecharacteristics during adolescence (Reilly et al., 2000; Williams & Reilly, 2000; Elferink-Gemser et al., 2004a; Veayens et al., 2007; Mohamed et al., 2009). There has been little effortin following the development of these players into adulthood. In order to gain a more comprehensive understanding of which factors most influence future success in sport, it is important to measure skill related characteristics of talented players, not only during adolescence, but also through any subsequent level of performance in adulthood. To achieveexpertise, youth players must invest many hours of intensive training over at least 10 years(Ericsson, 1990; Ericsson et al., 1993; Helsen et al., 1998). By the time soccer players are seventeen or eighteen years of age, they move up from youth competitions through adultcompetition and in the subsequent years it becomes apparent whether they actually reach thestatus of professional soccer players or not.

At the elite level, it appears that differences between players are less related to physicaland physiological characteristics, and more to tactics, motivation and specific technical skills(Reilly et al., 2000; Elferink-Gemser et al., 2004a). Tactical expertise is a prerequisite for expertperformance in sports (Janelle and Hillman, 2003) and studies on tactical skills underline thatathletes at a higher performance level consistently outscore players at a lower performancelevel on these measurements (e.g., Helsen & Starkes, 1999; Reilly et al., 2000; Elferink-Gemser et al., 2007; Vaeyens et al., 2007; Kannekens et al., 2009b).

Tactical skills refer to the quality of an individual player to perform the right action atthe right moment and must therefore be distinguished from strategy, which refers to choicesdiscussed in advance with the trainer in order for the team to organize itself (Gréhaigne &Godbout, 1995). Unlike physiological based predictors of high level performance, tacticalskills rely primarily on cognitive skills that are typically categorized as declarative (‘knowingwhat to do’) or procedural knowledge (‘doing it’) (e.g., Anderson, 1982; Thomas & Thomas,1994; Turner & Martinek, 1999; McPherson & Kernodle, 2003). Declarative knowledge hasbeen defined as the knowledge of the rules and goals of the game (French & Thomas, 1987;McPherson, 1994; Williams & Davids, 1995), whereas procedural knowledge refers to the selection of an appropriate action within the context of game play (McPherson, 1994). Theseelements of tactical skills, particularly in invasion games such as soccer, can be further classifiedinto “on-the-ball” and “off-the-ball situations” (Oslin et al., 1998). Furthermore, it is

Positioning and deciding: Key factors talent

71

important to consider that different skills are necessary for the different positions on the field(e.g., defenders, midfielders and attackers) and that each position has qualitatively differenttechnical and tactical aspects associated with it (Grehaigne et al., 1999; Taylor et al., 2004).As such, players with certain specific skill sets are selected for certain positions to perform thespecific tasks required of that position (Gil et al., 2007).

Thus, given the importance of tactical skills as a predictor of future success, this studyassessed the tactical skills of elite youth soccer players at an early stage of their sport careerand made a comparison between the players who reached professional performance level inadulthood and those who became amateurs.

6.2 Materials and method

SubjectsParticipants were elite youth soccer players of Dutch premier league clubs who

participated in their club’s talent development program. They competed, on average, in onematch per week at the highest level of performance and belonged to the top 0.5% of all soccerplayers in their age group (Royal Netherlands Football Association; KNVB, 2008). Overall,the level of Dutch soccer is high with the national team currently ranked in the third place inthe FIFA world ranking (http://www.fifa.com). Data were collected in the period 2001-2008.Data from players who reached adulthood (21 years or older in 2009) were analyzed in thisstudy resulting in a final data set of 105 players (age range: 16-18 years; mean age: 17.81±0.92years). These players were divided into two groups based on their adult performance level.Amateurs (n = 53) are players who play for an amateur club (second division national leagueor lower). Professionals (n = 52) are players who are playing either with a Premier league clubor on the first team of the first division of the national league. The amateurs comprised of 22

Chapter 6

Table 6.1. General characteristics (means, SD) of elite youth soccer players (N=105) specified per performance

level in adulthood and field position

TotalAmateurs Professionals n=53 n=52

Age (years) 17.72 (0.96) 17.89 (0.88) Accumulated organized soccer experience (years) 10.90 (2.54) 11.71 (2.13) Soccer practice (hours/week)** 9.58 (1.58) 11.24 (2.33)* Non-specific sport practice (hours/week) 2.30 (2.11) 2.78 (3.80)

* Significant difference between amateur and professional performance level (P<0.05)** Soccer practice is includes soccer matches

72

defenders, 15 midfielders and 16 attackers. The professional group was comprised of 19 defenders, 16 midfielders and 17 attackers. Table 6.1 presents the general characteristics concerning age and sport practice.

ProcedureAll players were informed of the procedures to be used in the study, following which

they provided their informed consent. All of the soccer clubs and trainers also gave their permission for the study. All procedures were in accordance with the standards of the ethicscommittee of the University Medical Center Groningen and the University of Groningen.

Tactical Skills Inventory for SportsThe Tactical Skills Inventory for Sports (TACSIS; Elferink-Gemser et al., 2004b), with

subscales of declarative and procedural knowledge, was used to assess the players’ tacticalskills. The scales Knowing about ball actions (e.g., “I know exactly when to pass the ball to ateammate or when not to”) and Knowing about others (e.g., “I know quickly how the opponentis playing”) contain four and five items respectively that were related to declarative knowledge.The scales Positioning and deciding (e.g., “My getting open and choosing positions is”) andActing in changing situations (e.g., “My interception of the opponents is”) consisted of nine andfour items respectively that gauged procedural knowledge. Questions in the scales Knowingabout ball actions and Positioning and deciding concern attacking situations (i.e., situations inwhich the team possesses the ball) whereas questions in the scales Knowing about others andActing in changing situations are related to defensive situations (i.e., situations in which the opposing team possesses the ball). When rating their soccer performance, the players wereinstructed to compare themselves with the top players in the same age category, scoring theitems on a 6-point Likert scale, ranging from ‘Very poor’ to ‘Excellent’ or from ‘Almost never’to ‘Always’.


Defenders Midfielders AttackersAmateurs Professionals Amateurs Professionals Amateurs Professionalsn=22 n=19 n=15 n=16 n=16 n=17

17.80 (0.90) 17.95 (0.88) 17.68 (0.91) 17.55 (0.93) 17.68 (1.11) 18.14 (0.78) 11.14 (1.89) 11.95 (2.54) 11.27 (1.97) 11.50 (1.71) 10.22 (3.67) 11.65 (2.09)

9.90 (1.89) 11.26 (1.93)* 9.93 (1.49) 11.84 (2.79)* 8.84 (1.64) 10.65 (2.25)* 2.36 (2.33) 2.42 (2.16) 1.53 (1.46) 2.38 (1.63) 2.94 (2.18) 3.56 (6.10)

73

In a previous research with competitive youth field hockey and soccer players (mean age:15.90±1.60), the TACSIS was shown to have good psychometric characteristics with internalconsistency coefficients (Cronbach’s alphas) of all four subscales ranging from 0.72 to 0.89(Elferink-Gemser et al., 2004b). The Intraclass Correlation Coefficients (ICC) for repeatedmeasures was 0.76 for Knowing about others and 0.88 and 0.82 for Positioning and deciding, andActing in changing situations, respectively. The scale Knowing about ball actions had an ICC of0.60 (Elferink-Gemser et al., 2004b). For detailed information about the development of theTACSIS, see Elferink-Gemser et al. (2004b). In the current study, the internal consistency ofthe four TACSIS scales indicates good internal consistency (Nunally, 1978) with Cronbach’salpha of 0.84 for Knowing about ball actions, 0.74 for Knowing about others, 0.88 for Positioningand deciding, and 0.72 for Acting in changing situations.

Data analysisA logistic regression analysis was performed using SPSS (version 16) to identify the

tactical skills (i.e., the four TACSIS subscales) that contribute to the professional performancelevel of soccer in adulthood. The tactical variables were checked on linearity of the logits. Ifthe logits were not linear, the variables were split into categories at a stationary point on thelogit curve. To check the linearity of the logits, the predictor variables were divided intogroups, creating dummy variables. Thereafter, a logistic regression analysis was performedwith the lowest group as a reference point. The midpoints of the groups, on the x-axis, wereplotted against the regression coefficients (the B of the reference group being 0). Where appropriate, the variables were split at a cut-off point where the curve of the logits showed aclear decrease or increase. From this plot, the shape of the curve (e.g., linear, quadratic) couldbe derived. This robust eye-ball method was considered accurate (see also Hosmer &

Chapter 6

Table 6.2. Tactical skills (means, SD) of elite youth soccer players (N=105) specified per performance level in

adulthood and field position

TotalAmateurs Professionals n=53 n=52

Declarative knowledgeKnowing about ball actions (attack) 4.30 (0.68) a 4.42 (0.61) Knowing about others (defense) 4.00 (0.56) 4.08 (0.56) Procedural knowledgePositioning and deciding (attack) 3.76 (0.56) b 4.00 (0.56)* b Acting in changing situations (defense) 4.20 (0.73) 4.16 (0.71)

* Significant difference between amateur and professional performance level (P<0.05) a One missing value. b Two missing values.

74

Lemeshow, 1989; Frankena & Graat, 1997). After the logits were checked, the tactical variables were divided in three categories (low, moderate, high). Spearman’s correlation coefficients between the predictor variables were examined. All variables were included in thefurther analysis as the variables were not highly correlated (rho < 0.60; Hosmer & Lemeshow,1989). The logistic regression analysis was performed using the enter procedure. The accuracyof the model was assessed with the Hosmer and Lemeshow test (P>0.05; Hosmer &Lemeshow, 1989). The percentage of correct classification was calculated for the differentfield positions.

6.3 Results

Table 6.2 presents the means and standard deviations of the four TACSIS subscales forthe amateur and the professional performance levels and field position.

The Hosmer and Lemeshow test was not significant (P=0.57), which indicated that themodel fitted well. The professional performance level was associated with the TACSIS subscale Positioning and deciding (Table 6.3). For players scoring moderate (Fair – Good) theodds ratios indicated a 3.52 times greater chance of becoming a professional than players scoring low (Very Poor – Poor - Fair), whereas for players scoring high (Good – Very Good– Excellent) this chance is even 6.60 times greater.

The correct classification of the adult performance level based on tactical skills was 69.2%for the defenders, 80.0% for the midfielders, while the attackers were classified correctly in75.0%.


Defenders Midfielders AttackersAmateurs Professionals Amateurs Professionals Amateurs Professionalsn=22 n=19 n=15 n=16 n=16 n=17

4.39 (0.61) a 4.25 (0.57) 4.17 (0.81) 4.53 (0.60) 4.31 (0.65) 4.50 (0.65) 4.05 (0.61) 4.09 (0.55) 4.05 (0.55) 4.25 (0.62) 3.90 (0.53) 3.89 (0.49)

3.80 (0.52) b 3.82 (0.49) 3.76 (0.66) 4.40 (0.59)* a 3.70 (0.54) 3.83 (0.44) a

4.49 (0.64) 4.58 (0.67) 4.00 (0.90) 4.05 (0.58) 4.00 (0.56) 3.81 (0.65)

75

Table 6.3. Logistic regression analysis of tactical skills with odds ratio’s for professional performance level in

adulthood of elite youth soccer players (N=105)

Range of scores OR P-valueKnowing about ball actionsLow 1.00-3.50 Almost Never-Mostly Not-Sometimes 1.00 -Moderate 3.51-4.00 Sometimes-Mostly 2.12 0.35High 4.01-6.00 Mostly-Almost Always-Always 1.33 0.69Knowing about othersLow 1.00-3.50 Almost Never-Mostly Not-Sometimes 1.00 -Moderate 3.51-4.00 Sometimes-Mostly 2.58 0.17High 4.01-6.00 Mostly-Almost Always-Always 0.88 0.87Positioning and decidingLow 1.00-3.50 Very Poor-Poor-Fair 1.00 -Moderate 3.51-4.00 Fair-Good 3.52 0.04High 4.01-6.00 Good-Very Good-Excellent 6.60 0.01Acting in changing situationsLow 1.00-3.50 Very Poor-Poor-Fair 1.00 -Moderate 3.51-4.00 Fair-Good 2.72 0.18High 4.01-6.00 Good-Very Good-Excellent 0.81 0.70

Figure 6.1 presents the percentage of amateurs and professionals in either the low, moderate, or high category on the TACSIS subscale Positioning and deciding for the differentfield positions. Overall, the percentage of professionals was higher than the percentage of amateurs scoring in the highest categories (28% of the amateurs versus 40% of the professionals). Defenders in both adult performance levels had a normal distribution aroundthe moderate category ‘Fair’ to ‘Good’ (50% of the amateurs and 47% of the professionals).For midfielders, the distribution of amateurs was normal around the moderate category (40%of the amateurs scored ‘Fair’ to ‘Good’). In contrast, data from the professional midfieldersrevealed a skewed distribution, with 67% of the professionals scoring ‘Good’ to ‘Excellent’.For the attackers, the distribution of amateurs was skewed to the lowest category, with 38%of the amateurs scoring ‘Very Poor’ to ‘Fair’. In contrast, scores from the professional attackersrevealed normal distribution around the moderate category with 56% of the professionalsscoring ‘Fair’ to ‘Good’ and 30% scoring ‘Good’ to ‘Excellent’.

Chapter 6

76

Figure 6.1. Percentage Amateurs and Professionals in different field positions for the Tactical Skills Inventory

for Sports (TACSIS) subscale Positioning and deciding (N=105 elite youth soccer players).

6.4 Discussion

The current study assessed the tactical skills of elite youth soccer players at an early stageof their sport career and made a comparison between the players who reached the professionalperformance level in adulthood and those who became amateurs. In contrast to most earlierstudies on talent identification and development (i.e., those that compare successful and lesssuccessful talented players at a specific moment during adolescence), this study followed theseplayers into adulthood (see also Figueiredo et al., 2009; Huijgen et al., 2009). The data fromthis study suggest that, at the age of eighteen, elite youth soccer players who scored in thehighest category (‘Good’ to ‘Excellent’) of the TACSIS procedural knowledge subscale Positioning and deciding, were almost seven times more likely to reach professional soccer thanthose scoring in the lowest category. Thus, even though all players had accumulated morethan 10 years of organized soccer experience and all had proven to be highly successful up tothis point in their sports career (top 0.5% of their age group), half of them did not reach professional status in the subsequent years. The results show that to reach professional performance level in adulthood, it is important to score at least ‘Fair’ to ‘Good’ on Positioningand deciding.

Tactical skills not only involves the ability to determine what action is most appropriatein a given situation, but also whether this action can be successfully executed. This means thatthe ability to make a certain tactical decision may not always equate with the ability to carryout or execute the decision (Janelle & Hillman, 2003). In line with this, studies show that automaticity and anticipation are very important for games like soccer (Williams, 2000; Huijgen et al., 2009). The current study shows that declarative knowledge did not differ between later amateurs and professionals (i.e., they all have sufficient knowledge about therules of the game, their teammates and opponents and ball actions). However, the proceduralknowledge involved in the interpretation of a specific situation and the ability to be at theright place at the right moment to make the right action (French & Thomas, 1987;


100% 100% 100%90% 90% 90%80% 80% 80%70% 70% 70%60% 60% 60%50% 50% 50%40% 40% 40%30% 30% 30%

20%20% 20%10% 10% 10%

0% 0% 0%Very Poor - Poor - Fair Fair - Good Very Poor - Poor - FairGood - Very Good - Excellent Fair - Good Fair - GoodGood - Very Good - Excellent Good - Very Good - Excellent

Position and decidingDefenders

Position and decidingMidÞelders

Position and decidingAttackers

Amateurs Profesionals

Very Poor - Poor - Fair

77

McPherson & Thomas, 1989; McPherson, 1994; Thomas & Thomas, 1994) appears to bethe factor that best differentiates between the more and less successful players in the future.

The percentages of correct classification (ranging from 69% to 80%) are considered extremely valuable and relevant because the prediction of long-term success in adolescentsoccer players is both particularly complex and the result of several factors from which tacticalskills are just one (Reilly et al., 2000; Williams & Reilly, 2000; Elferink-Gemser et al., 2004a).It is recommended for future research to combine these skills while relating them to futureperformance level to predict future performance level even better. Nevertheless, this studyshows that for players who are about to make the transition to the adult competition, tacticalskills can accurately predict, for roughly three out of four players, who will reach the professional level of performance. It remains unclear as to whether the same results hold foryounger players who are in the middle of their talent development program or at the start ofit.

The results of the current study confirm that tactical expertise is a prerequisite for expertperformance in sports (Janelle & Hillman, 2003). At 80%, the midfielders had an extremelyhigh correct classification rate of the adult performance level based on tactical skills and itseems impossible for a midfielder to reach professional performance level if his score is lowon Positioning and deciding. This scale contains questions related to procedural knowledge inattacking situations. The content of the questions is related to how to get open during a match,decision making about proceeding actions, overview, anticipation, and choosing positions.Midfielders are the link between the defenders and attackers, tasked with setting up attackingsituations and creating the best circumstances for the attackers. In addition, they assist defenders in their defensive task (Rienzi et al., 2000). This is in line with previous researchthat stated that the presenting part of tactical skills for midfielders is Positioning and deciding(Kannekens et al., 2009a).

In the current study, the TACSIS was used to measure tactical skills. This questionnaireis developed specifically for invasive game players, constructed with the help of expert trainersand is embedded in theory (Elferink-Gemser et al., 2004b). Nevertheless, self-report measuresare susceptible to a person’s self-confidence and, this might have affected the results (e.g.,Mahoney et al., 1987; Woodman & Hardy, 2003). However, all participants in this study hadalready accumulated more than 10 years of experience and they were all part of a talent development program of a professional soccer club of national prestige, which means thatthey have been confronted frequently with selection moments. Therefore, regardless of theirconfidence, elite youth soccer player are regarded to have a realistic perspective on their tactical skills.

One has to keep in mind that the participants of this study belonged to the top 0.5% ofall soccer players in their age group, and among these players, about half ultimately reach thetop. Hence, there is just a scarcity of players suitable for this research. Nevertheless, the resultswould have been even stronger if the logistic regression would have been applied not only

Chapter 6

78

with the total group but also within each position. Unfortunately, the number of players ineach position in the current study was too small to perform this analysis for each position separately. In addition, the literature is still lacking in data reporting goalkeepers and center-forward players. However, it is recommended to include them in future research because of their special positions in soccer. In the current study, the distinction between playersin terms of defenders, midfielders and attackers was kept quite robust for the purpose of thestudy. However, nowadays, soccer has become more and more of a two-sided game, forcingthe players’ attacking and defensive positions. For example, in the 3-5-2 system of play,wingers of full-backs could be attacker or defender. Basically, system of play (4-4-2; 4-3-3etc.) might affect the positioning and stacking, being an attacker or a defender. As a consequence, it is recommended in future research to adopt clear definitions of the playerspositions.

The implications of the current study for talent development is that during the adolescentyears, trainers and coaches have to pay attention to Positioning and deciding. If a soccer playerdoes not have the ability to develop this particular part of tactical skills, it is almost impossibleto become a successful player in the future. Therefore, it is recommended to consider Positioning and deciding as a major issue at selection moments, especially if a player scores inthe lowest category.

6.5 PerspectivesTo discriminate between players who are more successful versus their less successful

counterparts, it is necessary to follow elite youth players into adulthood. Positioning and decidingis the tactical skill that best predicts the performance level in adulthood, with a correct classification of over 70% in players who are about to make the transition from youth competitions to the adult competition. Players with a high score have an almost seven timesgreater chance to reach professional soccer than those scoring low. Especially for midfielders,it seems impossible to become a professional player if the score is low during adolescence.These findings highlight the importance of tactical skills and, especially, Positioning and decidingin talent development. The results may point toward a change in the content of talent development programs of elite youth soccer players that encourages trainers to pay major attention to tactical skills and especially to Positioning and deciding.


79

References

• Anderson, J.R. (1982). Acquisition of Cognitive Skill. Psychological Review, 89, 369-406.• Elferink-Gemser, M.T, Visscher, C, Lemmink, K.A.P.M., & Mulder, T. (2007).

Multidimensional performance characteristics and standard of performance in talentedyouth field hockey players: A longitudinal study. Journal of Sports Sciences, 25, 481-489.

• Elferink-Gemser, M.T., Visscher, C., Lemmink, K.A.P.M., & Mulder, T.W. (2004a). Relation between multidimensional performance characteristics and level of performancein talented youth field hockey players. Journal of Sports Sciences, 22, 1053-1063.

• Elferink-Gemser, M.T., Visscher, C., Richart, H., & Lemmink, K.A.P.M. (2004b). Development of the tactical skills inventory for sports. Perceptual and Motor Skills, 99, 883-895.

• Ericsson, K.A. (1990). Peak performance and age: An examination of peak performance insports. In Baltes, P.B, Baltes, M.M., editors. Successful aging: Perspectives from the behavioralsciences. Cambridge, England: Cambridge University Press, pp. 164-195.

• Ericsson, K.A., Krampe, R.Th., & Tesch-Römer, C. (1993). The role of deliberate practicein the acquisition of expert performance. Psychological Review, 100, 363-406.

• Figueiredo, A.J, Goncalves, C.E., Silva, M.J.C.E., & Malina, R.M. (2009). Characteristicsof youth soccer players who drop out, persist or move up. Journal of Sports Sciences, 27, 883-891.

• Frankena, K., & Graat, E.A.M. (1997). Multivariate analysis: Logistic regression. In Noordhuizen JPTM, Frankena K, van der Hoofd CM, Graat EAM, editors. Application ofquantitative methods in veterinary epidemiology. Wageningen: Wageningen Pers, pp 135-169

• French, K.E., & Thomas, J.R. (1987). The relation of knowledge development to children'sbasketball performance. Journal of Sport Psychology, 9, 15-32.

• Gil, S.M., Gil, J., Ruiz, F., Irazusta, A., & Irazusta, J. (2007). Physiological and anthropometric characteristics of young soccer players according to their playing position:Relevance for the selection process. Journal of Strength and Conditioning Research, 21, 438-445.

• Grehaigne, J.F., & Godbout, P. (1995). Tactical knowledge in team sports from a constructivist and cognitivist perspective. Quest, 47, 490-505.

• Grehaigne, J.F., Godbout, P., & Bouthier, D. (1999). The foundations of tactics and strategyin team sports. Journal of Teaching in Physical Education, 18, 159-74.

• Helsen, W.F., & Starkes, J.L. (1999). A multidimensional approach to skilled perceptionand performance in sport. Applied Cognitive Psychology, 13, 1-27.

• Hosmer, D.W., & Lemeshow, S. (1989). Applied logistic regression. New York: Wiley andSons.

• Huijgen, B.C., Elferink-Gemser, M.T., Post, W.J., & Visscher, C. (2009). Soccer skill development in professionals. International Journal of Sports Medicine, 30, 585-591.

Chapter 6

80

• Janelle, C.M., & Hillman, C.H. (2003). Expert performance in sport: Current perspectivesand critical issues. In: Starkes JL, Ericsson KA, editors. Expert performance in sports: Advancesin research on sport expertise. Champaign (IL): Human Kinetics, pp. 49-83.

• Kannekens, R., Elferink-Gemser, M.T., Post, W.J., & Visscher, C. (2009a). Self-assessedtactical skills in elite youth soccer players: A longitudinal study. Perceptual and Motor Skills,109, 459-472.

• Kannekens, R., Elferink-Gemser, M.T., & Visscher, C. (2009b). Tactical skills of world-class youth soccer teams. Journal of Sports Sciences, 27, 807-12.

• KNVB. KNVB Jaarverslag 2007/'08. Retrieved November 20, 2008, fromhttp://sportlink.com/manager/dms/index.php?option=com_content&view=article&id=21

• Mahoney, M.J., Gabriel, T.J., & Perkins, T.S. (1987). Psychological skills and exceptionalathletic performance. The Sport Psychologist, 1, 181-199.

• McPherson, S.L. (1994). The development of sport expertise - Mapping the tactical domain. Quest, 46, 223-240.

• McPherson, S.L., & Kernodle, M.W. (2003). Tactics, the neglected attribute of expertise.Problem representations and performance skills in tennis. In: Starkes JL, Ericsson KA, editors. Expert performance in ports: Advances in research on sport expertise.Champaign (IL):Human Kinetics, pp. 137-167.

• McPherson, S.L., & Thomas, J.R. (1989). Relation of knowledge and performance in boys'tennis - Age and expertise. Journal of Experimental Child Psychology, 48, 190-211.

• Mohamed, H., Vaeyens, R., Matthys, S., Multael, M., Lefevre, J., Lenoir, M., & Philippaerts, R. (2009). Anthropometric and performance measures for the developmentof a talent detection and identification model in youth handball. Journal of Sports Sciences,27, 257-266.

• Nunally, J.C. (1978). Psychometric theory. New York: McGraw-Hill.• Oslin, J.L., Mitchell, S.A., & Griffin, L.L. (1998). The game performance assessment

instrument (GPAI): Development and preliminary validation. Journal of Teaching in PhysicalEducation, 17, 231-243.

• Reilly, T., Williams, A.M., Nevill, A., & Franks, A. (2000). A multidisciplinary approachto talent identification in soccer. Journal of Sports Science, 18, 695-702.

• Rienzi, E., Drust, B., Reilly, T., Carter, J.E.L., & Martin, A. (2000). Investigation of anthropometric and work-rate profiles of elite South American international soccer players.Journal of Sports Medicine and Physical Fitness, 40, 162-169.

• Taylor, J.B., Stephen, D.M., & Nic, J. (2004). Behavioural comparisons of positional demands in professional soccer. 81-97.

• Thomas, K.T., & Thomas, J.R. (1994). Developing expertise in sport - The relation ofknowledge and performance. International Journal of Sport Psychology, 25, 295-312.


81

• Turner, A.P., & Martinek, T.J. (1999). An investigation into teaching games for understanding: Effects on skill, knowledge, and game play. Research Quarterly for Exerciseand Sport, 70, 286-296.

• Vaeyens, R., Lenoir, M., Williams, A.M., Mazyn, L., & Philippaerts, R.M. (2007). The effects of task constraints on visual search behaviour and decision-making skill in youthsoccer players. Journal of Sport & Exercise Psychology, 29, 147-169.

• Williams, A.M. (2000). Perceptual skill in soccer: Implications for talent identification anddevelopment. Journal of Sports Sciences, 18, 737-750.

• Williams, A.M., & Reilly, T. (2000). Talent identification and development in soccer.Journal of Sports Sciences, 18, 657-667.

• Williams, M., & Davids, K. (1995). Declarative knowledge in sport - A by-product of experience or a characteristic of expertise. Journal of Sport & Exercise Psychology, 17, 259-275.

• Woodman, T., & Hardy, L. (2003). The relative impact of cognitive anxiety and self-confidence upon sport performance: a meta-analysis. Journal of Sports Sciences, 21, 443-457.

Chapter 6

82


83

Chapter 7

General Discussion and Conclusion

7.1 General Discussion

How the performance of a talented youth team player progresses over time depends onthe interplay of three key factors: task-specific characteristics such as field position, the player’spersonal characteristics, among which his tactical skills, and the environment, most notablythe quality of the talent development programme. In invasive sports the task of a team playeris largely determined by the player’s position in the field with defenders, midfielders and attackers having to act differently depending on whether the team is attacking or being attacked. Different situations require different tactical expertise and skills. The collective studies presented in this thesis tried to gain more insight into the relationship between task-specific practice and experience and the players’ tactical skill levels, in terms of declarativeand procedural knowledge measured with the Tactical Skill Inventory for Sports (TACSIS),as well the relationship between performance or competitive level and the nature and extentof the tactical skills.

7.2 Task-specific practice and experience and tactical skilldevelopment

Although in virtually all sports and particularly in invasive games task specificity and experience are important prerequisites of tactical skill proficiency as illustrated in several ofthe reported studies, the study in Chapter 2 demonstrates that differences in skill proficiencyamong elite youth soccer players are also associated with the players’ level of formal education.In this study it is found that a higher educational level is associated with a higher score on thetotal construct of tactical skills. Players attending pre-university schools may benefit from thecognitive skills they acquire in their academic environment. In contrast, the players attendingpre-vocational schools have an advantage over their more academically trained peers on theprocedural knowledge scale of the TACSIS Acting in changing situations.

The study reported in Chapter 5 looked at the tactical skills in relation to different fieldpositions in elite youth soccer players. The results showed that different skills are more criticalthan others depending on the player’s field position. For defenders the skills represented inthe TACSIS scale Acting in changing situations, a declarative knowledge scale related to defensive situations during match play, is crucial. The insight in this scale is reflected by statements such as “My interception of the opponents is” and “If our team loses the ball duringa match, I quickly switch to my task as defender”. The qualifying factor for attackers is theTACSIS scale Knowing about ball actions, declarative knowledge that is related to the attack.Here a player’s insight was reflected by statements such as “If we receive the ball (getting ballpossession), I know exactly what to do” and “I know exactly when to pass the ball to a teammate or when not to”. For midfielders, the representing part is the TACSIS scale

Chapter 7

86

Positioning and deciding, a procedural knowledge scale related to the attack, which knowledgewas expressed by statements such as “My overview (in ball possession or in team’s ball possession) is” and “My getting open and choosing positions is”. The content of these scalesis closely related with the task of the matching field positions. The results regarding the midfielders was subsequently confirmed by the study reported in Chapter 6, where the datashowed that TACSIS scale Positioning and deciding is the best predictor of performance levelin adulthood. The correct classifications were especially high for midfielders. Since differentpositions rely on different tactical skills and in addition show different developmental patternsas demonstrated in Chapter 5, task-specific experience appears essential for the successfuldevelopment of tactical skills (see Figure 7.1). The findings are thus in line with earlier research concluding that a player’s understanding of a game has to be learned through practicewithin the context of that game (Turner & Martinek, 1999) and that experience through task-oriented practice in task-oriented contexts fosters a player’s tactical skills optimally(Davids & Myers, 1990; French & Thomas, 1987).

Figure 7.1. The relationship between task-specific experience, performance level and tactical skills. Font sizes

reflect proficiency levels.

7.3 Performance level and tactical skills

The accumulation of a player’s skills over time during training and competition is reflected in his performance level. Chapters 3, 4, and 6 assessed the progression of the tacticalskills of youth players competing at three performance levels, i.e. regional, sub-elite and elite.The performance level as well as the relative stage of the talent development programme theywere part of shift up through the chapters. The study reported in Chapter 3 compared thetactical skills of field hockey players playing at the various performance levels (i.e., regional,sub-elite and elite), while the study described in Chapter 4 compared these skills in the playersof a Dutch and an Indonesian national youth soccer team. Although the Indonesian nationalyouth team competed at the highest level in their age category in their country, the

General discussion and conclusion

Low Task-specific experience High

Performance level: Regional Sub-elite Elite

Tactical Skills

DeclarativeAttack/Defence

ProceduralAttack/Defence





87

performance levels are relative. In soccer Indonesia has a considerably lower world-rankingthan the Netherlands and players from their elite team would possibly be qualified as sub-elite or even lower in the Netherlands. In Chapter 6 the tactical skills of youth soccer playerscompeting at the elite youth level were assessed at the end of their talent development programme and the outcomes were related to their adult performance levels.

The Chapter 3 study showed that the youth players at the elite and sub-elite levels scoredhigher than the regional youth players on all four scales of the TACSIS, whereas the eliteplayers only achieved higher scores for the TACSIS scale Positioning and deciding relative tothe sub-elite players. The results reported in Chapter 4 showed that relative to the Indonesianteam, the Dutch, higher-ranked team had achieved higher scores on three TACSIS scales:Knowing about ball actions, Knowing about others and Positioning and deciding. Within the Indonesian team the results revealed a positive association between players with more playingminutes and players with less playing minutes in an international tournament on the TACSISscale Positioning and deciding. The evaluations in the study reported in Chapter 6 demonstratedthat the TACSIS scale Positioning and deciding best predicts the performance level a talentedyouth soccer player can achieve in adulthood: the players with the higher scores had significantly more chance to become a professional player.

The combined studies of this thesis underscore that there is a strong relationship betweena team player’s performance level and his tactical skills (see Figure 7.1): scores on the TACSISgauging the nature and extent of a player’s declarative and procedural knowledge, increasedwith increasing performance levels. Talented players demonstrated higher skill levels than regional players in every scale domain. However, the TACSIS did not detect differences onall aspects of tactical skills between players training and competing at the highest performancelevels. When we looked at talented youth players at the sub-elite and elite levels, the inventoryonly revealed differences in the players’ extent of procedural knowledge. Based on these outcomes it is concluded that at the highest performance level the skills captured by the TACSIS Positioning and deciding scale are the most determinative. At this level the tested players knew equally well what to do, but the sub-elite players and players who are not capableof making the final step from a youth talent to the professional performance level in adulthoodwere less skilled in acting upon it; they were less capable than the elite players at performingthe right action at the right time. At the higher performance levels, players usually have lesstime and space to perform their actions and it is therefore the best, the quickest and mostpractically skilled players in a team that will make it to the highest performance level. This isa crucial difference between players who make it to the top versus the rest.

Chapter 7

88

7.4 Limitations of the current investigations and directions for future research

To gain more insight into the relationship between the performance level and tacticalskills of talented youth athletes this thesis compared players performing at different levels andstages in their careers with each other by means of the TACSIS, a self-report instrument inwhich the players have to compare themselves with the top players in their age-category. Thecombined studies provided important information about the difference in and the progressionof these skills within the context of talent identification and talent development. However,the studies preclude assumptions about causality and the direction of the associations, i.e., itcannot distinguish whether a higher performance level fosters tactical skills or whether thehigh performance level is the result of a player’s superior skill proficiency.

The TACSIS was developed to make comparisons between (sub)elite youth players possible and to provide the players themselves with an external reference they are familiarwith. Since in the talent programmes of prestigious national clubs players all need to performat the highest level, they are expected to have a realistic view of what it takes to be the best intheir age category. What the TACSIS does not do is generate an absolute score for a player’stactical skills and this needs to be taken into account when interpreting the outcomes of thestudy about the development of tactical skills reported in Chapter 5. Although the study outcomes obtained with the TACSIS were relatively stable, the scores were measured compared to the top players in the same age category. Nevertheless, is plausible to assumethat the players’ actual progression in terms of their tactical skills indeed increased with age,which in the current study was supported by the judgments obtained from the trainers andcoaches involved in the players’ talent development programmes.

Self-reported measures are susceptible to the respondent’s self-confidence and, sinceconfidence in one’s own abilities is associated with elite performance in almost every sport,this personal factor, which is not gauged in the current thesis, might also have affected the results (Mahoney et al., 1987; Woodman & Hardy, 2003). However, all participants startedplaying competitive soccer or field hockey at a young age, which means they already had several years of experience in having to compare themselves with or being compared to theirpeers. In talent development programmes the progression of the players’ skills are evaluatedat different times throughout the season. Trainers, coaches, peers, and parents remind themof their strong and weak points on a regular basis for an extended period, so most players ultimately know how good (or bad) they are. Therefore, regardless of the extent of their self-confidence, elite youth players are thought to have a realistic perspective on their tacticalskills. Even so, this needs further investigation.

To establish whether the TACSIS captures the nature and extent of player’s tactical skillsadequately and whether it reflects actual skills, an accepted and objective reference instrumentis needed. Nevertheless, in the absence of such a validated method, the TACSIS was


89

constructed with the help of expert trainers and embedded in theory. The choice of the scale’sitems was based on logical or face validity and distinguishes between declarative and procedural knowledge in attack and defensive situations, supporting the notion that the inventory actually assesses the nature and extent of an athlete’s tactical skills (Thomas & Nelson, 1996). Still, the TACSIS is no objective measure and its outcomes solely reflect theplayer’s judgment of his own skills. Although talented players are assumed to have a realisticview of their skills and although the correlation between their subjective scores and their trainers’ judgments was high (see chapter 4), the current findings warrant a more objectivecorroboration and further investigation is needed. A possibility is to assess a player’s tacticalskills is in a laboratory setting using slides or video clips showing a variety of game situations(Allard et al., 1993). However, this method also has drawbacks as it lacks “real world” stimulisuch as the pressure of opponents, time constraints and stress to which a player normally needsto respond and act (McPherson & Kernodle, 2003).

The present thesis revealed important information about tactical skills of talented youthplayers in relation with talent identification and talent development. A relevant next step intothe process of talent identification and development is to conduct measurement for tacticalskills in a real-world setting in which it is possible to reliably test whether, besides the abilityto select the appropriate action in a given situation, a player can also execute these actionssuccessfully within the player’s constraints (McPherson, 1994; Starkes, 1993). It is importantto develop a real-world measurement in which different aspects such as technical and tacticalskills are linked to each other. A suggestion is to make video-registration of play-situationsduring training and matches and compare the decisions of a player with their technical skills.With this measurement it may be possible to establish if, when a particular aspect of a player’sskills is not well-developed, this weakness influences his decisions and actions. The methodwill also allow to see whether, when a player has more options, how and to what extent hisdecision and resultant action is affected by his lack of or proficiency in the required technicalskills and when their highly developed technical skills are negatively affected by a lack of tactical skills.

7.5 Conclusion and implications

The main purpose of this thesis was to gain more insight into the talent developmentprocess by examining the importance of tactical skills in the performance and success of talented youth team players over time. Based on the collective studies presented in this thesis,it can be concluded that, given that performance levels were positively related to tactical skilllevels, promoting tactical skills is essential when it comes to enhancing the performance ofelite youth soccer players in the Netherlands. The findings furthermore underscore that different field positions rely on different skills in terms of declarative and procedural

Chapter 7

90

knowledge and that for a successful development of a player’s tactical skills task-specific practice and experience is highly important. With higher performance levels, proceduralknowledge gains in importance. This was reflected by the higher scores on the items of theTACSIS Positioning and deciding scale indicating that the skills captured by this scale are themost determinative for player’s training and competing at the highest performance level during adolescence and the best predictor for success in terms of future professional careersin adulthood (see Figure 7.1).

As to the implications of the findings and pending further investigations and confirmation, for trainers, coaches, scouts, players, parents and other interested persons thefollowing relevant information can be given:

– Tactical skills should be a prominent factor in talent identification and developmentprogrammes;

– To foster tactical skills most effectively task-specific instruction, practice and experience is indispensable;

– Especially for midfielders the procedural knowledge gauged by the TACSIS scale Positioning and deciding is crucial for a successful career in soccer.

– Declarative knowledge of tactical play is relevant at all levels, but at the highest performance level procedural knowledge (Positioning and deciding) is the most distinguishing and crucial factor.


91

References

• Allard, F., Deakin, J., Parker, S. & Rodgers, W. (1993). Declarative knowledge in skilledmotor performance: byproduct or constituent? In Cognitive issues in motor expertise (editedby J. L. Starkes & F. Allard), pp. 95-107. Amsterdam: North Holland Elsevier science publishers.

• Davids, K. & Myers, C. (1990). The Role of Tacit Knowledge in Human Skill Performance.Journal of Human Movement Studies, 19, 273-288.

• French, K. E. & Thomas, J. R. (1987). The Relation of Knowledge Development to Children's Basketball Performance. Journal of Sport Psychology, 9, 15-32.

• Mahoney, M. J., Gabriel, T. J., & Perkins, T. S. (1987). Psychological skills and exceptionalathletic performance. The Sport Psychologist, 1, 181-199.


• McPherson, S. L. & Kernodle, M. W. (2003). Tactics, the neglected attribute of expertise.Problem representations and performance skills in tennis, In Expert performance in sports.Advances in Research on Sport Expertise (edited by J. L. Starkes & K. A. Ericsson), pp 137-167. United Kingdom: Human Kinetics.

• Starkes, J. L. (1993). Motor experts: Opening thoughts. In J.L. Starkes & F. Allard (Eds.),Cognitive issues in motor expertise (pp. 3-16). Amsterdam: Elsevier.

• Thomas, J. R., & Nelson, J. K. (1996). Research methods in physical activity. Champaign, ILHuman Kinetics.

• Turner, A. P. & Martinek, T. J. (1999). An investigation into teaching games for understanding: Effects on skill, knowledge, and game play. Research Quarterly for Exerciseand Sport, 70, 286-296.

• Woodman, T. & Hardy, L. (2003). The relative impact of cognitive anxiety and self-confidence upon sport performance: a meta-analysis. Journal of Sports Sciences, 21, 443-457.

Chapter 7

92


93

Summary

In sports, talent identification and talent development programmes have gained popularity in recent decades. Given the relevance and status of (team) sports in today’s societyand the money involved in elite sports, it has become increasingly important to know what ittakes for aspiring youth athletes to acquire the necessary skills and expertise to make it to thetop in their sport. This thesis focuses exclusively on the aspect of tactical skills within the context of team sports. In the investigations it is, of course, always kept in mind that to performwell in sports the multidimensionality of performance characteristics that cover no less thanfour different domains, i.e., the physiological, technical, tactical, and psychological domains,in combination with a player’s anthropometric characteristics, plays a fundamental role.

The studies invest the importance of tactical skills. These skills consist of two main components: declarative knowledge (‘knowing what to do’) and procedural knowledge (‘doingit’) in either attacking and defensive situations. In this thesis these skills are measured by waysof self-assessment bu the players using Tactical Skill Inventory for Sports (TACSIS). The inventory consist of four scales:

1. Knowing about ball actions;2. Knowing about others;3. Positioning and deciding;4. Acting in changing situations.

The scales 1 and 2 contain declarative knowledge in respectively attacking and defensive situations. The scales 3 and 4 refer to procedural knowledge in respectively attacking and defensive situations. The main purpose of this thesis was to gain more insight into the talent development processby examining the importance of tactical skills in the performance and success of talented youthteam players over time.

In Chapter 2, the constructs of declarative and procedural knowledge are addressed bystudying the tactical skills of 93 elite youth soccer players (mean age 17.3, standard deviation0.6) relative to their academic performance levels. The players were divided in two groupsbased on their type of education they attend: 45 players received pre-vocational educationand 48 players were studying at the pre-university educational level. Multivariate analyses ofcovariance including educational level as factor and field positions and repeating class in schoolas covariates, revealed that differences in skill proficiency among elite youth soccer playersare associated with the players’ level of formal education: a higher educational level is associated with a higher score on the total construct of tactical skills (P<0.05). Players attending pre-university schools may benefit from the cognitive skills they acquire in theiracademic environment. In contrast, the players attending pre-vocational schools have an advantage over their more academically trained peers on the TACSIS scale Acting in changingsituations (d=0.41).

Summary

96

The study in Chapter 3 compares the tactical skills of youth field hockey players (meanage 15.5, standard deviation 1.6) with an ‘average’ (regional), ‘high’ (sub-elite) and ‘very high’(elite) performance levels. Multivariate analyses of covariance including performance leveland gender as factor and age as covariate showed that the youth field hockey players at theelite and sub-elite performance level scored higher than the players at the regional performance level on all four scales of the TACSIS (P<0.05). In contrast, the elite players onlyachieved higher scores in the TACSIS scale Positioning and deciding relative to the sub-eliteplayers (P<0.05). Gender had no influence on the results (P>0.05).

To expand the knowledge of what makes a top player and to ultimately foster talent, thestudy in Chapter 4 compares the tactical skills of 18 players (age range: 18-20 years) of aDutch national youth soccer team and 19 players (age range: 18-23 years) of an Indonesiannational youth soccer team. Multivariate analyses of variance showed a positive relationshipbetween competitive level and tactical skills: the higher ranked Dutch players outscored theirIndonesian counterparts on the TACSIS scales: Knowing about ball actions (P<0.01), Knowingabout others (P<0.01) and Positioning and deciding (P<0.01). The analyses revealed no relationshipbetween tactical skills and playing time measured in an international tournament in the Dutchteam (P>0.05). Within the Indonesian team the results revealed a positive association betweenplayers with more playing minutes and players with less playing minutes measured in an international tournament on the TACSIS scale Positioning and deciding (d=0.99). These resultsshow that tactical skills are fundamental to high-level soccer performance. Ample, expert-ledtraining and match experience at a high competitive level, starting at a young age, and high-quality talent development programmes are suggested as key ingredients for the development of tactical skills.

The study reported in Chapter 5 features a longitudinal study that focuses on the tacticalskills among 191 elite youth soccer players from age 14 through 18 playing in different fieldpositions. A model to assess whether tactical skills change over time in defenders, midfieldersand attackers was developed using multilevel analysis. The models indicated that defendersand midfielders did not improve their tactical skills, whereas attackers increased their, relativeto top players in the same age category, tactical skills from age 14 to 18 years. The representingpart of tactical skills for defenders is the TACSIS scale Acting in changing situations; for midfielders TACSIS scale Positioning and deciding and the TACSIS scale Knowing about ball actions was the qualifying factor for attackers. Possible explanations for these differences intactical skills among elite youth soccer players are the selection procedures at a younger ageand task-specific experiences.

The studies in Chapter 3 and 4 showed that self-assessed tactical skills differ betweenperformance levels. To help establish which factors predict which talented players are mostlikely to make it to the top, it is important to follow them into adulthood. The study in Chapter 6 accordingly looked at the necessary tactical skills in adolescence to reach profsionalperformance level in adulthood. This study assessed the tactical skills of 105 elite youth soccer

Summary

97

players (mean age 17.8, standard deviation 0.8), playing at a different positions in the field(e.g., attackers, midfielders and defenders), at an earlier stage of their sport career. These skillswere related to their adult performance level, specifically whether they became amateur (n=52)or professional (n=53). A logistic regression analysis demonstrated that the TACSIS scale Positioning and deciding best predicts adult performance level (P<0.05). The percentage of correct classification was high for all positions, however this percentage was extremely highfor midfielders (80%). For players scoring high on the TACSIS scale Positioning and decidingthe odds ratios indicated a 6.60 times greater chance that a player became a professional thanplayers scoring low on this scale (P<0.05).

The present thesis revealed important information about tactical skills of talented youthplayers in relation with talent identification and talent development. A relevant next step intothe process of talent identification and development is to construct a measure for tactical skillsin a real-world setting that is not dependent on self-assessment. Besides the ability to selectthe appropriate action in a given situation, this measure needs to have the capability to measureif the player can also execute these actions successfully within the player’s constraints.

Based on the collective studies presented in this thesis, in Chapter 7 it is concluded that,given that performance levels were positively related to tactical skill levels, promoting tacticalskills is essential when it comes to enhancing the performance of elite youth soccer players.The findings furthermore underscore that different field positions rely on different skills interms of declarative and procedural knowledge and that for a successful development of aplayer’s tactical skills task-specific practice and experience is highly important. With higherperformance levels, procedural knowledge gains in importance. This was reflected by thehigher scores on the items of the TACSIS Positioning and deciding scale indicating that theskills captured by this scale are the most determinative for player’s training and competing atthe highest performance level during adolescence and the best predictor for success in termsof future professional careers in adulthood. The conclusions of this thesis give trainers andscouts additional criteria, which they can use to identify and develop talent.

Summary

98

Summary

99

Samenvatting

De enorm gegroeide status van sport in onze huidige maatschappij, de financiële belangen die daarmee gepaard gaan en de daaraan gekoppelde druk om te presteren, drageneraan bij dat de aandacht voor talentherkenning en talentontwikkeling steeds belangrijkerwordt in de wereld van de sport. NOC*NSF, sportbonden, Betaald Voetbal Organisaties,trainers en coaches investeren veel energie, tijd en geld in talentvol geachte jeugdige sporters.Maar wat is het rendement van deze investeringen? Hoeveel van deze geselecteerde sportershalen daadwerkelijk de top? Welke vaardigheden zijn bepalend om de top te bereiken?Voor een antwoord op deze vragen staat in dit proefschrift het belang van tactische vaardigheden bij talentontwikkeling centraal. Naast deze focus moet altijd rekening wordengehouden met andere prestatiebepalende kwaliteiten die een belangrijke rol spelen bij het leveren van een goede sportprestatie. Deze kwaliteiten vallen binnen de domeinen fysiologie,techniek en psychologie in combinatie met lichaamsbouw.

In dit onderzoek worden twee componenten van tactische vaardigheden onderscheiden:declaratieve kennis (‘weten wat te doen’) en procedurele kennis (‘het doen’), zowel in aanvallende als verdedigende situaties. Deze vaardigheden zijn gemeten met de Tactical SkillsInventory for Sports (TACSIS). De vragenlijst bevat vier schalen:

1. Inzicht in acties met de bal;2. Inzicht in anderen; 3. Positie kiezen en besluitvorming;4. Omgaan met veranderingen.

De schalen 1 en 2 hebben betrekking op declaratieve kennis in respectievelijk aanvallende enverdedigende situaties. De schalen 3 en 4 duiden op procedurele kennis, in respectievelijkaanvallende en verdedigende situaties.Het doel van dit proefschrift is meer inzicht te krijgen in de mate waarin tactische vaardighe-den van getalenteerde teamspelers een rol spelen in het proces van talentontwikkeling.

Voor het verkrijgen van inzicht in de relatie tussen meer algemene cognitieve kwaliteiten,weergegeven door het schoolniveau, en sportgerelateerde declaratieve en procedurele kennisworden scores op de TACSIS van 93 getalenteerde jeugdvoetballers (gemiddelde leeftijd 17.3,standaarddeviatie 0.6) in Hoofdstuk 2 gerelateerd aan schoolniveaus. De spelers zijn op basisvan hun schoolniveau ingedeeld in twee groepen: 45 spelers vinden we terug op het VMBOniveau en 48 spelers zitten op het HAVO/VWO niveau. Een multivariate analyse met schoolniveau als factor, blijven zitten en veldpositie als covariaat, laat zien dat verschillen invaardigheden van getalenteerde jeugdvoetballers worden gerelateerd aan het schoolniveauvan de spelers: een hoger schoolniveau (HAVO/VWO) hangt samen met hogere score op hettotale construct tactische vaardigheden (P<0.05). Dit zou samen kunnen hangen met de cognitieve vaardigheden die zij op school leren. Daar staat tegenover dat jeugdvoetballers opVMBO niveau beter scoren op de TACSIS schaal Omgaan met veranderingen (d=0.41). Wellichtis dit voordeel ook gerelateerd aan de wijze van kennisverwerving op school.

Samenvatting

102

Het onderzoek in Hoofdstuk 3 vergelijkt de tactische vaardigheden van 191 jeugdhockeyers (gemiddelde leeftijd 15.5, standaarddeviatie 1.6) met een ‘gemiddeld’ (breedte), ‘hoog’ (subtop) en ‘erg hoog’ (top) sportniveau. Multivariate analyse met sportniveau en geslacht als factor en leeftijd als covariaat laat zien dat de jeugdhockeyers meteen subtop- en topniveau hoger scoren op alle vier schalen van de TACSIS dan spelers meteen breedtesportniveau (P<0.05). Kijken we verder binnen de groep subtop en top dan zienwe dat de topspelers alleen beter scoren op de schaal Positie kiezen en besluitvorming (P<0.05)dan de subtopspelers. Geslacht heeft geen invloed op deze scores (P>0.05).

Om meer inzicht te krijgen in wat iemand tot een topsporter maakt en hoe het sportniveau samenhangt met tactische vaardigheden, worden in Hoofdstuk 4 de tactischevaardigheden van 18 spelers (leeftijdscategorie: 18-20 jaar) van een nationaal jeugdvoetbalteamuit Nederland vergeleken met 19 spelers (leeftijdscategorie: 18-23 jaar) van een nationaaljeugdvoetbalteam uit Indonesië. Uit de multivariate variantie analyse blijkt dat er een positieverelatie bestaat tussen competitieniveau en tactische vaardigheden: de spelers van het op dewereldranglijst hoger geplaatste Nederlandse team scoren beter dan de Indonesische spelersop de TACSIS schalen Inzicht in acties met de bal (P<0.01), Inzicht in anderen (P<0.01) en Positiekiezen en besluitvorming (P<0.01). De analyse laat binnen het Nederlandse team geen relatiezien tussen tactische vaardigheden en speelminuten gemeten op een internationaal toernooi(P>0.05). Binnen het Indonesische team zien we een positieve associatie (d=0.99) tussen spelersmet een verschillend aantal speelminuten gemeten op een internationaal toernooi op de TACSIS schaal Positie kiezen en besluitvorming. Tactische vaardigheden zijn van groot belangvoor voetbalprestaties op hoog niveau. Training van experts en wedstrijdervaring op hoogcompetitieniveau leveren mogelijk een bijdrage aan de ontwikkeling van tactische vaardigheden.

In Hoofdstuk 5 worden op basis van een longitudinale studie de tactische vaardighedenbeschreven van 191 getalenteerde jeugdvoetballers met verschillende posities in de leeftijdscategorie 14 tot 18 jaar oud. Door middel van multilevel analyse is een model ontwikkeld om de tactische vaardigheden van verdedigers, middenvelders en aanvallers in detijd te bekijken. De analyse laat zien dat, in vergelijking met de top in de eigen leeftijdscategorie, verdedigers en middenvelders gelijk scoren op hun tactische vaardighedenin die periode, in tegenstelling tot aanvallers die hun tactische vaardigheden verbeteren in deleeftijd tussen 14 en 18. Verdedigers scoren het best op Omgaan met veranderingen; de TACSISschaal voor procedurele kennis in verdedigende situaties. Middenvelders scoren het best opPositie kiezen en besluitvorming; de TACSIS schaal voor procedurele kennis in aanvallende situaties. Aanvallers scoren het best op Inzicht in acties met de bal; de TACSIS schaal voor declaratieve kennis in aanvallende situaties. Een mogelijke verklaring voor de verschillen intactische vaardigheden tussen de posities van getalenteerde jeugdvoetballers is de selectieprocedure op jonge leeftijd en taakspecifieke ervaring.

Samenvatting

103

Uit de Hoofdstukken 3 en 4 blijkt dat tactische vaardigheden een belangrijke factorzijn bij het sporten op verschillende niveaus. Om te kunnen voorspellen welke getalenteerdejeugdspelers de meeste kans maken op het bereiken van professioneel voetbal is het van belanghen te volgen tot het volwassen sportniveau. In Hoofdstuk 6 worden de tactische vaardigheden van 105 getalenteerde jeugdvoetballers (gemiddelde leeftijd 17.8, standaarddeviatie 0.8) aan het eind van hun talenttraject gerelateerd aan hun volwassen sportniveau (amateur of professional). Een logistische regressie analyse toont dat de TACSISschaal Positie kiezen en besluitvorming samenhangt met het bereiken van professioneel voetbal(P<0.05). De analyse laat zien dat spelers die hoog scoren op de schaal Positie kiezen enbesluitvorming een 6.60 maal grotere kans maken zich te ontwikkelen tot professionele voetballers dan spelers die laag scoren (P<0.05). Wanneer aan de hand van de tactischevaardigheden voorspeld wordt of spelers profvoetballer worden, blijkt het percentage juistingedeelde spelers voor alle posities hoog. Verdedigers (69%), aanvallers (75%) en vooral middenvelders (80%) kunnen op basis van hun scores op de TACSIS goed worden ingedeeldin hun toekomstige sportniveau: amateurs versus professionals. de groep latere amateurs versuslatere professionals.

Dit proefschrift onthult belangrijke informatie over het belang van tactische vaardigheden van talentvolle jeugdteamspelers in relatie tot talentontwikkeling. Een relevantevolgende stap om nog meer inzicht te verkrijgen in het proces van talentherkenning en talentontwikkeling is het ontwikkelen van een veldtest waarmee tactische vaardigheden in desportspecifieke situatie gemeten kunnen worden. Dit instrument zou naast de mogelijkheidom de juiste actie op het juiste moment te meten, bovendien de mogelijkheid moeten biedenom te bepalen of de speler de actie succesvol uitvoert binnen zijn mogelijkheden.

Gebaseerd op de onderzoeken in dit proefschrift wordt in Hoofdstuk 7 geconcludeerddat sportspecifieke ervaring zeer belangrijk is voor het ontwikkelen van tactische vaardigheden;hoe meer ervaring op een hoger sportniveau, hoe beter de tactische vaardigheden. Bovendienblijkt uit het onderzoek naar de verschillen tussen aanvallers, middenvelders en verdedigersdat taakspecifieke ervaring onmisbaar is voor de ontwikkeling van tactische vaardigheden. Deprocedurele kennis, die betrekking heeft op positie kiezen en besluitvorming, heeft een sterkerelatie met het toekomstige niveau van getalenteerde jeugdvoetballers. Tactische vaardighedenspelen een prominente rol binnen talentontwikkeling. Kwaliteiten zoals positie kiezen enbesluitvorming blijken cruciaal voor een succesvolle carrière in het professionele voetbal. Deresultaten van dit promotieonderzoek leveren in de praktijk aanvullende criteria op die voortrainers en scouts een belangrijke rol kunnen spelen in het herkennen en ontwikkelen van talent.

Samenvatting

104

Samenvatting

105

Dankwoord

Is het een verrassing dat ik voor mijn proefschrift de sport ben ingedoken? Neen, ik komnamelijk uit een sportief nest. Niet dat de Olympische Records bij mij thuis over de drempelrolden, integendeel zelfs, maar de stimulans om te gaan sporten en het beste uit jezelf te halenzijn de rode draad in mijn leven geworden. Tijdens mijn studie is daar nog een aspect bijgekomen: thuis lekker filosoferen over de sportinfrastructuur in ons land en zinvolle discussies over hoe onzorgvuldig met talent in ons land wordt omgegaan. Pa en ma, heel veeldank voor jullie inspiratie, steun en vertrouwen.

Ook op de RUG ben ik in een fijn nest beland en heb ik veel inspirerende mensen mogenontmoeten. Zonder iemand tekort te doen, blijft een aantal namen in mijn geheugen gegrift.

Chris Visscher heeft het mogelijk gemaakt dat ik mijn onderzoek binnen het Interfacultair Centrum voor Bewegingswetenschappen heb kunnen uitvoeren. Daartoe is zijnbijdrage niet beperkt gebleven. Op de momenten dat ik in details dreigde te verzanden, wistChris op geheel eigen wijze de grote lijnen te bewaken. Dank hiervoor en ik zal zijn methodiekniet snel vergeten.

Veel respect en bewondering draag ik over aan Marije Elferink-Gemser. Ik heb haar tijdens mijn afstuderen al mogen leren kennen. Haar vertrouwen in mij heeft mij in die periodeveel zelfvertrouwen gegeven. In de afgelopen vier jaar is het respect en bewondering voorhaar alleen maar groter geworden. Ondanks de bijzondere gebeurtenissen in haar privé-levenis Marije er altijd voor mij geweest en heeft zij mij tot een zelfverzekerd mens gemaakt. Ookhaar inhoudelijke bijdrage wil ik hier graag noemen. Over talentontwikkeling gesproken…

En dan zijn er de collega’s van bewegingswetenschappen. De collega’s bij wie ik mijnideeën kon toetsen of informeren naar hun werkwijze. Dank voor jullie luisterend oor.

Stoom afblazen; momenten van vertwijfeling; lachen en alle andere emotionele momenten deelde ik met de bewoners van de AIO-gang. Fijn dat jullie er waren en in het bijzonder moet ik Linda en Barbara noemen.

Onderzoek doen naar … kon in mijn geval niet zonder een lange rij van talentvol geachtejonge spelers van FC Groningen, SC Heerenveen, Jong Oranje en Jong Indonesië. Dank datik jullie gedrag mocht waarnemen. En dank voor de organisaties van FC Groningen en SC Heerenveen dat ik bij jullie mijn onderzoek mocht uitvoeren.

De vele telefoontjes van vrienden, familie en kennissen verlichtten menigmaal mijnwerkzaamheden voor het beeldscherm van mijn laptop. Wat was het fijn om weer even mijngedachten aan het alledaagse te kunnen wijden.

Nu mijn proefschrift de eindstreep is gepasseerd, staat mijn zus Annelies en mede AIO’erBarbara, ja inderdaad zij van de AIO-gang, weer voor mij klaar om de bijbehorende ceremoniële en feestelijke activiteiten in goede banen te leiden. Hoe dat allemaal is verlopen,is tijdens dit schrijven nog niet bekend. Toch durf ik te voorspellen dat het met hulp van beideneen succes zal zijn.

Dankwoord

108

Promoveren gaat gepaard met veel ups en downs, stress, tijdgebrek, teleurstellingen enemoties. Vraag Maarten maar! En toch vertrok hij geen spier, verloor hij nimmer zijn goedehumeur en wachtte hij geduldig op de momenten waarop ik nog enigszins gezellig kon zijn.Ik heb de komende tijd heel veel goed te maken…

Dankwoord

109

Curriculum Vitae

Rianne Kannekens is op 30 november 1981 geboren te Zwolle. In aansluiting op het behalen van haar VWO diploma aan de Thorbecke Scholengemeenschap te Zwolle in 2000is zij gestart met de opleiding Bewegingswetenschappen aan de Rijksuniversiteit Groningen.Na onder andere een semester aan de Univerity of Calcary (Canada) te hebben gevolgd, heeftzij in 2005 de opleiding afgerond met het doctoraalexamen in de richting ‘Sport’. Haar afstudeeronderzoek naar fysieke ontwikkeling in relatie tot voetbalspecifieke fysiologischekwaliteiten heeft plaatsgevonden binnen het talentonderzoek van het Centrum voor Bewegingswetenschappen in Groningen.

Na een klein jaar als Event Manager bij het TT Circuit in Assen gewerkt te hebben,keerde zij in het najaar van 2006 terug naar het Interfacultair Centrum voor Bewegingswetenschappen in Groningen. Tot 2010 verrichte zij daar als promovendus onderzoek naar de rol van tactische vaardigheden binnen talentontwikkeling.

Per 1 februari 2010 is zij werkzaam als docent bij het Instituut Sport en Bewegingsstudiesvan de Hogeschool Arnhem Nijmegen.

Curriculum Vitae

112

Curriculum Vitae

113

List of Publications

International publications (peer reviewed journals)

• Kannekens, R., Elferink-Gemser, M.T., Visscher, C. (submitted). Self-assessed tactical skillsof elite youth soccer players with different academic achievement. European Journal of SportScience.

• Kannekens, R., Elferink-Gemser, M.T., Visscher, C. (in press). Positioning and deciding:The key to professional soccer. Scandinavian Journal of Medicine & Science in Sports.

• Elferink-Gemser, M.T., Kannekens, R., Lyons, J., Tromp, E.J.Y., Visscher, C. (2010).Knowing what to do and doing it? The difference in self-assessed tactical skills betweenregional, sub-elite, and elite youth field hockey players. Journal of Sport Sciences, 28, 521-528.

• Kannekens, R., Elferink-Gemser, M.T., Visscher, C. (2009). Tactical skills of world-classyouth soccer teams. Journal of Sport Sciences, 27, 807-812.

• Kannekens, R., Elferink-Gemser, M.T., Post, W.J., Visscher, C. (2009). Self-assessed tactical skills in elite youth soccer players: A longitudinal study. Perceptual and Motor Skills,109, 459-472

• Kannekens, R., Elferink-Gemser, M.T. (2005). Talent has to grow: height in relation tosoccer-specific skills. Journal of Coimbra Network on Exercise Sciences, 2, 35-38.

Chapters in books• Elferink-Gemser, M.T., Kannekens, R., Huijgen, B.C.H., Tromp, E.J.Y, Jonker,

L., Toering, T.T., Visscher, C. (in press). Reading and writing the game: Tactical skills inteam sports. In: M. Coelho e Silva, A.J. Figueiredo, M.T. Elferink-Gemser, R.M. Malina(eds). Youth sports participation, trainability and readiness. Coimbra University Press, Portugal.

• Elferink-Gemser, M.T., Huijgen, B.C.H., Kannekens, R. (2010). Holland’s got talent. In:R. Bongers, S. Caljouw, E. Hartman and R. den Otter. Smart Movements. 25 jaar Bewegingswetenschappen Groningen Interfacultair Centrum voor Bewegingswetenschappen(ICBW), Rijksuniversiteit Groningen, Universitair Medisch Centrum Groningen, Nederland, 120-122.

Proceedings• Kannekens, R., Elferink-Gemser, M.T., Visscher C. (2008). Relationship between tactical

skills and performance level of expert youth soccer players. In: Proceedings of the 13thAnnual Congress of the European College of Sport Science, Estoril, Portugal, 34.

List of publications

116

• Kannekens, R., Elferink-Gemser, M.T., Visscher C. (2007). De rol van tactisch inzicht inde prestatie van jeugdige getalenteerde teamsporters. Congres ‘Talent Centraal’ NOC*NSF.Papendal, Nederland.

• Tromp, E.J.Y., Kannekens, R., Elferink-Gemser, M.T., Lyons, J., Visscher C. (2007). Decision-making in sports: a review about the underlying cognitive processes. SCAPPS,Windsor, ON (Canada), 77.

• Kannekens, R., Elferink-Gemser, M.T., Lemmink, K.A.P.M., Visscher C. (2006). Talentdevelopment in team sports. Talent, Training, Emotion and stress. FISA Junior CoachesConference. Groningen, the Netherlands.

• Kannekens, R., Elferink-Gemser, M.T., Lemmink, K.A.P.M., Visscher C (2006). Onderzoek naar talentontwikkeling. Voorlichting ouders en leerlingen van de LOOT-afdelingvan het Röllinkcollege. Groningen, Nederland.

List of publications

117

Documents

University of Groningen The importance of tactical skills ... · Visscher, C. (2010) Journal of Sport Sciences, 28, 521-528 Chapter 4 Tactical skills of world-class youth soccer teams