Main Article

Ger J Exerc Sport Res 2019 · 49:410–423https://doi.org/10.1007/s12662-019-00600-6Received: 5 July 2018Accepted: 12 June 2019Published online: 27 June 2019© The Author(s) 2019

Claudia Zuber · Achim ConzelmannUniversität Bern, Bern, Switzerland

Achievement-MotivatedBehavior in Individual Sports(AMBIS-I)—Coach Rating ScaleDevelopment and preliminary validation

Electronic supplementarymaterial

The online version of this article (https://doi.org/10.1007/s12662-019-00600-6) containssupplementarymaterial, which is available toauthorized users.

Talent research highlights the greatimportance of psychological variablesfor the successful development froma promising young to a successful topathlete (Coetzee, Grobbelaar, & Gird,2006; Johnston, Wattie, Schorer,&Baker,2018; MacNamara, Button, & Collins,2010). Achievement motivation, in par-ticular, seems to play a critical rolefor talent development and subsequentsuccess (e.g. Abbott & Collins, 2004;Coetzee et al., 2006; Zuber, Zibung, &Conzelmann, 2015). Unlike the differenttheories subsumed under the umbrellaterm achievement motivation, achieve-ment motivation itself is hardly definedprecisely (Elliot & Dweck, 2007). Elliotand Dweck (2007) therefore propose toput the construct competence, “definedas a condition or quality of effective-ness, ability, sufficiency, or success”(Elliot & Dweck, 2007, p. 5), at the coreof the achievementmotivation literature.

Nevertheless, theory building andresearch on achievement motivation hashad a huge impact in the last decades.The theories most currently discussed intalent research in sport are the achieve-ment motives hope for success and fearof failure (e.g. Elbe & Beckmann, 2006;Halvari & Thomassen, 1997; Höner &Feichtinger, 2016; Sagar, Busch,& Jowett,

2010; Zuber et al., 2015; Zuber&Conzel-mann, 2014), achievement goal orien-tations (e.g. Figueiredo, Gonçalves,Coelho e Silva, & Malina, 2009; Hel-landsig, 1998; Höner & Feichtinger,2016; Huijgen, Elferink-Gemser, Lem-mink, & Visscher, 2014; Reilly, Williams,Nevill,&Franks, 2000; Zuber et al., 2015)and the Self-Determination Theory (e.g.Gillet, Berjot, Vallerand, Amoura, &Rosnet, 2012; Gillet, Vallerand, & Paty,2013; Gillet, Vallerand, & Rosnet, 2009;Sarrazin, Vallerand, Guillet, Pelletier, &Cury, 2002; Zuber et al., 2015).

The achievement motive determineswhether individuals tend to approachachievement-related situations orwhether they tend to avoid them (Atkin-son, 1957). The positive connection be-tween hope for success and performancein sports has been empirically con-firmed in both cross-sectional (Coetzeeet al., 2006; Halvari &Thomassen, 1997)and longitudinal studies (Elbe & Beck-mann, 2006; Murr, Feichtinger, Larkin,O’Connor, & Höner, 2018; Unierzyski,2003; Zuber & Conzelmann, 2014). Fearof failure, on the other hand, is often as-sociated with a negative correlation withperformance (Halvari & Thomassen,1997; Sagar et al., 2010).

Whereas the achievement motiveinitiates actions aimed at attaining com-petence, achievement goal orientationsguide these actions towards certaingoals. It is differentiated between twoor three different goal orientations,which are either called task and egoorientation (Nicholls, 1984) or masteryand performance orientation (Ames &

Archer, 1988). Task/mastery orientationis aimed at improving one’s own skills,for which purpose an internal standardof comparison is used. Ego/performanceorientation, on the other hand, focuseson displaying one’s own superiorityto other people (Heckhausen & Heck-hausen, 2010). The third goal orientationcompetitiveness is characterized by “thedesire to enter and strive for success insport competition” (Gill & Deeter, 1988,p. 200). Therefore, it goes along witha need to compete and compare oneselfin a sporting competition. The conceptof achievement goals has fertilized a verylarge number of studies in sport sciences(Duda, 2007). Regarding the questiononthe relation between achievement goalorientations and achievement withinyouth elite sports, the major findingsby the most recent review (in soccer)exposed a mixed state of evidence forego/performance orientation with futureperformance. The majority of studiesconsidering task/mastery orientationfound a positive relation with futuresuccess (Murr et al., 2018).

In Self-Determination Theory, thereasons for motivated actions are dis-tinguished according to where theirperceived locus of causality is, or to whatextent they are self-determined (Deci &Ryan, 1985). The resulting motivationaltype lies on a continuum extendingfrom amotivation, a state with a com-plete absence of any motivation, throughextrinsic motivation, and ending withintrinsic motivation as the most self-determined form of motivation (Ryan &Deci, 2000). Intrinsic motivation is char-

410 German Journal of Exercise and Sport Research 4 · 2019

acterised by pleasure in performing theactivity itself. Extrinsic motivation, onthe other hand, pertains to actions whichare carried out because of the expectedconsequences, such as fame, honour orprize money. Four types of extrinsicmotivation are postulated, which arecharacterised by increasingly high levelsof self-determination or autonomy (foran overview, see Ryan & Deci, 2007).A high degree of self-determination hasbeen shown to be associated with higherlevels of performance in adolescents(Gillet et al., 2009; Gillet et al., 2012;Zuber et al., 2015). Conversely, lowlevels of self-determination appears tohamper a successful sports career in thesense of dropping out (Calvo, Cervelló,Jiménez, Iglesias, & Murcia, 2010; Pel-letier, Fortier, Vallerand, & Brière, 2001;Sarrazin et al., 2002).

As a result of (a) the empirical re-search pointing to the positive associ-ation between achievement motivationand performance, and (b), the experi-ences from sporting practice supportingthat motivational characteristics arehighly valued by coaches (Christensen,2009; Jokuschies, Gut, & Conzelmann,2017), as well as by elite athletes andtheir parents (MacNamara et al., 2010),the assessment of achievement motiva-tion was integrated in multidimensionaltalent identification programs (TID; cf.Vaeyens, Lenoir, Williams, & Philip-paerts, 2008; Germany: Feichtinger &Höner, 2014; Switzerland: Fuchslocher,Romann, Rüdisüli, Birrer, &Hollenstein,2011). For the field of motor diagnostics,tests are generally regarded to providea high degree of psychometric qualityand usability (Zibung, Zuber, & Conzel-mann, 2016), whereas the assessment ofpsychological variables, like achievementmotivation for real selection decisions incontrast to scientific research, is associ-ated with some serious problems, whichwill be described below.

To date, different methods to assessachievement motivation are in use. Onecan distinguish between (1) self-reportquestionnaires measuring the explicitachievement motive, (2) external as-sessment questionnaires (coach-ratings)measuring the explicit achievement mo-

tive, and (3)projective testsof the implicitachievement motive.

Self-report questionnaires—as thefirst method discussed here—becameincreasinglymore available andwere alsousedinsportscience(Feichtinger&Höner,2014) due to the high reliability andeconomical nature of self-reports. Ques-tionnaires that have often been used inthe context of achievementmotivation incompetitive sports are the AchievementMotivation Scale in Sport (Elbe, Wen-hold, & Müller, 2005b) to assess hopefor success and fear of failure, the SportOrientation Questionnaire (Elbe, 2004;Gill & Deeter, 1988) that focuses on theachievement goal orientations, and theSport Motivation Scale (Burtscher, Furt-ner, Sachse, & Burtscher, 2011; Pelletieret al., 1995; Pelletier, Rocchi, Vallerand,Deci, & Ryan, 2013) to measure self-determination. In addition to the advan-tages that have led to the frequent use ofself-reportquestionnaires, disadvantagesare also known: “Although this approachhas clear advantages, such as its high psy-chometric quality and ease of analysis,it also has its disadvantages. Responsesmay be biased by the tendency to presentoneself in a socially desirable light . . . ”(Brunstein & Heckhausen, 2010, p. 138).The more obvious the purpose of thequestionnaire, the more likely it is thatthe subject may intentionally or unin-tentionally influence the outcome, bothin a positive and in a negative direction(response bias). Particularly in person-ally significant situations, the risk ofsocially desirable responding appears toincrease: “Socially desirable respondingmight be more likely to occur in con-texts in which important consequenceshinge on the testing outcomes” (Furr &Bacharach, 2014, p. 280). In the contextof TID, in which the athletes wish to beselected for the higher squad, the ten-dency to give socially desirable answersbecomes particularly important. Theathlete forms an opinion as to whichanswers or which characteristic expres-sions might have a positive influence onthe selection, adjusting his/her answersin the corresponding direction. Assur-ing anonymity is recommended as onestrategy for preventing or minimizingthe bias of socially desirable responses

(Furr & Bacharach, 2014). This is pos-sible by using codes or pseudonyms forresearch purposes. Thus, it has beenshown that under anonymous condi-tions, youth footballers tended to scorelower on positively associated personal-ity traits than players under personalizedconditions (Feichtinger & Höner, 2014).However, as anonymity cannot be en-sured in selection-relevant contexts, theuse of self-assessment questionnaires forselection decisions should be viewedvery critically.

As this problem of socially desirableanswers in self-rating questionnaires iswell known, the use of external assess-ments or a coach’s rating—as the sec-ond method discussed here—is recom-mended in amendment to the self-reportquestionnaires (Anshel & Lidor, 2009).However, this is associated with furthermethodological problems as the externalraters have to evaluate statements thatare not directly accessible to them. Mo-tives have a low visibility as they oftenrefer to internal emotions and thoughtsand can therefore prove difficult to ob-serve (Connelly & Ones, 2010; Furr &Funder, 2010). Consequently they haveto be assessed with different indicatorsand using evidence from previous be-havior as an aid, both of which are as-sociated with the risk of memory effects(Buss&Craik, 1983). Accordingly, ques-tionnaires on the external assessment ofthe achievement motive—such as thoseusedbytheSwissNationalOlympicCom-mittee (SwissOlympic; Fuchslocheret al.,2011)—should only be used with greatcaution in TID.

As a third method, the assessmentof the implicit achievement motive isdiscussed. The implicit achievementmotive is not directly accessible andhence less affected by social desirability.For the assessment, projective methods,such as the Picture Story Exercise (PSE;Schultheiss & Pang, 2010) were used.However, assessment tools like theseare time-consuming and uneconomical,and therefore cannot be recommendedin TID.

Neither the explicit nor the implicitachievement motive seems to be assess-able problem-free in the context of se-lection decisions using up-to-date diag-

German Journal of Exercise and Sport Research 4 · 2019 411

nostic tools. Nevertheless, we must em-phasize that the tools mentioned are notwithout use per se, but that these as-sessment difficulties arise specifically tothe context of TID in sport. Insteadof assessing motives, it is also conceiv-able to rely on recording concrete be-havior, which, to put it very simplis-tically, results from the interaction be-tween persons (e.g., motives) and situa-tions(Heckhausen&Heckhausen, 2010).The advantage is that behavior can alsobe observed from the outside (e.g. bythe coaches), and does not need to beinvestigated invasively. It is argued byJohnson (1997) that observers are bet-ter judges of behavior, whereas for morecovert cognitions and feelings, self-re-ports are most accurate. Likewise, it canbe deduced from the general model ofdeterminants andcourseofmotivatedac-tion (Heckhausen & Heckhausen, 2010,p. 3) that achievement-motivated behav-iorpredicts future behavior and thus laterperformance better than a simple motiveassessed by self-report that ignores situ-ational factors. Besides, it seems reason-able to focus on decisions from coachesas they often form the basis for selectiondecisions in sport (Christensen, 2009).Coaches decide which athletes shouldmake the leap to the next highest squad,where they will receive optimal support.For this decision, expert coaches candraw on many years of experience, andthus possess different opportunities forthe comparisonof different athletes: “Ex-pert coaches canbase their assessmentsofpsychological characteristics on a repre-sentative sample of talented players theyhave worked with in the past . . . [and]are able to make inter-individual com-parisons and use this knowledge to eval-uate and predict a player’s current andfuture potential” (Musculus & Lobinger,2018, p. 2). In this way, talent charac-teristics identified by coaches seem tobe relevant for the prognosis of futureachievements (Jokuschies et al., 2017).Therefore, a suitable new tool for assess-ing achievement motivation in the con-textofselectiondecisionsinsportsshouldconsequentially be based on coaches’ rat-ings of achievement-motivated behavior.

But what overt behaviors can onethink of in terms of achievement-moti-

Abstract · Zusammenfassung

Ger J Exerc Sport Res 2019 · 49:410–423 https://doi.org/10.1007/s12662-019-00600-6© The Author(s) 2019

C. Zuber · A. Conzelmann

Achievement-Motivated Behavior in Individual Sports (AMBIS-I)—Coach Rating Scale. Development and preliminary validation

AbstractThe assessment of achievementmotivationin the context of selection decisions inelite youth sports is associated with seriousproblems (e.g. socially desirable responses).In order to counteract such problems, anexternal rating scale for the assessment ofthe Achievement-Motivated Behavior inIndividual Sports (AMBIS-I) from a coach’sperspective was constructed and checked forpsychometric quality in three consecutivesteps. The studies are based on four differentGerman-speaking samples, including 101experienced coaches from individual sportsand 26 sport psychologists. Multiple phasesof exploratory structural equationmodeling,item removal and cross-validation unveiled

a three-factorial model with 10 itemsdisplaying excellent fit indices, acceptableto good reliability, and evidence based oninternal structure. Relationshipswith athletes’performance level point to the instrument’sevidence for test-criterion relationship. Thesepreliminary results are promising whenconsidering the construction and show thepotential of the economical coach rating scalefor the scientifically sound assessment ofobservable achievement-motivatedbehaviorin individual sports.

KeywordsCoach rating scale · Youth sports · Talentidentification · Test construction · Psychology

Trainerbeurteilungsskala zum leistungsmotivierten Verhalten inIndividualsportarten. Entwicklung und erste Validierung

ZusammenfassungDie Diagnose der Leistungsmotivation imKontext von Selektionsentscheidungen imNachwuchsleistungssport ist mit ernsthaftenProblemen assoziiert (z. B. sozial erwünschtesAntwortverhalten). Um diesen Problemen zubegegnen, wurde in 3 aufeinanderfolgendenStudien ein Fremdbeurteilungsinstrumentfür die Erfassung von leistungsmotiviertemVerhalten in Individualsportarten ausTrainersicht (AMBIS-I) konstruiert undbezüglich seiner Testgütekriterien überprüft.Die Studien basieren auf 4 unterschiedlichendeutschsprachigen Stichproben mit 101erfahrenen Trainer(inne)n aus Individual-sportarten sowie 26 Sportpsycholog(inn)en.Nach mehreren Phasen von explorativerStrukturgleichungsmodellierung, Item-elimination und Kreuzvalidierung ergabsich ein 3-Faktoren-Modell, welches aus

10 Items besteht und eine hervorragendeAnpassungsgüte sowie akzeptable bisgute Reliabilität und faktorielle Validitätaufweist. Die gefundenen Zusammenhängemit dem Leistungsniveau der Athlet(inn)endeuten zudem auf die Kriteriumsvaliditätdes Instruments hin. Diese ersten Ergebnissesind im Hinblick auf die weitere Validierungvielversprechend und zeigen das Potenzialdieser ökonomischen Trainerbeurteilungsska-la zur wissenschaftlich fundierten Erfassungdes beobachtbaren leistungsmotiviertenVerhaltens in Individualsportarten auf.

SchlüsselwörterTrainerbeurteilungsskala · Nachwuchssport ·Talentidentifikation · Testkonstruktion ·Psychologie

vated behavior? Furr and Funder (2010)state that in general personality psychol-ogy we know a lot about persons, butmuch less about situations andbehaviors.In sport, to the best of our knowledge notheoretical approach exists that explic-itly deals with achievement-motivatedbehavior. The concept that seems closestto us in this regard is self-regulation orvolition. Volition is defined as “. . . meta-motivational processes required to sta-

bilize motivation in order to maintainthe intended action, especially whenrewards are not available immediately”(Elbe & Beckmann, 2006, p. 143). Itis according to the Rubicon Model ofAction Phases (at a glance: Achtziger &Gollwitzer, 2018) assuming that motiva-tional processes only lead to the decisionto act, and that volitional processes arethen needed to implement the favoredbehavior (Achtziger & Gollwitzer, 2018).

412 German Journal of Exercise and Sport Research 4 · 2019

Those processes are responsible for initi-ating an action and maintaining it untilthe goal has been reached, and includecognitive, motivational and emotionalcontrol strategies for not giving up whenthings get difficult, not letting oneself bedistracted, not losing one’s confidence,and staying positive (Elbe & Beckmann,2006). As in the course of a sportingcareer, long-term goals must be pursued,and high training loads have to be com-pleted, volition is of great importancefor elite youth athletes (Baron-Thiene &Alfermann, 2015; Elbe, Szymanski, &Beckmann, 2005a). Volition or self-regulation can be assessed using self-rating questionnaires like the VolitionalComponents in Sport Questionnaire(Wenhold, Elbe, & Beckmann, 2008),which is a sport-specific adaptationof thegeneral Volitional Components Ques-tionnaire (Kuhl & Fuhrmann, 1998).The analysis of the items (e.g. “I amoptimistic about most things in sports”;“In training, I often have to think aboutthings that have nothing to do with whatI’m doing”) displays that also this con-struct, which should be closer located tobehavior than the achievement motives,is not observable from the outside.

In summary, despite the very rich tra-dition of theories on achievement mo-tives, it is not possible to deduce con-creteperformance-motivatedbehavior insports from theory, which is whywe haveopted for an inductive approach for testconstruction (Smith, Fischer, & Fister,2003).

Present research

Toachieve our goal of constructing a reli-able, validandeconomical tool for thesci-entifically sound assessment of achieve-ment-motivated behavior in sports, ap-plicable in coaches every day practice(Horvath&Röthlin, 2018), webuilt uponacombinationofaprototype(Broughton,1984) and an inductive (Smith et al.,2003) test construction strategy. The act-frequency approach as a form of proto-type approach relies on the definitionsof constructs elaborated by psycholog-ical laypersons (Buss & Craik, 1983),and is used for item construction and to

provide evidence based on test content1(Studies 1 and 2). Despite the remark-able achievements in motivational psy-chology made in recent decades, little isknown about observable achievement-motivated behaviors. To close this gapand to ensure that the tool is applicablewithin the practice of TID, in Study 1(Act nomination) experienced coacheswere asked about manifest achievement-motivatedbehavior inconcrete situations(acts). In the subsequent phase (Study 2:Prototypicality rating), these acts werejudged by other coaches and sport psy-chologists to assess evidence based ontest content. In the third step (Study 3:Construction and initial validation), thefinal version of the rating scale was con-structed and then checked for reliabilityand evidence based on internal struc-ture and relation to a relevant criteria.The absence of an a priori theory onachievement-motivated behavior meansthat no factor or subscale structure is hy-pothesized in advance. Once the itemswere developed, the internal structureis uncovered, as when using inductivetest construction strategy (Smith et al.,2003). Because it must be assumed thatthe achievement-motivated behavior ofindividual and team athletes differs dueto their different settings (e.g. higher au-tonomy in individual sports, individualvs. team training), this study focuses onindividual sports only. The ethics com-mittee of the Phil.-hum. Faculty of theUniversity of Bern approved this study.

Study 1: Act nomination

Method

Participants and procedureThesamplesof the coacheswere recruiteddirectly through the sport federations viaSwiss Olympic. The sporting directors ofthe sport federations categorizedbySwissOlympic in the categories 1 to 3 (of 5),according to their national importance

1 We adopt the rather new standards forusing the concept of validity and thereforedo not use the historical nomenclatures likecontent or criterion validity (cf. AmericanEducational Research Association, AmericanPsychological Association, andNational CouncilonMeasurement inEducation,2014)

and achievement potential, were askedto send lists of all their coaches at the 1stor 2nd level of education (professionaltraining for elite sports or competitivesports). In total, we received the contactdetails of 438 coaches (15% female). Forthe studies in this publication, only Ger-man-speaking coaches were included.For act nomination, 36 coaches from18 sport federations were randomlyselected and invited to participate inStudy 1. They received an informationalletter with a link to the online survey.Overall, 20 coaches from 14 differ-ent sport federations (Mage = 46.0± 9.17years) took part and filled in the onlinesurvey. They had an average experienceof M= 16.15± 9.63 years as coaches intheir sport.

MeasuresIn order to collect acts that the expe-rienced coaches consider to representachievement-motivatedbehavior, thefol-lowing instructions were given in the on-line survey:

“Please think of one or more athleteswhom you consider to be particularlyachievement motivated with whom youare working with at the moment, or whomyou have trained in the past. Now namethree individual actions in specific situa-tions inwhich the achievementmotivationof these athletes you are thinking of wasclearly expressed.”

Data analysisIn this survey, the participants gen-erated a total of 67 situations. Thesewere minimally modified by the firstauthor, without changing the content ofthe respective statements (cf. Amelang,Herboth, & Oefner, 1991; Krüger &Amelang, 1995): Very long and detailedsituations were shortened (e.g. “He hascarried out regenerative activities [black-roll or stretching] without being asked todo so”→ “He/she has carried out regen-erative activities without being asked todo so”), or divided into two acts (e.g. “Heis very ambitious, sets himself goals andreacts emotionally to their achievementor failure”→ “He/she has set goals” and“He/she has reacted emotionally whenhe/she failed to reach a self-imposed

German Journal of Exercise and Sport Research 4 · 2019 413

Main Article

goal”). Errors in grammar or spelling,as well as colloquial expressions werecorrected. Repeatedly listed acts weredeleted. All the modifications were ex-tensively discussed with the coauthor toensure that no unjustified modificationswere made.

Results

Altogether, these adjustments resulted in58 acts, which were included in the sec-ond study, where they were tested withregard to being prototypical of achieve-ment-motivation behavior.

Study 2: Prototypicality rating

Thegoals of study 2were twofold: On theone hand, evidence based on test contentof the acts was assessed by a compar-ison between psychological laypersons(expert coaches) and experts in relationto achievement motivation (sport psy-chologists). On the other hand, by theexamination of the rated prototypicalityreferring to achievement-motivated be-havior of all acts, it should be avoidedthat especially low prototypical or inap-propriate acts were processed further.

Method

Participants and procedureFor the prototypicality rating, 40 coachesfrom 18 sport federations and 150 sportpsychologists were invited to partic-ipate in Study 2. The coaches wererandomly selected and contacted viathe contact lists mentioned earlier. Thesport psychologists were contacted viathe member list of the Swiss Associa-tion of Sport Psychology (SASP). Theyreceived an informational letter witha link to the online survey. In theend, 21 coaches from 12 different sportfederations (MAge = 41.48± 9.4 years;MExperience = 14.20± 5.67 years), who hadnot participated in Study 1, and 26 sportspsychologists(MAge = 43.23± 10.14years;MExperience = 9.42± 8.05 years) took part.

MeasuresThe participants were asked to rate the58 acts that had been constructed in

Study 1, based on the following instruc-tions:

Below you will see a series of acts thatmay describe achievement motivation toa greater or a lesser extent. For each ofthese behaviors, please rate on a scalefrom 1 (not at all) to 5 (very) to whatextent you consider this behavior to beachievement motivated, or whether inyour opinion this act has something to dowith achievement motivation.

In addition, the participants were askedto indicate their age, gender, vocationaltraining and experience as a coach orsport psychologist.

Data analysisThe acts were presented in a randomorder. The prototypicality ratings werechecked for mean differences betweenthe two samples of the coaches and thesport psychologists. As the null hypothe-sis is the requested hypothesis, the alphalevel was set to α= 0.20. Cohen’s d wascalculated to determine the size of theeffect.

Results

The results showed that the overall as-sessment leans towards fairly prototypicalin both groups with M= 3.75 (standarddeviation [SD]= 0.99). On the level of theindividual acts, the range of prototypical-ity is between M= 4.53 (“He/she has setgoals for himself/herself ”) and M= 2.74(“He/she cried when he/she did not windespite a good performance in the com-petition because he/she judged himself/herself better than the opponents”), withonly five of 58 acts lying slightly belowthe level of 3 (somewhat prototypical).Out of the 58 acts, 14 showed significantdifferences between the two groups withp< 0.20. Outof this, thegroup judgmentsdiffer for nine acts with a large (d> 0.8)or moderate (d> 0.5) effect: Seven actswere regarded as more prototypical bythe sports psychologists (e.g. “He/shehas asked how he/she could further de-velop themselves in sports”). Only twoacts were considered more prototypicalby the coaches (e.g. “He/she strived tobe the best at performance comparisonsin another type of sport”). For all other

acts, there are no or only small group dif-ferences in terms of the prototypicalityassessment.

Brief discussion of Study 2

The results show that, on average, theacts were rated as being fairly prototyp-ical and therefore appropriate for mea-suring the concept of achievement-mo-tivated behavior. For nine acts, the twogroups do not agree about the prototyp-icality with a meaningful different as-sessment (d> 0.5). These items were re-moved from the rating scale in a first stepbecause the results of the group compar-ison indicate that the evidence based ontest content of these acts is unclear.2 Interms of differences in the perception ofthe two samples, a trend became appar-ent that behavior pointing to the conceptof task orientation (Duda, 2007), that isstriving to achieve one’s own goals and toconstantly improve oneself, is regardedas more prototypical, and therefore ofhigher relevance for achievement mo-tivated behavior by the sports psychol-ogists than by the coaches. This indi-cates that the two groups might conceivethe concept of achievement motivationin a slightly different form, or that thecoaches make the everyday experiencesthat task orientation is not as importantthan it is considered to be in research(Elbe & Wikman, 2017).

As, to our knowledge, there exists nothreshold for sufficient prototypicality,and as only five acts showed prototypi-calities slightly lower than somewhat pro-totypical, all other acts were temporarilyretained in the item pool (but none ofthe lower prototypical acts was includedin the final rating scale). The upcomingvalidation phase had to examinewhetherthe remaining 49 acts could be arrangedto form a reliable and valid external rat-ing scale for assessing achievement-mo-tivated behavior.

2 The other five acts which showed smalldifferenceswere not removed in this stepbut allareexcludedfromthefinal instrument.

414 German Journal of Exercise and Sport Research 4 · 2019

Study 3: Construction andvalidation

In Study 3, the first step (t1) was to reducethe number of items, and to explore theunderlying factor structure of the itempool to such an extent that at least threeitems per scale (Smith et al., 2003) createan economically applicable instrumentfor themeasurementof theAchievement-Motivated Behavior in Individual Sports(AMBIS-I).The reductionof items seemsfor us to be a necessary precondition toensure that the coaches, as the poten-tial user of the coach-rating scale, willbe willing to use the instrument in fur-ther practice (Horvath & Röthlin, 2018).To check the dimensionality seems con-structive, as concerning the magnitudeof different theories related to achieve-ment motivation, a homogeneous 1-fac-tor structure cannot be expected, and asthe absence of an a priori theory withinthe inductiveapproachmeans thatnofac-tor structure is hypothesized in advance(Smith et al., 2003). The second step (t2/3)checks whether the designed instrumentfulfills the requirements towards relia-bility and internal structure. Finally, wechecked the evidence for test-criterionrelationship in terms of whether thereare differences in the achievement-mo-tivated behavior depending on the levelof performance.

Method

Participants and procedureThe sample for Study 3 consisted ofcoaches who were asked to rate theachievement-motivated behavior of theirathletes. Using the lists provided by thesport federations, 160 coaches were in-vited to participate in the investigation.69 coaches participated in the studyat t1, rating 288 athletes in individualsports with AMBIS-I. According to theguidelines of Swiss Olympic sports, dis-ciplines comprising of teamswith a smallnumber of athletes (e.g. tennis, rowing)are counted in addition to individualsports. In a next step of the pre-analysis,the assessments of those coaches whoindicated that they had known theirathletes for less than half a year, or thatthey did not feel certain in their assess-

ment, were removed from the dataset.The data were also checked for extremeresponding, where no evidence of oc-currence was found. The final sample ofthe coaches for Study 3at t1 is thereforecomposed of 278 ratings by 67 coaches(19 women, 28.4%, 48 men, 71.6%,Mage= 41.88± 11.96 years) from the fol-lowing sports: badminton, biathlon,curling, free skiing, golf, judo, artisticcycling, cross country, track and field,mountain biking, road cycling, sledding,rowing, swimming, alpine skiing, shoot-ing, tennis and vaulting. The coachesexhibit a high level of education, withmore than 50% having successfully com-pleted the highest or 2nd highest level.They have M= 16.19± 10.93 years ofprofessional experience and have knownthe athletes they judged for an average ofM= 4.11± 3.45 years. At t2 (6–8 weeksafter t1) and t3 (4 months after t1),46 coaches (16 women, 35%; 29 men,65%; Mage = 42.95± 11.67 years) partic-ipated and conducted ratings of 176athletes for the second time. For 52 ath-letes, the assessments of two coaches(mostly head and assistant coaches)are available at the first measurementpoint to assess inter-rater reliability.All data were collected using an inter-net-based questionnaire (LimeSurveyVersion 2.50). The coaches had the op-portunity to pause and save their ratingsto continue another day.

MeasuresAchievement-motivated behavior. ForStudy 3, a coach-rating version was cre-ated from each of the 49 acts. A 4-pointscale from 0 (never) to 3 (often) witha not able to respond option was used.For the second measurement time (t2/3),the answer format was extended by thecategory 4 (always), as we found ceilingeffects at t1 (see . Table 2). Because eachcoachhad submitted a listwith thenamesof athletes they were training at the mo-ment, we were able to request their ratingfor each athlete individually:

How often did athlete A [name of one ofthe coach’s athletes was inserted] displaythe acts mentioned below during the last12 months?

Further variables. As mentioned before,coaches were asked how certain they feltin their assessment of the respective ath-lete (not at all, a little, somewhat, fairlymuch), and how long (in years) they hadalready known the rated athlete. Finally,the educational level of the coaches wasassessed.

Performance level. As an external per-formance criterion to assess the evidencefortest-criterionrelationship,wecheckedat t1 whether the athletes rated by theircoachesonAMBIS-Iwerecurrentlyhold-ing a Swiss Olympic Card (SOC). SOCsdisplay achievements reached in compe-titionsandcanadditionallybeconsideredan expression of existing potential. Thenational sport associations award themaccording to their specific selection con-cept (0: no SOC; 1: local SOC; 2: regionalSOC; 3: national SOC; 4: international/elite SOC). Out of this criterion, we com-posed two groups: One group consistingofathletesonaregionalor lower level, andtherefore with a lower attributed poten-tial, and a secondgroup consistingof ath-letes on a national or international level,and therefore with a higher attributedpotential.

Data analysisThe analysis of the data was done usingIBM SPSS Statistics 24 and Mplus Ver-sion 8 (Muthén & Muthén, 2017). Inorder to construct a tool that can be ap-plied economically, a large part of theoriginal 49 items had to be removed. Forthis exploratory procedure and to gatherevidence based on internal structure, ex-ploratory structural equation modeling(ESEM) was chosen. We aimed to con-struct a tool thatunites at least three itemsper factor, features a goodmodel fit andatthe same timeavoids content redundancyas much as possible. The statistical crite-ria for the choice and elimination of theitems are based onAppleton, Ntoumanis,Quested, Viladrich, andDuda (2016)andPayne, Hudson, Akehurst, and Ntouma-nis (2013).

Exploratory structural equation mod-eling (ESEM).TheESEMmodel for t1wasestimated using the robust maximumlikelihood (MLR) estimatormethodwith

German Journal of Exercise and Sport Research 4 · 2019 415

Main Article

oblique geomin rotation, in line with theexploratory nature of the analysis (e.g.epsilon was fixed at 0.5). In doing so,missing values were estimated by meansof the full information maximum like-lihood (FIML) procedure. Overall, theproportionofmissingdatawas4.3%. Fol-lowing the recommendations of Scher-melleh-Engel, Moosbrugger, and Müller(2003), a good fit is indicated whenχ2/df≤ 2.00; CFI (comparative fit index)and TLI (Tucker-Lewis Index)≥ 0.97;and root mean square error of approxi-mation (RMSEA) and standardized rootmean square residual (SRMR)≤ 0.05. Totest the generalizability of the proposedmodel, the model was cross-validatedwith the data from t2 and t3. Because thefactor structure was already known atthat time, target rotation was chosen asa more confirmatory approach (Marsh,Morin, Parker, & Kaur, 2014).

Reliability and validity. To estimate thereliability of the indicators, we computedsquared multiple correlations (SMC). Toestimate the reliability of the constructs,we calculated the composite reliability(CR) (Bagozzi & Yi, 2012) and the aver-age variance extracted (AVE) (Fornell &Larcker, 1981). We used SMC≥ 0.40,CR≥ 0.70 and AVE≥ 0.50 as cut-offs forgood reliabilities. To estimate discrimi-nantevidenceof the tool, theFornell-Lar-cker criterion (Fornell & Larcker, 1981)was calculated. This criterion requiresthat the AVE of one factor should behigher than any squared correlation withanother factor, assuming discriminantevidence at the factor level. To deter-mine the test–retest reliability, referenceis made to the data at t2. The coaches ofthis partial sample were invited to par-ticipate again after 8 weeks. Altogether,64 estimates were made after an aver-age time of 3.2± 0.72 months. The in-ter-rater reliability between head coachand assistant coach was determined sep-arately for each training group (consist-ing of athletes assessed by the same twocoaches) using the intra-class coefficient(ICC 2-way random effects model, abso-lute agreement), and then averaged overeach training group. The interpretationis based on the recommendations of Kooand Li (2016), who classified ICCs lower

than 0.50 as poor, between 0.50 and 0.74as sufficient, between 0.75 and 0.89 asgood and higher than 0.90 as excellent.To compare the extent of achievement-motivated behavior between the two per-formance groups, an independent t-testwas performed using the values of thefactors of AMBIS-I as dependent vari-ables. The level of significance was set atp< 0.05 for all analyses.

Results

In a first step,we removed 24 itemswhichwereansweredby12to42%ofthecoacheswith not able to respond. It was assumedthat if a proportion of more than 10% ofall the coaches stated that they could notanswer, this is not an incidental finding,and that the reasons for not answeringmake such items not suitable in the rat-ing scale. The following reasons are con-ceivable: (a) not directly observable (e.g.“He/she went home after the training tocontinue practicing there”, 35% did notanswer), (b) unsuitable for certain sports(e.g. “He/she cried after a final defeat”,22%), or (c) improper evaluation period(e.g. “He/she decided to attend a special-ized sports school, even though his/herfriends continued to attend the regularone”; 15%; “He/she has moved his/herplace of residence closer to the traininglocation”, 20%). As the items were allconstructed by coaches without any ex-perience in item and test construction(study 1), those problematic points werenot addressed automatically, as wouldhave been addressed if the researchesthemselves would have constructed theitems.

Construction of AMBIS-IFor the remaining 25 items, six ESEMmodels with 1 to 6 factors were mod-eled and tested. The comparison of thedifferent models points to the 3 or 4 fac-tor model, as these two models have thelowest BIC (Schwarz’s Bayesian Informa-tion Criteria), and thus display a bettermodel fit taking into account the princi-ple of economy (Masyn, 2013). In boththe 3- and 4-factor versions, incremen-tal items were then removed if they fea-tured main loadings≤ 0.50, communal-ities≤ 0.40, high cross loadings or con-

tent redundancy. After multiple phasesofESEMmodeling and itemremoval, a 3-factor model with 12 items displaying anexcellent fit (CFI= 0.99, TLI= 0.97 andRMSEA= 0.04) and readily interpretablefactors emerged as the best model (see. Table 1, AMBIS-I-12t1).

For the cross-validation, thedata fromt2 and t3 was used to check the proposed12-item solution. The AMBIS-I-12 t2/3solution displays an inferior fit with thedata (cf. AMBIS-I-12 t2/3, . Table 1). Be-cause act 03 showed a low loading on itsfactors and therefore small communality,and as act 47 showedhighcross-loadings,wedecided to remove these two items, re-sulting in an excellentmatch between theestimated and the observed data at t1 (cf.AMBIS-I-10 t1, . Table 1), as well as att2/3. The coach-rating scale for achieve-ment-motivated behavior in individualsports (AMBIS-I) therefore consists of10 itemsformingthree factorswhichwereaccording to their content (see discus-sion), named proactivity, ambition andcommitment.

As shown by the descriptive statis-tics in. Table 2, proactivity is least likelyto be displayed, followed by ambitionand commitment. Additionally, com-mitment exhibits restricted variance, es-pecially at t1, and thus probably a ceil-ing effect. The higher values in all di-mensions at t2/3 compared with t1 areat least partly due to the change in thescaling from 0–3at t1 to 0–4at t2/3. Thefactor intercorrelations lie in the range0.22< r< 0.67, which is acceptable withregard to the differentiation of content(. Table 2). Due to the significant cor-relations of the three factors, especiallyat t2/3, the factors cannot be assumedto be uncorrelated. For this reason, thereliability of a total score—in the senseof a higher-order factor—will be exam-ined as well. The disadvantage of thelower content specificity of a total scorecould possibly be compensated for byimproved reliability (through the higherlength of the test) and thereby associatedwith higher validity (Furr & Bacharach,2014; Kingston, Scheuring, & Kramer,2013).

416 German Journal of Exercise and Sport Research 4 · 2019

Table 1 Goodness of fit statistics and information criteria for themodels estimated in Study 3

Models χ2 p (df) χ2/df CFI TLI RMSEA (CI 90%) SRMR

Accepted stan-dard

– >0.05 ≤2 <0.97 <0.97 ≤0.05 ≤0.05

AMBIS-I-12 t1 44.20 0.09 (33) 1.34 0.99 0.97 0.04 (0.00–0.06) 0.02

AMBIS-I-12 t2/3 64.64 <0.001 (33) 1.96 0.96 0.92 0.07 (0.05–0.10) 0.03

AMIBIS-I-10 t1 18.94 0.40 (18) 1.05 1.00 1.00 0.01 (0.00–0.06) 0.02

AMBIS-I-10 t2/3 26.09 0.10 (47) 1.45 0.99 0.97 0.05 (0.00–0.09) 0.05

CFI comparative fit index, TLI Tucker-Lewix index, RMSEA root mean square error of approximation,SRMR standardized root mean square residual

Table 2 Descriptive statistics and intercorrelations of the AMBIS-I factors and the total score att1 and t2/t3Factors Descriptive statistics Intercorrelations

M SD Proactivity Ambition Commitment

Proactivity t1 1.51 0.75 1 – –

t2/3 1.78 0.93 1 – –

Ambition t1 1.80 0.82 0.22 1 –

t2/3 2.24 0.95 0.55 1 –

Commitment t1 2.40 0.53 0.30 0.23 1

t2/3 2.81 0.68 0.67 0.39 1

Total scoreAMBIS-I

t1 1.91 0.53 – – –

t2/3 2.28 0.71 – – –

t1: Scale 0–3; t2/3: Scale 0–4,Mmean, SD standard deviation

Reliability and evidence based oninternal structureThe corresponding items of AMBIS-I,their factor loadings and communalities,and the factor reliabilities for t1 and t2/3are displayed in . Table 3. It should benotedthat the itemswereconstructedandchecked in German, and that the Englishtranslations of the items in. Table 3 havenot been validated and are only includedfor informational purposes. The originalGerman wording of the items is availablein Table 1 of the Electronic Supplemen-taryMaterial1. Thefactor loadingsonthemain factors lie between 0.48 and 0.84for all items. Five items exhibit crossloadings> 0.20. C1 has a somewhat lowindicator reliability at t1. All other itemsdisplay good communalities (>0.40). Thetwo factors proactivity and ambition dis-play good factor reliabilities at both timesof measurement. Only commitment hasfactor reliabilities that lie below the de-sired thresholds, especially at t1. In thecross validation at t2/3, their CR= 0.68and AVE= 0.49 values lie only just be-low the desired threshold values of 0.70and 0.50. The test–retest reliability rtt is

≥0.70 for all three factors and thuswithinan acceptable range over a comparativelylong period of 3.2± 0.72months. The to-tal score has the highest retest reliability,with rtt = 0.79. As the average varianceextracted for each factor is higher thanany squared correlationwith another fac-tor, all three factors meet the Fornell-Larcker criterion and therefore displaydiscriminant evidence at the factor level.

Inter-rater reliabilityThe intra-class coefficients (ICCs) of thecoaches averaged over all training groupsaredisplayedin. Table4. All ICCsexceptthat for proactivity lie in a satisfactoryrange (0.51< ICC<0.70). Generally, theaveragemeasure, i.e. ICCs which use themeanvaluesof the twocoachesasasourceof information, have higher values. Theinter-rater reliabilities of the total scoreshow the highest ICCs for both the av-erage and the single rater measures.

Evidence for test–criterionrelationshipInmattersofconcurrentevidencefortest-criterion relationship, all three factors, as

well as the total score, differentiate withsmall to medium effects (0.30< d< 0.62)between the two performance groups de-termined by the SOC (. Table 4). Ath-letes on the national or international levelwereratedasshowingmoreachievement-motivated behavior than those on thelower performance level.

Brief discussion of Study 3

In the first step of Study 3, we simul-taneously reduced the number of itemsto ten and checked the dimensionality.The final 10-item tool AMBIS-I, map-ping the three factors proactivity, ambi-tion and commitment, showed excellentfit indices andwas successfully replicatedby the data at t2/3.

The factor proactivity refers to gettinginvolved in training processes on one’sown initiative and for one’s own sake.The impulse to act comes from the per-son itself, and an action is carried outself-determined and does not require anexternal push (e.g. item P4: The ath-lete looks for opportunities to catch upon missed training content). Self-deter-mined means that motivation emanatesfrom the self (Ryan & Deci, 2000). In-trinsic motivation as the most self-de-termined motivation, and thus accom-panied with proactivity, is characterizedby the sensation of immanent pleasurestemming from performing an activityitself. Another theoretical link can bemade to the concept of achievement goalorientations: A proactive athlete pursueshis or her goals perseveringly (e.g. itemP2: Athletes to be first on the traininggrounds and practicing technical pro-cesses independently). The items of thefactor proactivity give no indication ofwhich goals are being closely pursued.The fact that those goals are being pur-sued persistently furthermore makes theconnection with volition obvious.

The factor ambition is characterizedby the absolute will to successfully pur-sue self-imposed goals in competitions.Ambitious athletes aim at winning com-petitions (e.g. item A2: The athleteclearly communicates before the com-petition that he/she wanted to win). Ifthose goals were not achieved, the ath-lete responds unsatisfied (e.g. item A1:

German Journal of Exercise and Sport Research 4 · 2019 417

Main Article

Table 3 Factor loadings, subscales and communalities of items, and factor reliability of subscales and retest reliabilities of subscales and total scoreScales/Items F1 F2 F3 SMC CR AVE Rtt

t1 t2/3 t1 t2/3 t1 t2/3 t1 t2/3 t1 t2/3 t1 t2/3

Total Score AMBIS-I 0.79

1. Proactivity 0.79 0.86 0.54 0.63 0.74

P1: He/she urged the coach to planfurther training sessions to improveeven further

0.64 0.83 0.21 – – – 0.48 0.62 – – – – –

P2: He/she was the first to be onthe training grounds and practicetechnical processes independently

0.74 0.76 – – – – 0.64 0.67 – – – – –

P3 He/she stayed on after trainingto continue practicing

0.73 0.78 – – – – 0.60 0.65 – – – – –

P4: He/she independently lookedfor opportunities to catch up onmissed training content

0.50 0.71 – – 0.32 – 0.46 0.57 – – – – –

2. Ambition 0.79 0.87 0.59 0.71 0.71

A1: He/she has acted annoyed whenhe/she did not finish the competi-tion in first place

– – 0.68 0.82 – – 0.51 0.64 – – – – –

A2: He/she clearly communicatedbefore the competition that he/shewanted to win today

– – 0.71 0.84 0.21 – 0.59 0.80 – – – – –

A3: He/she showed that he/she isnot satisfiedwith coming in 2ndplace

– – 0.77 0.78 – – 0.66 0.68 – – – – –

3. Commitment 0.62 0.68 0.40 0.49 0.70

C1: He/she oriented himself/herselftowards stronger athletes duringtraining

– – – 0.24 0.70 0.48 0.38 0.53 – – – – –

C2: In highly demanding exercises,he/she worked to the point of ex-haustion

– – – – 0.62 0.57 0.41 0.41 – – – – –

C3: He/she has shown an “active”stance in training

– – 0.26 – 0.59 0.75 0.42 0.51 – – – – –

F1–F3 Factor loadings; loadings< 0.20 are not displayed, SMC estimated squared multiple correlation or communality of the item, CR Composite reliability;AVE average variance extracted. The original German items are available in Table 1 of Electronic Supplementary Material 1

The athlete has acted annoyed when hedid not finish the competition in firstplace.) Thus, the conceptual proximityof the factor ambition to achievementgoal orientations competitiveness andgoal orientation (as the goals are self-imposed), and—to a lesser degree—egoorientation, becomes visible.

Inturn, the factorcommitment is againlocalized in the setting of training. Com-mitted athletes show that they are readyand willing to perform (e.g., item C3:The athlete shows an “active” stance intraining). He or she displays will to workhard to achieve a goal and tries to solvea task repeatedly, even in the face ofadversity (Scanlan, Carpenter, Simons,Schmidt, & Keeler, 1993). Hence, com-mitment is closely related to discipline

in training (e.g. item C1: In highly de-manding exercises, the athlete works tothe point of exhaustion) and volition, asvolition is especially important for “re-alizing long and intense training loadsduring the course of an athletic career orfor keeping up regular exercising” (Elbeet al., 2005a, p. 560). Because commit-ted athletes orient towards and comparethemselves to stronger athletes, a con-nection with competitiveness is also tobe assumed (Gill & Deeter, 1988).

In the second step, we checkedwhether the designed instrument fulfillsinitial requirements towards its psycho-metric properties. Overall, the resultspoint to satisfactory reliability and evi-dence based on internal structure. Thefactor reliability of commitment is a bit

more critical. However, as commitmentshows the highest width in terms ofcontent, and at the same time the lowestvariance (e.g. . Table 2), this is hardlysurprising. Contrasting these results,commitment showed the best valuesfor inter-rater reliability. Although thevalues for the inter-rater reliability arenot very high in the overview, the resultscoincide with coefficients of a meta-analysis of the inter-rater reliability ofsupervisor ratings in an occupationalcontext (Conway & Huffcutt, 1997). Be-sides, to increase the reliability of thestatements, the coach assessment shouldwhere possible be carried out by twocoaches (Musculus & Lobinger, 2018)and the values should then be averagedfor further interpretation. The findings

418 German Journal of Exercise and Sport Research 4 · 2019

Table 4 Intraclass correlationcoefficients (ICC)asmeasureof agreementbetweenheadandassistantcoachaveragedoverall traininggroups (n=13)for t1 (ICC 2-way randomeffectsmodel, absolute agreement) and descriptive statistics andbetween-group analyses of AMBIS-IScales Intraclass correlation Regional level & lower

(n=159)National level & higher(n= 119)

t (df) p d (95%CI)

ICC (singlerater)

ICC (mean of2 raters)

M SD M SD

Proactivity 0.42 0.55 1.32 0.77 1.77 0.65 5.23(270)

<0.001 0.62(0.38–0.87)

Ambition 0.51 0.65 1.69 0.83 1.93 0.78 2.44(268)

0.02 0.30(0.06–0.54)

Commitment 0.58 0.70 2.33 0.55 2.49 0.49 2.48(276)

0.01 0.31(0.07–0.54)

Total scoreAMBIS-I

0.60 0.72 1.78 0.55 2.07 0.46 4.65(276)

<0.001 0.57(0.32–0.81)

The 13 training groups (athletes and two coaches) comprise two to eight athletes. Nine training groups even comprise less than five athletesMmean, SD standard deviation, CI confidence interval

on test-criterion relationship point in theexpected direction, and are—in the caseof ambition and commitment—in linewith previous findings from a self-reportstudy on the difference between elite-and subelite-level athletes (Halldors-son, Helgason, & Thorlindsson, 2012) .The magnitude of the uncovered differ-ences—with small to medium effects—issubstantial considering performance insports is determined in a multidimen-sional manner. Thus, in addition topsychological characteristics, such asachievement-motivated behavior, fea-tures from various other areas are alsorelevant, including motor performance,social support or the characteristics oftraining (Meyer, Gnacinski, & Flechter,2018; Rees et al., 2016).

General discussion

The studies presented aimed to con-struct a reliable, valid and economic toolfor the scientifically sound assessmentof achievement-motivated behavior insports. This stemmed from the factthat the tools currently available forassessing achievement motivation dis-play methodological problems such associal desirability of self-rating question-naires, motives that were not directlyaccessible by coaches, and high costsif implicit motives should be assessed.Those three limitations in the field ofselection decisions in youth elite sportscan be reduced by using AMBIS-I. WithAMBIS-I, socially desirable respondingis less problematic as the assessment

is based on coach-ratings instead ofthe ratings of the athletes themselves,and the focus is switched from mo-tives to observable behaviors. The itemsgenerated in Study 1 refer to specificsituations, in which the coaches’ lay un-derstanding of achievement motivationwas clearly expressed in the behavior oftheir athletes.

In Study 2, these items were judged bysport psychologists as experts in regard toachievement motivation, and by coachesas psychological laypersons in terms oftheirbeingcontent-valid/prototypical forthe construct of achievement-motivatedbehavior. The results of Study 3 thenindicate that AMBIS-I is an economictool that meets many requirements ofreliability and validity. The results canbe summarized as follows: The externalrating scale AMBIS-I contains the threefactors proactivity, ambition and com-mitment based on ten forms of behavior,whose frequency of occurrence is ratedby the coaches. The three factors dis-play an excellent fit with the observeddata, reasonable to good reliability co-efficients and good evidence based oninternal structure. The first findings ontest-criterion relationship are promisingwhen considering the potential use ofAMBIS-I as an instrument in a batteryof tests for talent selection in the future.

Due to the larger number of items, thetotal score is somewhatmorereliable thanthe individual factors (Tavakol & Den-nick, 2011). The calculation of a totalscore also makes sense from the point ofview of content, as in Study 1 the coaches

were questioned about achievement-mo-tivated behavior in general and not aboutforms of behavior representing the ret-rospectively extracted factors proactivity,ambition or commitment. We thereforerecommend using the individual factorsas well as the overall score for inter-pretation. The individual factors pro-videmore detailed information, whereas,from a statistical point of view, the over-all score offers more reliable statements.AMBIS-I can therefore be described asa multidimensional test with correlatedfactors (according to the definition ofFurr & Bacharach, 2014).

Because future users were included inthe construction of the items and in as-sessing the quality of the items, it can beassumed that the final toolwill bewell ac-cepted in practice. We recommend usingAMBIS-I as part of a multidimensionaltest battery for theTID. From the point ofview of talent development and furthercareer planning, itmight also be interest-ing to compare the coaches’ ratings withthe self-ratings of the athletes, and dis-cuss possible differences or similarities(Musculus & Lobinger, 2018).

Despite the encouraging results so far,some critical pointsneed tobe addressed.The cross-validation with the same sam-ple at a later date is not the gold standard;however, due to the limited sample sizeand accessibility of the coaches respon-sible for selection decisions, this seemsto be an acceptable solution. Further-more, the need to adapt the scaling ofthe AMBIS-I items between the first andthesecondmeasurement time isnot ideal.

German Journal of Exercise and Sport Research 4 · 2019 419

Main Article

Nevertheless, the successful cross-valida-tion and the improved distribution of thevalues at t2/3 confirm the decision to im-plement this adaptation. As it is also notpossible for other test constructions, wecannot guarantee that the full breadth ofachievement-motivated behavior is cov-ered. However, therewerea largenumberof acts nominated in Study 1 and manyof themwere also overlapping each otherwhich suggests that the most importantsituations might be mentioned. Addi-tionally, Smith et al. (2003, p. 473) arguethat inductive test construction has theadvantage that “. . . investigators are likelyto be able to identify all or most elementsof the construct that are recognizedby thetarget individuals. The approach lessensthe possibility that a limited theoreti-cal perspective will cause investigatorsto omit important construct facets”.

In contrast to motor tests, there areno purely objective measured valueswhen recording psychological charac-teristics. Although the social desirabilityof the self-ratings can be reduced withthe coach-ratings used, not all possi-ble answering tendencies (e.g. recall orconfirmation bias; Althubaiti, 2016) canbe controlled for even though focus-ing on specific behavior might not be asstrongly affected by some response biasesas other forms of assessment (McCrae &Weiss, 2010). The fact that the numberof athletes to be rated by one coachis low speaks against a large effect ofsystematic errors caused by differencesin individual standards of the raters.The same is true concerning our resultsfor inter-rater reliability, which lie inan acceptable range and therefore speakagainst a high impact of individual effectsof—for example—confirmation bias. Toavoid primacy or recency effects, the or-der of athletes to be rated was randomlychanged between the first and the secondmeasurement point, which resulted inacceptable retest reliabilities. In addi-tion, we should consider the whole talentselection process, for which it is stronglyrecommended to use multidimensionaltest batteries (motor performance, socialsupport or the characteristics of train-ing), and therefore also different datacollection instruments as motor testsor performance results (Vaeyens et al.,

2008). Hence, selection decisions shouldnot be based on just one instrument orvariable. Therefore, the effect of possi-ble rating biases from AMBIS-I on theholistic selection decision can thus beconsidered small. However, wheneverpossible, it is recommended to have theratings of two coaches (cf. main andassistant coach) to mitigate subjectivepreferences and confirmation bias.

Despite the promising results basedon content, internal structure and test-criterion relationship, the next step has tobe the ongoing validation of the test. Onthe one hand, relationships between testscores and other measures must be car-ried out by checking whether AMBIS-I is related to other instruments mea-suring motivational and volitional con-structs and therefore refer to the un-derlying construct achievement motives.From a content perspective, connectionscan be assumed to exist between proac-tivity and intrinsic motivation (Deci &Ryan, 1985), or between ambition andcompetitiveness (Gill & Deeter, 1988).In addition, it seems reasonable to ex-amine the extent to which self-assess-ment regarding achievement-motivatedbehavior is related to the coaches’ assess-ment. Although high correlations arenot be expected (Conway & Huffcutt,1997), it seems important to review theoverall trend and to involve different per-spectives. On the other hand, there arestill important questions regarding test-criterion relationship. Thus, an instru-ment that is meant to predict athleticperformance or success should not onlybe examined using a cross-sectional de-sign, as was done in the present study,but also in a prognosticmanner (Vaeyenset al., 2008). Both coach assessment andobjective success criteria can be used asperformance criteria.

In addition to the extendedvalidation,it also makes sense to check the retest re-liability over a shorter time period. Par-ticipants took an average of more than3 months to complete the online ques-tionnaire, despite an early reminder forparticipation at the second date of mea-surement. Therefore the calculated relia-bility coefficient reflects, on the onehand,the degree to which measurements erroraffects the test and, on the other hand,

the amount of change in the true scores(Furr & Bacharach, 2014). It must there-fore be regarded as a mixture betweenthe examination of stability and reliabil-ity. In the literature, many test–retestanalyses are conducted over a period of2 to 8 weeks (Furr & Bacharach, 2014).

Besides curiosity about learning moreabout the psychometric properties, ad-vanced analyses of stability, change andage-relatedness of achievement-moti-vated behavior are of great interest.It can, for example, be assumed thatvery high values in proactivity are hardto achieve for the younger athletes, asthey might—due to external circum-stances—be unable to decide for them-selves when to arrive at training, andwill therefore be accompanied by par-ents. With the progressive developmentof behavioral autonomy, autonomousactions and decisions, higher values willbe possible as the athletes move towardyoung adulthood (Collins, Gleason, &Sesma, 1997). The question of age-de-pendent standardvalues should thereforebe addressed in the near future. In addi-tion, it is necessary to examine whetherthe much-expressed assumption of thehighest possible values in achievement-motivated behavior must be rated posi-tively in every case, or whether extrememanifestations, for example in proac-tivity, can be accompanied by negativephenomena, such as over-involvementor burnout (Gardner & Moore, 2006).Additionally, it should be emphasizedonce again that AMBIS-I has so far onlybeen tested for individual sports, andthat there has been no validation yet forteam sports. So, it is essential for futurestudies to check if the items of AMBIS-I are also suitable for team sports.

Based on the current state of knowl-edge, the newly constructed, economi-cally usable coach-rating scale AMBIS-I allows a reliable assessment of observ-able achievement-motivated behavior inindividual sports. Additionally, promis-ing initial indications of validity can bereported. On this basis, a first step hasbeen taken in identifying future success-ful athletes early on and accordingly pro-moting them more effectively.

420 German Journal of Exercise and Sport Research 4 · 2019

Corresponding address

Claudia ZuberUniversität BernBern, Switzerlandclaudia.zuber@ispw.unibe.ch

Compliance with ethicalguidelines

Conflict of interest C. Zuber andA. Conzelmanndeclare that theyhave no competing interests.

All procedures performed in studies involvinghu-manparticipants or onhuman tissuewere in accor-dancewith the ethical standards of the institutionaland/or national research committee andwith the1975Helsinki declaration and its later amendments orcomparable ethical standards. Informed consentwasobtained fromall individual participants included inthe study.

OpenAccessThis article is distributedunder the termsof the Creative CommonsAttribution 4.0 InternationalLicense (http://creativecommons.org/licenses/by/4.0/), which permits unrestricteduse, distribution,and reproduction in anymedium, provided yougiveappropriate credit to the original author(s) and thesource, providea link totheCreativeCommons license,and indicate if changesweremade.

References

Abbott, A., & Collins, D. (2004). Eliminating thedichotomybetweentheoryandpracticeintalentidentification and development: consideringtheroleofpsychology. Journal of Sports Sciences,22(5), 395–408.

Achtziger, A., &Gollwitzer, P.M. (2018). Motivationandvolitioninthecourseofaction. InJ.Heckhausen&H. Heckhausen (Eds.), Motivation and Action(3rdedn.,pp.485–527). Cham: Springer.

Althubaiti, A. (2016). Information bias in healthresearch: definition, pitfalls, and adjustmentmethods. Journal ofMultidisciplinary Healthcare,9, 211–217. https://doi.org/10.2147/JMDH.S104807.

Amelang, M., Herboth, G., & Oefner, I. (1991).A prototype strategy for the constructionof a creativity scale. European Journal ofPersonality, 5, 261–285. https://doi.org/10.1002/per.2410050402.

AmericanEducationalResearchAssociation,AmericanPsychologicalAssociation,&National Council onMeasurement in Education (2014). Standardsfor educational and psychological testing. Wash-ington, DC: American Educational ResearchAssociation.

Ames, C., & Archer, J. (1988). Achievement goals inthe classroom: students’ learning strategies andmotivation processes. Journal of EducationalPsychology, 80, 260–267. https://doi.org/10.1037/0022-0663.80.3.260.

Anshel, M., & Lidor, R. (2009). Talent detectionprograms in sport: The questionable use

of psychological measures. Journal of SportBehavior,35(3), 239–266.

Appleton, P. R., Ntoumanis, N., Quested, E., Viladrich,C., & Duda, J. L. (2016). Initial validation of thecoach-createdempoweringanddisempoweringmotivational climate questionnaire (EDMCQ-C).Psychology of Sport and Exercise,22, 53–65.

Atkinson, J.W. (1957). Motivational determinants ofrisk-takingbehavior. Psychological Review,64(6),359–372.

Bagozzi, R. P., & Yi, Y. (2012). Specification, evaluation,and interpretation of structural equationmodels. Journal of the Academy of MarketingScience,40(1), 8–34.

Baron-Thiene, A., & Alfermann, D. (2015). Personalcharacteristics as predictors for dual careerdropout versus continuation—A prospectivestudy of adolescent athletes from German elitesport schools. Psychology of Sport and Exercise,21, 42–49.

Broughton, R. (1984). A prototype strategy forconstruction of personality scales. Journalof Personality and Social Psychology, 47(6),1334–1346. https://doi.org/10.1037/0022-3514.47.6.1334.

Brunstein, J. C., &Heckhausen,H. (2010). Achievementmotivation. In J. Heckhausen & H. Heck-hausen (Eds.), Motivation and action (1st edn.,pp.137–183). Cambridge: UniversityPress.

Burtscher, J., Furtner, M., Sachse, P., & Burtscher,M. (2011). Validation of a German versionof the Sport Motiation Scale (SMS28) andmotivational analysis in competitive mountainrunners. Perceptual and motor skills, 112(3),807–820. https://doi.org/10.2466/05.06.25.PMS.112.3.807-820.

Buss, D.M., & Craik, K.H. (1983). The act frequencyapproach to personality. Psychological Review,90(2), 105–126.

Calvo, T. G., Cervelló, E., Jiménez, R., Iglesias, D., &Murcia, J. A.M. (2010). Using self-determinationtheory to explain sport persistence anddropoutin adolescent athletes. Spanish Journal ofPsychology,13, 677–684.

Christensen, M. K. (2009). “An eye for talent”: talentidentification and the “practical sense” of top-level soccer coaches. Sociology of Sport Journal,26(3), 365–382.

Coetzee,B.,Grobbelaar,H.W.,&Gird,C. C. (2006). Sportpsychological skills that distinguish successfulfrom less successful soccer teams. Journal ofHumanMovement Studies,51(6), 383–401.

Collins,W., Gleason, T., & Sesma, A. (1997). Internaliza-tion,autonomy,andrelationships:Developmentduringadolescence. In J. E. Grusec&L. Kuczynski(Eds.), Parenting and children’s internalizationof values. A handbook of contemporary theory(pp.78–99). NewYork: JohnWiley&Sons.

Connelly,B. S.,&Ones,D. S.(2010). Anotherperspectiveon personality: meta-analytic integration ofobservers’ accuracy and predictive validity.Psychological Bulletin, 136(6), 1092–1122.https://doi.org/10.1037/a0021212.

Conway, J.M., & Huffcutt, A. I. (1997). Psychometricproperties of multisource performance ratings:a meta-analysis of subordinate, supervisor,peer, and self-ratings. Human Performance,10(4), 331–360. https://doi.org/10.1207/s15327043hup1004_2.

Deci, E. L., &Ryan, R.M. (1985). Intrinsic motivation andself-determination in human behavior. NewYork:Plenum.

Duda, J. (2007). Motivation in sport: the relevanceof competence and achievement goals. In

A. J. Elliot & C. S. Dweck (Eds.), Handbook ofcompetence andmotivation (pp. 318–335). NewYork: Guilford.

Elbe, A.-M. (2004). Testgütekriterien der deutschenVersion des Sport Orientation Questionnaires[Psychometric properties of theGerman versionoftheSportOrientationQuestionnaire. Spectrumder Sportwissenschaft,16(1), 96–107.

Elbe, A.-M., & Beckmann, J. (2006). Essentialprocesses for attaining peak performance. InD. Hackfort & G. Tenenbaum (Eds.),Motivationalandself-regulatory factorsandsportperformancein young elite athletes (pp. 137–157). Aachen:Meyer&MeyerSport.

Elbe, A.-M., & Wikman, J.M. (2017). Psychologicalfactors indevelopinghighperformanceathletes.In J. Baker, S. Cobley, J. Schorer &N.Wattie (Eds.),Routledge handbook of talent identification anddevelopment in sport (pp.169–178).

Elbe, A.-M., Szymanski, B., & Beckmann, J. (2005a). Thedevelopment of volition in young elite athletes.Psychology of Sport and Exercise, 6, 559–569.https://doi.org/10.1016/j.psychsport.2004.07.004.

Elbe, A.-M., Wenhold, F., &Müller, D. (2005b). Zur Reli-abilität und Validität der Achievement MotivesScale-Sport [On the reliability and validity of theAchievementmotives scale—Sport]. Zeitschriftfür Sportpsychologie,12(2), 57–68.

Elliot, A. J., & Dweck, C. S. (2007). Competenceand motivation: Competence as the coreof achievement motivation. In A. J. Elliot &C. S. Dweck (Eds.),Handbook of competence andmotivation (pp.3–12). NewYork: Guilford.

Feichtinger, P., & Höner, O. (2014). Psychological diag-nostics in the talent development program ofthe German Football Association: Psychometricproperties of an Internet-based test battery.Sportwissenschaft, 44(4), 203–213. https://doi.org/10.1007/s12662-014-0341-0.

Figueiredo, A. J., Gonçalves, C. E., Coelho e, S.M., &Malina, R.M. (2009). Characteristics of youthsoccer playerswhodropout, persist ormove up.Journal of Sports Sciences,27, 883–891.

Fornell, C., & Larcker, D. F. (1981). Evaluating structuralequation models with unobservable variablesand measurement error. Journal of MarketingResearch, 18(1), 39. https://doi.org/10.2307/3151312.

Fuchslocher, J., Romann, M., Rüdisüli, R., Birrer,D., & Hollenstein, C. (2011). Das Talentiden-tifikationsinstrument PISTE—Wie die SchweizNachwuchsathletenauswählt [The talent identi-fication tool PISTE—how Switzerland identifiesjuniorathletes]. Leistungssport,4(2), 22–27.

Furr, R.M., & Bacharach, V. R. (2014). Psychometrics: anintroduction. LosAngeles: SAGE.

Furr, R.M., & Funder, D. C. (2010). Behavioral observa-tion. In R.W. Robins, R. C. Fraley & R. F. Krueger(Eds.), Handbook of research methods in person-ality psychology (pp. 273–291). New York,London: Guilford.

Gardner, F., & Moore, Z. (2006). Clinical sportpsychology. Champaign: HumanKinetics.

Gill, D. L., & Deeter, T. E. (1988). Development ofthe sport orientation questionnaire. ResearchQuarterly for Exercise and Sport,59(3), 191–202.

Gillet, N., Berjot, S., Vallerand, R., Amoura, S., &Rosnet, E. (2012). Examining the motivation-performance relationship incompetitivesport: acluster-analytic approach. International Journalof Sport Psychology,43, 79–102.

Gillet, N., Vallerand, R., & Paty, B. (2013). Situationalmotivationalprofilesandperformancewithelite

German Journal of Exercise and Sport Research 4 · 2019 421

Main Article

performers. Journal of Applied Social Psychology,43, 1200–1210. https://doi.org/10.1111/jasp.12083.

Gillet,N.,Vallerand,R.,&Rosnet, E. (2009).Motivationalclusters and performance in a real-life setting.Motivation and Emotion,33, 49–62.

Halldorsson, V., Helgason, A., & Thorlindsson, T.(2012). Attitudes, commitment andmotivationamongst Icelandic elite athletes. InternationalJournal of Sport Psychology,43, 241–254.

Halvari, H., & Thomassen, T.O. (1997). Achievementmotivation, sports-related future orientation,and sporting career. Genetic Social and GeneralPsychologyMonographs,123, 343–365.

Heckhausen, H., & Heckhausen, J. (2010). Motivationand action: Introduction and overview. InJ.Heckhausen&H.Heckhausen(Eds.),Motivationand action (pp. 1–9). Cambridge: UniversityPress.

Hellandsig, E. (1998). Motivation predictors of highperfomance and discontinuation in differenttypesofsportsamongtalentedteenageathletes.International Journal of Sport Psychology, 29,27–44.

Höner, O., & Feichtinger, P. (2016). Psychologicaltalent predictors in early adolescence and theirempirical relationship with current and futureperformance in soccer. Psychology of Sport andExercise,25, 17–26.

Horvath, S., & Röthlin, P. (2018). How to improveathletes’ return of investment: Shorteningquestionnaires in the applied sport psychologysetting. Journal of Applied Sport Psychology,30(2), 241–248.

Huijgen, B. C.H., Elferink-Gemser, M. T., Lemmink,K. A. P.M.,&Visscher,C. (2014).Multidimensionalperformance characteristics in selected anddeselected talented soccer players. EuropeanJournal of Sport Science, 14, 2–10. https://doi.org/10.1080/17461391.2012.725102.

Johnson, J. A. (1997). Units of analysis for thedescription and explanation of personality. InR.Hogan, J. Johnson&S. Briggs (Eds.),Handbookof personality psychology (pp.73–93). SanDiego:AcademicPress.

Johnston, K., Wattie, N., Schorer, J., & Baker, J. (2018).Talent identification in sport: a systematicreview. Sports Medicine, 48(1), 97–109. https://doi.org/10.1007/s40279-017-0803-2.

Jokuschies, N., Gut, V., & Conzelmann, A. (2017).Systematizing coaches’ ‘eye for talent’: playerassessments based on expert coaches’ subjec-tive talent criteria in top-level youth soccer.International Journal of Sports Science & Coach-ing, 12(5), 565–576. https://doi.org/10.1177/1747954117727646.

Kingston, N.M., Scheuring, S. T., & Kramer, L. B. (2013).Test development strategies. In K. F. Geisinger,B. A. Bracken, J. F. Carlson, J.-I. C. Hansen,N. R. Kuncel, S. P. Reise & M.C. Rodriguez(Eds.), Test theory and testing and assessment inindustrial and organizational psychology. APAhandbook of testing and assessment in psy-chology, (Vol. 1, pp. 165–184). Washington:AmericanPsychologicalAssociation.

Koo, T. K., & Li, M. Y. (2016). A guideline of selectingand reporting intraclass correlation coefficientsfor reliability research. Journal of ChiropracticMedicine,15(2), 155–163.

Krüger, C., & Amelang, M. (1995). Bereitschaft zuriskantemVerhaltenalsTrait-KonstruktundTest-Konzept: ZurEntwicklungeinesFragebogensaufder Basis des Handlungs-Häufigkeits-Ansatzes[Willingness for risk-taking behavior as trait

construct and test concept: The constructionof a questionnaire based on the act-frequencyapproach.Diagnostica,41, 35–52.

Kuhl, J., & Fuhrmann, A. (1998). Decomposingself-regulation and self-control: The volitionalcomponents inventory. In J. Heckhausen &C. S. Dweck (Eds.),Motivation and self-regulationacross the life span (pp. 15–99). Cambridge:CambridgeUniversityPress.

MacNamara, Á., Button,A., &Collins,D. (2010).The roleof psychological characteristics in facilitatingthe pathway to elite performance: Part 1:Identifying mental skills and behaviors. TheSport Psychologist,24, 52–73.

Marsh, H.W., Morin, A. J. S., Parker, P.D., & Kaur,G. (2014). Exploratory structural equationmodeling: An integration of the best featuresof exploratory and confirmatory factor analysis.Annual Review of Clinical Psychology,10,85–110.

Masyn, K. (2013). Latent class analysis and finite mix-turemodeling. InT.D.Little (Ed.),Oxford library ofpsychology. The Oxford handbook of quantitativemethods in psychology (pp. 551–611). NewYork:OxfordUniversityPress.

McCrae, R. R., & Weiss, A. (2010). Observer ratingsof personality. In R.W. Robins, R. C. Fraley &R. F. Krueger (Eds.), Handbook of research meth-ods in personality psychology (pp.259–272).NewYork, London: Guilford.

Meyer, B. B., Gnacinski, S. L., & Flechter, T. B. (2018).Talent identification. In J. Taylor (Ed.),Assessmentin applied sport psychology (pp. 169–182).Champaign: HumanKinetics.

Murr, D., Feichtinger, P., Larkin, P., O’Connor,D., & Höner, O. (2018). Psychological talentpredictors in youth soccer: a systematic reviewof the prognostic relevance of psychomotor,perceptual-cognitive and personality-relatedfactors. PLOS ONE,13, e205337.

Musculus, L., & Lobinger, B.H. (2018). Psycholog-ical characteristics in talented soccer play-ers—Recommendations on how to improvecoaches’ assessment. Frontiers in Psychology, 9,395.

Muthén, L. K., & Muthén, B.O. (2017). Mplus user’sguide. LosAngeles:Muthén&Muthén.

Nicholls, J. G. (1984). Achievement motivation:conceptions of ability, subjective experience,task choice, and performance. PsychologicalReview, 91(3), 328–346. https://doi.org/10.1037/0033-295X.91.3.328.

Payne, S.M., Hudson, J., Akehurst, S., & Ntoumanis, N.(2013). Developmentand initial validationof theimpression motivation in sport questionnaire-team: team. Journal of Sport and ExercisePsychology, 35(3), 281–298. https://doi.org/10.1123/jsep.35.3.281.

Pelletier, L. G., Fortier, M. S., Vallerand, R. J., & Brière,N.M. (2001). Associations among perceivedautonomysupport, formsof self-regulation, andpersistence: aprospective study.Motivation andEmotion,25, 279–306.

Pelletier, L. G., Rocchi,M.A., Vallerand, R. J., Deci, E. L., &Ryan, R.M. (2013). Validationof the revisedsportmotivation scale (SMS-II). Psychology of Sportand Exercise, 14(3), 329–341. https://doi.org/10.1016/j.psychsport.2012.12.002.

Pelletier, L. G., Tuson, K.M., Fortier, M. S., Vallerand,R. J., Briere, N.M., & Blais, M. R. (1995). Towarda newmeasure of intrinsicmotivation, extrinsicmotivation, andamotivation insports: TheSportMotivation Scale (SMS). Journal of Sport andExercise Psychology,17(1), 35–53.

Rees, T., Hardy, L., Güllich, A., Abernethy, B., Côté, J.,Woodman, T., Warr, C., et al. (2016). The GreatBritish Medalists Project: a review of currentknowledge on the development of the world’sbest sporting talent. Sports Medicine, 46(8),1041–1058. https://doi.org/10.1007/s40279-016-0476-2.

Reilly, T., Williams, M., Nevill, A., & Franks, A.M.(2000). A multidisciplinary approach to talentidentification in soccer. Journal of SportsSciences,18, 695–702.

Ryan, R.M., & Deci, E. L. (2000). Self-determinationtheoryandthefacilitationofintrinsicmotivation,social development, and well-being. TheAmerican Psychologist,55, 68–78.

Ryan, R.M., & Deci, E. L. (2007). Active human nature:Self-determination theory and the promotionandmaintenance of sport, exercise and health.In M. Hagger & N. Chatzis (Eds.), Intrinsicmotivation and self-determination in exerciseand sport (pp. 1–19). Champaign: HumanKinetics.

Sagar, S. S., Busch,B. K.,& Jowett, S. (2010). Successandfailure, fear of failure, and coping responses ofadolescent academy football players. Journal ofApplied Sport Psychology, 22, 213–230. https://doi.org/10.1080/10413201003664962.

Sarrazin, P., Vallerand, R., Guillet, E., Pelletier, L. G., &Cury, F. (2002). Motivation and dropout infemale handballers: a 21-month prospectivestudy. European Journal of Social Psychology,32,395–418. https://doi.org/10.1002/ejsp.98.

Scanlan, T. K., Carpenter, P. J., Simons, J. P., Schmidt,G.W., & Keeler, B. (1993). An introduction to theSport CommitmentModel. Journal of Sport andExercise Psychology,15, 1–15. https://doi.org/10.1123/jsep.15.1.1.

Schermelleh-Engel, K., Moosbrugger, H., & Müller,H. (2003). Evaluating the fit of structuralequation models: tests of significance anddescriptive goodness-of-fit measures. Methodsof Psychological Research Online,8(2), 23–74.

Schultheiss, O. C., & Pang, J. S. (2010). Measuringimplicit motives. In R.W. Robins, R. C. Fraley &R. F. Krueger (Eds.), Handbook of research meth-ods in personality psychology (pp.322–344).NewYork, London: Guilford.

Smith, G. T., Fischer, S., & Fister, S.M. (2003). Incre-mental validity principles in test construction.Psychological Assessment,15(4), 467–477.

Tavakol, M., & Dennick, R. (2011). Making senseof Cronbach’s alpha. International Journal ofMedical Education,2, 53–55.

Unierzyski, P. (2003). Level of achievementmotivationof young tennis players and their futureprogress. Journal of Sport Science andMedicine,2, 184–186.

Vaeyens,R.,Lenoir,M.,Williams,M.,&Philippaerts,R.M.(2008). Talent identification and developmentprogrammes in sport: currentmodels and futuredirections. Sports Medicine,38(9), 703–714.

Wenhold, F., Elbe, A.-M., & Beckmann, J. (2008).VKS: Fragebogen zur Erfassung der VolitionalerKomponenten im Sport [Questionnaire forrecording the volitional components insports]. http://www.bisp-sportpsychologie.de/SpoPsy/DE/Diagnostikportal/Motivation/Sportlerfrageboegen/vqs_2008/vks_einfuehrung.html?nn=3014646. Last access: June 24,2019

Zibung, M., Zuber, C., & Conzelmann, A. (2016). Themotor subsystem as a predictor of successin young football talents: a person-oriented

422 German Journal of Exercise and Sport Research 4 · 2019

study. PLOS ONE. https://doi.org/10.1371/journal.pone.0161049.

Zuber, C., & Conzelmann, A. (2014). The impact of theachievementmotive on athletic performance inadolescent football players. European Journal ofSport Science, 14, 475–483. https://doi.org/10.1080/17461391.2013.837513.

Zuber, C., Zibung, M., & Conzelmann, A. (2015).Motivational patterns as an instrument forpredicting success in promising young footballplayers. Journal of Sports Sciences, 33(2),160–168. https://doi.org/10.1080/02640414.2014.92882.

German Journal of Exercise and Sport Research 4 · 2019 423