This article was downloaded by: [Biblioteca. Universidad de Extremadura]On: 08 July 2014, At: 00:06Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Journal of Sports SciencesPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/rjsp20

Perceived coach-created and peer-createdmotivational climates and their associations with teamcohesion and athlete satisfaction: evidence from alongitudinal studyTomás García-Calvoa, Francisco Miguel Leoa, Inmaculada Gonzalez-Poncea, Pedro AntonioSánchez-Miguelb, Athanasios Mouratidisc & Nikos Ntoumanisd

a Faculty of Sport Sciences, University of Extremadura, Cáceres, Spainb Faculty of Teaching Training, University of Extremadura, Cáceres, Spainc Sport Sciences, University of KU Leuven, Leuven, Belgiumd School of Sport and Exercise Sciences, University of Birmingham, Birmingham, UKPublished online: 09 Jun 2014.

To cite this article: Tomás García-Calvo, Francisco Miguel Leo, Inmaculada Gonzalez-Ponce, Pedro Antonio Sánchez-Miguel,Athanasios Mouratidis & Nikos Ntoumanis (2014): Perceived coach-created and peer-created motivational climates and theirassociations with team cohesion and athlete satisfaction: evidence from a longitudinal study, Journal of Sports Sciences, DOI:10.1080/02640414.2014.918641

To link to this article: http://dx.doi.org/10.1080/02640414.2014.918641

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Perceived coach-created and peer-created motivational climates andtheir associations with team cohesion and athlete satisfaction: evidencefrom a longitudinal study

TOMÁS GARCÍA-CALVO1, FRANCISCO MIGUEL LEO1, INMACULADA GONZALEZ-PONCE1, PEDRO ANTONIO SÁNCHEZ-MIGUEL2, ATHANASIOS MOURATIDIS3 &NIKOS NTOUMANIS4

1Faculty of Sport Sciences, University of Extremadura, Cáceres, Spain, 2Faculty of Teaching Training, University ofExtremadura, Cáceres, Spain, 3Sport Sciences, University of KU Leuven, Leuven, Belgium and 4School of Sport and ExerciseSciences, University of Birmingham, Birmingham, UK

(Accepted 23 April 2014)

AbstractIn this longitudinal study, we examined the extent to which perceived coach- and peer-created motivational climates areassociated with athlete-group cohesion and satisfaction with participation among Spanish soccer players competing in theThird National Division. Multilevel modelling analyses showed that perceived coach-created task climate was positivelyrelated to perceived cohesion and players’ satisfaction with their participation within their team. Also, perceived peer-createdtask climate related positively to perceived cohesion. The results indicate the importance of considering peer-related aspectsof the motivational climate in addition to considering the coach-related aspects of the motivational climate when examiningmotivational group dynamics in sport.

Keywords: coach influence, peer influence, group cohesion, satisfaction, perceived motivational environment

An adaptive motivational climate within a team isessential for nurturing good relationships amongteam members and optimising group effectiveness(Beauchamp, 2007). Understanding how a team’smotivational climate is formulated requires consid-eration of the role of the coach and the team mem-bers. As the group dynamics literature has shown,both the leader and the members of a group (e.g.,the coach and the players of a sports team) create thesocial environment within which a group evolves andoperates across time (Levine & Moreland, 1990).

However, the examination of the joint role of thecoach- and peer-created motivational climate in sportshas been infrequent (for an exception, see Vazou,Ntoumanis, andDuda (2006)) and particularly in long-itudinal research designs (Ntoumanis, Taylor, &Thøgersen-Ntoumani, 2012). This is rather unfortu-nate because motivational climates are dynamic andcan evolve across time (Boyce, Gano-Overway, &Campbell, 2009; Reinboth & Duda, 2006). Knowing

how motivational climates fluctuate across time andhow such fluctuations predict variations in motivation-related outcomes and the relative importance of coach-and peer-created motivational climates may yieldimportant insights that have not been provided by pastcross-sectional studies.

Moreover, the extent to which motivation-relatedoutcomes are best predicted by between-persondifferences in the perceptions of motivational climatesor within-person changes in such perceptions is largelyunknown. For instance, finding that within-personvariability in motivational climate better predicts cer-tain motivational outcomes than do between-personindividual differences in such perceptions of the moti-vational environment may aid in the design of moreeffective interventions by focusing on how and whyperceptions of such an environment fluctuate overtime within a person. Further, identifying that peer-created motivational climates are more strongly andconsistently associated with outcomes across time thanare coach-createdmotivational climates would attest to

Correspondence. Tomás García-Calvo, Faculty of Sport Sciences, University of Extremadura, Cáceres, Spain. E-mail: tgarciac@unex.esPresent address for Francisco Miguel Leo is Faculty of Physical Activity and Sport Sciences, Catholic University of San Antonio, Murcia, Spain.Present address for Athanasios Mouratidis is Department of Psychology, Hacettepe University, Ankara, Turkey.Present address for Nikos Ntoumanis is School of Psychology & Speech Pathology, Curtin University, Perth, Western Australia.

Journal of Sports Sciences, 2014http://dx.doi.org/10.1080/02640414.2014.918641

© 2014 Taylor & Francis

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

the importance of cultivating a sustained positive peer-created motivational environment in sport.

Unlike previous studies on motivationalclimate which sampled amateur athletes, we recruitedsemi-professional athletes. In doing so, we aimed totest the role of the motivational climate in contextswhere competition is salient. Specifically, in our long-itudinal (season-long) study, we recruited a sample ofsemi-professional Spanish soccer players and investi-gated the extent to which different facets of coach- andpeer-created motivational climates are linked (addi-tively and interactively) to the players’ perceptions ofteam cohesiveness and their satisfaction with their par-ticipation within the team. We focused on team cohe-sion because it helps improve group performance(Carron, Colman, Wheeler, & Stevens, 2002), andonly a few longitudinal studies have investigated howmotivational climate is linked over time to groupdynamics in high-performance sport settings (Heuzé,Sarrazin, Masiero, Raimbault, & Thomas, 2006).Similarly, players’ satisfaction within their team hasreceived scant attention. This is an important oversightbecause players’ game-related participation satisfac-tion are closely connected with a team’s optimal func-tioning (Balaguer, Duda, Atienza, & Mayo, 2002;Beauchamp, Bray, Eys, & Carron, 2002). Weemployed a longitudinal research design because weconsidered team cohesiveness and player satisfactionwith participation to change across time (Paskevich,Estabrooks, Brawley, & Carron, 2001). In fact, pre-vious research has found that group cohesion changesduring the sport season (Heuzé et al., 2006; Leo,Sánchez-Miguel, Sánchez-Oliva, Amado, & García-Calvo, 2012). Therefore, besides testing the trajec-tories of individual variables across time, we were alsointerested in investigating the longitudinal variability ofthe associations between motivational climate, groupcohesion, and athlete satisfaction. In this sense,depending on the characteristics of the perceivedcoach- and peer-created motivational climate, a trendin the perceptions of group processes can be producedgiven the relationship between these variables (Heuzéet al., 2006; Leo, Sánchez-Miguel, Sánchez-Oliva,Amado, & García-Calvo, 2014).

Coach- and peer-created motivational climate

Coaches and peers are considered as two of the mainsocial agents within a sport team (Duda & Balaguer,2007). In fact, coaches and players’ behaviours havebeen shown to predict team cohesion (Turman,2003; Vincer & Loughead, 2010) and satisfactionwith sport participation (Loughead & Carron, 2004).

We use the classic achievement goal theory (i.e.,dichotomous approach) to define the motivational cli-mate of a team (Ames, 1992). This approach distin-guishes between two types of motivational climate, a

task-involving climate, where effort, mastery strivings,and individual improvement are encouraged, and anego-involving climate, where social comparisons, nor-mative ability, and interpersonal standards are pro-moted (Duda & Balaguer, 2007). Whereas task-involving motivational climate direct individuals’achievement strivings towards the development of per-sonal competence and challenging strivings (Adie,Duda, & Ntoumanis, 2008), an ego-involving climatedirects people’s strivings towards the demonstration ofnormative competence – a situation that makes themvulnerable to uncontrollable situations (see Dweck &Leggett, 1988).

Research in the sport context has predominantlyexamined the coach as the main shaping force oftask- and ego-involving goal climates (e.g., Conroy,Kaye, & Coatsworth, 2006; Olympiou, Jowett, &Duda, 2008). In congruence with achievement goaltheory, this research has shown that athletes who per-ceived that their coach created a task-involving climatereported more positive outcomes such as prosocialbehaviour (Boardley & Kavussanu, 2009; Boixadós,Cruz, Torregrosa, & Valiente, 2004), intrinsic motiva-tion (Newton, Duda, & Yin, 2000), and satisfactionwith their level of improvement (Balaguer et al., 2002).

However, in an effort to move the focus of attentionbeyond the coach, Vazou et al. (2005, 2006) and others(e.g., Harwood & Swain, 2001) have proposed thatpeers can also play an important role in formulatingthe motivational climate of a team. Indeed, consideringpeer-created motivational climate alongside the coach-created one has revealed that peer-created motivationalclimate uniquely contributes to the prediction of impor-tant variables such as sportspersonship, burnout, vitality(Ntoumanis et al., 2012), physical self-worth, anxiety,and effort (Vazou et al., 2006). Moreover, it has beenshown that perceived peer-created ego- and task-invol-ving motivational climates are only moderately relatedto the respective coach-created climate dimensions,suggesting that coach- and peer-created task and egoclimates are somewhat independent constructs. Teamcohesion is an important outcome variable that has notbeen examined so far by the few studies that have mea-sured both coach- and peer-created motivational cli-mates. We focused on cohesion in this paper becauseit contributes to a team’s optimal functioning andgrowth (Paskevich et al., 2001).

Team cohesion

Team cohesion has been defined as “a dynamicprocess that is reflected in the tendency for a groupto stick together and remain united in the pursuit ofits instrumental objectives and/or for the satisfactionof member affective needs” (Carron, Brawley, &Widmeyer, 1998, p. 213). This definition alignswith Carron’s model (Carron, Widmeyer, &

2 T. García-Calvo et al.

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

Brawley, 1985) which proposes that team membershold (a) collective beliefs about the group as a unit interms of its closeness, resemblance, and affinity(group integration; GI), and (b) individual beliefsregarding the degree to which the group attractsthem, thereby satisfying their needs and personalgoals (individual attraction to the group; ATG). Eachof these classes of beliefs is further divided into twocategories, depending on whether they revolvearound task-related (T) or social (S) issues (Carron& Brawley, 2000; Carron & Eys, 2012). Thus, fouraspects of team cohesion have been identified: (a)Group Integration-Task (GI-T) and (b) GroupIntegration-Social (GI-S) which reflect members’beliefs about the degree to which the group is unitedto reach its objectives and to have its members socia-lising, respectively; and (c) Individual Attraction to theGroup-Task (ATG-T), and (d) Individual Attractionto the Group-Social (ATG-S) which reflect members’judgments about the degree to which the group isattractive for its objectives and social relationships,respectively.

Cohesion is considered as one of the key attributesof a team (Paskevich et al., 2001). Not surprisingly,it has been associated with a diverse array of positiveoutcomes such as lower social loafing, more beha-vioural control, stronger efficacy beliefs (Paskevichet al., 2001), and higher performance (Carron et al.,2002; Mullen & Copper, 1994). More relevant to thepresent study, group cohesion has also been studiedwith regard to the motivational climate of a team.Studies have shown cohesion to relate positively tocoach-created task climate and negatively to coach-created ego climate (Balaguer, Castillo, Moreno,Garrigues, & Soriano, 2004; Eys et al., 2013;Horn, Byrd, Martin, & Young, 2012; Leo et al.,2014; Naylor, 1996). Such relations make concep-tual sense as a task-involving motivational climate (incontrast to an ego-involving climate) promotes coop-eration, emphasises the unique contribution of eachathlete to the team, fosters feelings of relatedness,and downplays interpersonal rivalry among athletes(Ntoumanis & Biddle, 1999). In one of the rarelongitudinal investigations in this area, Heuzé et al.(2006) showed that perceived coach-created task cli-mate positively predicted subsequent GI-T, whilecoach-created ego climate negatively predicted sub-sequent social cohesion (i.e., ATG-S and GI-S)among female players in professional basketball andhandball teams.

However, none of the studies has examined theassociations between peer motivational climate andteam cohesion. Further, the extent to which the linksbetween perceived motivational climate (coach- andpeer-created) and cohesion fluctuate across timewithin individuals is unknown. This is an importantissue that deserves further investigation because

perceived cohesion is by definition dynamic andundergoes changes across time (Paskevich et al.,2001). Moreover, given the relationship betweenthe two variables (Heuzé et al., 2006; Leo et al.,2014), the perception of a task-involving climatecan create a positive trend in group processes, buta perception of an ego-involving climate can create anegative trend. In this sense, the literature on groupdynamics indicates that the leader and members of agroup may exert varying degrees of influence ongroup affairs as the group develops (Carron & Eys,2012). Thus, we aimed to examine whether theassociations of perceived coach- and peer-created task and ego climate to cohesion vary acrosstime.

Satisfaction with participation

Besides group cohesion, we also examined howcoach- and peer-created motivational climatesrelate to satisfaction with participation. We optedfor satisfaction with participation because it isimportant for a team to function optimally. Whenplayers feel satisfied with their contribution to andfunctioning in their group, they are more likely todevelop friendship networks and cooperate withtheir teammates (Olmedilla et al., 2011; Weiss &Smith, 2002). The motivational climate within ateam can influence an athlete’s satisfaction withthe extent of participation within that team.Indeed, past research has shown a positive relation-ship between certain behaviours by coaches andpeers (e.g., support and positive feedback, emphasison individual effort, encouragement of cooperation)and sport satisfaction among athletes (Boixadóset al., 2004; Loughead & Carron, 2004;Loughead, Hardy, & Eys, 2006). In a task-involvingmotivational climate, where individual improve-ment and personal effort are encouraged and mis-takes are considered as part of the learning process,athletes are more likely to feel satisfied with theirsport participation regardless of their ability level.In contrast, in an ego-involving motivationalclimate, where interpersonal comparison, thedemonstration of the normative ability, and compe-tition among teammates are emphasised, athletesare likely to feel unsatisfied with their team partici-pation when they do not do as well as their team-mates (Balaguer et al., 2002; Boixadós et al., 2004;García-Calvo, Cervelló, Jiménez, Iglesias, &Santos-Rosa, 2005). In an ego-involving climateonly the most able athletes are expected to getample opportunities, whereas in a task-involvingclimate all players are expected to get similarchances to develop their potential and have a some-what unique role within a team (Boixadós et al.,2004; Weiss & Smith, 2002).

Motivational climate and cohesion 3

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

Present study

Our study extends past literature by employing alongitudinal (season-long) design to examine thedegree to which coach-created task and ego climatesare related to team cohesiveness and players’ satis-faction with participation within their team. Byexamining these associations across a sport season,we tried to shed light on the dynamics of theseassociations because we presumed that both themotivational climate and the cohesion of a team aswell as players’ satisfaction with participation fluctu-ate across time. In addition to examining theselongitudinal associations, we also examined twoimportant and closely related facets of a team’s moti-vational climate that have been relatively understu-died as a pair: the coach and peer climate.Furthermore, we tried to build on the prior scarceresearch that studied both coach and peer motiva-tional climate by separating within-person changesfrom between-person differences. To do this, wetested the main and interactive effects of coach-and peer-created climate facets on important vari-ables linked to team dynamics in semi-professionalathletes across a sport season, namely players’perception of group cohesion and satisfaction withtheir participation within the team.

First, we made no specific hypotheses regardingthe trajectories of the dependent variables becausethere are several factors that might lead to anincrease or decrease in a team’s cohesion andplayers’ satisfaction. Yet, we expected an upwardtrend for players perceiving their team to be moretask oriented (by either the peers, the coach, or both;Hypothesis 1a) and a downward pattern for playersperceiving their team to be more ego oriented (byeither the peers, the coach, or both; Hypothesis 1b).Second, we hypothesised that a perceived coach-cre-ated task and ego climate would, respectively, posi-tively and negatively covary across the threemeasurement waves with group cohesion (task andsocial) and satisfaction with participation; weexpected these associations to remain significantacross time (Hypothesis 2). Further, we predictedthat higher mean levels of perceived coach-createdtask climate would explain higher mean levels ofgroup cohesion and satisfaction with their participa-tion (Hypothesis 3a); we expected the oppositepattern for mean level of perceived coach-createdego climate (Hypothesis 3b). Additionally, we antici-pated that the task and ego facets of peer-createdclimate would account for the unique incrementalvariance of the dependent variables at both the intra-personal and interpersonal level (Hypothesis 4).Specifically, we expected that perceived peer-createdtask and ego climates would show patterns of rela-tionships to the dependent variables at both the

intrapersonal and between-athlete levels that wouldbe similar to the pattern that coach-created task andego climate would exhibit. Moreover, we exploredthe extent to which a likely fit or misfit betweencorresponding task and ego facets of coach- andpeer-created climates, as originally discussed byVazou et al. (2006), would moderate the hypothe-sised relationships. In particular, we tested, in anexploratory fashion due to the absence of priorempirical evidence, whether significant interactionswould emerge between (a) coach- and peer-createdtask climates, (b) coach- and peer-created egoclimates, (c) coach-created task and ego climates,or (d) peer-created task and ego climates. Suchhypotheses have not been previously tested, but wetentatively expected that the most positive outcomeswould emerge under high-task conditions for bothcoach and peer climate (Hypothesis 5).

Method

Participants

The participants were male semi-professional soccerplayers from 20 teams who participated in the XIVgroup of the Third Division of the Spanish soccerLeague. At the beginning of the season (Time 0), werecruited a total of 377 players, ranging in age from16 to 39 years with a mean age of 24.51 years(s = 3.73). At the middle of the season (Time 1), atotal of 339 players from the original sample wererecruited, ranging in age from 16 to 38 years with amean age of 24.41 years (s = 4.24). At the end of theseason (Time 2), there were a total of 303 playersfrom the original sample, ranging in age from 16 to39 years with a mean age of 24.58 years (s = 4.26).We ran a MANOVA to determine whether the par-ticipants who were absent from one or two measure-ments differed in some ways from those who filled inall the measurements in all three waves of assess-ment. These analyses suggested that the athletes ofthe two groups did not differ on the variables mea-sured (Time 0: Wilks’ lambda = .99, F4,361 = 1.79,P = .36, ηp

2 = .01; Time 1: Wilks’ lambda = .99,F4,328 = .32, P = .86, ηp

2 = .00; Time 2: Wilks’lambda = .98, F4,287 = 1.58, P = .18, ηp

2 = .02).From an original sample of 389 questionnaires

collected, 12 (3.08%) were deleted at wave 1 dueto invalid completion of the questionnaires. At themiddle and end of the season we further deleted 5(1.45%) and 6 questionnaires (1.94%), respectively,for the same reason.

Measures

Perceived coach motivational climate. The Spanishversion of Perceived Coach Motivational Climate in

4 T. García-Calvo et al.

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

Sport Questionnaire-2 (PMCSQ-2: Newton et al.,2000), developed by Balaguer, Guivernau, Duda,and Crespo (1997), was used. This questionnaireconsists of 33 items that measure six dimensions oftask- and ego-involving coach climate. In this paper,we were interested in the two higher-order dimen-sions and not in the lower-order dimensions. A con-firmatory factor analysis (CFA) of the data taken atthe beginning of the season offered support for ahigher-order structure with two factors(χ2 = 178.78; P < .01; df = 53; χ2/df = 3.37;CFI = .91; IFI = .93; RMSEA = .07;SRMR = .07). The CFA of the data from the middleand the end of the season showed similar fit indicesto those obtained at the beginning of the season.Players responded to all items on a 5-point scaleranging from strongly disagree (1) to strongly agree(5). The stem was “On this team…”. An exampleof task-involving climate item is “The coach tells usthat trying our best is the most important thing”.Ego-involving climate included items such as “Thecoach pays the most attention to the best players”.

Perceived peer motivational climate. To measure ath-letes’ perceptions of the peer-created motivationalclimate on their team, the Spanish version of the21-item Peer Motivational Climate in Youth SportQuestionnaire (PeerMCYSQ: Ntoumanis & Vazou,2005), adapted to the Spanish language by Morenoet al. (2011) was used. The PeerMCYSQ consists ofa task-involving and an ego-involving higher-orderdimension, each of which comprise of a number oflower-order factors. Again, in this paper we wereinterested in the higher-order factors. A CFA withour data offered support to a higher-order structurewith two factors (χ2 = 179.58; P < .01; df = 53; χ2/df = 3.38; CFI = .92; IFI = .92; RMSEA = .08;SRMR = .07). Participants responded to the stem“On this team, most athletes…”, using a 5-pointscale ranging from strongly disagree (1) to stronglyagree (5). An example item for the task-involvingclimate factor is “Encourage their teammates toimprove their weakness”. An example of an itemconstituting the ego-involving climate factor is “Tryto do better than their teammates”.

Group cohesion. The Spanish version of group envir-onment questionnaire (GEQ) (Carron et al., 1985),adapted by Iturbide, Elosua, and Yanes (2010), wasused to assess team cohesion. This inventory of 18items comprises four factors: GI-T (e.g., “Teammembers are united in their efforts to reach theirperformance goals in training sessions andmatches”), GI-S (e.g., “Team members would liketo spend time together in situations other than train-ing and games”), ATG-T (e.g., “On this team, I cando my best”), and ATG-S (e.g., “The team is one of

the most important social groups I belong to”). ACFA with our data showed acceptable model fit(χ2 = 128.59; P < .01; df = 48; χ2/df = 2.67;CFI = .91; IFI = .91; RMSEA = .06;SRMR = .05). Responses were rated on a 5-pointscale ranging strongly disagree (1) to strongly agree (5).

Satisfaction with participation. To assess satisfactionwith participation, we wrote three items that askedplayers whether they were satisfied with their parti-cipation within the team and with their playing time(i.e. “Are you satisfied with your participation withinthe team?”; “Are you satisfied with your playing timein matches?”; and “Are you satisfied with your con-tribution to the team?”). Responses were rated on a5-point scale ranging not at all (1) to very much (5).

Procedure

We used a longitudinal correlational design. Wecarried out three assessments at three time points:three weeks within the beginning of the sport season,at the middle, and at the end of the season, separatedby a 20–22 week interval between each measurementwave.

The study received ethical approval from theUniversity of Extremadura. All participants weretreated according to American PsychologicalAssociation ethics guidelines regarding consent, con-fidentiality, and anonymity of responses.Questionnaires were matched over time using a cod-ing system to protect confidentiality. Participantscompleted the questionnaires in the changing roombefore a training session. Participants completed thequestionnaires individually within 15–20 min, in theabsence of their coach, supervised by the researchassistants and under non-distracting conditions.

Data analysis

Multilevel regression analysis, employing Mplus(Muthén & Muthén, 1998–2012), was used toexamine changes in all variables over the three timepoints. This type of analysis is particularly usefulwhen there are missing observations since it doesnot assume equal number of measurement occasionsfor all individuals. Two level of analysis were speci-fied. Level 1 encompassed the repeated observationsof all variables. These observations were nestedwithin athletes, therefore the latter constituted level2 in the analysis. Although players were nestedwithin teams, we did not include a third level (i.e.,team) in our analyses because of the small number ofteams (see Maas & Hox, 2005). The analysis hadtwo parts. The first part examined whether therewere significant between-person variations in themeans (intercepts) and rates of change (growth

Motivational climate and cohesion 5

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

trajectories) of all variables under investigation. Thesecond part included a number of multilevel regres-sions with the repeated scores of cohesion and satis-faction with participation being the dependentvariables. The aim of these regressions was to ascer-tain whether within-person changes and between-person differences in perceptions of coach- andpeer-created task and ego climate were predictive ofwithin-person changes in each dependent variable.

Results

Descriptive statistics and Cronbach’s alpha coefficients

Means, standard deviations, and Cronbach’s alphacoefficients for each variable at each time point arepresented in Table I. Most scales, but GI-S andATG-S, at the first measurement wave demonstratedacceptable internal consistency. In light of this find-ing, GI-S was removed from further analyses. Incontrast, we kept ATG-S because its internal consis-tency coefficients at the remaining measurementwaves were within acceptable limits. In general,participants reported scores above the mid-point ofthe scale for coach- and peer-created task climate,peer-created ego climate, cohesion, and satisfactionwith participation. Participants also reported scoresfor coach-created ego climate which were close tothe mid-point of the scale.

Main analyses

We set up a multivariate multilevel model to test ourresearch hypotheses. Specifically, we estimated towhat extent ATG-S, GI-T, ATG-T, and satisfactionwith participation could be modelled as a function of(a) time effects (linear changes; wave 1 was coded as0), and coach- and peer-created task and ego climateat level 1, that is, the intrapersonal level, and (b)person-average scores of perceived coach- andpeer-created task and ego motivational climate atlevel 2, that is, the interpersonal level. The climatescores at level 1 were group mean centred and atlevel 2 were grand mean centred. At both levels wealso tested for interactions between coach-createdtask and ego climate, peer-created task and ego cli-mate, coach-created task and peer-created taskclimate, and coach- and peer-created ego climate.To test the interactions we followed the recommen-dations by Cohen, Cohen, West, and Aiken (2003),and for reasons of parsimony we retained in the finalmodel only the statistically significant interactions.Also, for the interest of model parsimony and com-putational efficiency, we fixed all the slopes which wefound to be not random (i.e., not significantlyvarying from person to person).

To examine the relative predictive effects of peer-created motivational climate over and above thecoach-created one, we set up the model in a stepwisefashion thereby entering in Step 1 time along withtask and ego coach-created motivational variables,followed in Step 2 by task and ego peer-createdmotivational variables and the statistically significantinteractions. We only display the final model inTable II, but we briefly present the results from theunconditional (i.e., without predictors) model andthose from Step 1 in which time as a predictor wasalso added.

Step 1: Coach-created motivational climate variables

Inspection of the unconditional model revealedconsiderable within-person variance for all thedependent variables (see Table II). When time (lin-ear) change (β10) and perceived coach-created task(β20) and ego (β30) climate were entered as predic-tors at the intrapersonal level, a significant declinewas found for GI-T (β10 = −.19, s�x = .03, P < .01)and ATG-T (β10 = −.10, s�x = .03, P < .01), whichsuggests that team cohesiveness on the task-relateddimension decreased throughout the sport season.

Nevertheless, the decrease of the ATG-T in thefinal model was moderated by the mean levels ofathlete perceptions of coach- and peer-created moti-vational climate (presented in the full model – seeTable II). Specifically, a significant cross-level inter-action was found between time and the mean levelsof peer-created task climate for ATG-T (β12 = .10,s�x = .05, P = .041).

Examination of the simple slopes indicated thatATG-T decreased across time for players who per-ceived that the peer-created task climate throughoutthe whole season was low (β12 [−1 s in mean peer task

climate] = −.12, s�x = .04, z = −3.10, P = .002) but notfor those who perceived that it was high (β12 [+1 s in mean

peer task climate] = −.02, s�x = .04, z = −.44, P = .66, ns).Regarding changes across time, the final model (see

Table II) revealed two statistically significant cross-level interactions for satisfaction with participation.The first concerned between time and mean levelsof perceived coach-created task climate (β11 = .08,s�x = .03, P = .02). Although slope was non-significant,the test of simple slopes indicated that satisfactionwith participation tended to decrease across time forathletes who perceived the coach-created task climateto be low (β11 [−1 s in mean coach task climate] = −.06,s�x = .04, z = −1.55, P = .12), while it tended toincrease for athletes who perceived the coach-createdtask climate to be high (β12 [+1 s in mean coach task

climate] = .04, s�x = .04, z = 1.01, P = .31).Taken together, these findings provided partial

support to Hypothesis 1a because they showed thatsome of the negative trajectories occurred only in the

6 T. García-Calvo et al.

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

Tab

leI.Bivariate

correlations

(two-taile

d)an

dinternal

relia

bilities(onthediagon

al)of

thestud

yvariab

les.

Variables

12

34

56

78

910

1112

1314

1516

1718

1920

2122

2324

2526

27M

s

Tim

e1

1.Coa

ch-created

ego

clim

ate

.85

2.31

.92

2.Coa

ch-created

task

clim

ate

−.39*

*.84

4.20

.65

3.Peer-createdego

clim

ate

.34*

*.05

.69

3.44

.79

4.Peer-createdtask

clim

ate

−.20*

*.61*

*.28*

*.88

4.25

.57

5.Group

integration-

Soc

ial

.01

.20*

*.12*

.31*

*.51

3.88

.67

6.Attractionto

the

grou

p-Soc

ial

−.02

.22*

*.14*

*.29*

*.41*

*.46

3.75

.75

7.Group

integration-

Task

−.06

.43*

*.07

.51*

*.25*

*.26*

*.73

4.04

.76

8.Attractionto

the

grou

p-Task

−.19*

*.49*

*.04

.38*

*.27*

*.34*

*.46*

*.68

3.94

.77

9.Satisfactionwith

Participa

tion

−.20*

*.19*

*−.06

.05

.04

.07

.05

.31*

*.77

3.09

.60

Tim

e2

10.Coa

ch-created

ego

clim

ate

.47*

*−.27*

*.21*

*−.09

−.07

.02

−.08

−.24*

*−.20*

*.85

2.49

.89

11.Coa

ch-created

task

clim

ate

−.24*

*.42*

*−.04

.28*

*.13*

.19*

*.27*

*.37*

*.07

−.38*

*.83

3.95

.70

12.Peer-createdego

clim

ate

.18*

*.06

.38*

*.13

.03

.14*

.10

.08

−.10

.33*

*.10

.67

3.29

.70

13.Peer-createdtask

clim

ate

−.12

.32*

*.08

.43*

*.20*

*.21*

*.15*

*.16*

*−.04

−.12*

.61*

*.29*

*.91

4.03

.69

14.Group

integration-

Soc

ial

−.06

.32*

*.10

.33*

*.36*

*.48*

*.12

.21*

*−.02

−.06

.33*

*.18*

*.42*

*.58

3.59

.85

15.Attractionto

the

grou

p-Soc

ial

−.04

.26*

*.11

.33*

*.36*

*.12*

.19*

*.11

−.09

−.08

.37*

*.17*

*.44*

*.49*

*.76

3.92

.81

16.Group

integration-

Task

−.01

.39*

*.15*

.31*

*.14*

.17*

*.20*

*.38*

*.13*

−.17*

*.52*

*.12*

.43*

*.35*

*.28*

*.71

3.60

.89

17.Attractionto

the

grou

p-Task

.04

.30*

*.06

.31*

*.14*

.11

.39*

*.09

−.14*

−.07

.49*

*.18*

*.58*

*.37*

*.40*

*.51*

*.77

3.75

.83

18.Satisfactionwith

Participa

tion

−.12

.12*

.04

.10

.12*

.01

.11

.23*

*.47*

*−.17*

*.22*

*−.06

.00

.07

.03

.26*

*−.04

.78

3.01

.69

Tim

e3

19.Coa

ch-created

ego

clim

ate

.45*

*−.19*

*.23*

*−.07

−.01

.02

−.10

−.16*

−.16*

*.55*

*−.40*

*.13*

−.15*

-.03

−.09

−.16*

−.06

−.23*

*.83

2.59

.84

20.Coa

ch-created

task

clim

ate

−.21*

*.36*

*.01

.20*

*.14*

.07

.28*

*.27*

*.04

−.37*

*.60*

*.00

.41*

*.17*

*.23*

*.43*

*.40*

*.20*

*−.43*

*.86

3.64

.81

(contin

ued)

Motivational climate and cohesion 7

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

absence of a perceived task climate (either coach-created or peer-created).

In support of Hypothesis 2, a perceived coach-created task climate was found to positively relateto all the dependent variables, namely ATG-S(β30 = .22, s�x = .07, P < .01), GI-T (β30 = .32,s�x = .07, P < .01), ATG-T (β30 = .38, s�x = .06,P < .01), and satisfaction with participation(β30 = .19, s�x = .04, P < .01). In further support ofHypothesis 2, a perceived coach-created ego climatewas negatively associated with GI-T (β20 = −.11,s�x = .05, P = .023). Including time and a perceivedcoach-created task and ego climate at the intraperso-nal level explained 2.4% of the variance for ATG-S,41.0% of the variance for GI-T, 25.6% of the var-iance for ATG-T, and 24.7% of the variance forsatisfaction with participation.

When mean-level scores of perceived coach-createdtask (β01) and ego (β20) climate were entered asbetween-player predictors, higher perceptions of aver-age coach-created task climate were, in support ofHypothesis 3a, positively associated with ATG-S(β01 = .45, s�x = .05, P < .01), GI-T (β01 = .70,s�x = .05, P < .01), ATG-T (β01 = .70, s�x = .04,P < .01), but not with satisfaction with participation(β01 = .07, s�x = .05, P = .20). Compared to averageperceived coach-created task climate, coach-createdego climate was, in support of Hypothesis 3b, lessstrongly associated with some of the dependent vari-ables, such as ATG-S (β02 = .13, s�x = .04, P < .01),GI-T (β02 = .15, s�x = .04, P < .01), ATG-T(β02 = .10, s�x = .03, P < .01) while it was negativelyrelate to satisfaction with participation (β02 = −.13,s�x = .04, P < .01). Including the average scores ofperceived coach-created task and ego climate asbetween-player predictors explained 32.0% of var-iance for ATG-S, 67.4% for GI-T, 77.3% for ATG-T, and 10.4% for participation satisfaction.

Step 2: Coach- and peer-created motivational climatevariables

Entering perceived peer-created task and ego motiva-tional climate as predictors at both the intrapersonaland interpersonal levels revealed several noteworthyfindings (see Table II). Specifically, at the intraperso-nal level, perceived peer-created task climate waspositively associated with the three cohesion-relatedvariables; interestingly, the previously significant asso-ciation between perceived coach-created task climateand ATG-S became non-significant once perceivedpeer-created task climate was considered (seeTable II). Taken together, these findings providedsome support to Hypothesis 4.

In addition, two significant interactions werefound. First, perceived peer-created task climateinteracted with the respective ego one in the

Tab

leI.(C

ontin

ued).

Variables

12

34

56

78

910

1112

1314

1516

1718

1920

2122

2324

2526

27M

s

21.Peer-createdego

clim

ate

.15*

.10

.32*

*.11

.11

.08

.11

.13

−.07

.16*

*.11

.35*

*.17*

*.15*

.17*

*.09

.10

−.04

.26*

*.17*

*.65

3.32

.64

22.Peer-createdtask

clim

ate

−.13

.38*

*.07

.37*

*.25*

*.12

.28*

*.29*

*.01

−.18*

*.44*

*.04

.57*

*.23*

*.29*

*.39*

*.48*

*.07

−.08

.59*

*.35*

*.91

3.82

.68

23.Group

integration-

Soc

ial

−.14*

.26*

*.05

.27*

*.28*

*.39*

*.06

.17*

−.04

−.16*

*.32*

*.07

.33*

*.54*

*.34*

*.35*

*.29*

*.06

−.14*

.38*

*.20*

*.45*

*.61

3.66

.77

24.Attractionto

the

grou

p-Soc

ial

−.12

.29*

*.02

.26*

*.37*

*.27*

*.14*

.24*

*−.03

−.10

.18*

*.00

.24*

*.45*

*.47*

*.32*

*.32*

*.07

.00

.28*

*.20*

*.42*

*.47*

*.80

3.70

.85

25.Group

integration-

Task

−.05

.23*

*−.07

.17*

.13

.03

.18*

*.29*

*.11

−.10

.41*

*−.01

.26*

*.20*

*.17*

*.51*

*.31*

*.17*

*−.18*

*.52*

*.18*

*.48*

*.40*

*.26*

*.67

3.44

.80

26.Attractionto

the

grou

p-Task

−.09

.26*

*.00

.27*

*.11

.02

.30*

*.15*

−.01

−.18*

*.36*

*−.06

.30*

*.15*

.20*

*.36*

*.48*

*.04

−.22*

*.53*

*.15*

.48*

*.40*

*.42*

*.45*

*.81

3.49

.86

27.Satisfactionwith

Participa

tion

−.15*

.09

.00

.09

.16*

.11

.04

.21*

*.39*

*−.28*

*.26*

*−.09

.07

.10

.10

.24*

*.04

.57*

*−.30*

*.28*

*−.07

.04

.06

.15*

*.26*

*.16

.81

3.00

.75

Note:

*P<.05.

**P<.01.

8 T. García-Calvo et al.

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

prediction of GI-T (β60 = −.17, s�x = .07, P < .05).Probing this interaction – see Figure 1 – followingthe recommendations made by Bauer and Curran(Bauer & Curran, 2005; see also Preacher, Curran,& Bauer, 2006), revealed that the relation betweenperceived peer-created task climate and GI-T wasweaker among athletes who perceived high (β60 [+1

s in peer ego climate] = .21, s�x = .10, z = 2.18, P = .029)as compared to low peer-created ego climate (β60 [−1

s in peer ego climate] = .45, s�x = .08, z = 5.74, P < .01).The second significant interaction was that

between perceived coach- and peer-created task

climate (β50 = .16, s�x = .06, P < .01) in the predic-tion of ATG-T (see Figure 2). This interaction pro-vided partial support to Hypothesis 5 and indicatedthat the relation between coach-created task climateand ATG-T was positive when peer-created taskclimate was high (β50 [+1 s in peer task climate] = .35,s�x = .09, z = 4.02, P < .01) but statistically non-significant when peer-created task climate was low(β50 [−1 s in peer task climate] = .12, s�x = .07, z = 1.67,P = .10). Collectively, the inclusion of perceivedpeer-created task and ego climate at the intraperso-nal level along with the statistically significant

Table II. Regression coefficients and standard errors of the multilevel models exploring perceptions of the motivational climate as predictorsof the outcome variables.

ATG-Social GI-Task ATG-TaskSatisfaction withparticipation

Fixed effectsIntercept 3.71 (.04) 3.86 (.03) 3.81 (.03) 3.03 (.03)

Intrapersonal predictorsTime (linear changes) .08* (.03) .16** (.03) .07* (.03) .01 (.02)Coach-created variablesEgo climate .03 (.06) .13** (.05) .05 (.05) .04 (.03)Task climate .08 (.08) .16* (.08) .24** (.07) .22** (.05)Peer-created variablesEgo climate .02 (.05) .03 (.06) .01 (.06) .05 (.04)Task climate .30** (.08) .33** (.07) .34** (.08) .05 (.05)InteractionsPeer-task X peer-ego .17* (.07)Peer-task X coach-task .16** (.06)Interpersonal predictorsCoach-created variablesMean ego climate .07 (.05) .12** (.04) .07* (.04) .12** (.04)Mean task climate .18* (.07) .49** (.07) .49** (.06) .19** (.07)Peer-created variablesMean ego climate .05 (.05) .00 (.05) .02 (.04) .05 (.05)Mean task climate .40** (.08) .34** (.07) .24** (.07) .21** (.06)InteractionsPeer-task X peer-egoPeer-task X coach-task .16** (.06)Cross-level interactionsTime X mean coach-task .08* (.03)Time X mean peer-task .10* (.05)

Variables Intercorrelations

ATG-Social – .01 .02 .01GI-Task .02 – .06** .03ATG-Task .09** .06* – .01Satisfaction new .02 .03* .03* –

Random effects Variance components

Intercept .11** (.02) .08** (.03) .05* (.02) .17** (.02)Time – .03* (.01) .03* (.01) .04** (.01)Coach-created task – .21** (.08) − –

Peer-created task – – .27** (.10) –

Residual variance .44 .31 .32 .18

Notes: *P < .05.**P < .01.Intercorrelations at the lower and upper diagonal refer, respectively, at the within- and between-person level. The significant negative effectsof peer-created task climate on the satisfaction with participation variable at the intrapersonal level are due to suppression effects.

Motivational climate and cohesion 9

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

interactions explained 3.1% of variance for ATG-S,9.5% for GI-T, and 17.9% for ATG-T.

At the between-person level, person-mean scoresof perceived peer-created task climate positively pre-dicted mean levels of cohesion variables over andabove perceived coach-created task climate (seeTable II). Unexpectedly, person-mean scores ofpeer-created task climate negatively predicted meanlevels of satisfaction with participation. This associa-tion however was statistical artefacts (due to suppres-sion effects); when we removed coach-createdclimate from the equation, the associations betweenpeer-created task climate and satisfaction with parti-cipation variables was positive but non-significant. Inaddition, we found two significant interaction effectsinvolving peer-created task climate but these werenot further interpreted due to the suppression effectsinvolving this variable.

Discussion

In this study, we examined the extent to which per-ceptions of coach- and peer-created task and egomotivational climates among semi-professional soc-cer players can explain intrapersonal variation and

interpersonal differences in team cohesion andplayers’ satisfaction with their participation withintheir team over the course of a sport season. To thebest of our knowledge, this is the first longitudinalstudy that looks at both within-person change andbetween-person differences in motivational climatesand underscores the unique role of peer-createdmotivational climates in predicting important teamdynamic aspects in competitive level athletes (i.e.,semi-professional soccer players) over and abovecoach-created climates.

Overall, there was a significant decrease in themean levels of the variables over time (Heuzéet al., 2006; Leo et al., 2012). Only ATG-S exhib-ited a short increase in the second wave, but itdecreased by the end of season. However, whenthe interactions between time and the mean levelsof coach- and peer-created task climates were ana-lysed, it was observed that a decrease over time forATG-T and satisfaction with participation occurredonly in players who perceived low levels of taskclimate. Therefore, the downward trajectories ofcertain variables were most likely due to the lowlevels of task climate. These findings prompted usto examine how motivational climate predicts fluc-tuations of cohesion and satisfaction over time. Wefound that task-involving motivational climate andego-involving motivational climate, both coach-and peer-created, positively and negativelycovaried, respectively, with cohesion and satisfac-tion over time. Therefore, these results confirmedHypothesis 2.

To investigate these relationships further, weexamined the longitudinal predictive effects ofcoach and peer motivational climates on the depen-dent variables at various levels of analysis. Whenonly the coach dimensions of the motivational cli-mate were included as predictors, the task-involvingdimension emerged as the strongest predictor ofboth intrapersonal fluctuation and the mean levelsof cohesion. Similar positive relationships among acoach task-involving climate and ATG-S, GI-T, andATG-T scores were reported in a cross-sectionalstudy by Balaguer, Castillo, and Duda (2003),which suggests that a coach-created task climatemight play a key role in fostering group cohesion(Eys et al., 2013; Horn et al., 2012). Further,coach-created ego climate was negatively related tosatisfaction with participation. Likewise, Balagueret al. (2002) and Boixadós et al. (2004) showedthat the promotion of a coach-created task climateand the downplaying of a coach-created ego climatecan promote greater satisfaction among players.Taken together, such findings indicate that themore players perceive their coaches to praise indivi-dual effort, mastery strivings, and personal improve-ment, the more they are attracted by their team

4,4

4,2

4,0

3,8

3,6

3,4

3,2

3,0

Low (–1 s) High (+1 s)

Perceived coach-created task climate

Low peer taskclimate (–1 s)

High peer taskclimate (–1 s)A

TG

-T

Figure 2. The within-person relation between perceived coach-created task climate and ATG-T at high and low levels ofpeer-created task climate.

4.4

4.2

4.0

3.8

3.6

3.4

3.2

3.0

Low (–1 s) High (+1 s)

Perceived peer-task climate

Low peer taskclimate (–1 s)

High peer taskclimate (–1 s)

GI-

T

Figure 1. The within-person relation between perceived peer-created task climate and GI-T at high and low levels of peer-created ego climate.

10 T. García-Calvo et al.

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

(Balaguer et al., 2003, 2004), and the less playersperceive their coaches to encourage social compar-isons and normative ability, the less they are satisfied(Balaguer et al., 2002; Boixadós et al., 2004).

Surprisingly, mean levels of coach-created ego cli-mate were positively associated with the cohesionvariables at the interpersonal level. At first glance,this finding may tempt someone to suggest that coa-ches who set up an ego-involving climate can facil-itate team cohesion (Horn et al., 2012). We believethat this is most likely not the case for two reasons.First, perceived coach-created ego climate was nega-tively related or not related to cohesion across time(i.e., at the intrapersonal level). Second, mean levelsof coach-created task climate were far stronger pre-dictors of between-athlete differences in GI-T andATG-T. Taken together, these findings supportedHypothesis 3 and indicate that the best avenue toenhance cohesion is via promoting task- rather thanego-involving coach motivational climate.

In subsequent analyses, we predicted the sameoutcome variables by including perceived peer-cre-ated task and ego motivational climate alongside thecoach-created task and ego climate. A perceivedpeer-created task climate was positively associatedwith task and social aspects of cohesion, whichexplains unique variance over and above a coach-created task climate at both the intrapersonal andinterpersonal level and supports Hypothesis 4.Further, the regression coefficients for peer climatewere in general stronger than those for coach cli-mate. These findings indicate that peers, in additionto coaches, play an important role in shaping themotivational atmosphere of a team and, thus peerinfluence should be taken into account when exam-ining the links between motivational climate andteam cohesion.

When predicting satisfaction with participation, aperceived peer-created motivational climate failed toaccount for any additional variance over and abovecoach-created climate. The task facet of the coach-created task climate predicted satisfaction with parti-cipation at the interpersonal level. An ego-involvingcoach-created climate was weakly but negativelyrelated to satisfaction with participation at the inter-personal level. These results could be explained bybearing in mind that a task-involving coach climategives the opportunity for all players to contribute insome important and unique ways within their team(Balaguer et al., 2002; Boixadós et al., 2004). Theabsence of incremental effects for the peer-createdclimate variables over and above the coach-createdones might reflect the possibility that athletes’ satis-faction with participation are primarily determined bytheir coach and to a much lesser extent by their peers.

Further evidence for the need to consider a peer-created motivational climate alongside a coach-created

climate emerged from the interaction that showed aperceived coach-created task climate to relate positivelyto ATG-T across time and more strongly so when aperceived peer-created task climate was also high (seeFigure 2). Nevertheless, it should be underscored thatcoach- and peer-created climates may not operatecompletely independently. For instance, coaches, asleading authorities on their team, can have an impacton the peer-created climate. Therefore, coach-createdclimates may be associated not only directly with teamcohesion and players’ satisfaction but also indirectly viaa peer-created climate.

An additional noteworthy finding that highlightshow a peer-created motivational climate relates tocohesion is evidenced by the interaction that showsthat the task facet of this motivational climate wasmore strongly associated with GI-T when the egofacet of this climate was low (see Figure 1). Previouswork on coach-created motivational climates hasindicated that a combination of high task-involvingand low ego-involving facets is related to positivecognitive and affective outcomes in sport (Balagueret al., 2002; Boardley & Kavussanu, 2009; Boixadóset al., 2004).

Conclusions, limitations, and future research directions

The present results indicate that perceptions of bothcoach- and peer-created motivational climates at theinterpersonal and intrapersonal levels can predictchanges in the social and task cohesion levels aswell as satisfaction with participation within a teamamongst semi-professional players. However, theobserved effects differed as a function of the mea-surement wave, level of analysis, and variable underinvestigation. Future research that employs qualita-tive methods (e.g., interviews) is needed to identifythe potential reasons for this. Research on the impor-tance of a peer motivational climate and how thecoach climate might set the tone for the peer climateis still in its infancy.

A limitation of our study was that the findings,although longitudinal with three measurementsacross the sport season, were correlational, and nocausal inferences can be drawn as to the relation-ships between motivational climates and satisfactionwith participation. Nevertheless, our results are con-sistent with theoretical predictions and previousempirical research concerning the associationbetween motivational climates and team dynamicsvariables (Balaguer et al., 2004; Boixadós et al.,2004; Heuzé et al., 2006). Moreover, it should benoted that the results pertaining to ATG-S should beviewed with caution because this variable showed alow reliability at the first measurement wave.Furthermore, the scale we used to assess players’satisfaction with participation has not been

Motivational climate and cohesion 11

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

previously validated. Nevertheless, the apparent facevalidity of the items, the high internal consistency ofthe three-item scale, and their modest inter-correla-tions across all three measurement waves provide uswith some confidence in our operational definition.

Another limitation of this study was that it reliedexclusively on self-reports, and thus our findings aresubject to potential influences of shared method var-iance to some extent. Future longitudinal research inthis area will do well to assess objective markers ofmotivational climate and team cohesion (e.g., obser-vation instruments). Nevertheless, there are pub-lished studies that suggest that a task-involvingmotivational climate is associated with group cohe-sion and satisfaction (Balaguer et al., 2004; Boixadóset al., 2004; Heuzé et al., 2006; Leo et al., 2014);thus, it is unlikely that our findings can be solelyattributed to shared method variance. Finally, thegeneralisation of our findings to other populationsamples and sports should be made with cautionbecause our sample comprised only males from aparticular sport (i.e., soccer) and from a particularcountry (i.e., Spain).

Despite the aforementioned limitations, we believethat this work makes a unique contribution to theliterature by examining the concurrent predictiveeffects of peers’ and coaches’ motivational climateon a variety of important team dynamics variables insemi-professional sport at two different levels over a12-month period. Future research should examinethe development and interconnectedness of coachand peer motivational climates over time.Furthermore, future research can build upon thiswork by incorporating measures of coaches’ reportsof the motivational climate they create as well astheir perceptions of the peer climate within theirteam, and such research can subsequently comparevariations between athletes’ and coaches’ percep-tions of both types of climates (Keegan, Spray,Harwood, & Lavallee, 2010). This approach wouldrepresent an important development because similarresearch in school physical education has identifiedlarge discrepancies between students’ and teachers’ratings of the teacher motivational climate (Taylor &Ntoumanis, 2007).

References

Adie, J. W., Duda, J. L., & Ntoumanis, N. (2008). Achievementgoals, competition appraisals, and the psychological and emo-tional welfare of sport participants. Journal of Sport & ExercisePsychology, 30, 302–322.

Ames, C. (1992). Classrooms: Goals, structures, and studentmotivation. Journal of Educational Psychology, 84, 261–271.

Balaguer, I., Castillo, I., & Duda, J. L. (2003). Interrelacionesentre el clima motivacional y la cohesión en futbolistas cadetes[Interrelationships between motivational climate and cohesionin cadet football players]. Edupsykhé, 2(2), 243–258.

Balaguer, I., Castillo, I., Moreno, Y., Garrigues, V., & Soriano, L.(2004). El clima motivacional y la cohesión en equipos defútbol [Motivational climate and cohesion in soccer teams].Encuentros En Psicología Social, 2(1), 152–156.

Balaguer, I., Duda, J. L., Atienza, F. L., & Mayo, C. (2002).Situational and dispositional goals as predictors of perceptionsof individual and team improvement, satisfaction and coachratings among elite female handball teams. Psychology of Sportand Exercise, 3, 293–308.

Balaguer, I., Guivernau, M., Duda, J. L., & Crespo, M. (1997).Análisis de la validez de constructo y de la validez predictiva delcuestionario de clima motivacional percibido en el deporte(PMCSQ-2) con tenistas españoles de competición [Analysisof the construct and predictive validity of the perceived motiva-tional climate in sport (PMCSQ-2) with Spanish competitiontennis players]. Revista de Psicología del Deporte, 11, 41–58.

Bauer, D. J., & Curran, P. J. (2005). Probing interactions in fixedand multilevel regression: Inferential and graphical techniques.Multivariate Behavioral Research, 40, 373–400.

Beauchamp, M. R. (2007). Efficacy beliefs within relational andgroup contexts in sport. In S. Jowett & D. Lavallee (Eds.),Social psychology in sport (pp. 181–193). Champaign, IL:Human Kinetics.

Beauchamp, M. R., Bray, S. R., Eys, M. A., & Carron, A. V.(2002). Role ambiguity, role efficacy, and role performance:Multidimensional and mediational relationships within inter-dependent sport teams. Group Dynamics: Theory, Research, andPractice, 6, 229–242.

Boardley, I. D., & Kavussanu, M. (2009). The influence of socialvariables and moral disengagement on prosocial and antisocialbehaviours in field hockey and netball. Journal of Sports Sciences,27, 843–854.

Boixadós, M., Cruz, J., Torregrosa, M., & Valiente, L. (2004).Relationships among motivational climate, satisfaction, per-ceived ability, and fair play attitudes in young soccer players.Journal of Applied Sport Psychology, 16, 301–317.

Boyce, B. A., Gano-Overway, L. A., & Campbell, A. L. (2009).Perceived motivational climate’s influence on goal orientations,perceived competence, and practice strategies across the ath-letic season. Journal of Applied Sport Psychology, 21, 381–394.

Carron, A. V., & Brawley, L. R. (2000). Cohesion: Conceptualand measurement issues. Small Group Research, 31, 89–106.

Carron, A. V., Brawley, L. R., & Widmeyer, W. N. (1998). Themeasurement of cohesiveness in sport groups. In J. L. Duda(Ed.), Advances in sport and exercise psychology measurement (pp.213–226). Morgantown, WV: Fitness Information Technology.

Carron, A. V., Colman, M. M., Wheeler, J., & Stevens, D.(2002). Cohesion and performance in sport: A meta-analysis.Journal of Sport & Exercise Psychology, 24, 168–188.

Carron, A. V., & Eys, M. A. (2012). Group dynamics in sport (4thed.). Morgantown, WV: Fitness Information Technology.

Carron, A. V., Widmeyer, W. N., & Brawley, L. R. (1985). Thedevelopment of an instrument to assess cohesion in sportteams: The group environment questionnaire. Journal of SportPsychology, 7, 244–266.

Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Appliedmultiple regression/correlation analysis for the behavioural sciences(3rd ed.). Mahwah, NJ: Erlbaum.

Conroy, D. E., Kaye, M. P., & Coatsworth, J. D. (2006).Coaching climate and the destructive effects of mastery-avoid-ance goals on situational motivation. Journal of Sport & ExercisePsychology, 28, 69–92.

Duda, J. L., & Balaguer, I. (2007). The coach-created motiva-tional climate. In D. Lavalee & S. Jowett (Eds.), Social psychol-ogy of sport (pp. 117–130). Champaign, IL: Human Kinetics.

Dweck, C. S., & Leggett, E. L. (1988). A social-cognitiveapproach to motivation and personality. Psychological Review,95, 256–273.

12 T. García-Calvo et al.

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4

Eys, M. A., Jewitt, E., Evans, M. B., Wolf, S., Bruner, M. W., &Loughead, T. M. (2013). Coach-initiated motivational climateand cohesion in youth sport. Research Quarterly for Exercise andSport, 84, 373–383.

García-Calvo, T., Cervelló, E. M., Jiménez, R., Iglesias, D., &Santos-Rosa, F. L. (2005). La implicación motivacional dejugadores jóvenes de fútbol y su relación con el estado deflow y la satisfacción en competición [Youth football players’motivational involvement and their relationship with flow stateand satisfaction with competition]. Revista de Psicología delDeporte, 14, 21–42.

Harwood, C., & Swain, A. B. J. (2001). The development andactivation of achievement goals in tennis: I. Understanding theunderlying factors. The Sport Psychologist, 15, 319–341.

Heuzé, J. P., Sarrazin, P., Masiero, M., Raimbault, N., &Thomas, J. P. (2006). The relationships of perceived motiva-tional climate to cohesion and collective efficacy in elite femaleteams. Journal of Applied Sport Psychology, 18, 201–218.

Horn, T., Byrd, M., Martin, E., & Young, C. (2012). Perceivedmotivational climate and team cohesion in adolescent athletes.Sport Science Review, 21, 25–48.

Iturbide, L. M., Elosua, P., & Yanes, F. (2010). Medida de lacohesión en equipos deportivos. Adaptación al español delgroup environment questionnaire (GEQ) [A measure of teamcohesion in sport. Spanish adaptation of group environmentquestionnaire (GEQ)]. Psicothema, 22, 482–488.

Keegan, R., Spray, C., Harwood, C., & Lavallee, D. (2010). Themotivational atmosphere in youth sport: Coach, parent, andpeer influences on motivation in specializing sport participants.Journal of Applied Sport Psychology, 22, 87–105.

Leo, F. M., Sánchez-Miguel, P. A., Sánchez-Oliva, D., Amado,D., & García-Calvo, T. (2012). Evolution of perceived cohe-sion and efficacy over the season and their relation to successexpectations in soccer teams. Journal of Human Kinetics, 34,139–147.

Leo, F. M., Sánchez-Miguel, P. A., Sánchez-Oliva, D., Amado,A., & García-Calvo, T. (2014). Análisis de los procesos gru-pales y el rendimiento en fútbol semiprofesional [Analysis ofthe group process and the performance in semiprofessionalsoccer]. Revista Internacional de Medicina y Ciencias de laActividad Física y del Deporte, 14, 153–168.

Levine, J. M., & Moreland, R. L. (1990). Progress in small groupresearch. Annual Review of Psychology, 41, 585–634.

Loughead, T., Hardy, J., & Eys, M. (2006). The nature of athleteleadership. Journal of Sport Behavior, 29, 142–158.

Loughead, T. M., & Carron, A. V. (2004). The mediating role ofcohesion in the leader behavior–satisfaction relationship.Psychology of Sport and Exercise, 5, 355–371.

Maas, C. J. M., & Hox, J. J. (2005). Sufficient sample sizes formultilevel modelling. Methodology, 1, 85–91.

Moreno, J. A., Conte, L., Martínez, C., Alonso, N., González-Cutre, D., & Cervelló, E. (2011). Propiedades psicométricasdel peer motivational climate in youth sport questionnaire(PeerMCYSQ) con una muestra de deportistas españoles[Psychometris properties of the peer motivational climate inyouth sport questionnaire (PeerMCYSQ) with a Spanish ath-letes sample]. Revista de Psicología del Deporte, 20, 101–118.

Mullen, B., & Copper, C. (1994). The relation between groupcohesiveness and performance: An integration. PsychologicalBulletin, 115, 210–227.

Muthén, L. K., & Muthén, B. O. (1998–2012). Mplus user’s guide(7th ed.). Los Angeles, CA: Muthén & Muthén.

Naylor, S. L. (1996). The relationship among achievement goalorientation, perceived motivational climate, and cohesion in sportteams (Unpublished doctoral dissertation). California StateUniversity, Long Beach.

Newton, M., Duda, J., & Yin, Z. (2000). Examination of thepsychometric properties of the perceived motivational climatein sport questionnaire-2 in a sample of female athletes. Journalof Sports Sciences, 18, 275–290.

Ntoumanis, N., & Biddle, S. J. H. (1999). A review of motiva-tional climate in physical activity. Journal of Sports Sciences, 17,643–665.

Ntoumanis, N., Taylor, I. M., & Thøgersen-Ntoumani, C.(2012). A longitudinal examination of coach and peer motiva-tional climates in youth sport: Implications for moral attitudes,well-being, and behavioral investment. DevelopmentalPsychology, 48, 213–223.

Ntoumanis, N., & Vazou, S. (2005). Peer motivational climate inyouth sport: Measurement development and validation. Journalof Sport and Exercise Psychology, 27, 432–455.

Olmedilla, A., Ortega, E., Almeida, P., Lameiras, J., Villalonga,T., Sousa, C., … García-Mas, A. (2011). Cohesión ycooperación en equipos deportivos [Cohesion and cooperationin sports teams]. Anales de Psicología, 27, 232–238.

Olympiou, A., Jowett, S., & Duda, J. L. (2008). The psychologicalinterface of the coach-created motivational climate and thecoach-athlete relationship. The Sport Psychologist, 22, 423–438.

Paskevich, D. M., Estabrooks, P. A., Brawley, L. R., & Carron, A.V. (2001). Group cohesion in sport and exercise. In R. N.Singer, H. A. Hausenblas, & C. Janelle (Eds.), Handbook ofsport psychology (pp. 472–494). New York, NY: Wiley.

Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006).Computational tools for probing interactions in multiplelinear regression, multilevel modeling, and latent curve ana-lysis. Journal of Educational and Behavioral Statistics, 31,437–448.

Reinboth, M., & Duda, J. L. (2006). Perceived motivational cli-mate, need satisfaction and indices of well-being in teamsports: A longitudinal perspective. Psychology of Sport andExercise, 7, 269–286.

Taylor, I. M., & Ntoumanis, N. (2007). Teacher motivationalstrategies and student self-determination in physical education.Journal of Educational Psychology, 99, 747–760.

Turman, P. D. (2003). Coaches and cohesion: The impact ofcoaching techniques on team cohesion in the small groupsport setting. Journal of Sport Behavior, 26, 86–104.

Vazou, S., Ntoumanis, N., & Duda, J. L. (2005). Peer motiva-tional climate in youth sport: A qualitative inquiry. Psychology ofSport & Exercise, 6, 497–516.

Vazou, S., Ntoumanis, N., & Duda, J. L. (2006). Predictingyoung athletes’ motivational indices as a function of their per-ceptions of the coach- and peer-created climate. Psychology ofSport and Exercise, 7, 215–233.

Vincer, D. J. E., & Loughead, T. M. (2010). The relationshipamong athlete leadership behaviours and cohesion in teamsports. The Sport Psychologist, 24, 448–467.

Weiss, M. R., & Smith, A. L. (2002). Friendship quality in youthsport: Relationship to age, gender, and motivation variables.Journal of Sport & Exercise Psychology, 24, 420–437.

Motivational climate and cohesion 13

Dow

nloa

ded

by [

Bib

liote

ca. U

nive

rsid

ad d

e E

xtre

mad

ura]

at 0

0:06

08

July

201

4