THE RELATIONSHIP BETWEEN PERCEIVED COACH LEADERSHIP BEHAVIOURS AND TEAM COHESION … · 2019. 9. 18. · LEADERSHIP BEHAVIOURS AND TEAM COHESION AMONG ROMANIAN ATHLETES Marius Crăciun

REVISTA DE IBEROAMERICANA DE PSICOLOGÍA DEL EJERCICIO Y EL DEPORTEVol. 4, nº 2, pp. 217-232 ISSN: 1886-8576

ABSTRACT: Cohesion is an important aspect of performance groups. Therefore it is impor-tant to investigate the antecedents of team cohesion, such as leadership. Although the relations-hip between coach leadership behaviors and team cohesion has been extensively studied, it is notso clear. In the present study, we examined the relationship between the perceived coach leaders-hip behaviors and group cohesion. The 81 participants have completed the Leadership Scale forSport (LSS; athlete perceptions version) and Group Environment Questionnaire (GEQ). Theresults show that perceived cohesion of the basket team members is significantly lower than theperceived cohesion of the volleyball, rugby and handball team members. Most of the coach lea-dership behaviors were negatively associated with the overall team cohesion and its dimensions.It was found that the coach leadership behaviors did not predict the overall team cohesion, whencontrolling the effects of gender and the type of sport. Theoretical and practical implications arediscussed.

KEYWORDS: Leadership, team cohesion, basket, coach

RESUMEN: La cohesión es un aspecto importante en el rendimiento de los grupos. Por con-siguiente, es importante investigar los antecedentes de la cohesión del equipo, tales como el lide-razgo. A pesar de que la relación entre las conductas de liderazgo percibidas del entrenador y lacohesión del equipo se ha estudiado bastante, no está suficientemente clara. En el presente tra-bajo, estudiamos la relación entre las conductas de liderazgo percibidas en el entrenador y la

217

THE RELATIONSHIP BETWEEN PERCEIVED COACHLEADERSHIP BEHAVIOURS AND TEAM COHESION

AMONG ROMANIAN ATHLETES

Marius Crăciun y Claudia-Lenuţa RusBabes-Bolyai University, Department of Psychology, Cluj-Napoca,

Romania

cohesión de equipo. 81 participantes han completado el Cuestionario de Liderazgo Deportivo(LSS, del inglés Leadership Scale for Sports) -la versión de la percepción del atleta- y elCuestionario del Ambiente Grupal (GEQ, del inglés Group Environment Questionnaire). Losresultados muestran que la cohesión percibida por los jugadores de baloncesto es significativa-mente más baja que la percibida por los de voleibol, rugby y balonmano. La mayoría de las con-ductas mostraban una asociación negativa con la cohesión total del equipo y sus dimensiones.Se detectó que las conductas de liderazgo del entrenador no predecían la cohesión total del equi-po, al controlar los efectos del género y el tipo de deporte. Se analizan las implicaciones teóri-cas y prácticas.

PALABRAS-CLAVE: Liderazgo, cohesión de equipo, baloncesto, entrenador

RESUMO: A coesão é um aspecto importante no rendimento dos grupos. Nesse sentido, éimportante investigar os antecedentes da coesão das equipas, tais como a liderança. Apesar de arelação entre os comportamentos de liderança percebidos do treinador e a coesão da equipa tersido já bastante estudada, não é suficientemente clara. Neste trabalho, estudamos a relação entreos comportamentos de liderança percebidos do treinador e a coesão da equipa. Os 81 partici-pantes preencheram a Escala de Liderança para o Desporto (LSS, do inglês Leadership Scale forSports), na versão correspondente à percepção dos atletas, e o Questionário do Ambiente deGrupo (GEQ, do inglês Group Environment Questionnaire). Os resultados revelaram que acoesão percebida pelos jogadores de basquetebol é significativamente mais baixa do que a per-cebida pelos jogadores de voleibol, rugby e handebol. A maioria dos comportamentos de lide-rança do treinador associou-se negativamente à coesão global da equipa e às suas dimensões.Verificou-se que os comportamentos de liderança do treinador não prediziam a coesão globalda equipa, quando controlados os efeitos do gênero e do tipo de desporto. As implicações teó-ricas e práticas são discutidas.

PALAVRAS-CHAVE: Liderança, coesão de equipa, basquetebol, treinador.

218 Revista Iberoamericana de Psicología del Ejercicio y el Deporte. Vol. 4, nº 2 (2009)

Marius Crăciun y Claudia-Lenuţa Rus

INTRODUCTIONThe relationship between leader

behaviors and cohesion in naturalgroups has received empirical attention.In a military context, the cohesiveness ofcombat groups has been investigated inrelationship to the behavior of unit lead-ers (Bartone & Kirkland, 1991). In thetherapeutic context, data suggest thatleaders who are less directive and exhib-it more personal warmth have groups

with higher cohesiveness (Antonuccio,Davis, Lewinsohn, & Breckenridge,1987). In an organizational context,group cohesiveness has been shown toincrease when leaders reward productiv-ity (Podsakoff & Todor, 1985).

The leadership perspective empha-sizes coaches’ interpersonal behavior,more specifically, the impact of coaches’behaviors and actions on athletes’ physi-cal and psychosocial aspects such as per-

219

The relationship between perceived coach leadership behaviours and team cohesion...

Revista Iberoamericana de Psicología del Ejercicio y el Deporte. Vol. 4, nº 2 (2009)

formance, satisfaction and self-esteem.This perspective has largely definedcoaches’ behaviors or coach leadershipas “the behavioral process of influenc-ing individuals and group toward setgoal” (Barrow, 1977). In essence, leader-ship is a characteristic of the coach andleading is what the coach does to theathlete and originates with the coach.

One of the major approaches to thestudy of coach leadership behavior insport teams has been most frequentlystudied in light of Chelladurai andCarron’s (1978) Multidimensional Modelof Leadership. This model focuses onthree aspects of coaching behavior:actual leader behavior (i.e. behavior thatoriginates from a leader’s personality,ability, and experience), preferred leaderbehavior (i.e., behavior that reflectscharacteristics of the members), andrequired leader behavior (i.e., behaviorthat is dictated by the demands and con-straints of the situation) (Beauchamp,Bray, Eys, & Carron, 2005). The modelsuggests that athletes’ performance andsatisfaction are attributable to the degreeof congruence among the three aspectsof leader behavior.

Leadership Scale for Sports (LSS;Chelladurai & Saleh, 1980) was devel-oped to measure five coaching behaviorsfrom both the coach’s (perceived) andthe athlete’s (perceived and preferred)perspectives. According to Chelladurai(1993), the autocratic and democraticbehaviors represent the decision stylecoach commonly used. The training andinstruction behaviors correspond to thetask-oriented coach behavior, which

aims at improving athletes’ perform-ance. The social supportive behaviorscorrespond to the relationship-orientedbehavior, where the coach emphasizespersonal concern for individual athletes.Finally, coaches’ positive feedback isfocused on providing credit, apprecia-tion, and reward for athletes’ good per-formance, which serve as a motivator.

The issue of group cohesion in thecontext of team sports has received con-siderable attention (Widmeyer, Carron,& Brawley, 1993), specially in the sportpsychology domain (Ntoumanis &Aggelonidis, 2004). Group cohesion isconsidered an important variable in themodels of effectiveness and perform-ance (Høigaard, Säfvenbom, &Tønnesses, 2006). The high interest inthe concept of cohesion stems, in part,from the belief that team cohesivenessplays an important role in team per-formance, a belief that has been largelysupported (Widmeyer et al., 1993). Overthe years, group cohesion has been oper-ationalised in different ways (Chand &Bordia, 2001). Initially, it was defined asthe degree of group cooperation towarda goal (Weinberg, 1979), then as a traitvariable (Bormann, 1990) or groupattraction (Cragan & Wright, 1995;Rozell & Gundersten, 2003).

Defining the group cohesion as adynamic process which is reflected inthe tendency for a group to stick togeth-er and remain united in the pursuit of itsinstrumental objectives and/or for thesatisfaction of member affective needs,Carron (1982) proposed aMultidimensional Model of Group



Cohesion, in which leadership is identi-fied as an important antecedent. It wasdeveloped specifically for sport teamsand recent studies have highlighted thechallenges of adapting the GEQ formeasuring cohesion in work teams(Carron & Brawley, 2000). According tothis author, the concept of cohesion isbest conceptualized as involving bothtask and social dimensions, encompass-ing both individual and group aspects(Carron & Brawley, 2000; Hardy, Eis, &Carron, 2005). Carron‘s theoreticalmodel of group cohesion stimulated thedevelopment of the GroupEnvironment Questionnaire (GEQ;Carron, Widmeyer, & Brawley, 1985).This questionnaire is currently theinstrument most widely used to assesscohesion among sport teams due to theempirical support received in the past 20years by Multidimensional Model ofGroup Cohesion (Hardy, Eis, & Carron,2005; Rovio, Eskola, Kozub, Duda, &Lintunen, 2009). GEQ has proven to bea valid instrument that has value in thelaboratory and in the field (Carron at al.,1985) and has been adapted for differentcultural settings and different languages.The GEQ is a useful tool at several lev-els when groups of athletes are broughttogether to form a team.

The complex interaction betweencoach and players appears to influencethe development of cohesion. A consid-erable body of research has examinedthe association between leader decisionstyle and cohesion (Carron &Chelladurai, 1981; Westre & Weiss,1991). Results from these studies: a par-

ticipative style of decision-making isrelated to greater perceptions of cohe-siveness. The type of leadership behav-ior exhibited by the coach also has beenfound to be associated with the develop-ment of cohesion. However the optimaltype of leader behavior is not clear. Forexample, Westre and Weiss (1991) foundthat high level of training and instruc-tion behavior, social support behaviorand positive feedback with high schoolfootball players were associated withhigher levels of task cohesion. In a studyconducted with high school basketballteams, however, Kozub et al. (1993)found that only higher level of trainingand instruction behavior and social sup-port behavior were related to greatertask cohesion. These authors found apositive relationship between every LSSscale except autocratic behavior (whichcould not be assessed because of poorinternal consistency) and task cohesion.In both studies, the social cohesionscales of the GEQ had to be discardedbecause of poor internal reliability.Gardner et al.(1996) found all fivecoaching behaviors to be related to boththe task and social dimensions of teamcohesion, and the relationships wereespecially strong when team-level analy-ses were conducted.

In the present study, we investigatedthe relationship between leadershipbehavior and group cohesion within thecontext of Romanian team sport.According to our knowledge, there areno Romanian studies which have studiedthis relationship using the model ofcoach leadership behavior proposed by

221



Chelladurai (1980) and the group cohe-sion model developed by Carron et al.(1998).

METHOD

ParticipantsIn the present study were included 81

participants from 6 sport teams. Themean age of the participants is 23.1,(SD= 1.85). 16 athletes (19.8%) derivedfrom two basketball teams (9 male, 7female), 10 female athletes (12.3%) fromone volleyball team, 30 athletes (37%)from two handball teams (14 male, 16female) and 25 male athletes from onerugby team. All the athletes participatedvoluntarily in the study.

InstrumentsIn order to evaluate the dimensions

of leader behavior, the perception ver-sion of the Leadership Sport Scale (LSS;Chelladurai & Saleh, 1978, 1980) wasused. The 40 items of this scale weregrouped in 5 scales measuring five dif-ferent coaching behaviors. The firstscale, Training and Instruction, evaluat-ed through 13 items (α= .90) the coach-ing behavior aimed at improving the ath-letes` performance by emphasizing andfacilitating hard and strenuous training;instructing them in the skills, techniquesand tactics of the sport; clarifying therelationship among the members and bystructuring and coordinating the mem-bers activities. The DemocraticBehavior Scale (9 items, α= .69), meas-ured the coaching behavior which allowsgreater participation by the athletes in

decisions, pertaining to group goals,practice methods, and game tactics andstrategies. The Autocratic Behavior wasevaluated through 5 items (α= .53). Itwas conceptualized as the coachingbehavior which involves independentdecision making and stresses personalbehavior authority. The Social SupportScale evaluated the coaching behaviorcharacterized by a concern for the wel-fare of individual athletes, behavior pos-itive group atmosphere and warm inter-personal relations with members (8items, α= .61). The Positive FeedbackScale evaluated the coaching behaviorwhich reinforces an athlete by recogniz-ing and rewarding good performancefeedback (5 items, α= .81). The partici-pants rated the frequency of their actualcoach`s behavior using a 5-point Likert-scale ranging from always (1) to never(5). The scores for each scale were com-puted by summing the items responsesand by dividing this sum by the numberof items. High scores reveal a high per-ceived frequency of the two decisionmaking styles (democratic and autocrat-ic behavior), motivational tendencies(social support and positive feedback)and of the instructional behavior of thecoach.

The team cohesion was evaluatedusing The Group EnvironmentQuestionnaire (Carron, Brawley &Widmeyer, 2002). This instrument con-sists of 18 items grouped in 4 scaleswhich measure four different aspects ofteam cohesiveness. The GroupIntegration Task Scale (GI-T) evaluatedthe participants` feelings about the simi-



larity, closeness and bonding within theteam as a whole around the group task(5 items, α= .58). The Group IntegrationSocial Scale (GI-S) was used to measurethe individual team members` feelingsabout the similarity, closeness, andbonding within the team as a wholearound the group as a social unit (4items, α= .56). The IndividualAttractions to the Group Task Scale(ATG-T) measured the individual teammembers feelings about his/ her person-al involvement with the group task, pro-ductivity, goals and objectives (4 items,α= .57). The Individual Attraction to theGroup Social Scale (ATG-S) evaluatedthe individual team members` feelingsabout his/ her personal acceptance andsocial interactions with (5 items, α= .50).The participants had to check a numeri-cal response for each question abouttheir team sport experience, using aLikert-type scale with 9 points rangingfrom strongly disagree (1) to stronglyagree (9). In order to compute the scoresfor each scale, all the individual itemresponses were summed and then divid-ed by the number of items (in order tocompare the four aspects of cohesive-ness between them). Higher scores rep-resent a stronger perception of cohe-siveness.

ProcedureThe 2 questionnaires were adminis-

tered at one meeting at the end of atraining session. The participants filledin the perceived version of the LSS andthen the GEQ. Every participant com-pleted the questionnaires individually,

being assured in what concerned theanonymity and the confidentiality oftheir ratings. In order to compute theresults we used SPSS 15.

RESULTSThe results of the univariate analysis

for each scale used are presented inTable 1. It was computed the means,standard deviations and the skewnessindicators because all the variables wereevaluated on numerical scales.

The participants consider that theircoaches express lower behaviors such asimproving the athletes` performance byemphasizing and facilitating hard andstrenuous training, instructing them inthe skills, techniques and tactics of thesport, clarifying the relationship amongthe group members, structuring andcoordinating the team members` activi-ties (M = 2.05, SD = .70). The evaluatedcoaches are perceived by the participantsin the terms of giving a low social sup-port (M = 2.68, SD = .62) and positivefeedback (M = 2.15, SD = .80).

Further, the value of skewness indi-cators for the training and instruction,autocratic behavior and positive feed-back variable reveals that most of theparticipants have given low scoresregarding these three dimensions ofcoach behavior. In order to establish thesymmetry of the scores distributions fora variable, it was used the z thresholdvalue (1.96) proposed by Field (2000).

223



The analysis of the means of GEQscales shows that the sample included inthis study is more cohesive regarding thesocial and task individual attractions tothe group (M = 7.12, SD = 1.33; M =6.78, SD = 1.45) and less cohesiveregarding social and task group interac-tion (M = 5.56, SD = 1.56). It seems thatparticipants have stronger perceptionabout the personal motivations acting toattract and to retain the individual in thegroup and stronger feelings about thegroup compared to their perceptionabout the closeness, similarity and bond-ing within the group as a whole, as wellas the degree of unification of the

group field.In order to set if these aspects of

group cohesion can be included in theregression analysis, the score distribu-tion and the reliability of each scale wereexamined. It was found that all fourscales had a poorer reliability, lower than.70. This means that these scales can notbe used in regression analysis as a crite-rion variable. So, we computed the relia-bility for the 2 components of teamcohesion specified in the conceptualmodel used in this study: individualattraction to the group and group inte-gration. A total score for the groupcohesion was also computed (see Table


Marius Crăciun y Claudia-Lenuţa Russ

2). The results show that the reliability ofindividual attraction to the group scalewas lower than .70. Even if the reliabili-ty of the group integration scale wasacceptable (.71), it was preferred toinclude in the regression analysis as a cri-terion variable only the overall scale ofteam cohesion.

The results of bivariate analysis of allvariables measured are presented inTable 3. The results show that all dimen-sions of coaching behavior are positive-ly interrelated, excepting the associationof autocratic behavior with training andinstruction (r = .12, p > .05), respective-ly with democratic behavior (r = .03, p >.05). The autocratic behavior is inde-pendent form these two types of coach-ing behavior.

The training and instruction behav-ior is associated negatively with the mostof the cohesion dimensions, except thesocial individual attraction to group (r =-.04, p > .05). This means that a highcoach`s focus on improving the athletes`performance is associated with low feel-ings of athletes about their personalinvolvement with the group task, pro-ductivity, goals and objectives (r = -.49,p < .01), about the similarity, closenessand bonding within the team as a wholearound the task of the group (r = -.36, p< .01) and around the group as a socialunit (r = -.22, p < .05). Similar, a highcoach`s focus on training and instruc-tion of the athletes is negatively associ-ated with the individual attractions tothe group (r = -.31, p < .01), group inte-gration (r = -.34, p < .01) and overallcohesion (r = -.38, p < .01).

The coach` s democratic behaviorseems to be independent from all thewhole? group cohesion variables. Theautocratic behavior is associated onlywith the social group integration (r = -.22, p < .05) and the composite groupintegration (r = -.23, p < .05). Theparticipants that perceive a high level ofcoach`s autocratic behavior will tend toexpress low feelings about the closeness,similarity and bonding within the groupas a whole, specially around the group asa social unit.

The social support offered by thecoach of the team is negatively associat-ed with the individual attractions to thegroup (r = -. 22, p < .05), the attractionof the group to individual task (r = -.22,p < .05) and the group integration (r = -.28, p < .01), particularly the task aspect(r = -.30, p < .01). Similarly, the socialsupport correlates negatively with theoverall group cohesion (r = -.29, p <.01). The direction of these associationsreflects that participants who rated theircoach as giving a high social support tothe team members have expressed lowerlevels of feelings about their personalinvolvement with the group task, pro-ductivity, goals and objectives, feelingsabout the similarity, closeness and bond-ing within the team as a whole aroundthe group task and about the team as aunit.

As it happens with the training andinstruction behavior, the positive feed-back is associated negatively with themost of the cohesion dimensions, com-posite cohesion dimensions and overallcohesion, except the social individual



225


Revista Iberoamericana de Psicología del Ejercicio y el Deporte. Vol. 4, nº 2 (2009)226

attraction to group (r = .00, p > .05).The participants who perceive that theircoach gives them a positive feedbackregarding their performance expresslower levels of feelings about their per-sonal involvement with the group task,productivity, goals and objectives (r = -.39, p < .01), about the similarity, close-ness and bonding within the team as awhole around the task of the group (r =-.34, p < .01) and around the group as asocial unit (r = -.27, p < .05). A highcoaching behavior which reinforces theparticipants by recognizing and reward-ing a good performance is negativelyassociated with the individual attractionsto the group (r = -.22, p < .05), groupintegration (r = -.36, p < .01) and over-all team cohesion (r = -.34, p < .01).The associations between the fouraspects of team cohesion (social andtask individual attractions to the group,social and task group integration), thetwo major subfactors of team cohesion(individual attractions to the group andgroup integration) and the overall teamcohesion are all positive.

In order to examine the power of thecoaching behaviors in predicting theteam cohesion when controlling theeffects of gender and the type of sport,we conducted a multilinear regressionanalysis with 2 steps:Step 1: includes gender (male, female)and sport type (basket, volleyball, hand-ball and rugby).Step 2: includes gender (male, female),sport type (basket, volleyball, handballand rugby) and the five dimensions ofleader behavior in sport (training and

instruction, democratic behavior, auto-cratic behavior, social support, positivefeedback).

In the first step, the gender and thesport type significantly estimate theoverall team cohesion, F(4, 76) = 5.70, p< .01 and explain 23% of its variance.Instead of this, at an individual level ofpredictors, only dummy volleyball vari-able is a significant predictor of overallteam cohesion (β = .61, t = 4.71, p <.01). This means that the team volleyballmembers are more cohesive than thebasket team members.

In the second step, the demographicvariables and the dimensions of leaderbehavior in sport estimated significantlythe team cohesion, F(9,71) = 4.44, p <.01. This model is significantly betterthan the model based only on the demo-graphic variables, F(5,71)= 2.86, p < .01and additionally explains 12% of theteam cohesion. The variables regardingthe sport type were the only predictorsof the criterion variable. The perceivedcohesion of the basket team members issignificantly lower than the perceivedcohesion of the volleyball team mem-bers (β = .50, t = 3.80, p < .01), rugbyteam members (β = .41, t = 2.36, p <.05) and the cohesion of the handballteam members (β = .29, t = 2.14, p <.05). None of the coaching behaviorssignificantly predicts the team cohesion.


227

The results show that the dimensionsof leader behavior in sport do not sig-nificantly predict the team cohesion,when controlling the effects of genderand type of sport.

DiscussionsIn this study we investigated the rela-

tionship between leadership behaviorsand group cohesion of some sportteams. The athletes perceive that theircoaches are very concerned about thewelfare of the individual athletes, posi-tive group atmosphere and warm inter-personal relationship with the teammembers. In addition, these coaches are

considered as not reinforcing an athleteby recognizing and rewarding its goodperformance. The results show that per-ceived cohesion of the basket teammembers is significantly lower than theperceived cohesion of the volleyball,rugby and handball team members.Most of the coach leadership behaviorswere negatively associated with the over-all team cohesion and its dimensions.The inspection of the hierarchical multi-ple regression analysis show that differ-ent leadership behaviors are not signifi-cant predictors of team cohesion, whencontrolling the effects of the gender andtype of sport. We controlled the effectsof gender because some studies suggest-


ed that gender may influence the factori-al structure of cohesiveness model pro-posed by carron. Scultz et al. (1994) havefound that the analysis of factorial struc-tures of GEQ across gender showeddifferent factorial structures, but noneof these structures had an model ade-quate fit (apud. Ntoumanis &Aggelonidis, 2004).

Generally, these results are not inaccordance with the results of someother studies. Firstly, this discrepancycan be explained through the poorerreliability of some of the scales thatevaluated the coach behaviors. Further,in a review, Chelladurai and Riemer(1998) showed that the scale‘s reliabilityestimates have not been consistent oradequate. In other words, certain behav-ioral dimension (i.e. autocratic behavior,social support and positive feedback) inboth the “athletes` preferences” and“athletes’ perceptions” versions haverepeatedly failed to record acceptableinternal consistency estimates. The mod-erately low level of explained variancereports in the original LSS validationsuggest that the scale may not measureall or majority of the aspects involved inrelationship coach-athletes.

The reliability of the GEQ scaleswas also lower. Even if the reliability ofthe GEQ scale was above .70, accordingto some authors the criterion variable ina predictive regression analysis is betterto be above .80 (Sava, 2004). Some ofthe studies from the sport literaturerevealed that the factorial structure ofthe GEQ is a problematic one. Forexample, Ntoumanis and Aggelonidis

(2004) found in a study of psychometricevaluation of the Group EnvironmentQuestionnaire, that the very high factorcorrelations rendered problematic thediscriminant validity of the question-naire. But in this context, it is importantto note that Bawley and Carron (2003)have suggested some sport teams maynot exhibit every dimension of cohesionmeasured by the questionnaire.

Secondly, the team cohesion wascompute by summing the social and taskcohesion. Some studies had shown thatdifferent coach leadership set of behav-iors explained differently the variance ofthe social and task cohesion (Jowett &Chaundy, 2004). For example, Turman(2003) identified that specific leadershipstrategies such as the behaviors that pro-mote instruction, can potentiallyenhance the level of task cohesion insport teams. Further, there are inconclu-sive results regarding the factorial validi-ty of the GEQ. For example, Leesonand Fletcher (2005) using elite femalenetball players, reported a strong factori-al invariance for social and task cohesionscores, whereas differential stability wasachieved only by the task cohesionscores. These authors also found thatthe latent mean stability of GEQ wasnot established.

The relation between coach behaviorand team cohesion cannot be consideredwithout noticing some limitations ofthis study. Because the subjects partici-pated voluntarily, we cannot discuss thegeneralizability of the results. The par-ticipants derived form different sportteams. As the regression analysis





229

showed, there were some effects of thetype of sport on team cohesion.

In the present study, only the versionof the actual coach behaviour was used.Further studies can use all the versionsof the Leadership Sport Scale in orderto predict the team cohesion.

REFERENCESAntonuccio, D. O., Davis, C.,

Lewinsohn, P. M., & Breckenridge, J.S. (1987). Therapist variables relatedto cohesiveness in a group treatmentfor depression. Small Group Behavior,18, 557-564.

Barrow, J.C. (1977). The variable ofleadership: A review and conceptualframework. Academy of ManagementReview, 2, 231-251.

Bartone, P. T., & Kirkland, F. R. (1991).Optimal leadership in small armyunits. In R. Gal & A. D.Mangelsdorff (Eds.), Handbook ofmilitary psychology (pp. 393-409).Chichester, England: Wiley.

Beauchamp, M. R., Bray, S. R., Eys, M.A., & Carron, A. V. (2005).Leadership behavior and multidi-mensional role ambiguity percep-tions in teams sports. Small GroupResearch, 36, 5-20.

Brawley, L.R. & Carron, A.V. (2003).Caution in InterpretingConfirmatory Factor Analysis of theGroup Environment Questionnaire:A Response to Sullivan, Short, andCramer (2002). Perceptual and MotorSkills, 97, 315–18.

Carron, A. V. (1982). Cohesiveness in

sport groups: Interpretations andconsiderations. Journal of SportPsychology, 4, 123-128.

Carron, A.V., & Chelladurai, P. (1981).Cohesion as a factor in sport per-formance. International Review of SportSociology, 16, 2-41.

Carron, A. V., Widmeyer, N., & Brawley,L. (1985). The development of aninstrument to assess cohesion insport teams: The GroupEnvironment Questionnaire. Journalof Sport and Exercise Psychology, 10,244-266.

Carron, A. V. & Brawley, L. R. (2000).Cohesion: Conceptual and measure-ment issues. Small Group Research, 31,89-106.

Chang, A. & Bordia, P. (2001). A multi-dimensional approach to the groupcohesion-group performance rela-tionship. Small Group Research, 32,379-405.

Chelladurai, P., & Carron, A. V. (1978).Leadership. Canadian Associationfor Health, Physical Education andRecreation Sociology of Sport MonographSeries A, Calgary, AB: University ofCalgary.

Chelladurai, P. (1993). Leadership. In R.Singer, M. Murphey, & L. K. Tennant(Eds.), Handbook of research onsport psychology (pp. 647-671). NewYork: Macmillan. Chelladurai, P., &Saleh, S. (1978). Preferred leadershipin sports. Canadian Journal of AppliedSport Sciences, 3, 85-92.

Chelladurai, P., & Saleh, S. (1980).Dimensions of leader behavior in

sports: Development of a leadershipscale. Journal of Sport Psychology, 2, 34-45.

Chelladurai, P., & Riemer, H.A. (1998).Measurement of leadership on sport.In J.L.Duda (Ed), Advances in sport andexercise psychology measurement (pp. 227-253). Morgantown, WV: FitnessInformation Yechnology.

Gardner, D. E., Shields, D. L. L.,Bredemeier, B. J. L., & Bostrom, A.(1996). The relationship betweenperceived coaching behaviors andteam cohesion among baseball andsoftball players. The Sport Psychologist,10, 367-381.

Hardy, J., Eys, M. A., & Carron, A. V.(2005). Exploring the potential disa-vantages of high cohesion in sportsteams. Small Group Research, 36, 166-187.

Høigaard, R., Säfvenbom, R., &Tønnessen, F. E. (2006). The rela-tionship between group cohesion,groups norms, and perceived socialloafing in soccer teams. Small GroupResearch, 37, 217-232.

Jowett, S. & Chaundy, V. (2004). Aninvestigation into the impact ofcoach leadership and coach-athleterelationship on group cohesion.Group Dynamics: Theory, Research, andPractice: 8, 4, 302-311.

Kozub, S.A. and Button, C.J. (2000). Theinfluence of a competitive outcomeon perceptions of cohesion in rugbyand swimming teams. InternationalJournal of Sport Psychology, 31, 82-95.

Ntoumanis. N. & Aggelonidis, Y. (2004).

A psychmetric evaluation of theGroup Environment Questionnairein a sample of elite and regional levelGreek volleyball players. EuropeanPhysical Education Review, 10, 261-278.

Podsakoff, P. M., & Todor, W. D. (1985).Relationships between leader rewardand punishment behavior and groupprocesses and productivity. Journal ofManagement, 11, 55-73.

Rovio, E., Eskola, J., Kozub, S. A., Duda,J. L., & Lintunen, T. (2009). Can highgroup be harmful? A case study of ajunior ice-hockey team. Small GroupResearch, 40, 421-435.

Rozell, E. R. & Gundersen, D. E.(2003).The effects of leader impressionmanagement on group perceptionsof cohesion, consesus, and commu-nication. Small Group Research, 34,197-222.

Shields, D. L. L., Gardner, D. E.,Bredemeier, B. J. L., & Bostro, A.(1997). The relationship betweenleadership behaviors and groupcohesion in team sports. The Journal ofPsychology, 13(2), 196-210.

Turman, P. D. (2003). Coaches andcohesion: The impact of coachingtechiniques on team cohesion in thesmall group sport setting. The Journalof Sport Behavior, 26, 86-104.

Westre, K., & Weiss, M. (1991). The rela-tionship between perceived coachingbehaviors and group cohesion inhigh school football teams. The SportPsychologist, 5, 41-54.



Widmeyer, W. N., Carron, A. V., &Brawley, L. R. (1993). Group cohe-sion in sport and exercise. In R.Singer, M. Murphey, & L. K. Tennant(Eds.), Handbook of research on sportpsychology (pp. 672-692). New York:Macmillan.



231

Documents

THE RELATIONSHIP BETWEEN PERCEIVED COACH LEADERSHIP BEHAVIOURS AND TEAM COHESION … · 2019. 9. 18. · LEADERSHIP BEHAVIOURS AND TEAM COHESION AMONG ROMANIAN ATHLETES Marius Crăciun