Coach Effectiveness Trainingz A Cognitive­ Behavioral Approach to Enhancing Relationship Skills in Youth Sport Coaches RONALD E. SMITH, FRANKL. SMOLL, AND BILL CURTIS University of Washington Toward the mid-1970s, sport psychologists, in particular in North America, became increasingly interested in the developments in children’s sport. More specifically, in the preceding decades, youth sport had been formalized and organized, with an evidently broadening range of community participation. Parents, coaches, and other relevant adults had also become more involved in youth sport, and the level of complexity steadily increased around this sociocultural system. However, the desirability of the growing number of organized youth sport programs had also been subject to substantial controversy. The central issue of debate had been whether the outcomes of children’s athletic experiences gained through their participation in organized sport is indeed favorable (e.g., for their psychosocial development). As a result, sport psychologists started to direct their attention and effort to promoting the psychological welfare of young athletes involved in organized sport. In a series of classic investigations, two sport psychologists with a background in clinical psychology, Ron Smith and Frank Small, made an immense contribution in this area. Theyfirst developed a system for the behavioral assessment of athletic coaches, widely known as CBAS (Coaching Behavior Assessment System): CBAS is an observationally based instrument, which directly observes and codes overt coaching behavior. It has well-established psychometric properties (i.e., reliability and validity) gairied through extensive empirical research. Over the years, CBAS has become the most commonly used instrument for measuring coaching behavior. It involves assessing the quality, frequency, and type of feedback provided to athletes by their coaches in contests and practices. After the development of CBAS, Smith, Small, and their associates presented and examined the effectiveness of a training pro­ gram for coaches. This program was derived from the principles of the cognitive-behavioral approach and was intended to assist coaches in improving their relationship skills and make the entire training process more effective for both children and coaches. In this framework, Smith and Smoll presented a list of behavioral guidelines for coaches in youth sport, which were formulated in terms of “dos and don’ts,” and constituted the basis of the to-be-developed positive approach to coaching. This article presents both the eBAS and the major guidelines for enhancing the positive interactions between coaches and players. After about three decades, in which it has become a milestone in understanding coaching behavior in particular and sport leadership in general, Smith and Small’s work seems to have lost nothingfrom its relevance to the current reality of children and youth in organized sport both in and outside North America. Little League Baseball coaches were exposed to a preseason training program designed to assist them in relating more effectively to children. Empirically derived behavioral guidelines were presented and modeled, and behavioral feedback and self-monitoring were used to enpance self-awareness and to encourage compliance with the guidelines. Trained coaches differed from controls in both overt and player-perceived behaviors in a manner consistent with the l;>e�vioral guidelines. They were also evaluated more positively by their players, and a higher level of intrateam attraction was found on their teams despite the fact that they did not differ from controls in won-lost records. Chil­ dren who played for the trained coaches exhibited a significant increase in general self-esteem compared with scores obtained a year earlier; control group children did not. The greatest differences in attitudes toward trained and control coaches were found among children low in self-esteem, and such children appeared most sensitive to variations in coaches’ use of encouragement, punishment, and technical instruction. T he present investigation involved the development and assessment of an experimental training program designed to enhance the ability of Little League Baseball coaches to relate more effectively to their players. The behavioral guidelines communicated to coaches in the training program were empirically derived from a prelimi­ nary investigation which involved 51 Little League coaches and 542 of their players (Smith, Smoll, & Curtis, 1978). In this investigation, a behavioral assessment system was used to categorize.the behaviors of the coaches during an Adapted, by permission, from R.E. Smith, F.L. Smoll, and B. Curtis, 1979, “Coach effectiveness training: A cognitive-behavioral approach to enhancing relationship skills in youth sport coaches,” Journal of Sport Psychology I: 59-75: Smith, Smoll, and Curtis average of nearly four complete games. At the conclusion of the season, the children were interviewed individually in their homes to obtain measures of their perception and recall of their coaches’ behaviors and of their evaluative reactions to the coach, teammates, and other aspects of their athletic experience. On the basis of empirical relationships between observed coaching behaviors, players’ perceptions and recall of such behaviors, and player attitudes, a series of behavioral guidelines were developed (Smoll, Smith, & Curtis, 1977). In the present study, an attempt was made to transmit these guidelines to coaches and to promote their utilization. The intervention program was conceptualized within a cognitive-behavioral framework (cf. Bandura, 1977). The techniques chosen were designed to make coaches more aware of their behaviors, to create expectancies concern­ ing the likely consequences of various coaching behaviors, to increase their desire to generate certain consequences rather than others, and to develop or enhance their ability to perform desirable behaviors effectively. It was expected that cognitive changes of this nature would promote and mediate positive changes in overt coaching behaviors. The training package involved a number of techniques. In addition to verbal and written presentation of behav­ ioral guidelines, modeling, behavioral feedback, and self-monitoring were employed. These methods, singly and in combination, have proven to be effective behavior change procedures in a variety of intervention contexts (e.g., Edelstein & Eisler, 1976; Gottman & McFall, 1972; McFall & Twentyman, 1973). The effects of the train­ ing program on coach behaviors and player perceptions, attitudes, and self-esteem were assessed. Additionally, the role of self-esteem as a moderator variable affecting children’s reactions to trained and untrained coaches was investigated. In a previous investigation (Smith, Small, & Curtis, 1978), it was shown that children low in general self-esteem were most responsive to differences in coaching behaviors in terms of their attitudes toward their coaches. Since the behavioral differences to which they were respon­ sive were an explicit focus of the training program, it was hypothesized that differences in attitudes toward trained versus untrained coaches would be most pronounced for low self-esteem children. Method Subjects The initial sample consisted of 34 Seattle-area male Little League Baseball coaches. All of the coaches were involved at the major (10- to 12-year-olds) and senior (13- to 15year-olds) levels of the program. They coached in three leagues that had participated in the earlier investigation of r elationships between coaching behaviors and players’ reactions to their Little League experience (Smith et al., 1978). Eighteen coaches were randomly assigned to an experimental group and 16 were assigned to a no-treat­ ment control condition. The unequal group sizes were to allow for a sufficiently large experimental group in case of no-shows for the training program. All of the experimen­ tal coaches attended the training session. However, three coaches in the control condition were lost during the course of the season due to team mergers or changes in residence. The final design, therefore, consisted of an experimental group of 18 coaches and a no-treatment control group of 13. The mean age of the coaches was 36.10 years (SD = 9.99). They had an average of 8.37 years of coaching experience (SD= 6.11). Training Procedures The coaches in the experimental group were contacted by telephone and invited to participate in an evening training session. They were told that the results of the previous study conducted within their leagues would be described and that coaching guidelines would be presented and discussed. The training session lasted approximately 2 hours and was conducted by the authors. The relationships that had been found between specific coaching behaviors and children’s attitudes toward their coach, teammates, and other aspects of their sport involvement were described, as was the relationship that had been found between winning and player attitudes. The research results had shown that winning percentage was essentially unrelated to liking for the coach and desire to play for him again, but that certain coaching behaviors were highly related to these measures. The coaches were told that these behavioral findings were the basis for the guidelines. The behavioral guidelines were presented both verbally and in written materials given to the coaches. The verbal presentation was supplemented by modeling of both desir­ able and undesirable methods of responding to specific situations (e.g., player mistakes) by the experimenters. In general, the guidelines stressed the desirability of reinforce­ ment, encouragement, and technical instruction designed to elicit and strengthen desirable behaviors. The explicit goals of the guidelines were to increase positive interactions between coach and players, as well as among teammates, and to reduce fear of failure among players. The following, excerpted from the written materials given to the coaches, summarizes the major guidelines: I. Reactions to player behaviors and game situations: A. Good plays Do: REWARD!! Do so immediately. Let the players know that you appreciate and value their efforts. Reward effort as much as you do results. Look for positive things, reward them, and you’ll see them increase. Remember, whether the kids show it or not, the positive things you say and do stick with them. Don’t: Take their efforts for granted. Coach Effectiveness Training: A Cognitive-Behavioral Approach to Enhancing Relationship Skills in Youth Sport Coaches B. Mistakes, screw-ups, boneheaded plays, and all the things the Cincinnati Reds seldom do Do: ENCOURAGE immediately after mistakes. That’s when the kid needs encouragement most. Also, give corrective INSTRUCTION on how to do it right, but always do so in an encouraging manner. Do this by emphasizing not the bad thing that just happened, but the good things that will happen if the kid follows you instruction (the “why” of it). This will make the player positively self-motivated to correct the mistake rather than negatively motivated to avoid failure and your disapproval. Don’t: PUNISH when things are going wrong. Punishment isn’t just yelling at kids; it can be any indication of disapproval, tone of voice, or action. Kids respond much better to a positive approach. Fear of failure is reduced if you work to reduce fear of punishment. C. Misbehaviors, lack of attention Do: Maintain order by establishing clear expecta­ tions. Emphasize that during a game all members of the team are part of the game, even those on the bench. Use REWARD to strengthen team partici­ pation. In other words, try to prevent misbehaviors from occurring by using the positive approach to strengthen their opposites. Don’t: Get into the position of having to constantly nag or threaten the kids i!l order to prevent chaos. Don’t be a drill sergeant. If a kid refuses to coop­ erate, quietly remove him from the bench for a period of time. Don’t use physical measures (e.g., running laps). The idea here is that if you establish clear behav­ ioral guidelines early and work to build team spirit in achieving them, you can avoid having to repeat­ edly KEEP CONTROL. Remember, kids want clear guidelines and expectations, but they don’t want to be regimented. Try to achieve a healthy balance. II. Getting positive things to happen: Do: Give INSTRUCTION. Establish your role as a teacher. Try to structure participation as a learning experience in which you’re going to help the kids develop their abilities. Always give INSTRUC­ TION in a positive fashion. Satisfy your players’ desire to become the best ball players they can be. Give instruction in a clear, concise manner and, if possible, demonstrate how to do it. Do: Give ENCOURAGEMENT. Encourage effort, don’t demand results. Use it selectively so that it is meaningful. Be supportive without acting like a cheerleader. Do: Concentrate on the game. Be “in” the game with the players. Set a good example for team unity. Don’t: Give either INSTRUCTION or ENCOUR­ AGEMENT in a sarcastic or degrading manner. Make a point, then leave it. Don’t let «·encourage­ ment” become irritating to the players_ To supplement these guidelines, a written brochure given to the coaches contained concrete suggestions for communicating effectively with players, gaining their respect, and relating to parents. The importance of sensitiv­ ity and being responsive to individual differences among players was also stressed. In addition to the information-modeling portion of the training program, behavioral feedback and self-monitoring procedures were employed to increase self-awareness and to encourage compliance with the coaching guidelines. Both procedures were designed to increase the coaches’ awareness of their own behavioral patterns and to repeat­ edly focus their attention on the guidelines. Behavioral feedback was provided in terms of a 12-category behav­ ioral assessment system known as the Coaching Behavior Assessment System (CBAS) (Smith, Smoll, & Hunt, 1977a). The CBAS is described in greater detail below. The coaches were observed during the first 2 weeks of the season by trained coders and were then mailed behavioral profiles reflecting their behavioral patterns during two complete games. The profiles, based on a mean of 219 behaviors per game, indicated the percentage of behaviors falling into each of the 12 CBAS categories. The feedback sheet also included norms derived from the previous year’s behavioral data of 51 coaches across 202 games. One explicit behavioral goal was communicated: Coaches were urged to increase their reinforcement rate to 25% of their responses. In addition to receiving behavioral feedback, coaches also monitored their own desirable behaviors, Given the impracticality of asking coaches to monitor and record their behaviors during games, the coaches were given a brief self­ monitoring form which they completed immediately after each of their first 10 games of the season. On these forms, they indicated approximately what percentage of the time they engaged in the recommended behaviors in relevant sit­ uations. Self-monitoring was restricted to desired behaviors in light of evidence that the tracking of undesired behaviors can be detrimental to effective self-regulation (Cavior & Marabotto, 1976; Gottman & McFall, 1972; Kirschenbaum & Karoly, 1977). The self-monitoring forms were returned in stamped envelopes provided by the experimenters. The coaches were contacted periodically to remind them to complete and return the self-monitoring forms. All of the coaches returned their completed forms. Evaluation Procedures In order to assess the effects of the training program on coaches and their players, the experimental and controi coaches were compared in terms of observed behaviors during games, players’ perceptions of their behaviors, and Smith, Smoll, and Curtis player attitudes toward themselves, the coaches, teammates, and the sport. Observed behaviors: Overt coaching behaviors were assessed by means of the CBAS. This behavioral assess­ ment procedure was developed to permit the direct obser­ vation and coding of coaches’ behavior during practices and games. The CBAS encompasses two major classes of behaviors. Reactive behaviors are responses to immediately preceding player or team behaviors, whereas spontaneous behaviors are initiated by the coach and are not a response to an immediately preceding event. These classes are roughly analogous to the distinction between elicited behaviors (responses to identifiable stimuli) and emitted behaviors (behaviors that do not have clear-cut antecedents). Reactive behaviors are responses to either desirable performances, mistakes, or misbehaviors on the part of the players, while the spontaneous class is subdivided into game-related and game-irrelevant behaviors initiated by the coach. The system thus involves basic interactions between the situa­ tion and the coach’s behavior. In utilizing the CBAS, observers stationed themselves at a point from which they could observe the coach in an unob­ trusive manner. Observers did not introduce themselves to the coach, nor did they indicate in any way that they would be observing him. They were unaware of which coaches were trained and which were controls. Observations were recorded by writing the behavioral codes (e.g., R, P, TIM) on code sheets designating half-innings as the behaviors occurred. Each coach was observed during four complete regular season games. Although CBAS data may be utilized in a number of ways, we have found the most useful and reliable behavioral index to be the percentage of behaviors within each coding category. The mean percentages across all observations (games) served as the units of analysis. The observers were 16 undergraduates who were trained over a 4-week period. The training program included: (a) extended study of a training manual containing instructions for use of the CBAS and a programmed learning module (Smith, Smoll, & Hunt, 1977b); (b) group instruction in use of the scoring system, including viewing and discussion of an audio-visual training module (Smith, Smoll, Hunt, & Clarke, 1976); (c) written tests in which trainees were required to define the CBAS categories and score behav­ ioral examples; (d) the scoring of videotaped sequences of coaching behaviors; and (e) extensive practice in the use of the CBAS in actual field settings. A high degree of demonstrated expertise in the use of the CBAS was required before an observer was permitted to collect research data. In their field codings, the observers demonstrated a median inter-rater reliability coefficient of .94. Player; perceptions and attitudes. The players’ percep­ tion/recall of the coaches’ behaviors and their attitudes toward the coach and other aspects of their participation were assessed in structured interviews conducted at the con- clusion of the season. A total of 325 male players (82% of those who played for the experimental and control coaches) were individually interviewed in their homes by trained interviewers who had not been involved in the behavioral observations of the coaches and who were unaware of the experimental conditions. Children and their parents were assured that their data were confidential and that coaches would be given no information about the data obtained from their teams. The measure of the players’ perception of the coach’s behavior was presented as a recall test in the hope of minimizing distortions in reporting: “We’ve observed your coach, and now we want to see how well you and your team­ mates can remember what he did.” The player was given a description and examples of each of the 12 CBAS behaviors and indicated on a 7-point scale ranging from Never to Almost always, how frequently his coach had engaged in that behavior in situations like those described. Following the recall section of the interview. the children indicated reactions to their participation and ability-related perceptions. The children were given a clipboard and recorded their own responses on a series of 7-point scales in such a way that the interviewer could not see them. The following questions were asked: 1. How much do you like playing baseball? 2. How much did you like playing for your coach? 3. How much would you like to have the same coach again next year? 4. How much does your coach know about baseball? 5. How good a baseball teacher is your coach? 6. How well did the players on your team get along? 7. How good are you in sports? 8. How good are you in baseball? 9. How good does your coach think you are in base­ ball? 10. How good do your teammates think you are in baseball? 11. How good do your parents think you are in baseball? The scales relating to liking ranged from Dislike a lot to Like a lot. Item 4’s scale ranged from Almost nothing to Almost everything, item 5’s from Very poor to Excellent, and that for item 6 from Very poorly to Very well. Items 7 through 11 were rated on scales ranging from Very poor to Excellent. Assessment of self-esteem. As part of the postseason interview, the players were administered an adaptation of Coopersmith’s (1967) Self-Esteem Inventory as a mea­ sure of general or “global” self-esteem. It consisted of 14 descriptive statements, each of which was rated on a 5-point scale ranging from Not at all like me to Very much like me. Six of the items referred to positive attributes (e.g., “I’m Coach Effectiveness Training: A Cognitive-Behavioral Approach to Enhancing Relationship Skills in Youth Sport Coaches pretty sure of myself’; “I’m proud of myself’), whereas 8 were negative self-evaluative statements (e.g., “I have a low opinion of myself’; “I’m a failure”). The scale was designed to provide a maximum range of scores (14 to 70) with a relatively small number of items. The scale has adequate interitem reliability (alpha coefficients ranging from .63 at ages IO to 12 to .70 at ages 13 to 15). Test-retest reliability coefficients over 12 months were .60 at ages l O to 12 and .74 at ages 13 to 15. Results Comparability of Experimental and Control Coaches In order to assess the preseason comparability of the experi­ mental and control group coaches, they were compared on a number of variables which might possibly affect their behavior and their players’ perceptions and attitudes. No significant differences were found between the two groups of coaches in age, number of years of total coaching experi­ ence, and number of years coaching baseball. Data were also available from the previous season on all the dependent variable measures for 13 coaches in the experimental group and 6 members of the controi group. At the level of observed CBAS behaviors, the two groups of coaches did not differ on any of the behaviors except General Communication, in which the experimental coaches engaged relatively more frequently (p < .05). On the corresponding player-perceived behaviors, there were no significant differences between the experimental and control coaches on any of the 12 behavior categories. Finally, the two coach groups did not differ significantly on any of the player attitude measures from the previous season or on won-lost record. Given that the remaining coaches on whom previous year’s data were not available were also randomly assigned to the treatment conditions, we may assume that the experimental and con­ trol groups were quite comparable. Observed CBAS Behavior Differences A total Qf 26,412 behaviors were coded during four game observations of the experimental and control coaches. Each coach averaged 213.19 codable behaviors per game. Since the time duration of the games varied, an initial analysis of rate scores was conducted. The frequency data within the CBAS categories were converted to rate scores by dividing the total behavior frequencies by the number of minutes observed across the four games. The rate scores for the experimental and control group coaches did not differ significantly on any of the 12 behavior categories, nor did they differ on the total of the combined categories. Because the two groups did not differ in their level of activity, sub­ sequent analyses focused on the distribution of behaviors within the categories. In order to identify the combination of behaviors which discriminated optimally between the experimental and control coaches, a stepwise discriminant analysis was conducted (Cooley & Lohnes, 1971). Behavior categories were thus entered into the analysis on the basis of their dis­ criminating power. Maximum separation between the group centroids (multivariate group composite scores) occurred with a.weighted linear combination of the following seven behaviors: Reinforcement, General Communication, Keeping Control, Nonreinforcement, Mistake-contingent Encouragement, Organization, and Punishment (Wilks’ Lambda = .52, p < .02). However, it is worth noting that after Reinforcement was entered on the first step and sig­ nificance was attained, differences between group centroids were not significant for the step-down multivariate F ratios on Steps 2 through 5. Given the small number of subjects relative to the number of variables being combined thereaf­ ter, the most conservative and appropriate conclusion is that Reinforcement was the major discriminator between the two groups. This conclusion is consistent with the results of univariate F tests which yielded a significant group dif­ ference only for the Reinforcement category. Differences in Players’ Perceptions of Coaching Behaviors A stepwise discriminant analysis of the behavioral ratings made by the two groups of children revealed a significant difference.in group centroids based on the 1 2 behaviors (Wilks’ Lambda= .91, p < .002). Follow up univariate ANOVAs yielded significant group differences on six of the perceived behavior categories. Experimental group coaches were rated as more frequently engaging in Reinforce­ ment, Mistake-contingent Encouragement, and General Technical Instruction, and as less frequently engaging in Nonreinforcement, Punishment, and Punitive Technical Instruction. All of these differences were consistent with the behavioral guidelines. Player Attitudes and Self-esteem Evaluative reactions to coach and teammates. These data indicate that the children who played for the trained coaches did not differ in liking for baseball. On the other hand, they indicated greater enjoyment in having played for their coaches and a stronger desire to play for them in the future. They also rated the trained coaches as better teachers of baseball. Finally, children who played for trained coaches evaluated the relationships which existed among teammates more positively. Postseason self-evaluations. Postseason measures of general self-esteem and athletic self-evaluations of children who played for trained and untrained coaches were compared by means of one-way ANOVAs. On the measure of general self-esteem, no significant group dif­ ference was found, F(l, 323) = .78. Likewise the children’s evaluations of their own baseball ability did not differ, F(l, 316) = .0 I. There were, however, significant differences in the children’s perceptions of their coaches’ and team­ mates’ evaluation of their skills. Children who played for Smith, Smoll,and Curtis the trained coaches felt that both their coach F(l, 317) = 4.73, p < .05, and their teammates, F(l, 315) = 6.05, p < .02, evaluated their skills more highly. No difference was found in the children’s perception of their parents’ evalu­ ations of their skills. Pre-post self-esteem changes. As reported above, the total samples of children (N = 325) who played for the trained and untrained coaches did not differ in postseason self-esteem scores. However, self-esteem data obtained in similar interviews conducted the previous year were available for 75 of the children who played for the control group coaches and 112 of those who played for the trained coaches. It was therefore possible to examine changes in self-esteem scores for these subsamples. An analysis of covariance of the posttreatment scores, using baseline self-esteem scores as covariate, revealed a significantly higher level of self-esteem in the children who had played for the trained coaches, F(I, 184) = 6.43, p < .01. As an additional test of treatment effects, changes in self-esteem scores for the two groups were assessed separately by means oft tests for correlated means. These tests revealed a significant increase in scores for the chil­ dren who played for the trained coaches, t( 111) = 2.07, p < .05. The control group children exhibited no significant change in scores,t(74)- 1.70. Self-esteem as a moderator variable. A question of both applied and theoretical significance concerns the impact of the trained and untrained coaches on the evaluative reac­ tions of children who differ in general self-esteem. Given that the total groups (including the subsamples on whom scores from the previous year were available) of children who played for the two groups of coaches did not differ in postseason self-esteem,the use of the self-esteem scores as a moderator variable was statistically justifiable. Likewise, within the total sample self-esteem was unrelated to any of the attitudes toward the coach. The mean self-esteem score for the entire player sample was 51.87 (SD= 6.79). The total player sample was divided into high, moderate, and low self-esteem groups; the low self-esteem group was comprised of players having scores of less than 48, while the highs had scores of 55 or above. The numbers of low, moderate, and high self-esteem children who played for the trained coaches were 44, 71, and 70, respectively. The corresponding cell sizes for the untrained coaches were 46, 41, and 52. The self-esteem levels were crossed with the coach groups to create a 3 x 2 factorial design. Because of unequal cell sizes, ANOVAs for unweighted means were computed. The four attitude items concerning enjoyment in playing for the coach, desire to play for him again, and evaluation of his knowledge and teaching skill all intercorrelated above .80 and were therefore combined into a single dependent variable measure reflecting overall evaluation of the coach. The ANOVA yielded a significant main effect for coach groups,F(1, 317) = 11.84, p< .00 l . Neither the main effect for self-esteem nor the self-esteem x coach groups interac- tion was significant. Because of the a priori prediction that attitudinal differences between children who played for trained as opposed to untrained coaches would be great­ est at the low self-esteem level, F tests of simple group effects were computed at each self-esteem level using the procedure recommended by Winer ( 1971, p. 387). These analyses revealed a significant groups effect only at the low self-esteem level, F(1, 88) = 8.54,p< .01. It is interesting to note that a corresponding pattern of results occurred on the player-perceived measures of Mistake-contingent Encour­ agement, Punishment, and General Technical Instruction. That is, as noted above, significant main effects for coach groups were obtained for all three perceived behavior categories. But the simple effects at each self-esteem level were significant only in the case of the low self-esteem groups (p < .0 l for Mistake-contingent Encouragement; p’s< .05 for Punishment and General Technical Instruction when tested with the Newman-Keuls procedure). ANOVAs of the other attitudinal items (intrateam attrac­ tion and liking for baseball) yielded significant effects only on the intrateam attraction variable. On the measure of how well teammates got along with one another, significant main effects were found for both coach groups, F( l, 318) = 9.33, p< .01,and self-esteem, F(2, 318) = 3.72,p< .05. Children who played for the trained coaches rated their teams as higher in intrateam attraction, as did children high in self­ esteem. The coach group x self-esteem interaction was not significant. However, Newman-Keuls comparisons of the intrateam attraction means for the two coach groups at each level of self-esteem disclosed that the ratings of the chil­ dren who played for the trained coaches were significantly higher than were those of children who played for untrained coaches only at the low self-esteem level (p < .05). Team Records Given the strong attitudinal differences found between children who played for the two groups of coaches, it seemed important to examine the potential influence of won-lost records. The trained coaches had a mean winning percentage of 54.5% (SD = 24.72), whereas the control coaches won 44.7% of their games (SD= 26.91). The dif­ ference in winning percentages did not approach statistical significance. As the large standard deviations indicate,both groups contained both successful and unsuccessful coaches in terms of winning games. Discussion The results of the present study indicate that the experi­ mental training program exerted a significant and positive influence on overt coaching behaviors, player-perceived behaviors, and children’s attitudes toward their coach, teammates, and other aspects of their athletic experience. There is also evidence that positive changes in self-esteem occurred in children who played for the trained coaches and on whom previous scores were available. Coach Effectiveness Training: A Cognitive-Behavioral Approach to Enhancing Relationship Skills in Youth Sport Coaches Despite the fact that there was no significant difference between the won-lost records of trained and untrained coaches, children who played for the trained coaches evalu­ ated their coach and the interpersonal climate of their teams more positively. These evaluative differences were related to observed and player-perceived behavioral differences between the two groups of coaches which were consistent with the behavioral guidelines and with coaching behavior­ attitude relationships found in earlier research (Smith et al., 1978). The fact that the trained and untrained coaches on whom data from the previous year were available did not differ at that time on any of the guideline behaviors (either observed or player-perceived) or on player attitudes strongly suggests that the training program was responsible for the group differences obtained in the present study. Of the observed behavioral differences between the trained and untrained coaches, only the relative frequency of Reinforcement was significant. This was the behavior that was most highly emphasized and urged in the train­ ing program. As noted above, maximum discrimination between the trained and untrained coaches occurred with a combination of seven behaviors, six of which were direc­ tionally consistent with the training guidelines. However, because of the large number of variables being combined and the relatively small number of coaches, these latter results are best viewed as suggestive and deserving of replication with a larger coach sample. The player-perception data yielded more imposing group differences, perhaps because they were based on a far larger sample of the coaches’ 6ehavior. In line with the behavioral guidelines, trained coaches were seen by their players as more reinforcing in response to desired behav­ iors, more encouraging and less punitive in response to mistakes, and as more technically instructive. All of these behaviors have previously been shown to relate to attraction toward coach and teammates (Smith et al., 1978). In the present study, players who played for the trained coaches evaluated both the coach and the team’s interpersonal cli­ mate more positively. The latter finding suggests that the “positive approach” to relating to players and developing team cohesion resulted in more positive interactions among players. It is also interesting to note that the children who played for the two coach groups did not differ in liking for baseball. Evidently, attitudes toward the sport were fairly well established and not readily affected by coach­ ing behaviors. Self-esteem was found to be an important moderator of attitudinal responses to coaching behaviors. Consistent with predictions derived from previous research (Smith et al., 1978), low self-esteem children exhibited by far the greatest difference in attitudes toward the trained and untrained coaches. They also perceived the greatest differ­ ence between trained am;I untrained coaches in frequency of Mistake-contingent Encouragement, Punishment, and Gen­ eral Technical Instruction, suggesting that low self-esteem children may be particularly sensitive to variations in such behaviors. Whether these perceived differences reflect more sensitive perception of such behaviors or greater impact of such behaviors upon them, or both, is a question worthy of future empirical attention. Whatever the factors involved, we find this pattern of results to be noteworthy, since it is the low self-esteem child who is probably in greatest need of a positive athletic experience and who -appears to respond most favorably to desirable coaching practices and most unfavorably to negative practices. The role of self-esteem in mediating responses to coaching (and other adult) behaviors and the manner in which specific kinds of athletic experiences affect self-esteem are issues deserving of further inquiry. Previous research (Smith et al., 1978) has shown that children who have played for highly reinforcing and encouraging coaches have significantly higher levels of postseason self-esteem that do those exposed to coaches who do not behave in this manner. In the present study, it was shown that children who played for coaches trained to behave in this fashion evidenced significant increases in self-esteem scores as opposed to those who played for untrained coaches. Given that these changes were shown only in a subsample on whom scores from the previous year were available, replication of these results is highly desirable, as is an assessment of the stability of such changes. Nonetheless, these results suggest that training programs designed to assist coaches, teachers, and other adults occupying leadership roles in creating a positive and supportive environment can influence children’s personal­ ity development in a positive manner. The cognitive-behavioral approach used in the present study had a number of treatment components-informa­ tion, modeling, self-monitoring, and behavioral feedback. As noted above, each of these individual components has been demonstrated in other research to be an effective means of producing positive behavior change and skill acquisition with a wide range of populations in diverse settings. The present attempt was to develop a potentially effective training package and to assess its effects. Thus, the training components are necessarily confounded. Future research is needed to assess the relative efficacy of the components singly and in combination with one another as well as the cognitive processes which underlie their effects. From a practical point of view, for example, it would be desirable if behavioral feedback were not a necessary train­ ing component, since this was the most costly (in time and effort) aspect of the present program. Practical considerations (namely, the limited number of coaches available to study) precluded the use of an atten­ tion-placebo control condition in the present study. Thus, we cannot know the extent to which nonspecific factors might have combined with the treatment components to produce the obtained differences. In future studies, credible attention-placebo control conditions in combination with unconfounding of treatment components would permit more definitive and theoretically meaningful conclusions Smith,Smoll,and Curtis to be drawn concerning the efficacy of the treatment components. However, the lack of an attention-placebo control group does not detract from the practical reality that the training program produced desirable behavior and attitudinal effects. References Bandura, A. Social learning theory. Englewood Cliff, N.J.: Prentice-Hall,1977. Cavior,N.,& Marabotto,C.M. Monitoring verbal behaviors in a dyadic interaction. Journal of Consulting and Clinical Psychol­ og� 1976,44,68-76. Cooley,W.W.,& Lohnes,P.R. Multivariate data analysis. New York: Wiley, 1971. Coopersmith, S. The antecedents of se/f­ esteem. San Francisco: Freeman, 1967. Edelstein,B.A.,& Eisler,R.M. Effects of modeling and modeling with instructions and feedback on the behavioral components of social skills. Behavior Therapy, 1976,7,382-389. Gottman, J.M., & McFall, R.M. Self-monitoring effects in a program for potential high school dropouts: A time series analysis. Journal of Consulting and Clinical Psychology. 1972, 39,273-281. Kirschenbaum, D.S., & Karoly, P. W hen self-regulation fails: Tests of some preliminary hypotheses. Journal of Consulting and Clinical Psychology, 1977,45,1116-1125. Little League Baseball. Official rules. Williamsport,Pa.: Author, 1977. McFall, R.M., & Twentyman, C.T. Four experiments on the relative contribuations of rehearsal,modeling,and coaching to assertion training. Journal of Abnormal Psychology, 1973,81, 199-218. Smith, R.E., & Smoll, F.L. Sport and the child: Conceptual and research perspectives. In F.L. Smoll & R.E. Smith (Eds.), Psychological perspectives in youth sports. Washington,D.C.: Hemisphere,1978. Smith, R.E., Smoll, F.L., & Curtis, B. Coaching behaviors in Little League Baseball. In F.L. Smoll & R.E_ Smith (Eds.), Psychological perspectives in youth sports. Washington,D.C.: Hemisphere,1978. Smith,R.E.,Smoll,F.L.,& Hunt,E. A system for the behavioral assessment of athletic coaches. Research Quarterly, 1977, 48, 401-407. (a) Smith, R.E., Smoll, F.L., & Hunt, E.B. Training ��nual for the Coaching Behavior Assessment System. JSAS Catalog of Selected Documents in Psychology, 1977,7, 2. (Ms. No 1406) (b) Smith,R.E.,Smoll,F.L.,Hunt,E.B.,& Clarke,S.J. CBAS audio visual training module. Seattle: University of Washington, 1976. (Film) Smoll,F.L.,Smith,R.E.,& Curtis,B. Coaching roles and rela­ tionships. In J.R. Thomas (Ed.), Youth sports guide for parents and coaches. Washington,D.C.: American Alliance for Health, Physical Education,and Recreation,1977. Winer,B.J. Statistical principles in experimental design (2nd ed). New York: McGraw-Hill, i 971.  KINE4000  Dr. Scott Martin 1 Project 1: Article Evaluation Form 1=Completely Incompetent, 2= Poor, 3=Mediocre, 4=Good, 5=Excellent Characteristics 1. Problem is clearly stated 2. Hypotheses are clearly stated 3. Problem is significant 4. Assumptions are clearly stated 5. Limitations of the study are stated 6. Important terms are defined 7. Relationship of the problem to previous research is made clear 8. Research design is described fully 9. Research design is appropriate for the solution of the problem 10. Research design is free of specific weaknesses 11. Population and sample are described 12. Method of sampling is appropriate 13. Data-gathering methods or procedures are described 14. Methods or procedures are appropriate to the solution of the problem 15. Data-gathering methods or procedures are utilized correctly 16. Validity and reliability of the evidence gathered are established 17. Appropriate methods are selected to analyze the data 18. Methods used in analyzing the data are applied correctly 19. Results of the analysis are presented clearly 20. Conclusions are clearly stated 21. Conclusions are substantiated by the evidence presented 22. Generalizations are confined to the population from which the sample was drawn 23. Article is clearly written 24. Article is logically organized 25. Tone of the article displays an unbiased, impartial scientific attitude Adapted from the Handbook in Research and Evaluation, Isaac and Michael (1983) 1 2 3 4 5
Purchase answer to see full attachment