" “3342' '2. 'o '1 1‘1“./.,,‘ ,‘fzh-H \.:. Lf.‘ :I";?,; ,IQIP'VY ’0? 11 . F1" ,‘Miy .In""- a v. . . _1 :I '1.“ magi. (1:. It; hf“, ' < ’L‘ 39- 1 19103-1 fi-‘Q- L- Q o 53:3; ‘ 1. 4.- V 0'.‘ - '41,). 1:1,, $4.11,... - 1A3: ‘1" 1 ‘1‘ “31' a k- . . ,1.- 1w, .. I 3"1:,1”"‘,-n,<)' ’i' 315:9?“ .3; - 1-1’ 1.30.4..- 4.:— ..-. l?4-v-'v“ ‘1 11" '- ‘f'fnu .4,’ "Q". 1 . Ir- 2 ' - p . c‘ ' 1 . ‘ .LJ ‘4 V’Q:"‘ ' - ' ‘ ' H y- . .. .‘ , , 1” - v_ .‘,|a,\fi,rc‘ “-1 V" 1 '1 4 ’ ' "' Mr '1. 32“,", :1 ,' ‘.1. V . . 1 : . , , . "5” ~, 0:... .w .. 1 . 1 ”'1 . . \ . '(sz ., P. .Q: . ’ UR}. ' '1' ’ ' J ' \ .‘ (”pt 2‘. u‘. .5» "V .4 " - .2 1 .1" ,4-1N" '1 ‘1 3r 1 .. . («\1’1' 7‘ 13.4%, , . 1 a- .- ‘4.’ 291 ""‘,;.",, .541.“ I . 4‘ '31:.I, ‘1- 1' . '1 I- , 2.“. f’ ‘ .- . 1 111‘2' . 1 -..1 ,h 2.1 .3, ‘41 . ".r- E" III: 1:21:13 '9'; ' . - XQN'H‘IIC ‘ Am?) 3714 1.’ U." f. . ,, , 1:11;“. 'r " lift! g” L. 0'4 111;:‘1 -- 41 i 1-?" W" '3 ' ’ {1’5" 1. . 1"" ”k f- . 1} .1 . 4 '9}? 13.61 ".9. .1 O, {1, .1 ""' .. ’5‘”, / ., 1 (g r , 1 , . A 1‘ p‘ - J. 22:): , 04,1 . (1,, p 2 , 1 Thai/1?“ ‘ 4.5%?" . ‘1: or"; 9;. 5‘35”: ' 5’ mil 1 ‘a-v‘i. 2,11,? fi‘q’l’h‘ '7.“ .-. 7/ /.1' n if“ r .3. (Cl/”1' 1 (av, :36 11:, .~ '1,” Eh? “GIN” V12 1' (Qt/2W '_1 "a 'J.“" '2' ~ ; 0 ) 11 » . 1 I 9691645711,: 4 . - , 1- 1,. , 11,132,. 2.,vz'1. geifizétifqfl" 1., "‘ ‘ 4291‘“. Wed . }. 4:4“ 334,-.."htzf4191i: a}? 1le.,“‘ 'I “‘7,” - , hi“; z"‘ n. 1,1, A (“'1“) \ v {40‘210' air- ‘l '/. “1,11,,” 4 0,! ..".'.' #32325} ." .i .4 ,, . ‘ iu' ., l- ‘1' "iofié‘a u "1“,; kiwi r "‘(f l 1 I 1%,)" 'I' "I." ““J‘. ' I1 ‘ ‘ - -- 1"" (2;, l 1‘}. ("9" "r 7' ‘5'" ‘ - " (231;) '1 QJ?';':‘:\¥'-'1'|’--Q". V. l." / "ZN-35131, . I. v.12“ $7,} ,1 2", ’E’g:“‘r. » j,“ 9, 1/1,? ' ’.;.'"11'1}(" Q_ 1 [1:5H;:.(f ' [‘1‘ . 1 . .- ‘A‘é/rfll/ . NBC #1. 9 ;-’E10' ) "f?" 1"!" . IA,~ , $533X1: {2: ) ‘~)‘5?;|‘ '1‘?J"§"7 “131), r" '-"£Q r J {4 ,‘r at}? ‘, ‘41:. - ”a," 'b , ., 1- fl‘fiifi, !‘ n r I}??? “1".4'4'. .5151} , kg.” ‘ A {‘2‘ 2;"“1’45-{9 ",2 “FF 7“;qu Ari-8‘.- fii{51::y',':,,.1 '} ?/ ,1, l ”A? 31". 3'4} \‘44 £7, .2 ,1; .., "'¢. ’9 ' ‘ _ ' ‘ l 1} W; (if; gl-l‘ .flt’j‘f .{ .~ 1 a? {1 :51 4%” 1" M A. d‘ .. .-‘ a¢..,1,,fifl.. ” 4353:? Hfldfii *1 314.5%? $3115.11mgifl', /‘ v. M42}: .1 M1, 3..., 1L .9 ., ,2»? A? . ,4’ ‘3‘ 3;]:- ‘J ' "21513”? 1?; #1235“ I I: .. ,‘.' ;‘*‘1 .. 33%? ,1“; 1.] mg (1,, an,” ,1, 4,. :1 ' - , 1 . 42' egg-1‘ f 1 ’ ., ’1‘, . 1'. 2 . . ' ' 4 {,m .M ’. *1- ' _ ' a. l HIGAN SYATE l lllunlljljlifilllllllllm} mm by. [fill 93(’) A.— (Ill LIBRARY ““5 Michigan “on Unlvonlty This is to certify that the thesis entitled In LC‘\‘1710.+1'L‘ HQ 1 VII [u C a C t-V/4LM H7/VC' firmer/xvii INFLMELI £1 0F r’lzms‘c’ MUD BLAME CIV [7L/KCUH 77c‘k’5~ 0 F AGIL /TY “U ’Ifif §Pcir SETTING presented by GHRY ALMLL/n/VlSoAJ has been accepted towards fulfillment of the requirements for ("4 5' degree in 6/“ ( A f / out; 724%; 49 Mg; Major professor Date fl‘flé‘5’? 0-7639 MS U i: an Afi'num'w Action/Equal Opportunity Institution )V1ESI_J RETURNING MATERIALS: Place in book drop to LIBRARIES remove this checkout from .‘nunzs-I-L. your record. FINES will be charged if book is returned after the date stamped below. “39-15:.“ .~ ~ -—' ‘ 9m 8 3332. 'Z ‘ " _ .2 ‘4' 3i ,. '-«;:‘..'. see 2 719931 . ‘ fl 1 ’1,“ ~_' It ! ‘ { ($511 Him INFORMATIONAL VALUE OF EVALUATIVE BEHAVIOR INFLUENCES OF PRAISE AND BLAME ON PERCEPTIONS OF ABILITY IN THE SPORT SETTING by Gary Williamson Thesis Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS School of Health Education. Counseling Psychology. and Human Performance 1987 ABSTRACT INFORMATIONAL VALUE OF EVALUATIVE BEHAVIOR OF PRAISE AND BLAME ON PERCEPTIONS OF ABILITY IN THE SPORT SETTING By Gary Williamson This study investigated the degree to which praise and blame provide information and affect childrens self -pereeptions of ability. The purpose of this study was to replicate a previous experiment conducted by Meyer, Bachmann, Biermann, Hempelmann, Ploger and Spiller (1979) and to extend their experiment number three to include (a) a hypothetical sport task and (b) children aged 9 to 15 years. One hundred and forty subjects were given descriptions of two students, one of which they had to imagine was themselves, who obtained identical results in performing easy and difficult motor tasks. One of the students received neutral feedback and the other student was praised for success or criticized for failure. The results failed to replicate the findings of Meyer et al. (1979). Specifically, the results of this study found. that praise after success led to the conclusion that the acting person’s ability was viewed as high. Blame after failure led to the perception that the acting person’s ability was low. However, support was found for Meyer et al.’s (1979) findings in experiments number one and two which reported differences in childrens’ self -perceptions of ability with increasing age. The present study found that children aged 9, 10 and 11 years tended to rely on outcome and teacher feedback in judging self -perceived ability. Older children aged 12, l3, l4 and 15 years tended to rely more on outcome and task difficulty in assessing perceived ability. These differences in the ratings of perceived ability with age are discussed in terms of Nicholls’ (1978) developmental theory of perceived ability. Informational Value of Evaluative Behavior Influences of Praise and Blame on Perceptions of Ability in the Sport Setting by Gary Williamson We approve the thesis of Gary Williamson Date of Signature: M/Z/f?’ ”fa/(:42, 6069M r Martha E. Ewing, Ph.D. cfi/I Thesis Adviser, Committee air Assistant Professor School of Health Education, Counseling Psychology and Human Performance Michigan State University a; iii“? W ails Deborah L. Feltz, Ph.p? Committee Member Associate Professor School of Health Education, Counseling Psychology and - Human Performance * ni:::gan State University 41/” V/Jze/l/M Committee Member Professor Director, Youth Sports Institute Michigan State University II ACKNOWLEDGEMENTS I wish to thank my advisor, Dr. Marty Ewing for her conscientious support in this project. I appreciate the time she has spent on the constant reviewing of each chapter. Marty was always available and ready to offer help and advice despite my constant interruptions. Most of all I would like to thank Marty for her infectious enthusiasm for sport psychology and making it possible for me to persue my masters degree in this country. I also wish to thank Dr. Vern Seefeldt, Director of the Institute for the Study of Youth Sports located at Michigan State University, for providing me with a research assistantship which enabled me to pursue my graduate education. I wish to thank my other committee member, Dr. Deborah Feltz for her guidance in the research process and valuable feedback. I am also grateful to Mr. Richard Smith, the principal at Augres Simms School, and teachers who gave their time and went out of their way to assist me in the data collection. I especially want to thank all of the students for their cooperation in filling out the questionnaires. iii CHAPTER Acknowledgements List of Tables List of Figures TABLE OF CONTENTS INTRODUCTION ................................................. Nature of the Problem .................................... Statement of the Problem ................................. Hypotheses ............................................... Definitions .............................................. Limitations .............................................. Implications of the Study ................................ REVIEW OF THE LITERATURE ..................................... Effective Social Reinforcement ........................... Teacher Expectation ...................................... Perceived Ability-A Development Model .................... Informative Function of Evaluative Feedback .............. METHODOLOGY .................................................. Subjects ................................................. Design ................................................... Instrumentation .......................................... Procedure ................................................ Data Collection .......................................... Limitations .............................................. RESULTS ...................................................... Analysis 1: Replication and Extension ................... Discussion ............................................... Analysis 2: Developmental Differences ................... Hypothesis 1 ............................................. Hypothesis 2 ............................................. Hypothesis 3 ............................................. Groups ......................... Analysis 3: Perceived Perceived Perceived Perceived Perceived Perceived Perceived ability ability ability ability ability ability ability Specific Age ratings ratings ratings ratings ratings ratings ratings of of of of of of of nine year old children ...... ten year old children ....... eleven year old children.... twelve year old children.... thirteen year old children.. fourteen year old children.. fifteen year old children... Summary .................................................. iv Page iii vi vii 10 11 ll 12 l3 13 18 23 28 34 34 34 34 35 35 36 37 37 43 44 54 55 55 56 56 58 6o 62 64 67 69 72 CHAPTER Page 5. DISCUSSION, SUMMARY AND RECOMMENDATIONS .................. 73 Summary .............................................. 85 Recommendations ...................................... 86 REFERENCES ....................................................... 88 APPENDICES A.‘ Perceived Ability Questionnaires ........................ 93 B. Parental Consent Form ................................... 97 C. Analysis 2: Means and Standard Deviations for Non- Significant Interactions ................................ 98 D. Data Coding ............................................. 102 E. Raw Data.......................................... ...... 103 TABLE 10. 11. 12. 13. 14. LIST OF TABLES Teacher expectations and their influence on Children in a sport setting ........................ Levels of differentiation of concept of ability from those of difficulty and effort ............... Summary of analysis of Variance for Perceived AbiIity by Outcome, Task Difficulty and Feedback... Means and Standard Deviations for the Perceived Ability Ratings .................................... Means and Standard Deviations for Perceived Ability by Task Difficulty by Outcome .............. Means and Standard Deviations for Perceived Ability by Outcome and Feedback .................... Means and Standard Deviations for Perceived Ability by Task Difficulty and Feedback ........... Means and Standard Deviations for Perceived Ability for Outcome, Task Difficulty and Feedback .......................................... Analysis of Variance for Perceived Ability by Age, Outcome, Task Difficulty and Type of Feedback .......................................... Means and Standard Deviations for the Seven Age Bands by Outcome, Task Difficulty and Feedback ..... Means and Standard Deviations for Age by Feedback.. Means and Standard Deviations for Task Difficulty by Age ............................................. Means and Standard Deviations for Each Age Across Questionnaires ..................................... Means and Standard Deviations for the Young and Old Children for Each Situation .................... vi Page 20 26 38 4O 41 41 41 42 46 47 48 SO 51 S3 FIGURE 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. LIST OF FIGURES Ratings of ability for outcome by feedback by task difficulty ................................................ Ratings of ability for age by feedback .................... Ratings of ability for age by task difficulty ............. Ratings of ability for success outcome by feedback by task difficulty for young and old children ................ Ratings of ability for failure outcome by feedback by task difficulty for young and old children ................ Ratings of ability for success outcome by feedback by task difficulty for the 9 year old children ............... Ratings of ability for failure outcome by feedback by task difficulty for the 9 year old children ............... Ratings of ability for success outcome by feedback by task difficulty for the 10 year old children .............. Ratings of ability for failure outcome by feedback by task difficulty for the 10 year old children .............. Ratings of ability for success outcome by feedback by task difficulty for the 11 year old children .............. Ratings of ability for failure outcome by feedback by task difficulty for the 11 year old children .............. Ratings of ability for success outcome by feedback by task difficulty for the 12 year old children .............. Ratings of ability for failure outcome by feedback by task difficulty for the 12 year old children .............. Ratings of ability for success outcome by feedback by task difficulty for the 13 year old children .............. Ratings of ability for failure outcome by feedback by task difficulty for the 13 year old children .............. Ratings of ability for success outcome by feedback by task difficulty for the 14 year old children .............. Ratings of ability for failure outcome by feedback by task difficulty for the 14 year old children .............. Ratings of ability for success outcome by feedback by task difficulty for the 15 year old children .............. Ratings of ability for failure outcome by feedback by task difficulty for the 15 year old children .............. vii Page 39 39 51 54 56 57 57 59 59 61 61 63 63 65 65 67 67 7O 7O CHAPTER I INTRODUCTION Participation in competitive sport has the potential to exert either a positive or negative effect on the psychological development of the young child depending on the quality of the program itself (Alley, 1974; Martens, 1978; Small & Smith, 1979). A key factor in determining whether sport participation will be a positive or negative experience is the quality of the adult leadership. Both Martens (1978) and Gould (1980) have emphasized the role which coaches can play in influencing young athletes’ cognitive perceptions and attitudes towards athletic participation. Gould (1980) further recommended that coaches should actively plan and initiate coaching strategies designed to facilitate children’s perceptions of competence and self-worth in athletic endeavors. Research in the area of developmental psychology has indicated that the behaviors of parents, teachers and other significant adults influence the development of such attributes as achievement motivation and self -concept in young children (Harter, 1978). Several studies in the athletic setting have highlighted that coaches can, and do, influence the psychosocial development of their young athletes (Horn, 1985; Smith, Smoll & Hunt, 1977). From a practical standpoint, there exists a need to identify which coaching behaviors are most facilitative of the child’s growth. Theoretically, it is necessary to ascertain how these particular coaching behaviors influence the child’s development. -2- One category of instructional behavior which has been examined in relation to its effect on students’ attitudes and affective responses involves teachers’ reinforcement. In an academic setting, Flanders (1970) and Rosenshine (1973), for example, reported on the association between teachers’ feedback and students’ attitudes. They found that teachers’praise is significantly related to students’ positive attitudes and that critical statements by teachers are negatively correlated with such attitudes. A series of studies conducted by Smith and Small and their associates has provided evidence that coaching behaviors influence the cognitions and attitudes of young male athletes (Smith, Smoll & Hunt, 1977; Small & Smith, 1979). The results of these studies revealed that those coaches who exhibited the highest frequencies of behaviors categorized as ”technical instruction," ”reinforcement," and “mistake-contingent reinforcement“ were evaluated more positively by their players than those coaches who scored low in these . categories. Furthermore, players of these highly reinforcing and instructive coaches attained significantly higher scores on tests of post-season self -esteem and had more positive attitudes toward participation than did players of coaches who did not consistently exhibit these behaviors (Smith et al., al.,l977). The delivery of feedback has powerful evaluative implications. Neither the delivery nor the reception of feedback is simply an objective transfer of information; both arouse strong emotions. The quality of the feedback may vary depending on a teacher’s relationship with a pupil. Furthermore, strong emotions aroused in the recipients of feedback have the potential of interfering with or facilitating the reception of feedback. Thus, information communicated by a teacher to a child may directly or indirectly tell each child what the teacher thinks of him/her. .3. Feedback from a teacher has the potential to influence a child’s perception of his/her ability as it relates to actual ability. Although perceptions of one’s ability appear to be based on years of self evaluation and comparison, several studies (Dweck & Repucci, 1973; Haas & Maehr, 1965; Maehr, Mensing & Nafzger, 1962) have demonstrated that the accuracy of self -perceptions for children of the fifth grade level and older may be signifcantly affected by feedback from others, at least temporarily. In these studies, subjects received positive or negative feedback from significant others on a short term basis. Results indicated that a significant change in perceptions of ability ‘ could be of f ected, both positively and negatively, depending on the type of feedback given. In an athletic setting a number of studies have highlighted that there are considerable individual differences in the way in which children interpret self-perceptions of ability based on the feedback provided by adults (Horn & Hasbrook, 1984; Weiss, Bredemeier & Schewchuk, 1981). Moreover, such individual differences have been found, in an academic environment, to indicate a developmental trend (Hatter, 1978; Nichols 6; Miller, 1984). To date very little research has been conducted to examine whether there are developmental differences in childrens’ perceptions of ability, as indicated by their interpretation of positive and negative feedback, in the athletic setting. NATURE OF THE PROBLEM A considerable amount of research has been conducted in the academic classroom for the purpose of identifying the teaching behaviors that are most conducive to children’s academic achievement. Based on the classic study by -4- Rosenthal and Jacobson (1968), the ‘pygmalion’ effect predicts that a teacher’s expectancy of student ability triggers a series of verbal and non-verbal interactions which inadvertently control the student’s subsequent achievement behavior. Research supporting the ‘pygmalion’ effect suggests that high-expectancy students (compared to low-expectancy students) are provided with a warmer emotional climate and given more praise (Cooper & Baron, 1977; Meichenbaum, Bowers & Ross 1969), less criticism and have more opportunity to learn novel material (Brophy & Good, 1974). In contrast, considerably less information has been acquired concerning teaching effectivness in regard to childrens’ psychosocial development in the athletic setting. Both Martinek and Johnson (1979) and Re jeski, Darracott and Hutslar (1979) found that physical education teachers gave more reinforcement and praise to high-expectancy students. However, a number of studies have observed that low-expectancy students received more reinforcement (Horn, 1985; Kleinfeld, l975; Weinstein, 1979). Despite this contradiction in the literature, positive social reinforcement is believed to be more beneficial to children than negative social reinforcement, as it may foster interpersonal relations between teacher and pupil (Martens, I975), create an enjoyable climate (Smith et al., 1977) and simply be more credible than negative feedback (Jacobs, Jacobs, Feldman a Caviar, 1973). Continuous positive reinforcement, however, may distort the perceptions of ability attributions made by students. For example, low-expectancy students may be receiving indiscriminate reinforcement for success at a very easy task. If this is the case, these students might perceive from the teacher’s feedback, because it is given indiscriminatly, that they are indeed viewed as a low-expectancy student. In support of this, Kirschenbaum and Smith (1983) found that continuous negative or positive feedback led to decrements in performing an underhand free throw compared to mixed feedback and control conditions. -5- Indiscriminate use of social reinforcement and its possible effects on children has been detailed by Harter (1981) in her theory of competence motivation. Harter (1981) specified that the evaluation of a child’s performance attempts by significant adults, contributes to the child’s perceptions of internalization of control. Perceptions of performance control in a particular achievement situation directly, and causally, influence a child’s perceived competence (Harter, 1981). If a child has been given clear, consistent and realistic evaluation about his or her performance, then the child will develop consistent and realistic internalization structures and will perceive an internal source of performance control. Inconsistent or unclear evaluation of performance will lead the child to perceive that control for his or her performance lies with powerful others. Are adults solely responsible for determining childrens’ self -perceptions of competence or ability? Harter’s (1981) model of competence motivation argues that evaluative information provided by adults determines a child’s perceived competence. However, another developmental model has been proposed by Nichols (1978) which postulates that children are actively involved, and cognitively process information in forming self-perceptions of ability. The way in which children cognitively process information is believed to develop with age. Nicholls (1978) reasons that to perceive ability and effort as distinct causal factors should lead to a more accurate assessment of one’s own competence. He discovered that children below the age of 12 were unable to make perceptive inferences of their ability because they were unable to distinguish between ability and effort in relation to task difficulty. After the age of 12 Nicholls (1978) reported that children were able to distinguish between effort and ability. -5- In the educational setting teachers have an important role to play in shaping a child’s self -perceptions. Research reported in the educational psychology literature (Meyer, Bachman, Biermann, Hempelmann, Ploger & Spiller, 1979) suggest that it is possible to perceive estimates of one’s ability from the informational value of evaluative behavior provided by a teacher. This group of researchers reported that a person who is praised or blamed is provided with information about how the teacher who is praising or blaming perceives his/her ability. The study by Meyer et a1. (1979) involved a series of six experiments which demonstrated that differential feedback given by teachers to two performers for the same level of performance provides information to subjects concerning the performers’ ability. Ability is judged according to the expenditure of effort. For example, if two people obtained identical successes and one of them was praised more for it than the other, it is assumed that this individual was perceived to have expended more effort on the task and, is thus judged to have less ability. Conversely, if two people both experience identical failures and one of them is blamed for it more than the other, it could be concluded that this person was perceived to have expended less effort on the task and consequently has the most ability (Meyer et al., 1979). A second theoretical assumption of the Meyer et a1. (1979) study relates to the amOunt of effort necessary for obtaining a solution to a task, with ability and task difficulty taken into account. If the task is perceived to be very easy and the ability is perceived to be high, then little effort is considered necessary for the solution. If task difficulty is very high, then in the case of high ability, high effort is thought to be necessary to obtain success; in the case of low ability even maximal effort is thought to be pointless (Meyer et al., 1979). .7. Hence, if praise and blame allow us to draw conclusions about the extent to which an obtained outcome was ascribed to effort expenditure, and if the difficulty level of the task is known, then a person’s "cognitive evaluative structure” formulates conclusions about how the person’s ability was estimated. However, Nicholls and Miller’s (1984) model of perceived ability further demonstrates that the cognitive processes associated with perceptions of ability, as it relates to effort expenditure and task difficulty, deve10p with age. Children below the age of 12 find it difficult cognitively to distinguish between effort and ability. Related to this, the first experiment conducted by Meyer et a1. (1979) found that there were developmental divergencies in the cognitive processing of identically perceived information of children aged eleven to thirteen years compared to high school pupils and adults. The younger age groups of children perceived that for two pupils who achieved the same performance the one who received the most praise by the teacher was the , one whom the teacher ranked higher in ability. Children below the age of 12 who were possibly not capable of distinguishing between effort and ability properly, perceived that a pupil who was praised, compared to a pupil receiving neutral feedback for the same performance, was the more able student. However, older students and adults perceived the exact opposite of the above situation. Perhaps the younger subjects in Meyer et al.’s (1979) experiment number one did not realize that both students were said to have obtained identical results. The possibility that the older and younger subjects interpreted the instructions differently was tested by Meyer et a1. (1979) in experiment number two. Furthermore, additional children aged eight, nine and ten were included in the study. Similar to the first experiment, experiment number two found developmental divergencies when comparing the young and older subjects. .3. Specifically, with decreasing age students indicated that the teacher thought the student whom he prasied was the brighter one, whereas neutral feedback was rated higher than praise by the majority of students from age fourteen to nineteen. These results may be explained in terms of Nicholls and Miller’s (1984) model of perceived ability. Subjects aged twelve and over who were cognitively mature enough to understand the effort/ability distinction, together with an appreciation of task difficulty, concluded that praise compared to neutral feedback, in this case, was indicative of low ability. Here the informational value of the evaluative behavior indicated that the student receiving the neutral feedback was perceived the more able because this student did not need to exert as much effort to achieve the same performance as the student who was praised. Using adult subjects in their third experiment, Meyer et a1. ( 1979) looked specifically at praise and blame compared to neutral feedback together with two levels of outcome (success, failure) and two levels of task difficulty (easy, difficult). For both levels of task difficulty, the results indicated that the student who was praised after success was rated lower in ability than the student who received neutral feedback. In the failure condition, the student receiving neutral feedback was rated lower in ability than the student who was blamed. Meyer et al. (1979) chose not to investigate further the observed developmental divergencies among the ability ratings found in experiment two. In their third study this group of researchers continued their series of experiments with adult subjects only. Moreover, the age differences in the ratings of perceived ability wereilef t unexplained. Thus, this study failed to address the issue of why developmental differences occur in perceptions and interpretation of feedback. In light of these findings, the present study was designed to replicate the third study of Meyer et a1. (1979) and extend it to the context of sport, to see further if these results are applicable to children. Social reinforcement theory encourages the use of praise or positive feedback, but fails to explain how such teaching behavior may affect childrens’ self -perceptions of ability. Nicholls and Miller’s (1984) developmental model of perceived ability may provide such an understanding. By knowing the interrelationships of these variables, teachers will be in a better position to interpolate their criticisms with their praise to maximize its impact and acceptance. STATEMENT OF THE PROBLEM The purpose of this investigation was to partially replicate and extend the third experiment conducted by Meyer et a1. (1979) by looking at the effects of teacher evaluative behavior on young children’s perceived ability, as it relates to sport. This study investigated developmental differences in perceptions of ability of identically perceived information across four dimensions: age of subjects (9 to 15 years); outcome (success, failure); task difficulty (easy, difficult); and evaluation (praise or blame versus neutral). A secondary aim of this study was to link the findings above to test Nicholls and Miller’s (1984) developmental model of perceived ability for children 9 to 15 years. -10- HYPOTHESES The following hypothesis will be investigated in this study. 1. For success and failure at an easy and difficult sport task, children aged 9, 10, and 11 years will give higher ratings of ability for a child who is praised and lower ratings of ability for a child they see blamed than children aged 12, 13, 14 and 15 years. For both an easy and difficult sport task, children aged 12 and above will give lower ratings of ability than children aged 9, 10, and 11 years when they see a child praised after success and themselves receiving neutral feedback after failure. For both an easy and difficult sport task, children aged 12 and above will give higher ratings of ability than children aged 9, 1-0 and 11 years when they see a child blamed after failure and themselves receiving neutral feedback after success. -11- DE'” Definitions of the following variables will aid in understanding the design of this study. I. Blame; For this variable the teacher’s response to a subject’s performance on the test item was read aloud as: ”What have you done there? I am displeased. That is not a proper forward roll/somersault.” W: This variable was defined as a somersault in the air in the hypothetical situation on the test item. mm This variable was defined as a forward roll on the floor in the hypothetical situation on the test item. m1: For this variable the teacher’s response on the test item was read aloud as: "Good” for success at the easy and difficult tasks and ”No, that is not the correct way to perform the skill” for failure at the easy and difficult task. 111113;: For this variable the teacher’s response on the test item was read aloud as: ”You have done very fine; I am very pleased.” W: This variable denoted children’s perceptions of another persons’s ability, or their own ability, to perform gross motor tasks as reflected in their responses to test items. I... This study will be limited to boys ages 9 to 15 years attending Augres Elementary/Middle High School in the State of Michigan. -12- IMPLICATIONS OF THE STUDY In view of the research presented to date, the use of praise as an appropriate or inappropriate instructional behavior in physical education or the athletic setting needs to be examined. Feedback in these settings is given to a child in front of his/her peers and may result in different ratings of I ability than feedback given to a child in the absence of his/her peers. Research suggests that children of varying ages differ in their cognitive abilities to process and perceive evaluative information. Thus, effective teaching skills require knowledge of optimal instructional strategies for children of different age levels. However, very little of the research in the sport psychology literature has not examined the effects of adult feedback on children’s perceptions of their sport ability. Research of this nature will help identify the teaching behaviors that are most conducive to children’s , psychosocial development in the athletic setting. CHAPTER 2 REVIEW OF LITERATURE One of the primary objectives of physical education is to facilitate the acquisition of motor skills through effective instruction. Related to this a coach or teacher has a major influence in shaping a pupil’s psychosocial development. Both skill acquisition and attributions of self -worth are shaped by the information or feedback a teacher provides to a child after performance. Social reinforcement is one form of feedback which is under the control of significant others (Martens, 1975). In sport and physical activity, social reinforcement can be viewed as all the positive and negative comments or reactions made by coaches, teachers, parents and peers. This information can be transmitted overtly or covertly through praise, encouragement, favorable gestures, smiles, frowns, reproof, sarcastic comments, direct criticism or disinterest. Adults continually use social reinforcers, both reward and punishment, to change children’s behavior (Carron, 1984; Martens, 1975) It is presumed that social reinforcement may function in a number of ways by providing information, motivation, and/or reinforcement (Ammons, 1956). Thus, social reinforcement may inform individuals about what they should or should not do, it may motivate individuals, it may reward individuals for correct performance, or punish them for incorrect performance. The nature of the interaction of these variables is not clearly known. The effects of social reinforcement may differ depending on the task complexity. For simple motor responses, it appears that the reinforcement is motivational in that reinforcement increases the rate of responding on many simple motor tasks (Zajonc, 1965). For complex motor tasks Martens (1975) .13- -14- contends that when knowledge of results and kinesthetic feedback are available, information provided via social reinforcement will be redundant. From a different perspective, social reinforcement may play a more important role in shaping a childs’ psychosocial development. Social reinforcement may encourage good interpersonal relations between the teacher and child, which may indirectly facilitate learning (Martens, 1975). Conversely, neither the delivery nor the reception of social reinforcement is simply an objective transfer .of information, both are capable of arousing strong emotions. Moreover, strong emotions aroused in the recipients may interfere or facilitate reception of the response provided. A considerable body of research has evolved which has investigated the specific effects of the interpersonal behaviors of teachers and the differential forms of reinforcement given. For example, in the classroom setting it is generally accepted that praise is significantly related to positive student attitudes and that critical statements by teachers are negatively correlated with such attitudes (Flanders, 1970; Rosenshine, 1973). In the physical education instructional setting Martens (1975), on the basis of previous research, contended that punishments tend to be a more effective short-term reinforcer than rewards, especially for very simple tasks. For more complex tasks, reward is viewed as more beneficial. Martens (1975) concluded that giving only punishment or reward may be detrimental over a long period of time. Both positive and aversive social reinforcement underlie what has been termed the positive approach and the negative approach to coaching (Smoll & Smith, 1979). The positive approach is designed to strengthen desired -15- behaviors by motivating players to perform by rewarding them when desirable behaviors occur. The negative approach involves attempts to eliminate unwanted behaviors through punishment and criticism (Smith & Smoll, 1979). Observational studies of coaches indicate that most coaches use a combination of positive and aversive control (Horn, 1982; Smith & Small, 1979). The limited research, which examines this particular issue, has produced conflicting results, providing support for both these styles of coaching philosophy. To understand more fully the effects of social reinforcement on childrens’ psychosocial development, Smith et al. (1977) devised the Coaching Behavior Assessment System (CBAS) to measure the behaviors of coaches toward their players in both contest and practice situations. The CBAS consists of a number of behavioral categories which were derived on the basis of empirical observation of coaches, as well as on theoretical principles identified in the sport psychological and social psychological literature. Following the development and validation of the CBAS, Smith, Smoll and Curtis (1979) conducted a two-phase investigation examining the relationship between coaching behaviors and players’ attitudes and self -concept. In the project’s first phase, the researchers observed and categorized the coaching behaviors of 51 little league coaches over the course of an entire playing season. At the end of the season, the 542 male players, ages 8 to 15, were interviewed and their attitudes toward their teammates, the sport, and their coaches were measured. In addition, measures of both general and athletic self -concept were taken. Data analyses revealed that those coaches who exhibited the highest frequencies of behaviors categorized as ”technical -l6- instruction,” ”reinforcement,“ and “mistake-contingent reinforcement" were evaluated more positively by their players than those coaches who scored low in these categories. Furthermore, players of these highly reinforcing and instructive coaches had more positive attitudes toward participation than did players of coaches who did not consistently exhibit these behaviors (Smith et al., 1979). Based on the results of their systematic research, Small and Smith (1979) advocated a positive approach to coaching, because this would make the sports experience for children more enjoyable. This research provided very persuasive evidence encouraging the frequent use of positive feedback and the near elimination of negative feedback. Other investigators, however, have cautioned against an over emphasis on positive social reinforcement. In a classroom setting, Kennedy and Willicutt (1964) noted that for young children frequent social reinforcement facilitates a humanistic environment, but there is loss of an incentive value because children receive no information on performance effectiveness. Moreover, Dunkin and Biddle (1974) caution that teachers’ praise is not indiscriminately associated with positive affect, but that the use of discriminatory, performance-contingent, positive instructional reinforcement may be influential in increasing students’ attitudes. In contrast to the positive approach to coaching advocated by Smoll and Smith (1979), a unique study on teaching effectiveness was conducted by Tharp and Gallimore (1976). Identifying then U.C.L.A. basketball coach, John Wooden, as one of the most successful coach-teachers in the history of college athletics, Tharp and Gallimore (1976) attempted to identify the teaching behaviors commonly employed by this ”master teacher.“ Coding and analyzing -17- behaviors exhibited by Wooden during practice sessions, Tharp and Gallimore recorded more than 2,000 acts of teaching over the course of 30 hours of observation. Data analyses revealed that the teaching behaviors coded as ”instructions” accounted for 50% of the total exhibited behaviors. In addition, it was found that at least 75% of Wooden’s teaching acts carried information relating to skill development-information that tended to be highly repetitive. These researchers (Tharp 8r Gallimore, 1976) also found that praise as a teaching behavior was rarely given by Wooden. In fact, he was coded to give twice as many ”scolds“ as “praise.“ However, the majority of Wooden’s scolds were accompanied by remarks containing additional instruction and were not equally distributed to the various athletes. Although the results of Tharp and Gallimore (1976) are based on the . observation of only one teacher-coach, descriptive evidence is provided concerning specific instructional behaviors which have been associated with consistent gains in player’s performance. However, Tharp and Gallimore (1976) emphasize that the generalizability of this information is certainly limited by the case study approach, as well as by the use of elite, college-level athletes who are both highly motivated and highly skilled. In summary, although the effects of positive or negative social reinforcement are not fully understood, if used frequently and/or indiscirminately they may lose their informational and/or motivational impact (Martens, 1975). Carron (1984) advises that coaches and teachers have to assess the extent to which they want to use social reinforcement for performance purposes and/or social psychological purposes. The study by Tharp and Gallimore (1976) suggests that positive social reinforcement provided -13- sparingly may facilitate performance benefits. While positive reinforcement provided liberally, as advocated by Smoll and Smith (1979), would be more beneficial to social psychological development, a mixture of both positive and negative reinforcement is likely to produce the best results (Martens, 1975; Kirschenbaum & Smith, 1983 ). Brophy and Evertson (1976) indicated that the effectiveness of a teacher’s reinforcement may be dependent on the teacher’s intention in administering such reinforcement. Walnut Considerable research has been conducted in both the academic and athletic setting to determine whether teachers’ reinforcement patterns (evaluative feedback) towards individual students are influenced by their expectations. These studies consistently show two major trends: a) differential teacher student interactions are related to the level of expectation a teacher has for a particular child, and b) teacher perceptions of students’ ability can ultimately affect students’ self -perceptions, which in turn influence behavior (Cooper, 1979; Horn, 1985). In the classroom, several studies have shown that teachers form expectations concerning the academic abilities of pupils and that these expectations influence the behavior and interactions that teachers have with individual students (Brophy & Good, 1974; Cooper, 1979; Good, 1981). For example, teachers’ reinforcement patterns (or evaluative feedback) towards individual students is one cateogry of instructional behavior which has been found to be influenced by their expectations. Generally, these studies have reported that teachers tend to provide more praise per correct response for high expectancy students and more criticism per incorrect response for low -19- expectancy students (Brophy & Good, 1970; Cooper & Baron, 1977; Meichenbaum, Bowers, & Ross, 1969). Hence, this evidence suggests that the evaluative feedback that teachers provide to students may be dependent on their expectations concerning the students’ ability. The same processes would appear to be operating in the physical education and athletic setting. A study by Crowe (1977) and another by Martinek and Johnson (1979), after observing student/teacher interactions, both reported that physical education instructors gave more praise and encouragement to high expectancy students in the elementary schools studied. Expectancy effects have been studied further in the athletic setting. Rejeski et a1. (1979), after observing individual coach-athlete interactions in practice and game settings, compared high and low expectancy athletes and the type of feedback received. From the results, high expectancy athletes received more reinforcement from their coaches than low expectancy athletes. Secondly, coaches also showed a significant tendency to provide more general technical instruction to low expectancy athletes compared to their high expectancy peers. However, a study by Horn and Hasbrook (1984) reported conflicting findings to the study by Rejeski et al. (1979). Here, athletes who were perceived by the coach to be low in ability were given a higher proportion of reinforcement and mistake contingent criticism than players considered to be high in ability. Teacher expectations, apart from influencing the type and quality of reinforcement, are predicted also to determine students’ self -perceptions and behavior. Differential communication, transmitted by the teacher and observed by students, has the potential of informing students just how competent the teacher thinks they are. This may have adverse effects. For example, low expectancy students who percieve the teacher’s lower perceptions of their -20- potential based on the quality of the feedback given in relation to others, may exert less effort, accept failure more quickly, and subsequently aspire to lower achievement standards than their high expectancy classmates. This phenomenom has been labeled the ’Self Fulfilling Prophecy’ or ‘Pygmalion Effect’ and has been found to operate in the classroom (Rosenthal & Jacobson, 1968) and the physical education setting (Horn, 1985). Horn (1985) describes a sequence of events whereby teacher expectations may indeed become reality in shaping childrens’ self -perceptions and behavior in sport. This model consists of four sequential steps outlined in Table 1. 131213.]. Teacher expectations and their influence on children in a sport setting Step 1. The coach develops an expectation for each athlete. Step 2. The coach’s expectations influence his or her treatment of individual athletes. Step 3. Differential communication tells each athlete how competent the coach thinks he or she is. This information affects the athlete’s performance and self -concept,' achievement motivation, and level of aspiration. Step 4. The athlete’s behavior and performance conform to the coach’s expectations. This behavioral conformity reinforces the coach’s original expectations and the process continues. This model highlights that information communicated by a teacher to a child directly or indirectly informs the child what the teacher thinks. A child’s self -perceptions are shaped by a sequence of cognitive appraisals, triggered by the instructional interchange between teacher and child. First, a teacher’s perceptions of a child’s ability are based on observation and previous knowledge. -21- Secondly, children are given information about the teacher’s perceptions of their abilities. Thirdly, this in turn influences a child’s actual self -perceptions of ability--which may affect future behavior. Hence, the feedback that a teacher provides may have a major influence in shaping a pupil’s perceptions‘of ability, self -image and self -conf idence. Phase 2 of the research project conducted by Smith et a1. (1979), although not direct support of this model, does provide evidence that coaching behavior can affect the psychosocial development of young athletes. This group of researchers utilized an experimental approach by manipulating the behavior of some of the coaches and then measuring the subsequent effects of this manipulation on players’ attitudes and perceptions. An educational program, administered to the experimental group of coaches, utilized three behavior-modification techniques, and was designed to teach coaches to exhibit the effective coaching behaviors identified in the previous study. Following this instruction period, behavioral data were collected. Data analyses revealed that significant differences were evident between the coaching behaviors of the experimental and the control groups, with the experimental coaches exhibiting more of the desired behaviors (encouragement, reinforcement, technical instruction). The athletes who played for the experimental coaches also rated their coaches significantly higher in knowledge and teaching technique and expressed a greater degree of enjoyment than did players of control group coaches. Finally, children who played for the trained coaches evidenced significant increases in self -esteem scores over the course of the playing season, while players of control coaches did not show comparable changes. Thus, in this study the level of the athletes’ self -concept was found to be influenced by specific types of behaviors exhibited by the coaches (Smith et al., 1979). -22. Although the results of the Smith et al. (1979) multi-year research project did contribute valuable information concerning the relation between coaches’ behavior and athletes’ self -perceptions, the focus in these studies was on the team as a group. However, recent research conducted in the athletic setting has demonstrated that there is considerable within team variation in the frequency and quality of coaches’ communications directed to individual players (Horn, 1985; Rejeski, et al., 1979). In view of obtaining a more complete assessment of coaching effectiveness, Horn (1985) used the individual player as the observational unit by recording coaches’ behaviors to individual athletes. Horn (1985) provided further support for the contention that changes in players’ perceptions of competence were influenced by their coaches’ feedback. However, Horn (1985) also discovered that this influence varied depending on the contextual situation in which the feedback was administered. Differences were noted between practice and game behaviors. And secondly, contrary to the findings and philosophy advocated by Smith and Smoll (1979), the data revealed that coaches’ responses to players’ successful performance (positive reinforcement) were identified as negative contributors to players’ development of perceived competence. Criticism, quite surprisingly, was associated with gains in perceived competence (Horn, 1985). In attempting to explain these controversial findings, Horn (1985) suggested that low coaching expectations of players’ performance resulted in non-contingent praise which in turn induced negative self -perceptions in regards to the skill competence of young athletes. Meyer et a1. (1979) has previously reported how differential feedback can affect estimates of perceived -23- ability. These researchers consistently demonstrated that differential feedback given by teachers to two performers for the same level of performance provided information to subjects concerning the performers’ ability. Excessive praise given to a performer for success at an easy academic task (in contrast to the neutral feedback given to a co-performer who exhibits an identical level of performance) led to perceptions of lower ability for the praised performer. Criticism given after failure at an academic task (in comparison to a co-performer who received neutral feedback for the same performance) induced higher perceptions of ability for the criticized performer. Hence, the differential feedback patterns reported by Horn (1985) are consistent with the theory and research discussed by Meyer et a1. (1979). That is, athletes whom coaches perceived to be of lower ability were given a higher proportion of reinforcement and a lower frequency of mistake-contingent criticism (punishment) than were their perceived high-ability teammates. These results raise the question: Why should self -perceptions of ability, which in part are influenced by differential feedback, be so important to the child? .v . . . v A number of researchers have provided evidence to support the contention that the perception of ability is the most important determinant of achievement behavior (Covington & Omelich, 1981; Nicholls, 1978; Roberts & Pascuzzi, 1979; Spink J: Roberts, 1980). Several investigators have theorized that attributions of ability and the self -concept of ability play the central role in mediating motivation (Bandura, 1977; Griffin & Keogh, 1982; Harter, 1981; Kukla, 1978; White, 1959). -24- What is the exact role of perceived ability in mediating motivation? In relation to sport, Roberts (1985) argued that in order to understand the motivation and subsequent achievement behavior of individuals, it is necessary to understand the subjective meaning of achievement to the achiever that is, the achievement goal of the athlete. Related to this, Nicholls, (1984) postulated that the achievement goal of individuals is to demonstrate and/or develop high ability and to avoid demonstrating low ability. In sport, there is a growing body of evidence to support Nicholls’ (1984) contention that perceptions of ability are the essential mediating construct of achievement behavior. This has been found with attributions to ability by winners and losers (Roberts & Pascuzzi, 1979), sport participation (Roberts, Kleiber & Duda, 1981), sport dropouts (Feltz & Petlichkof f, 1983) outcome of success and failure (Spink & Roberts, 1980) and, demonstration of high versus low ability (Roberts, Kleiber & Duda, 1981). Nicholls’ model (1984) also makes specific predictions about the theoretical antecedents and consequences of perceived ability and how they affect task choice. If participation enhances the perceived ability of young athletes they should continue to have positive expectancies about future success and place greater value on continued participation. However, as soon as participation threatens young athletes’ perceived ability, they should develop negative future expectancies and devalue participation. Thus, this theoretical approach suggests that inadequacies in the sport experience turn young athletes off and prompt them to search elsewhere for activities that fulfill their achievement needs (Burton & Martens, 1986). In addition to this, perceptions of ability are believed to mature and change with age (Nicholls & Miller, 1984). -25- From a deve10pmental viewpoint Nicholls and Miller (1984) view the concept of ability being differentiated as a function of age. Younger children use different criteria to judge whether they have demonstrated and developed ability compared to older children. Perceptions of control of ability involve a complex interaction of inferences which relate to the performance of others, the amount of effort employed and the constraints of one’s own actual capacity to perform the task in question. In the youth sports setting this social comparison process involves three assessments: l) the ability of the opponent in relation to all other opponents, 2) own ability compared to that of the opponent’s, and 3) amount of effort applied by self and/or opponent (Roberts, 1985). This developmental notion of the differentiation of ability matures through three levels of ability related to difficulty and four levels of ability related to effort (Nicholls & Miller, 1984). (See Table 2) The three levels of the differentiation of the concepts of difficulty and ability proposed by Nicholls and Miller (1983) are egocentric, objective and normative. This differentiation between ability and difficulty develops with age, from a subjective appraisal or certainty of task completion, to a focus on task characteristics from simple to difficult, and finally to a social comparison of performance norms. Nicholls and Miller (1983) viewed this development of perceived ability to be sequential in nature because mastery of later levels incorporates mastery of the previous levels. ~-26- IIethl Levels of differentiation of the concept of ability from thoee of didiculty and effort __Jiss, lZflfissflDLee oaatiflkni 2-4 Egocentrlc: Own expectations of success the basis for judging task difficulty and the level of ability by task outcomes. 4/54 Objective: Perceptions of task difficulty based on properties of task (complexity) and the level of ability indicated by taak outcomes. 8-1 Normative: Task difficulty and ability judged in 7-9 9-10 10/11-13 relation to the performance of others. Task that few can do are hard, and success on those indieatee high ability. I Effort is the cause of outcomes. Equal aflort is expected to lead to equal outcomes. Ability is partially differenti- ated (as a cause of outcomes) from effort. Complete differentiation: Ability conceived as capacity which limits the effect of effort on performance. -27- Another progression, in terms of development, is the distinction between the perception of effort and ability. Herc ability is inferred from social comparison, that is, judged relative to the effort and performance of others. The four levels progress from: level 1, effort is indicative of ability; level 2, effort equals outcome; level 3, effort and ability are partially differentiated; and level 4, ability is the amount of individual capacity. Nicholls (1978) suggested that higher levels of congnitive development include lower levels in the form of hierarchical integrations. This developmental notion of perceived ability and its relationship to effort and difficulty, within the context of social comparison, has important implications for children taking part in sport. Significantly with increasing age, level 2 (being more able, dependent on effort) and normative conception (social comparison), integrate and develop simultaneously (Nicholls & Miller, 1984). By the age of 12 children realize that higher effort implies lower ability (Nicholls & Miller, 1984). In other words, one has to work harder to do as well as others which implies that one has less ability than others. This developmental change in the concepts of effort and ability is accompanied by changes in perceptions of the relative importance of effort and ability for performance. This could provide a possible explanation for why a large proportion of children drop out of sport at the age of 12 (Seefeldt, Blievernicht, Bruce & Gilliam, I978). The proposition that children below the age of 12 do not understand the distinction between cf fort and ability in competitive-oriented environments has been investigated in the youth sports setting. Ewing, Roberts and Pemberton -23- (1984) studied the developmental achievement orientations of children, aged 9 to 14 years, in the sports of baseball and softball. The results were consistent with those reported by Nicholls and Miller (1984), because children only demonstrated a full understanding of the cf f art/ability differentiation in these sports after the age of 11. This shift in a child’s perception of ability, which evolves around the age of 12, is especially applicable to competence motivation for children taking part in sport. A child with low perceptions of ability realizes that to compensate for lack of ability he/shc will need to exert more effort. This, in turn, will indicate to others that they lack ability, which may lead to ”social embarassment" and result in the child dropping out of sport (Ewing et al., 1984). Once an individual has acquired differentiated conceptions of ability after 12 years of age, situational factors will determine whether an individual will rely more or less on differentiated conceptions of ability (Nicholls & Miller, 1984). The type of reinforcement a child receives from the teacher is one such situational factor which informs a child of the differentiation between effort and ability. 1!: .VE I. [5"] I.E " I Children formulate their self-perceptions of ability from a number of sources. Theory and research from the psychological literature suggest that the susceptibility of children to a particular information source is age-dependent (Verof f, 1969). In early childhood, for example, children tend to base their judgements of competence on autonomous performance standards (i.e., successful completion of a task) in combination with the feedback of significant adults (Stipek, 1981). During the elementary school years however, -29- children show an increasing tendency to use peer performance as a means of judging their own competence (Cook & Stingle, 1974). The saliency of information pertaining to social comparison appears to reach its highest intensity during the late childhood and early adolescent years (Cook & Stingle, 1974). Finally, some researchers have speculated that the maturational changes in cognitive or intellectual functioning which occur during the adolescent years may be reflected in the expansion of judgements of competence, from simple comparison with peers to an integrated use of many systems of information and the establishment of self -detcrmined or internal standards of performance (Harter, 1978; Verof f, 1969). In the sport setting the dependence on adult feedback has'been found to decline with age, while the use of peer comparison increases. Using the statistical technique of exploratory factor analysis, Horn and Hasbrook (1984) were able to highlight six information sources that children use to assess their competence. Factor 1 reflected social comparison (i.e., using the performance of teammates and opponents to judge their own ability). Factor 2 included evaluative information from coaches, peers and spectators. Factor 3 referred to internal sources of information indicating that young athletes may be using such cues as a) the amount of effort they exert, b) the degree of improvement in their skill performance, and c) the ease with which they acquire new skills to judge their ability. Factor 4 included evaluation provided by parents and spectators. Factor 5 suggested that game outcome is important while factor 6 typified a reliance on affect (i.e., degree of liking for the sport) as a basis for judgements of ability. The results of this study clearly indicate that the cognitive processes underlying the competence judgements of younger children are qualitatively different from those of older children in sport. -30- In order to understand how children use the evaluative information, it is helpful to apply Nicholls’ (1978) cognitive developmental model of perceived ability. According to Nicholls (1978) adolescents and older children begin to judge their past performances relative to peer performance. If this comparison leaves them lower than average, they expect to demonstrate low ability. If perceptions indicate that their own abilities are higher than similar others, positive outcomes are expected from performance. Thus, the individual’s assessment of the probability of the outcome inf luenccs subsequent behavior. As well as the influence of outcome, Nicholls and Miller (1983, 1984) drew further links on self-perceptions of ability to effort and task difficulty. For example, high effort which results in f ailurc to perform an easy task clearly exposesa basic lack of ability resulting in low perceptions of ability. For a person with low perceived ability, high effort, which results in failure, clearly exposes a basic lack of ability. Conversely, for a person with high ability, failure at an easy task typifies a lack of effort. Hence, Nicholls’ (1978) postulated that the interaction of outcome, effort and task difficulty shape self -perceptions of ability. In addition to this, the preceding-discussion has emphasized that a relationship exists between differential evaluative feedback and perceived estimates of ability. A series of laboratory studies by Meyer (1979, 1982) has consistently demonstrated that differential feedback given by teachers to two performers for the same level of performance provides information to subjects concerning the performer’s ability. Meyer et al. (1979) specifically studied the effects of praise and blame on perceived ability estimates investigating -31- the variables of outcome, effort and task difficulty, as Nicholls (1978) did to explain his findings. The difference being that Meyer et al. (1979) used these variables to account for the psychological structures and processes that mediate the linkages between praise or blame as estimates of ability. Based on the results of several studies Meyer (1982) argued that perceived effort expenditure is among the main determinants of administered reward and punishment. Moreover, praise and blame are maximized when an outcome is ascribed to high or low effort. For example, success is especially rewarded when a recipient is perceived as low in ability and as having expended high effort, whereas failure is especially punished, given high ability and low cf fort. Meyer’s (1982) contentions are based on the premise that linkages between praise-blame and estimates of perceived ability relate to the amount of effort necessary or sufficient for success at a task. Furthermore, praise and blame enables a person to draw inferences on how the acting person’s ability is estimated. These assumptions were tested by Meyer and his associates (1979) in a series of experiements. The research varied in that some of the studies had teachers making judgements while others analyzed the judgements of students of varying ages. Some investigations directly assessed estimates of ability, whereas others assessed more indirect indexes, such as effort expenditure, classifying differential evaluative feedback and the f amiliarity/non- familiarity of a teacher to the students (Meyer et al., 1979; Meyer, 1982). In spite of these variations, the general pattern of data was quite similar for adult subjects. In one of these studies (Meyer et al., 1979) teachers served as subjects. Each subject received one of four versions of a short -32- questionnaire. 1n the questionnaires, two students were described as solving an arithmetic problem that was characterized as very easy or very difficult. It was also conveyed that both students either succeeded or failed at the task. The evaluative feedback given by the teacher was portrayed as different for both students although their performance was identical. Given success, the teacher’s neutral reaction to one of the students was ”Yes, 32 is the correct number”; his praise for the other student was "You have done very well; I’m very pleased.” Given failure, the teacher’s neutral reaction was described as "No, 35 is not correct"; his blaming response was “What have you done there! 35 is wrong.” After reading the scenario, subjects indicated for the two students on separate scales how the teacher would estimate their abilities. The results revealed that the student praised for success was rated lower in ability than the student receiving neutral feedback. In the failure condition, the student receiving neutral feedback was rated lower in ability than the student who was blamed. These effects were present at both levels of task difficulty. In another study Meyer et al. (1979) had each subject imagine himself to be a student who, together with a fellow student, received evaluative reactions from a teacher. Moreover, the teacher was described as familiar or unfamiliar with the students. The reasoning behind this variation in familiarity between teacher and student was that praise and blame were not expected to influence the ability ratings unless the person distributing feedback was ascribed as having knowledge about the recipient’s ability. Therefore, only praise and blame from the familiar teacher were expected to influence estimates of ability. These expectations were supported; the results revealed that praise and blame from an unfamiliar teacher did not influence perceptions of ability as greatly as they did from a teacher who was familiar with the students’ ability. .33- To date no research has been conducted in the athletic setting to test these assumptions concerning the informative function of evaluative behavior. Moreover, while the above findings were typical of adults, the study of Meyer et a1. (1979) reported that children differ in their interpretation of teacher feedback. These developmental differences need to be tested in a sporting context, so that teachers can devise optimum teaching strategies based on the age of the children, the complexity of the task and the outcome of performance. CHAPTER 3 METHODOLOGY Salim Males were chosen as the subjects for this study. The sample was comprised of 140 boys divided into seven groups by age. The ages of the boys ranged from 9 through 15 years with the mean age being 12 years. The rationale for choosing seven age groups was based on the assumption that there would be more extreme ability ratings the greater the age difference. This assumption was based on Nicholls and Miller’s (1984) developmental theory of perceived ability. There were 20 boys in each of the seven age groups. The children were drawn from Augres Elementary/Middle/High School located in Central Michigan. Permission was obtained from the director of the school to invite participation in the study. A consent form was signed by the parent/guardian of all of the boys in the school who indicated a willingness to participate in the study. Drain This study employed a 7x2x2x2 (age by task difficulty by feedback by outcome) factorial design, and similar to Meyer et a1. (1979), with repeated measures on the final three factors. There were seven levels of age (9 - 15), two levels of task difficulty (easy, difficult), two levels of teacher feedback (praise or blame versus neutral), and two levels of outcome (pass or fail). The last three factors were combined to produce eight experimental conditions. All children experienced all experimental conditions in the same order. The order in which the questionnaires were answered was easy success. easy failure, difficult success, difficult failure. Inmmmatinn The instrument used was a modified version of the questionnaires in the Meyer et. a1. (1979) study. The modification involved responses to a -34- -35- hypothetical sport situation. Each of the four questionnaires (see Appendix A) was distributed to every subject and read aloud in turn, followed by the subjects marking their responses on a 9-point scale. The 9-point scale was ordered at odd numbers with the following labels: . I-very low ability, 3-low ability, 5-average ability, 7-high ability, and 9-very high ability. The use of two additional instruments, Harter’s Perceived Physical Competence Scale (1979) and a demographic questionnaire, had to be eliminated from the study. The director of the school and the classroom teachers believed that too much time was required to complete the other questionnaires. Procedure Boys enrolled at Augres Elementary/ Middle/High School, between the ages of 9 and 15 years, were selected for this study. Contact with the director of the school was arranged to discuss the nature of the school’s involvement and purpose of the study. Following this meeting, the director agreed to provide his cooperation to allow the boys in his school to serve as subjects and to contact the parents of the children to inform them that this study met with his approval. Printed materials describing the nature of the study were presented and time and facility commitments were discussed. A letter explaining the study, and a parental permission form (see Appendix B) were sent to parents of each child. The teachers randomly selected 20 boys within an age group, whose parents had given permission to participate in the study. Children selected for the study gave their assent to participate. W The questionnaires were distributed by the teacher during classroom time. The subjects were seated opposite the teacher while testing took place. The study was held in the classroom for each of the seven age bands (i.e., 9, 10, ll, 12, 13, 14 and 15 years). Each subject received a questionnaire containing four situations (easy/success, easy/failure, dif f icult/success, -35- dif f icult/f ailure). The teacher read aloud each situation waiting for all the children to finish answering their questionnaire before reading the next questionnaire. The four situations involved a description of two students who performed a gross motor skill. The task was presented as very easy (or very difficult), and subjects were told that both students had performed the skill correctly (or incorrectly). For each of the four conditions the subjects who participated in this study had to imagine they, themselves, were performing the skill described in the hypothetical sport situation. Moreover, for each of the conditions the subjects received neutral feedback. For success at an easy and difficult task, teacher feedback was praise and neutral. For failure of an easy and difficult task, feedback was blame and neutral. . Following presentation of each situation, subjects were asked to circle the name of the child the teacher thought had higher ability, and also how the teacher would rate each child’s ability on a 9-point scale. Administration time was approximately 20 minutes per group of children. After completing the test, children were thanked for their participation. I . . . A limitation of this study was a possible order cf f ect in answering the questionnaires in the sequence easy success, easy failure, difficult success and difficult failure. An additional limitation was that the subjects only experienced receiving neutral feedback from the teacher. CHAPTER 4 RESULTS This study was designed to address two issues in relation to childrens’ self -perceptions of ability in the sport setting. First, a replication of Meyer et al.’s (1979) third experiment which studied, specifically, the effects of outcome, task difficulty and feedback on childrens’ self -perceptions of ability. Secondly, an extension of this replication involved investigating the additional factor of age to discover if any developmental trends were evident. Hence, do the factors of outcome, task difficulty and feedback affect childrens’ self -perceptions of ability differently depending on their age? The results of the replication and the extension will be discussed separately. 5 l . l‘ B 1' . I E . The first part of this study was concerned with replicating Meyer et al.’s (1979) third experiment. However, this study deviated from the original study in several ways. Firstly, the subjects chosen were children as opposed to adults. This study also had a much larger sample size, i.e., 140 compared to 17 subjects in Meyer et al.’s orginal study. Thirdly, unlike the Meyer et al. (1979) study, boys only were tested. Fourthly, subjects were asked to imagine themselves as one of the students in each of the four hypothetical situations. And finally, the scenarios read by each subject were related to the context of sport. Ratings of perceived ability on the four questionnaires were analyzed within a 2x2x2 (outcome x feedback x task difficulty) ANOVA with repeated measures on each factor. This analysis revealed a significant main effect for the factors of outcome, task difficulty and feedback. None of the two way interactions were significant. There was, however, a significant three way interaction for outcome by feedback by task difficulty. A summary of all the main effects and interactions are reported in Table 3. -37- -33- Table 3 Summary of Analysis of Variance for Perceived Ability by Outcome, Task Difficulty and Feedback Source df as E Outcome 1 5.86 4.44* Task Difficulty 1 67.52 22.26** Feedback 1 2644.50, 266.83** Outcome x Task Difficulty 1 2.51 1.85 Outcome x Feedback 1 1.22 0.95 Task Difficulty x Feedback 1 6.15 2.40 Outcome x Task Difficulty 1 92.00 22.95** x Feedback Error 139 6.37 ‘ p<.05 “ p< .001 In order to interpret the results of the three-way interaction, eight groups were created. These groups were easy failure blame, easy success praise, difficult failure neutral, difficult success neutral, easy failure neutral, easy success neutral, difficult failure blame and difficult success praise. A f ollow-up Schef f e’ test on the eight groups yielded a significant result, F(7,lllZ)-lO7.9378, p,(.001. The means and standard deviations for the three-way interaction are presented in Table 4. A diagramatic view of the three-way interaction compared to Meyer et al. (1979) is presented in Figure l. - 39- Abthy x 0.30:: O mmm< m J I 932: 222 2. m: . mmmv. 28;: m». m: \\\‘x V I m I I - m M k l I w I N I I A ll _ _ _ _ 29:2: mime 222m. , w_m3m mcooomm maze.» WEE sea—Eu 3 32:: 2:. 233:5 E. Banana—A 3. Eur 232:2. -40- A closer examination of the three way interaction revealed that all the easy and difficult success conditions were rated significantly higher from all the easy and difficult failure conditions. Furthermore, the easy and difficult/success/praise groups differed significantly from the easy/success / neutral group. Groups which did not differ significantly from one another included all the easy and difficult failure conditions with accompanying feedback, with means ranging from 3.26 to 4.09. There was no significant difference between the easy and diff icult/success/neutral groups with means of 5.95-and 6.80 respectively. Furthermore, the dif f icult/success/neutral group and easy and difficult/success/praise group did not differ significantly, with means of 6.83, 7.26 and 7.23 respectively. Table 4 Means and Standard Deviations for the Perceived Ability Ratings Task Difficulty Success Failure N "a u ral Bl§_§ :1 SD M SD M SD M SD EASY 5.95 (1.65') 6.83 (1.87) 3.66 (1.70) 3.26 (2.26) (Meyer et a1.) 5.53 (- ) 2.59 ( - ) 3.18 (- ) 5.65 ( DIFFICULT 6.80 (1.57) 7.26 (1.88) 4.09 (1.95) 3.53 (2.42) (Meyer et al.) 6.65 ( - ) 4.47 ( - ) 4.77 ( - ) 6.24 ( - The surface means and standard deviations for the two-way interactions are reported in Tables 5, 6 and 7. -41- Table 5 Surface Means and Standard Deviations for Perceived Ability .by Task Difficulty and Outcome Success Failure M. .52 L1 5.1.). Easy 6.39 (1.82) 3.81 (2.21) Difficult 7.03 (1.74) 3.46 (2.01) Table 6 Surface Means and Standard Deviations for Perceived Ability by Outcome and Feedback Success Failure M $.12 11. SD. Praise 7.04 (1.88) -- -- Neutral 6.38 (1.66) 3.40 (2.34) Blame -- -- 3.88 (1.84) Table 7 Surface Means and Standard Deviations for Perceived Ability by Task Difficulty and Feedback Easy Difficult M £2 M £2 Praise/Blame 5.05 (2.73) 5.39 (2.86) Neutral 5.44 (2.23) 4.81 (2.03) -42- The significant main effects indicated that subjects rated students who were successful higher in ability than students who failed. Subjects rated students working on a difficult task as higher in ability than students working on an easy task. And students who were praised or blamed were perceived bu the subjects as having greater ability than students who received neutral feedback. The means and standard deviations for the main effects are reported in Table 8. Table 8 Means and Standard Deviations for Perceived Ability for the Variables of Outcome, Task Difficulty and Feedback Group M ED Outcome Success 6.71 1.81 Failure 3.64 2.12 Task Easy 4.93 2.41 Difficulty Difficult 5.42 2.56 Feedback Praise/Blame“ 5.22 2.80 Neutral 5.13 2.15 Looking at the means for the main effects, it is evident that for outcome there was a significant difference between the conditions of success and failure. Overall success had a mean of 6.71 while failure had a mean of 3.64. Hence, subjects rated students who were successful as having higher ability than students who failed. -43- For task difficulty there was not as great a difference between the easy and difficult groups. Performance on an easy task yielded a mean perceived ability rating of 4.93 whereas the mean perceived ability rating on a difficult task was 5.42 Thus, perceived ability was rated higher on the difficult tasks than the easy tasks. The feedback conditions had a mean rating of 5.22 for praise or blame and a mean of 5.13 for neutral. This was not surprising because subjects rated the student who was praised as being more able than the student who received neutral feedback. In contrast, the student who was blamed was rated lower in ability compared to the recipient of neutral feedback in the same condition. When viewed as a main effect this contrast in ratings of perceived ability was neutralized. 12' . The results of this anlaysis failed to replicate the findings of Meyer et al. (1979). In their original study Meyer et al. reported no significant main effects for outcome or feedback but they did report significant two way interactions between outcome and feedback as well as outcome and task difficulty. Moreover, Meyer et al. (1979) did not find the three way interactionof outcome by feedback by task difficulty to be significant. The only result which was replicated was the main effect for the factor of task difficulty. The means and standard deviations reported in the Meyer et al. (1979) study are included in Table 4 for comparison with the present findings. Under the easy and difficult success conditions the neutral ratings did not differ very much from the Meyer et al. (1979) ratings. However, the easy and difficult success praise ratings were considerably higher than those -44- reported by Meyer et al. (1979). The mean rating for easy success praise was 6.83 (1.87) compared to Meyer et al.’s. rating of 2.59 ( - ). The difficult success praise had a mean rating of 7.26 (1.88) while Meyer et al. reported a rating of 4.47 ( - ). For the easy and diff icult/failure conditions, again the neutral ratings did not differ greatly from Meyer et al.’s (1979) findings with the means for both studies ranging from 3.66 to 4.77. However, for easy f ailure/blame this study reported a lower mean of 3.26 (2.22) compared to Meyer et al.’s (1979) 5.65 ( - ) rating. For the diff icult/f ailure/blame rating, this study reported a mean of 3.53 (2.42) while Meyer et al. (1979) reported a mean rating of 6.24 ( - ). Unfortunately, Meyer et al. (1979) did not give any standard deviation for each of the conditions, which would indicate how much each group varied compared to the present findings. In sum, the present study reported similar ratings of perceived ability to Meyer et al. (1979) for the easy and diff icult/success and f ailure/neutral conditions. However, subjects in this study had higher mean ratings for the easy and dif f icult/success/praise conditons and lower ratings for easy and dif f icult/f ailure/blame conditions, than the mean ratings of the subjects in the Meyer et al. (1979) study. Perhaps one explanation for the failure to replicate Meyer et al.’s study was the age of this study’s subjects. EI'Z'DVI HIKE The main purpose of this second analysis was to extend the third experiment of Meyer et al. (1979) by studying the additional factor of age on the factors of outcome, task difficulty and feedback. The rationale for investigating possible developmental differences was twofold. First, Meyer et al. (1979) reported that children differed in their interpretation of teacher feedback. However, these differences were not explored in relation to outcome, task difficulty and type of feedback. Secondly, Nicholl’s (1978) .45- cognitive developmental model of perceived ability posits that children’s perceptions of their own ability may be influenced by outcome, effort and task difficulty. By including age as another variable in the original design, it should be possible to detect when developmental differences in perceptions of ability occurred and what factors were most influential in each age group’s rating of ability. The influence of age was investigated by means. of a 2x2x2x7 (outcome x task difficulty x feedback x age) ANOVA. Results of this analysis replicated the main effects found in the previous three-way ANOVA for the factors outcome, task difficulty and feedback. However, this analysis did reveal three important results. Firstly, there was no significant main effect for age. Secondly, there were significant interaction effects for age by task difficulty as well as age by feedback but age by outcome was not significant. And thirdly, a f our-way interaction was evident for age by outcome by task difficulty by feedback. A summary of all the main effects and interactions are reported in Table 9 on the following page. The significant f our-way interaction, while unanticipated, supercedes interpretation of significant main effects and two- and three-way interactions. The means and standard deviations of the perceived ability ratings by the seven age groups across each condition are presented in Table 10. (See page 47). The f ollow-up Schef fee’ test for the four-way interaction required the comparison of 56 group means. Unfortunately, the SPSS statistical package places a limit of 50 on the number of cells it can accomodate in a post hoc analysis. To facilitate the interpretation of the data relative to the hypotheses, it was necessary to combine subjects by age into younger (9, 10 and 11 years) and older (12, l3, l4 and 15 years) groups. Because meaningful information was potentially lost, the significant two-way interactions will be presented first followed by the revised testing of the f our-way interaction. ~46- 'Tabk:9 Analysis of Variance for Perceived Ability by Age, Outcome, Task Difficulty and Type of Feedback Source gr fli E Age 6 8.23 1.31 Outcome 1 5.86 4.50* Task Difficulty 1 67.52 27.37** Feedback 1 2644.50 278.25** Age x Outcome 6 1.71 1.31 Age x Task Difficulty 6 15.57 6.31** Age x Feedback 6 18.93 1.99* Outcome x Task Difficulty ' 1 2.51 1.87 Outcome x Feedback 1 1.22 0.92 Task Difficulty x Feedback 1 6.15 2.40 Age x Outcome x Task Difficulty 6 1.69 1.26 Age x Outcome x Feedback 6 0.40 0.30 Age x Task Difficulty x Feedback 6 2.72 1.06 Outcome x Task Difficulty x Feedback 6 92.00 25.79** Age x Outcome x Task Difficulty 6 13.79 3.87* x Feedback Error 133 6.29 ‘ 24.05 ” {24 .Ol “hm-.3 no.0 Cmfiv n~.0 3W: o—.n 30.: .K.~ 3~.C 3.0 30.3 mnfi. 30.: 00.0 30.: and 9 30.3 2.0 £53 8.0 CE: 8...... 30.3 or...» «3.: 3K 30.3 no.0 2.3.: 8K 20.: 3.0 0.. _ CN.~0 2.0 £43 9.0 2.”: 36 80.3 2...» an”: m0.~. 3a.: 3.x. 35.3 006 30.3 3.0 n— 7 A... 3.33 8.0 3...: mud 30.: m0.n 30.: £4. 35.: no.0 3a.: 3.5 3n...» 3.0 Ann-.5 an.» N. 30.: o~.0 «3&0 m~.n 2.0.3 m0.n 30.: o~.n 393 3:0 30.: 00.» a0“: 8.0 n0n._-v and :- 373 00.0 353 m~.0 3E: m0.0 «3&0 o~.0 $0.5 8.0 35.3 8.0 23.: and «8.3 2.0 Q— So.: m0.n 3a.: mo.~ «no...» n0.n AON.~v 8.~ 30.: m0.0 30.: 3.x. 30.: o0.n 30.: 8.5 o a m a a ma m a m a m .fl a a a a m .9530: 0.3 .9530: a... .9530: memo-E .9530: 3.0.5 wu< agis. ms; :5 «mun-03m mag 58:53 55 53:39 >05 seen—goon us. 33333 .3: .2833 .5 39¢ :33 05 03 83:38 venue-um use file: 2 030» .43.. A Scheffe’ post hoc test was conducted for the significant two-way interaction for age by feedback, 1503,1106) - .7643, 2405. The results of this analysis failed to produce a significant difference between any two groups. The reason why no two groups differed may be explained in that praise and blame were coded the same in the original data. This procedure was followed in order to replicate the Meyer et al. (1979) study. The means and standard deviations for the perceived ability ratings in the f eedback conditions are reported in Table ll. Table ll Means and Standard Deviations for Age by Feedback AGE PRAISE/BLAKE NEUTRAL H 3.9 if! $.12. 5.18 3.33 4.89 2.12 10 5.49 .80 .20 .88 11 5.39 .74 .31 .12 12 4.98 .80 .15 .24 13 5.26 .50 .07 .32 14 5.35 .79 .55 .29 15 4.90 .60 .70 .02 For both of the feedback groups the means and standard deviations did not differ greatly across age, with ratings ranging from 4.70 to 3.55. However, the 15 year old children had the lowest mean of all the age groups for praise/blame feedback and the neutral feedback conditions. Moreover, the 12 and 14 year old differed from the other age groups by rating students who -49- 5.6 — X Neutral 5.5 —‘ o Praise/Blame 5.4 -— 5.3 '— ,. 5.2 .— § 5.: __ < 5.0 -— 4.9 --4 4.8 —- 4.7 -— l l l l l l T 0101112131415 Age Em; Ratings of ability for age by feedback. received neutral feedback higher in perceived ability than students receiving praise and blame. The means for the two feedback conditions across age are plotted in Figure 2. A Schef f e' post hoc test was conducted for the significant two-way interaction for age by task dif f iculty, E(l3,l lOO) - 2.6476, p(.OI. Although this follow-up test was significant, no two groups differed significantly from each other. The means and standard deviations are reported in Table l2. .50. Table 12 Means and Standard Deviations for Task Difficulty by Age AGE EASY DIFFICULI u an n 9 59 9 5.04 2.71 . 5.02 2.88 10 5.22 2.21 ' 5.46 2.55 11 5.34 2.48 5.36 2.42 12 4.96 2.48 5.16 v2.60 13 4.64 2.24 5.70 2.46 14 5.31 2.51 5.59 2.59 15 3.97 1.92 5.62 2.41 The results indicated that the 9 year old boys were the only group to rate the easy tasks higher than the difficult tasks. The rest of the age groups rated difficult tasks higher than easy tasks. The means for the 9, 10, and 11 year old boys were similar for both easy and difficult task conditions. After age 11 there was a notable split between the ratings of the easy tasks compared to the difficult tasks for the older children. This difference in the ratings of the older children is responsible for the two-way interaction. The means for the easy and difficult tasks are plotted in Figure 3. -51- 5.0 '— X Otlllcull 5.6 "‘ 0 Easy 5.4 -" 5.2 _l 5.0 — 4.0 —4 Ablllty 4.8 — 4.4 '- ‘.2 —" ‘oo _ 3.. — lllllll O l01112 13 N 15 Ann M Ratings of ability for age by task difficulty. The additional variable of age showed no significant main effect. The means and standard deviations of ability ratings for the seven ages across all conditions are reported in Table 13. Table l3 Means and Standard Deviations of Ability Ratings for each Age Across" Questionnaires M 812 9 2.80 2.26 l0 . 4.20 2.43 l l 3.20 L“ :2 2.15 1.94 l 3 3.70 2.49 n 4 3.50 2.69 15 2.70 l.84 -52.. Although these results do not support the hypothesis which posits developmental differences, age differences were apparent when combined with the other variables of outcome, task difficulty and feedback. No follow-up tests were necessary for the two-way interaction of age by outcome because there were no significant differences in the ratings of success and failure by age groups. The remaining two-way interactions of outcome by task difficulty, outcome by feedback and task difficulty by feedback were not significant. Furthermore, the three-way interactions of age by outcome by task difficulty, age by outcome by feedback and age by task difficulty by feedback were not significant. The means and standard deviations for all the non-significant findings are presented in Appendix C. As stated previously, in order to test the proposed hypotheses it was necessary to conduct a more direct test. Hence, the seven ages were combined into younger and older groups. Nine, 10 and ll year old boys were combined to form the younger group and 12, l3, l4 and 15 year old boys formed the older group. This yielded the two groups defined in the hypothesis. A discriminate analysis was chosen to test the proposed hypotheses because of the multivariate nature of the problem. The three independent sets of variables combined to form eight sets of conditions. In addition it was necessary to investigate this three-way relationship of the variables together because the hypotheses were stated in the form of an interaction. -53- The discriminant function analysis was significant,xz(8)-29.72, n .001, and accounted for 20% of the variance. All eight combined conditions were significant in discriminating the two age groups. The means and standard deviations for the ability ratings are presented in Table 14. Table 14 Means and Standard Deviations for the Young and Old Children for each Situation Young 010 ' 500* E 59 8 £2 Difficult Failure Blame 3.18 (2.31) 3.79 (2.47) -0.402 Easy Failure Blane 3.40 (2.26) 3.16 (2.27) 0.544 Easy Failure Neutral 4.05 (1.89) 3.33 (1.50) 0.382 Difficult Failure Neutral 4.12 (1.92) .4.06 (1.98) -0.053 Easy Success Neutral 5.98 (1.54) 5.93 , (1.74) 0.012 Difficult Success Neutral 6.38 (1.66) 7.11 (1.42) -0.690 Easy Success Praise 7.37 (1.90) 6.43 (1.74) 0.562 Difficult Success Praise 7.45 (2.04) 7.11 (1.76) 0.376 ‘ SDC - Standardized Discriminant Coefficients Note: '-" Sign favors older children group. The means for the younger and older children are plotted in Figures 4 and 5 on page 54. These plots highlight the points of interaction between the two groups on the variables investigated. Ablllly -54- 9 — {’ 1‘ Difficull young 0 _ ° (asg young ‘ ‘“ 0 lfllcull old A esg old ’ -' 7 ._. 6 -" 6 up Ablllty .3 l 3 -—1 3 -u E Neutral Praise Neutral Blame Success; Failure m East-.5. Welabllylerouteemebyleemaekby Ratitoeelabilitylorcueem W mmmmmngwoumm leekdilieulqlermendelbd'm" In order to test the proposed hypotheses the mean perceived ability ratings of younger children were compared to the mean perceived ability ratings of the older childern on each set of conditions. Wham The first hypothesis proposed that for both an easy and difficult task younger children, aged 9, IO and ll years, would have higher perceptions of ability than older children, aged 12, l3, l4 and 15 years, when receiving praise after success and have lower perceptions of ability when receiving blame after failure. Comparison of the means indicated that the younger children had higher perceptions of ability than the older children for the easy/success/praise and dif f icult/success/praise conditions. Moreover, the younger children had lower perceptions of ability than the older children for -55- the dif f icult/f ailure/blame condition. However, the younger children reported a higher rating for the easy/failure/blame condition compared to the older children’s rating (see Table 14). Therefore, the first hypothesis was only partially supported. W The second hypothesis predicted that for both an easy and difficult task older children aged 12 and above would have lower perceptions of ability than younger children when receiving praise after success and neutral feedback after failure. Comparison of the means indicated that the easy/success / praise, easy/failure/neutral and difficult/success/praise were rated lower by older children. Moreover, although the diff icult/failure/neutral was rated slightly higher by the older children, this difference was only marginally significant (see Table 14). Partial support was theref ore, provided for the second hypothesis. 112201135113. The third hypothesis stated that for both an easy and difficult task children aged l2 and above would have higher perceptions of ability than children aged 9, 10, and 11 years when receiving blame after failure and neutral feedback after success. The difficult/failure/blame and difficult / success/neutral situations resulted in higher ratings of ability by the older children compared to the younger children. However, the easy f ailure/blame and easy/success/neutral situations were rated lower by the older children compared to the younger children (see Table 14). Hence, the third hypothesis was only partially supported. -55- E l . 3. S .E. E Q In order to understand the developmental sequence of the ability ratings it was necessary to look at each age group individually in greater detail. By combining the ages in Analysis 2 unique variations within an age group may have been disguised. Hence, it was necessary to look at each age separately in order to compare the different age groups to each other. 11' I I l . For the 9 year old children there was a significant effect for the combined cell means, £0,152) - 30.89, n<.001. Groups which differed significantly from each other were all the easy and dif f icult/success/praise conditions which were rated higher than the easy and dif f icult/f ailure/blame conditions. The two groups - dif f icult/f ailure/neutral and easy/failure/neutral - were rated significantly lower than both the easy and dif f icult/success/praise and the dif f icult/success/neutral groups. The easy/ success/neutral group had a lower significant mean rating than both the easy and dif f icult/success/praise groups. See Figures 6 and 7 which illustrate the three-way interaction for the nine year old subjects. Groups whose means did not differ significantly "from each other include all the failure conditions; the easy and dif f icult/f ailure/neutral and easy success/neutral; both the easy and dif f icult/success/neutral groups; and the easy and diff icult/success/praise and dif f icult/success/neutral groups. -57- 9 — ° — x Difficult 0 Easy 8 I. 8 d 7 -l 7 -—l 6 —4 6 -l O 3 9 —4 3 4 -—l < < 3 — 3 I'— 2 - 2 —l l l l 1 Neutral , Praise Neutral M Success Fellure flout-J. m1. mamweueaeemmeby WNMMWMM leeaaekbytaekdllaflylermyeereld mummmwmw children children With the exception of the easy/success/neutral condition, 9 year old children indicated a clear distinction for outcome with failure conditions resulting in lower ratings than success. In addition to outcome, 9 year old boys considered the type of feedback received in their rating of perceived ability. This group of children appeared especially sensitive to both praise and blame conditions. The mean ratings for perceived ability for successful praise conditions were the highest compared to the failure/blame conditions which have the lowest mean perceived ability rating. The neutral feedback conditions receiVed more average ratings and were placed between the blame and praise feedback. Although not statistically meaningful, it is interesting to note the order of perceived ability ratings with regard to task difficulty. Specifically, easy/failure resulted in higher ratings than failure at -53. difficult tasks. However, under success conditions, difficult tasks resulted in higher ratings of ability. Apart from this contradiction on the reversal of ratings of task difficulty, successful praise resulted in high perceptions of ability whereas f ailure/blame resulted in very low perceptions of ability, for the 9 year old children. In summary, nine year old children included all of the available information in their ratings of perceived ability. Praise from the teacher following success at either an easy or difficult task resulted in the highest ratings of perceived ability. Blame following failure of an easy or difficult task resulted in very low ratings of perceived ability. Finally, a success on an easy task to which a teacher gave neutral feedback resulted in a lower rating of ability than success on either an easy or difficult task followed by praise. Thus, the type of feedback a teacher provided may have an additive effect on a 9 year old subjects’ rating of perceived ability. This was particularly apparent for children who failed and received blame, children who succeeded and were praised, and for children who succeeded on an easy task but received neutral feedback. mm For the 10 year old children the Schef f e’ post hoc test produced a significant overall effect, [(7,152) - 5.10, 24 .001. However, in contrast to the results for the 9 year old boys, no two groups were significantly different from each other across all conditions. The three-way interaction for nine year old boys’ perceived ability ratings are presented in Figures 8 and 9. Ablllty .59- ° -‘ ° -" x Oilticutt 0 Easy 8 «- 8 -l 7 u / 7 fl 0 " / e — 5 -l ,, s - 4 _ g 4 _. 3 < 3 —I 3 -—1 2 — 2 _ l _ t ._l l 1 l 1 Neutral Praise Neutral Blame Success Failure m EDI-1M Mopseldlltylereucceeeeuteemeby - Ratinedebileylorlailureelaeemeby mummwmmou tmwmmeymmyuou children children These results are puzzling. Although there were no significant dif f erenccs statistically between the groups, the pattern established by the 9 year old subjects was followed. For the outcome factor, the success conditions were rated higher than the failure conditions. For feedback, blame conditions resulted in the lowest perceived ability ratings followed by neutral f ecdback, with praise rated the highest. The most notable dif f erencc, in ordering of perceived ability, occurred with task difficulty under failure conditions. The 9 year old children rated ability lowest for subjects in a difficult/failure/blame situation while the 10 year old boys rated subjects in the casy/failure/blame condition lowest. In contrast to 9 year old boys, 10 year old boys were less critical of ability following failure. Likewise, 10 year old: were more critical of ability given success. This may explain part of the overall f our-way interaction. In summary, the relationship between outcome, task difficulty, feedback and ratings of perceived ability by ten year old boys remains unclear. The fact that the overall Schef f e’ test was significant but no two groups differed significantly was discouraging. However, in comparison to the 9 year old boys, these 10 year old children rated ability higher following failure and lower following success. For this group of 10 year old subjects, failure was not so terrible and success was not so great. This difference between 9 and 10 year old subjects may suggest some cognitive differences in the use of available information between the two age groups, as proposed by Nicholls and Miller (1984). W The Schef f e’ post hoc test was significant across all conditions for the eleven year old children, [(7,152) - 26.134, p ( .001. For this group of children all the success conditions differed from all the failure conditions. No other differences within the groups were evident and the ratings for the failure conditions were not significantly different from each other. Moreover, there were no differences within the success conditions. The three-way interaction for the 11 year old boys perceived ability ratings are plotted in Figures 10 and 11. ~61- 9 - 9 "' x Oillicult 0 Easy d — a e— 7 "" / 7 — 6 -'l a — 2‘ 3: n n 5 < < ‘ — ‘ fl 6% 3 — a -—-l 2 —- 2 —1 ‘ — 1 — I l l T Neutral Praise Neutral Blame Success Failure Elms-.111 mu Ratlngsclwiltylereucceesetaeomeby Ratingeelabilkylorlaluteeutcomeby leedsackbytaakdl'llcultylorelevenyearcld leerbaekbytaak ditlieulylorelevenyeareld children chilaen The mean perceived ability ratings for the 11 year old boys are ranked similarly to those of the 9 and 10 year old boys for the factors Outcome and Feedback. That is, the failure conditions were rated lower than the success conditions. And for feedback, blame was ranked lowest followed by the neutral conditions, with praise rated the highest. The 11 year old subjects did differ from the 9 and 10 year old boys however, in their assessment of task difficulty. Comparison of the 11 year boys’ perceived ability ratings for task difficulty to that of the 9 and 10 year old subjects revealed differences in the way these children interpreted the situations presented. For instance, the 10 and 11 year old children rated the easy task accompanied with failure -52- and blame the lowest in contrast to the 9 year old boys who rated the difficult task together with failure and blame as the lowest perceived ability score. Moreover, whereas the 10 and 11 year old children rated the easy task under failure with neutral feedback below the dif f icult/f ailure/neutral condition the opposite was true for the 9 year old boys. However, the 11 year old boys ranked the easy task above the difficult task under success/neutral conditions whereas the reverse was reported for the 9 and 10 year old boys. In sum these three age groups showed a similar pattern in the way the factors outcome and feedback were ordered. They were dissimilar in that task difficulty was evaluated differently by each age group. As the age group increased from 9 to 11 years, apart from the 10 year old boys, a difficult task was rated at some point below an easy task. This result was found in the failure blame conditions for the 9 year old boys and the success/neutral conditions for the 11 year old boys. Apart from these difference in ratings of task difficulty, the younger age groups (9, 10, and 11 year old children) recorded high perceptions of ability for success/praise and low perceptions of ability for failure blame conditions. These results provide partial support for hypothesis 1. mm: For the 12 year old children the Schef f e’ post hoc test produced a significant overall effect, {(7,152) . 27.81, p,< .001. Similar to the eleven year old boys, this group of children rated all the success conditions significantly different from all of the failure conditions. No other differences were evident within the groups. Furthermore, none of the failure groups differed significantly from each other; neither did any of the groups in the success conditions differ significantly. The three-way interaction for the perceived ability ratings of the twelve year old children are plotted in Figures 12 and 13. -53- 9 " 9 — x Otilicull 0 Easy 8 — 8 '- 7 — ~74 7 " g ‘ —i i 4 —i < < 2 — 2 -—l ‘ — ‘ — l l I 1 Neutral Praise Neutral Blame Success Failure w m Rathgeelabiliyloreueeeesoutcomeby Ratingetoitltylerlalhteouteomeby teeaaek by task diliculy for twelve year old leerback by task dillicuity tor twelve year old children children Once again, the order in which the 12 year old children have rated perceived ability is very similar to those of the 9, 10 and 11 year old boys on the factors of outcome and feedback. Failure conditions are rated lower than success conditions. Feedback conditions were rated from low‘ to high with blame, neutral and praise in that order. The ratings based on task difficulty however, indicated a few notable similarities and differences compared to the ratings for the previous age groups. The 12 year old boys were similar to the 10 and 11 year old boys but different from the 9 year old boys in two respects. First, the easy task with f ailurc and blame had the lowest perceived ability score. And secondly, the easy task with failure and neutral was rated below difficult/failurc/neutral. -54. For the easy and difficult tasks combined with success and neutral conditions, the 12 year old children only dif f ercd from the 11 year old children. Similar to the 9 and 10 year old boys the 12 year old boys ranked easy/success/neutral below dif f icult/success/neutral in the order of ratings. The 12 year old boys differed from all the younger age groups in regard to rating the easy task with success and praise above the dif f icult/success/ praise task which would appear to be a contradictory finding. Apart from this the 12 year old boys, like the younger age levels, valued praise as an indicator of high ability; blame on the other hand was typical of low ability, and neutral teacher evaluative behavior reflected average ability. In summary, 12 year old children used outcome as the basis for their rating of ability. Children who experienced success were perceived to have greater ability than children who experienced failure. I] . I H l . For the thirteen year old children the Schef f e’ post hoc test was significant, £0,152) - 14.464, n< .001. Several groups differed significantly from each other for this age group. All the success conditions were rated significantly higher than the easy/failure/neutral condition. The easy and diff icult/success/praise and difficult/success/neutral were rated significantly higher than the easy/failure/neutral and blame ratings. Both the difficult/success/praise and neutral ratings were also rated higher than the dif f icult/f ailure/neutral and blame conditions. Several groups did not differ statistically from one another. All the failure group means were relatively close as were the success group means. The diff icult/failure/blame, easy/success/neutral and diff icult/failure/ neutral were not significantly different. Furthermore, the difficult/failure -65- with neutral and blame were not different from the easy/success/neutral and praise, perceived ability ratings. The three-way interaction for the perceived ability ratings of the thirteen year old boys are plotted in Figures l4 and 15. ° - 9 -* x Difficult 0 Easy 8 -1 8 —-l 7 .4 N 7 .— z. 5 —1 r 5 —l 2: ‘ E .. n 2 — 2 ‘ .1 f/ 3 — 3 —-l 2 — 2 — 1 —i 1 —1 r l . T 1 NOW?" Praise Neutral Blame Success Failure Hill—L4 M Ratings at abilly for success outcome by Ratings at ability tor lailure outcome by leedsack by task dilliculty tor thirteen year old leedback by lack dilliculy lcr thirteen year old children children The means for the various conditions indicated that the easy/failure/neutral group had the lowest mean. For difficult situations success with its accompanying feedback was rated far higher than f ailurc and its feedback. For thirteen year old children praise appeared to be more credible than blame. However, neutral feedback under a difficult success condition had the highest rating. Comparison of the 13 year old children to the younger age groups revealed a similarity on the factor of outcome in that all the failure conditions were rated lower than the success conditions. However, notable differences were evident in the order of the ratings based on feedback and task difficulty. -66- For the factor task difficulty the 13 year old boys, in contrast to the 9, 10, 11 and 12 year old subjects, ordered the easy conditions close to each other and the difficult conditions close to each other, depending on the outcome. The order of the ratings had the two/easy failure conditions ranked lowest followed by the two difficult failure conditions; next were the two easy/success conditions and finally the two dif f icult/success conditions were rated the highest. For the factor of feedback the 13 year old boys differed from the 9, 10, 11 and 12 year old subjects under several conditions. This age group had easy/failure/ncutral as the lowest rating compared to easy/failure/blame for the 10, 11 and 12 year old boys and difficult/failure/blame for the 9 year old boys. Easy/failure/blamc was ranked next by the 13 year old boys which was similar to the 9 year old boys rating but, different from the dif f icult/f ailure/blame of the 10, 11 and 12 year old boys. The third lowest rating was dif f icult/f ailure/neutral as it was for the 9 year old boys whereas the 10, 11 and 12 year old boys had easy/failure/neutral ordered third in the ratings. The 13 year old boys rated dif f icult/f ailure/blame higher than dif f icult/f ailure/neutral, yet lower than easy/success/neutral. This was in contrast to the perceived ability scores according to feedback of all the younger age groups. Finally, the diff icult/success/neutral condition was rated highest, above the two difficult/praise conditions, which again was unique for this age group of children. In summary, the difference in the results for the feedback condition for the 13 year old boys compared to the younger age groups may have contributed to the two way interaction between age and feedback. This factor does not follow the same pattern reported for the 9, 10, 11 and 12 year old boys. This older group of children began to show signs of lowering perceptions of -67- ability for easy and difficult/success/praise conditions and higher perceived ability rankings for the easy and dif f icult/success/neutral conditions and the dif f icult/f ailure/blame condition. This trend partially supports hypotheses 2 and 3. W The Schef f c’ post hoc test for the 14 year old children yielded a significant result, E(7,152) - 14.64, pe .001. Results for this age group were similar to the findings for the 11 and 12 year old boys. All of the success conditions differed significantly from all of the failure conditions. Groups which did not differ significantly from one another included all the failure conditions and all the success conditions. The three-way interaction for the fourteen year old boys are illustrated in Figure 16 and 17. e — ' " x Ditties! 0 Easy 0 — I — ‘ - Q — E 5 - 3‘ s "i 3 _ 3 _. 2 - 2 -I l _ l —-i I l l I "00"“ Prabe Neutral Blame Success Failure m m MktosctabiDylcrsucceescutccmeby Batingsetebldtyicrlalmeeutcenteby leeaeekbytaskdllieuaylerleuneenyearold leedsackbytaskdflicutylcrlctn‘teenyeeretd children children -53- The 14 year old boys were similar to the younger age groups in rating failure outcome conditions lower than success outcome conditions. Certain similarities and differences were evident for the 14 year old boys’ ratings compared to those of the 9, 10, ll, 12 and 13 year old boys on the factors task difficulty and feedback. The success conditions were ordered difficult, easy, easy, difficult. However, for the 14 year old boys, neutral feedback resulted in higher perceived ability ratings than did praise in contrast to the 11 year old boys. For the factor task difficulty the 14 year old boys, like the 13 year old boys, rated the two easy tasks under failure outcome as the lowest followed by the two difficult tasks under failure outcome. This was different from the results reported for the 9, 10, 11 and 12 year old subjects’ perceived ability scores. Other similarities related to task difficulty for the 14 year old boys with the ratings for the 11 year old subjects, which together differed from those of the 9, 10, 12 and, 13 year old subjects. For the factor feedback no clear pattern was visible for the 14 year old boys. This age group rated easy /fai1ure/blame as the lowest, similar to the 10, 11 and 12 year old boys but in contrast to the ratings of 9 and 13 year old children. Second in the ordering was easy/failure/neutral which was different from that of all the younger age groups, as was the third perceived ability rating of diff icult/failure/blame . Difficult/failure/neutral was ranked fourth as it was for the 10, 11 and 12 year old boys. Both praise conditions were ranked next, which no other age group shares with the 14 year old boys. The two highest ratings were successful neutral conditions, which again was different from that of every age group except the 13 year old boys who had diff icult/success/neutral rated highest, also. -59- Clearly the 14 year old boys have contributed to the f our-way interaction of age by outcome by task difficulty by feedback as well as the two, two-way interactions of age by task difficulty and age by feedback. Similar to the 13 year old boys, in comparison with the younger age groups, the 14 year old boys had higher perceptions of ability for dif f icult/f ailure/blame and neutral as well as neutral being rated the highest under easy and dif f icult/success conditions. Praise was not ranked as highly by the 14 year old children. This evidence provides further partial support for hypotheses 2 and 3. ET I I l . The Schef f e’ post hoc analysis for fifteen year old children resulted in a significant effect for the ability ratings, £0,152) - 17.40, p_< .001. Several groups were significantly different in their mean perceived ability ratings. The easy and dif f icult/success/praise and diff icult/success/neutral groups were rated higher than both the easy/failure/blame and easy/failure/neutral groups. The difficult/success/ neutral and praise groups were rated significantly higher than the dif f icult/ f ailure/neutral and blame conditions. Finally, the difficult/success/praise group differed from the easy/success/neutral group which had a far lower rating. The groups easy and dif f icult/f ailure with neutral and blame feedback along with easy/success/neutral indicated no differences between their mean ratings. There were no differences between the dif f icult/f ailure/neutral and blame and the casy/success/neutral and praise. Furthermore, the easy/success/neutral and praise together with the diff icult/success/neutral did not differ from each other. Finally, the diff icult/success/neutral and praise and the easy/success/praise indicated no difference in their mean ratings. The three-way interaction for the fifteen year old boys perceived ability ratings are plotted in Figures 18 and 19. -70. 9 T 9 1 x Dcttlcult 0 Easy a -—l B -—i 1 ._ /x 7 -+ 6 —‘ 6 -—l i 4 _ g 4 _ M < 4 2 — 2 — ‘ — 1 — l l l 1 Neutral Praise Neutral Blame Success Failure figural. W Ratinp st ability tor success outcome by Ratings at $in tor tailors «deems by lesaaek by task dltlculty tor titteen year old testbed: by task dltleuly tor tittssn year old children children The results for the 15 year old children were similar to the findings for the 13 year old children. For difficult success, with praise or neutral feedback, higher ratings were given than for difficult failure conditions with neutral or blame feedback. Furthermore, easy tasks with a successful outcome were rated higher than for the easy tasks with teacher blame or neutral feedback. Once again for the 15 year old children, praise was valued more highly than blame in regard to teacher’s responses. .71- Of the three factors, outcome, task difficulty and feedback, the 15 year old boys showed a similar trend to all the other age groups by rating failure outcomes lower than success outcomes. Notable similarities and differences to the other age groups again were apparent on the variables of task difficulty and feedback. For task difficulty the, 15 year old boys had exactly the same order as the 13 year old boys with two easy task difficulty ratings followed by two difficult task ratings for failure and success outcomes. The task difficulty ordering by 13 and 15 year old boys was different from that of all the other age groups. There were several notable findings for the factor of feedback for this age group. Similar to the other ages, except for the 13 year old boys, blame feedback received the lowest rating for an easy task. Similar to the 14 year old boys and different from all the other age groups, the two neutral feedback under failure conditions were ranked next. The 15 year old boys also paralleled the order for blame under difficult/failure followed by neutral and then praise for easy/success conditions to that of the 13 year old boys. Moreover, unlike the other age groups, the 15 year old boys had neutral followed by praise feedback as the two highest ratings under diff icult/success conditions. These similarities and differences for the 15 year old boys clearly contributed to the interaction effects reported previously. A similar trend to the 13 and 14 year old boys was apparent for this age group with neutral feedback under success conditions and blame in the difficult failure condition being indicative of higher ability; however this group did rate difficult success praise very highly. These results provided partial support for hypotheses 2 and 3. -72- Summau Three separate analyses were computed from the data collected in this study. Analysis 1 was a replication of a study done by Meyer et a1. (1979). The results of this study produced three significant main effects for outcome, task difficulty and feedback, together with a significant three-way interaction between these variables. These results failed to replicate the findings reported by Meyer et a1. (1979). Analysis 2 was similar to analysis 1 with the additional variable of age being investigated. Here the results yielded no significant main effect for age, a significant f our-way interaction for age by outcome by feedback by task difficulty, a two-way interaction for task difficulty by age, and a two-way interaction for feedback by Age. Analysis 3 involved detailed comparisons of each individual age group to understand more fully the f our-way interaction. The most direct test of the hypotheses was accomplished via a discriminant analysis performed with the subjects grouped according to younger and older children. The discriminant analysis provided partial support for the hypotheses. CHAPTER 5 DISCUSSION, SUMMARY AND RECOMMENDATIONS One purpose of this investigation was to replicate a study conducted by Meyer et al. (1979) which was concerned with the question of whether, under certain conditions, a person who ispraised or blamed is provided with information about how the person who is praising or blaming perceives their ability, which in turn will influence the self -perception of ability. This replication was specifically designed to determine the effects of praise and blame on self -perceptions of ability involving a sport task. The question which guided this replication was to see if similar results would be found to those reported by Meyer et al. (1979), who studied adults in an academic setting, by further investigating children in a hypothetical sport setting. The present results revealed a different set of findings and failed to support the results of Meyer et al. (1979). In their original study Meyer at al. (1979) reported that neutral evaluative behavior was superior to praise following successful mastery of both easy and difficult tasks in facilitating self-perceptions of ability. The present study found praise to be more conducive to formulating self -ability perceptions than neutral feedback. For failure at an easy and difficult task Meyer at al. (1979) posited blame feedback to result in higher self -perceptions of ability compared to neutral feedback. These results of Meyer et a1. (1979) were found for adults who probably identified with the teacher. The present study, however, found the opposite to be true with children. Both the present study and the Meyer et al. (1979) study reported conflicting results. One possible explanation for this discrepancy between the mean perceived ability ratings may be that this study had greater statistical power owing to the larger sample size of 140 compared to the 17 -73. .74. subjects in the study of Meyer ct al. (1979). However, the results should still be in the same direction if power was the only reason. Moreover, the accessible population for this investigation was boys, whereas Meyer et al.’s (1979) sample was comprised of both male and female subjects. However, Meyer et al. (1979) reported that there were no significant differences between the answers of male and female subjects to the four types of questionnaires. Another possible explanation for the obtained divergencies in the ability ratings could be an artifact of the experimental method. The questionnaires used in the third experiment by Meyer et al. (1979) provided information about three persons - two adult unknown performers who were evaluated and a teacher who provided differential feedback. Hence, the subjects were outsiders in this situation. In the present study each subject was asked to imagine himself to be a student who was one of the performers experiencing the differential feedback given by the teacher who was an adult. These situations. may have been interpreted differently in the two studies owing to the personalized nature of the response. The fact that this study involved a hypothetical sport situation, as opposed to an academic situation used in the Meyer et al. (1979) study, may also provide further insight into the divergent ratings of the two studies. There is a major difference between performing cognitive and physical tasks. This striking distinction may imply differences in the development of the concept of ability, based on the teacher’s evaluative behavior in sport versus it’s development in the classroom. In support of this proposition, Roberts and Pascuzzi (1979) argued that the nature of ability, effort, and task difficulty in sport are more revealing than in cognitive tasks. For example, in the present study it was probably easier for the subjects to visualize that the two performers were trying hard in the difficult success condition -75- involving the performance of a somersault. Compare this to the subjects in the Meyer et al. (1979) study who had to imagine that the two students were performing a difficult academic task. Hence, the demonstration of effort would appear much more apparent in the present study. The very natue of the performance process and outcome is more obvious in sport activities than academic tasks. It could be argued because of this that the subjects in the Meyer et al. (1979) study relied on the teacher’s evaluation more than the subjects of the present study in the assessment of ability. Perhaps praise is valued more in sport compared to academic settings and blame is more indicative of low ability in classroom settings. Related to this, Horn (1985) suggested that because children place a high value on their competence in sport activities, adult feedback may be especially influential in the athletic context in which children’s motor competence is continually being evaluated. In the sport setting this relationship between coaches’ feedback and children’s psychological responses has been tested by Smith and Smoll and their associates (Smith et al., 1979; Smoll et al., 1978). Similar to the findings of the present study, these researchers found that positive coaching behaviors were more effective in facilitating childrens’ self -perceptions than negative coaching behaviors. A more plausable explanation for the different results reported by Meyer et al. (1979) contrasted to the present findings concerns the age of the subjects in each study. Meyer et al. (1979) tested adults who were second and third year undergraduate students, whereas the sample in this study consisted of children aged 9 to 15 years. In the first two experiments conducted by -76- by Meyer et al. (1979), in which these researchers were interested in the self-perceptions of children and adults to praise and neutral feedback, they found that younger and older subjects differed in their cognitive structures based on the way they processed identically perceived information. Young children (aged 9) indicated that the teacher thought the student whom he praised was the brighter one; children aged 10 to 14 choose the options praise and neutral in approximately equal numbers whereas subjects from 15 to 19 years of age predominantly thought that the teacher gave the neutral feedback to the student whom he considered brighter. These results support the findings of the present study in that praise feedback was associated with the more able student for the younger children (age 9, 10 and 11 years) whereas there was a trend with increasing age (12 to 15 years) for neutral feedback to indicate higher ability. Alternatively, this could mean that younger children use teacher feedback in formulating self ~perceptions of ability whereas older children place greater emphasis on task difficulty. In the third experiment by Meyer et al. (1979) however, the observed developmental divergencies among the ability ratings, reported in experiment one and two, are not investigated further. Instead this group of researchers choose to investigate the additional factors of outcome, task difficulty and the effects of evaluative feedback on adult subjects, only. The second purpose of the present study was to replicate and extend the third experiment of Meyer and his colleagues (1979) by investigating further the reported developmental differences in perceptions of ability. The developmental theory of achievment motivation proposed by Nicholls (1984) and further refined by Roberts (1984) to the context of sport, provided a theoretical basis for studying the factors of outcome, task difficulty and perceived ability. -77. This theoretical model posits that young children compared to older children use different criteria to judge self-perceptions of ability. That is, a child’s perception of personal competence relates to an assessment of outcome judged relative to the performance of peers, effort required and the level of task difficulty. This assumption is supported by the research of Meyer et al. (1979; 1982). Hence, the question guiding the second part of this study was whether developmental differences were apparent in self -perceptions of ability on the factors of outcome, task difficulty and feedback. A discriminant analysis, which combined the younger age levels of 9, 10 and 11 years and the older age levels of 12, 13, 14 and 15 years, found support for developmental differences in self -perceptions of ability based on the factors of outcome, task difficulty and feedback. However, only partial support was provided for the proposed hypotheses because developmental differences were not evident across all conditions. The first hypothesis stated that for success and failure at an easy and difficult task young children (age 9, 10 and 11 years) would have higher perceptions of ability for praise and lower perceptions of ability for blame than older children (age 12, l3, l4 and 15 years). This hypothesis was supported for the following conditions: easy/success/praise, difficult / success/'praiseand diff icult/f ailure/blame. However, the condition of easy / f ailure/blame was not supported by these subjects. These results indicated that the younger children appear to be very sensitive to positive and negative evaluation received from a teacher. Blame .73. in a failure situation was very detrimental, resulting in extremely low perceptions of ability. In contrast, praise following success was conducive to very high perceptions of ability. These results taken together lend support to the first two experiments by Meyer et al. (1979). The research of Smith and Small and their associates (1979, 1978) has also indicated that in a sport setting positive coaching behavior enhances young childrens’ self -perceptions, while negative coaching behavior resulted in lowering childrens’ self -perceptions. For the young children these high and low perceptions of ability as a consequence of praise and blame can be explained in terms of Nicholls’ (1984) developmental theoretical perspective on perceived ability. Up to the age of 12 children attribute performance outcomes to both ability and effort. Conceptions of ability and effort are only "partially differentiated." Unlike the older children, the younger children are not as capable of perceiving the , antecedents of evaluative teacher feedback to imply certain inferences about ability. Teacher praise implies high ability when effort and ability are undifferentiated. The differentiation from effort appears necessary for the inference of low ability from higher praise (Nicholls 1: Miller, 1984). For the performer who was praised in the success situation the young children may have perceived that the teacher thought this student had done well by trying hard. For these children high effort implied high ability. Hence, the high ability ratings for the student praised following performance. In the case of the student who was blamed for failure the younger children perceived that this student did not try hard enough as a result of the teacher’s reaction who was annoyed by the lack of effort. Here, low effort implied low ability owing to the children’s inability to separate the concepts of effort and ability. -79- It is not clear why the easy failure blame condition did not follow this line of reasoning. A tentative explanation for this non-support could be that the young children sympathized with the failing performer and it was wrong for the teacher to blame the student in this manner. However, Nicholls and Miller (1984) stressed that an inference of valid ability demands evidence that optimum effort was employed. That is, ability will not be revealed if effort is low. Hence, the teacher’s blaming may have indicated to the younger children that the performer was not trying hard enough, causing the teacher to be upset. Moreover, the younger children’s inability to perceive cues of ability and effort from the teacher’s evaluative behavior may have interfered with their assessment of ability because maximum effort was not exerted by the performer of the easy task. The second hypothesis claimed that for both an easy and difficult task, older children would experience lower perceptions of ability than younger children when receiving praise after success and neutral feedback after failure. This hypothesis was supported for the conditions easy success praise, difficult success praise and easy failure neutral. No support was provided for the difficult failure netural group. However, the mean difference was minute. For the older children overt praise following successful mastery attempts and neutral evaluation by the teacher after failure on the easy task was detrimental to self -perceptions of ability. These results support the findings of Horn (1984) who indicated that coaches’ responses to players’ successful performance via praise were not facilitative of players’ perceived competence. Horn (1984) reasoned that inappropriate or noncontingent praise may have established lower expectations for players’ performance, thereby inducing negative self -perceptions in skill competence for the nineth grade children in her study. .30. Moreover, this negative consequence of praise has been found to occur in classroom settings (Kennedy & Willicutt, 1964; Dunkin & Biddle, 1974) and with elite athletes (Tharp & Gallimore, 1976). The reason why praise should negatively influence self ~perceptions of ability for children age 12 to 15 in this study can be explained in terms of Meyer’s (1979; 1982) previous research and Nicholls’ (1984) developmental model. Success of the performer who was praised was probably ascribed to effort expenditure. Supporting this notion, Meyer et a1. (1979) reasoned that through cognitive appraisal we assume easy tasks require more effort when ability is lower. Similar conclusions about differing amounts of effort and ability are drawn in the case of a difficult task, although the abilities of both performers will be rated higher. Here again, cognitive appraisal would assume that success at a difficult task can only be obtained if the ability is comparatively high. Hence, the older subjects in'this study were able to make such cognitive appraisals owing to their maturational capacity to differentiate between effort and ability, which develops at the age of 12 (Nicholls, 1984). These results support Nicholls and Miller’s (1984) contentions and imply that with development it becomes increasingly difficult for teachers to use praise to raise perceived ability. Support was found also for the easy failure neutral condition. That is, the older children were associated with lower ability ratings than the younger children. Adults have been found to react with sympathy or pity for lack of ability (Nicholls & Miller, 1984). Thus, for the older subjects failure at an easy task coupled with the neutral teacher response was indicative of low ability, their interpretation may have been that the teacher was sympathetic after failure of an easy task. -31- The difficult failure neutral group did not support the assumption of lower perceived ability ratings by the older children compared to the younger age groups. One possible explanation for this could be that for young children, with undifferentiated effort/ability conceptions, high ability is implied by sucess at a task they are uncertain of being able to complete (Nicholls & Miller, 1984). Hence, failure at the difficult task resulted in the lower ability ratings of the younger age groups compared to the older. The third hypothesis stated that for both an easy and difficult task older children would have higher perceptions of ability than younger children when receiving blame after failure and neutral feedback after success. Support for this hypothesis was provided for the difficult failure blame and difficult success neutral conditions only. The easy failure blame and easy success neutral conditions did not support the hypothesis. For the older children task difficulty and feedback were important factors in their appraisals of ability. These results can partly be explained by the assumptions proposed by Meyer et al. (1979). For the two performers who experienced identical failures, the one who was blamed for the outcome more than the other led the older subjects to conclude that the former was perceived to have expended less effort on the task. Through cognitive appraisal the higher the ability the more the result was ascribed to low effort expenditure, because very low ability will lead to failure no matter how much effort was expended (Meyer et al., 1979). Apparently this assumption was only true for the difficult failure blame condition and not the easy -32. failure blame condition. Perhaps the older subjects perceived the task of a forward roll to be so easy that in spite of the teacher’s critical reaction to one performer, they were unable to convince themselves that this performer had high ability and was not expending enough effort to perform this simple task. The notion that adults react with anger to perceived lack of effort is not supported by the older children’s lower ratings of perceived ability compared to the higher ability ratings by the younger children for the easy failure blame condition. The results reported here suggest that task difficulty is a more valid predictor of ability than teacher’s feedback for the older children. . The older children may have preferred high ability rather than high effort. A study by Harari and Covington (1981) found that up to thirteen years of age both high effort and high ability are valued. After thirteen years of age ability is valued more highly. In line with Nicholls and Miller’s (1984) reasoning, older children would prefer ability more than effort because it implies a “capacity” to do well. The shift in emphasis after twelve years of age from teacher feedback to task difficulty, found in the present study, may support this notion. Again, a differentiated ef f ort/ability capacity of the older children would result in the difficult failure blame condition being accounted as high cf fort, whereas the difficult success praise condition would occur as a consequence of high ability. The difficult success neutral condition also received higher ratings of ability from the older children. One reason for this could be that the teacher response ‘Good’, which comprised neutral feedback, was not neutral but qualitatively evaluative. The success neutral condition was paired with difficult success praise on the questionnaire. Hence, another possible -33- explanation for this high rating again relates to perceived effort expenditure as it relates to ability. Because the two performers Obtained identical successes and one was praised more for his outcome than the other, from the research of Meyer et a1. (1979) and the theoretical postulations of Nicholls (1984), the success of the former was ascribed to ef f ort-expenditure. Hence, the performer receiving neutral feedback was perceived as exerting less effort than his peer by the older respondants. And, on the basis of Nicholls’ (1984) developmental theoretical postulations these older children were able to differentiate between effort and task difficulty in comparison of peers’ via ability norms. The easy success neutral condition did not support the third hypothesis. Perhaps one of the reasons why the older subjects did not rate this condition higher than the younger subject’s ratings was that they perceived the teacher to favor the performer who was praised. In spite of the questionnaire V condition stating that the teacher liked both students equally well, the older subjects may not have been'convinced. Here again the qualitative wording of the neutral response ‘Good’ by the teacher, may have influenced the older subjects’ response. Moreover, as discussed previously, task difficulty was a good predictor of ability for the older children because more ability was required for successful mastery of difficult tasks. And for older children ability is preferred to effort (Harari & Covington, 1981) because it implies an innate capacity to achieve (Nicholls & Miller, 1984). -34- The results of this study provide partial support for Nicholls’ (1984) theoretical development model of perceived ability together with Meyer et al.’s (1979) predictions that a teacher’s evaluative behavior may influence children’s self-perceptions of ability. Based on these results, because young children judge performance outcomes with an undifferentiated conception of ability in successful situations, praise results in perceptions of high ability. To these children effort was indicative of ability; hence, effort should be rewarded positively. In failure situations blame, however, is indicative of low effort which is comparable to ability and should thus be avoided by teachers of young children. In line with Nicholls and Miller (1984), the younger children based their judgements of ability with reference to self rather than a social comparison. The personalized nature of the teacher’s response resulted in an over-estimation of ability following praise and an under-estimation following blame. For older children aged 12 to 15 years teachers and coaches should still strive to give support and encouragement through positive feedback. However, owing to the capacity of these children to differentiate between the conception of effort from ability, praise should be used with caution as it may imply low ability. Moreover, while blame may lead to increasing inferences about ability it should be used sparingly as it is not as facilitative as praise. The older subjects may have relied more on social comparison of peers in judging their own ability in relation to performance norms. This may account for the two-way interaction for age by task difficulty. Ability was rated highly on tasks which were difficult and could only be performed by a few. SUMMARY This study investigated the degree to which teacher praise and blame may affect children’s self-perceptions of ability in a sport setting. The present study had two specific purposes: (a) to replicate experiment number three conducted by Meyer et al. (1979), and (b) to investigate possible developmental differences in ability ratings of children (9 to 15 years of age) as a result of differential feedback provided by a teacher. The results of this study failed to replicate Meyer et al. (1979). Specifically, praise after success led to the conclusion that the acting person’s ability was perceived as high. Blame after failure led to the inf erencc that the acting person’s ability was low. However, developmental differences were apparent in the interpretation of teacher feedback based on the different ability ratings of the young compared to the older children. The present study found that children aged 9, 10 and 11 years tended to rely on outcome and teacher feedback in judging self -perceived ability. Older children aged 12, l3, l4 and 15 years tended to rely more on outcome and task difficulty in assessing perceived ability. These developmental differences in childrens’ interpretation of differential teach‘er feedback, as indicated by the ability ratings, provided support for Nicholls’ (1978) developmental theory of perceived ability. With increasing age children were more capable of to differentiating the concepts of ability/cf fort and ability/task difficulty and were thus less likely to rely solely on the teacher’s evaluation in formulating their ability appraisals. -35- W This study has provided partial support for Nicholls’ (1978; 1984) developmental theory of perceived ability. Age related changes in evaluative responses, when combining the variables of outcome, task difficulty and feedback were noted. However, the complex nature of the interaction between these variables has perhaps raised more questions than answers. Further replications are needed to test Meyer et al.’s (1979) study in classroom settings as well as the present study in sport contexts. The results of this study should be treated with caution as the differences between mean perceived ability ratings of the older and younger children did not differ by a great amount. More research is needed on the effects of evaluative behavior in the sport setting. In addition to the complex interaction of the various factors involved, specific contests in which evaluative behavior is administered needs to be investigated. It is difficult to determine conclusively from the findings reported here which factors are the best predictors of perceived ability estimates. These results suggest outcome may be the best predictor for all age groups with both task difficulty and feedback serving as additive factors for younger children. For older children, task difficulty may be more useful in determining perceptions of ability. Future investigators would be advised to employ Harter’s (I981) Perceived Physical Competence Scale. This may highlight further what type of child is prone to high or low perceptions of ability on the factors investigated in this study. Mcthodologically, future studies should test for any possible order effects in (a) the order in which the questionnaires are answered and (b) whether the subjects should receive neutral or praise/blame feedback. -37- Moreover, due to the complexity of the design in the present study, the additional factor of gender was not investigated. Further research should look at possible gender differences, in perceived ability ratings, as this may have important implications for instruction. Additional studies which test the theoretical link between evaluative behavior and Nicholls’ (1978; 1984) developmental theory of perceived ability will help identify the teaching behaviors that are most conducive to children’s psychological development. REFERENCES W Alley, S. E. (1974). Athletics in education: The double-edged sword. W 515.. 102- 105 Ammons, R. B. (1956). Effects of knowledge of performance: A survey and tentative theoretical formulation. MW 53, 279-299. Bandura, A. (1977). Self-cf f 1cacy Toward a unifying theory of behavioral chaNsc W M 191-215 Brophy. J. & Evertson. C. (1976). Wm W Boston: Allyn & Bacon, Inc. Brophy. J. & Good. T. (1974). Wins: Cams-91.111111 consequences. New York: Holt, Rinehart 81 Winston. Brophy, J. a Good, T. (1970). Teachers’ communication of differential expectations for children’ s classroom performance. Some behavioral data W 6.1- 365- 374 Burton, D. a Martens, R. (1986). Pinned by their own goals: An exploratory investigation into why kids drop out of wrestling. W Eur-111212111. 3. 183-197 Carton. AN. (1984). WWW London, Ontario, Canada: Sports Dynamics. Cook, H. a Stingle, S. (1974). Cooperative behavior in children. Eur-WM 11- 918-933 Cooper, H. M. (1979). Pygmalion grows up: A model for teacher expectation communication and performance influence. W 12, 389-410. Cooper, H. M. & Baron, R. M. (1977). Academic expectations and attributed responsibility as predictors of professional teachers’ reinforcement behavior MW 52. 409-418 Covington, M. V. & Omelich, C. L. (1981). As failures mount: Affective and cognitive consequences of ability demotion in the classroom. M W 13. 796-808 Crowe. P. (1977). W W 91m Unpublished doctoral dissertation, University of North Carolina at Greensboro. Dunkin, M., & Biddle, B. (1974). W; New York: Holt, Rinehart & Winston. -33- -39- Dweck, C. S. & Repucci, N. D. (1973). Learned helplessness and reinforce- ment responsibility in children. WW Wm 25.. 109- 116 Ewing, M- E., Roberts, G. C., & Pemberton, C. L. (1984). A developmental look at children’s goals for participating in sport. L111 niv 9:31;! of Illinois at W. Unpublished manuscript. Feltz, D. L. & Peklichkof f , L. (1983). Perceived competence among inter- scholastic sport participants and dropouts. WM mm 8- 231-235. Flanders, M. A. (1970). W. Reading, Mass: Addison Wesley. Good, T. (1981). Teacher expectations and student perceptions: A decade of research Washin- 18- 415-422 Gould, D. (1980). W. East Lansing, MI, Institute for the Study of Youth Sports. Griffin, N.S., & Keogh, J. F. (1982). A model for movement confidence. In J.A.S. Kelso & J. E. Clark (Eds.), v 1 v r 1 W New York: John Wiley. pp. 213-238. Haas, H., & Maehr, M. (1965). Two experiments on the concept of self and the reaction of others MW 1.. 100-105. Harari, O. & Covington, M. V. (1981). Reactions to achievement behavior from a teacher and student perspective. A developmental and student perspective. A developmental analysis. MW 101ml. 1.8. ”-28 Harter, S. (1978). Ef f ectance motivation reconsidered: Toward a develop- mental model. W 21, 34-64. Harter.S.(l979). WWW Denver. CO.: University of Denver. Harter, S. (1981). The development of competence motivation in the mastery of cognitive and physical skills: Is there still a place for joy? In G. C. Roberts & D M Landers (Eds) W 1.28!!- Champaign, Illinois. Human Kinetics. Horn, T. S. (1982). The influence of coaching behaviors on young athletes’ perceptions of competence and control. W W. Horn, T. S. (1985). Coaches feedback and changes in children’s perceptions of their physical competence. W31. 21 (21. 174-186. -90- Horn, T. S. & Hasbrook, C. (1984). Information components influencing children’ s perceptions of their physical competence. W V - . Jacobs, M., Jacobs, A., Feldman, G, & Cavor, N. (1973). Feedback 11- The "Credibility Gap" Delivery of positive and negative and emotional and behavioral feedback in sroups Wan]. 1251121191931. 11. 215- 223 Kennedy, W. A, & Willcutt, H. C. (1964). Praise and blame as incentives. W 52. No (5) 323- 332. Kirschenbaum, D. S., a Smith, R. J. (1983). A preliminary study of sequencing effects in simulated coach feedback. MW 5, 332-342. Kleinfeld, J. (1957). Effective teachers of Eskimo and Indian students. Wear. 83.. 301-344. Kuckla, A. (1978). An attribution theory of choice. In L. Berkowitz (Ed. ), meol. 2). New York: Academic Press. Maehr, M-, Mensing, J., & Nafzger, S. (1962). Concept of self and the reaction of others. W 21, 352-357. Martens. R. (1978). MW. Champaign, Illinois: Human Kinetics Publishers. Martens. R. (1975). W. New York. NY: Harper & Row. Martinek, T. & Johnson, S. (1979). Teacher expectations: Effects on dyadic interactions and self -concept in elementary age children. 83531.91]. 91111121111. 59. 60-70- Meichenbaum, C., Bowers, K., & Ross, R. (1969). A behavioral analysis of teacher expectancy effects WW 1}, 306-316. Meyer, W. U. (1982). Indirect communications about perceived ability estimates. WWW. 14-(6). 888 897 Meyer, W. U., Bachman, M., Biermann, U., Hempelmann, M., Ploger, & F. O. Spiller, H. (1979). The informational value of evaluative behavior: Influences of praise and blame on perceptions of ability. 1911mm Whales!- fl. 342- 346. Nicholls, J. G. (1978). The development of the concepts of effort and ability, perception of own attainment, and the understanding that difficult tasks require more ability. W 12, 800-814. Nicholls, J. (1984). Achievement motivation: Conceptions of ability, subjective experience, task choice, and performance. W 8mm 2.1- 328— 346. -91- Nicholls, J. G., & Miller, A. T. ( 1983). The differentiation of concepts of difficulty and ability. Wm 54, 951-959. Nicholls, J. G. & Miller, A. T. (1984). Development and its discontents: The differentiation of the concepts and ability. In J. G. Nicholls (Ed.) WW 3.. Greenwich CT: JAI. pp. 185- 218. Re jeski, W., Darracott, C., & Hutslar, S. (1979). Pygmalion in youth sport: A field study. WW 1.. 311-319. Roberts, G. C. (1985). Toward a new theory of motivation in sport: The role of perceived ability. In J. M. Silva and R. S. Weinburg (Eds. ), WM Champaign, IL. Human Kinetics, 1985. Roberts, G. C., Kleiber, D. A., & Duda, J. L. (1984). Motivation in sport: The mediating role of perceived ability. 193W §_, 312-324. Roberts, G. C. a Pascuzzi, D. (1979). Causal attribution in sport: Some theoretical implications. Wm 1, 203-211. Rosenshine, B. (1973). Teacher behavior and student attitudes revisted. Wm 05.. 177-180. Rosenthal. R. & Jacobson. L. (1968). W W. New York: Holt. Rinehart, & Winston. Seefeldt, V., Blievernicht, D. Bruce, R. & Gilliam, T. (1978). Joint Wm. East Lansina. MI: Youth Sports Institute. Smith, R. E. Smoll, F. I... & Curtis, B. (1979). Coach Effectiveness Training: A cognitive-behavioral approach to enchancing relationship skills in youth sport coaches. Wm L 59-75. Smith, R. E., Smoll, F. L... & Hunt, E. (1977). A system for the behavioral assessment of athletic coaches. W 41, 401-407. Smoll. F. L. & Smith. R. E. (1979). W W East Lansing, Michigan: Institute for the Study of Youth Sports. Spink, K. S. 81 Roberts, G. C. (1980). Ambiguity of outcome and causal attributions. MW 2. 237-244. -92- Stipek, D. J. (1981). Children’s perceptions of their own and their class- mate’ s ability W 7.3.. 404-410. Tharp, R. G., & Gallimore, R. (1976). What a coach can teach a teacher. Mainly—101“ 2. 74-78. Verof f J. (1969). Social comparison and the development of achievement motivation. In C. P Smith (Ed). Achienmemmlamotimin. 9111151311. New York: Russell Sage Foundation. ' Weinstein, R. (1979). Reading group membership in first grade: Teacher behaviors and pupil experience over time. MW Emboloax. Ll- 3- 25. Weiss, M. R., Bredemeir, B. J. & Shewchuck, R. M. (1981). The dynamics of perceived competence, perceived control, and motivational orientation in youth sports. A casual analysis. Whammy. WW July 19-26, 1984, Eugene, Oregon. White, W. (1959). Motivation reconsidered: The concept of competence. W 6.6. 297 333. Zajonc, R. (1965). Social facilitation. Science, 152, 269-274. APPENDICES APPENDIX A QUESTIONNAIRES if" i -95- Imagine the following situation: During a physical education class on gymnastics, the teacher asks the children to perform a somersault in the air. The skill is very difficult and only a few children are able to do it. The children are given two minutes time to practice doing somersaults into a landing area. While the children perform somersaults, the teacher watches carefully two children (whome the teacher particularly likes). Both students, You and Gary, can do a somersault quite well even though it is a difficult task to perform. The teacher says to You, "Good." The teacher says to Gary, "You have done very fine. I am very pleased." Which of the two pupils does the teacher consider to have higher ability, You or Gary? How would the teacher rate You and Gary on the scales below. You 1* very low ability 2 3- low ability (4. 5- average ability 6 7- high ability 8 9- very high ability Gary 1 - very low ability 2 3 low ability 4 5- average ability 6 7— high ability 8 9- very high ability APPENDIX B Parental Consent Form HICNICAN STATE UNIVERSITY Health Education, Counseling Psychology and Human Performance Parental Consent Form Dear Parents: 1 am a graduate student with the flichigan Youth Sports Institute, established in 1978, which is dcdlcgcgd to helping parents, coaches and adult leaders provide positive and beneficial sport experiences for children. One of the ways in which we have attempted to accomplish this objective is through continuing research programs designed to provide us with information about young athletes and the effects of sport participation on their PhYSKCIl and psychological development. I am currently working on a study entitled 'Informational Value of Evaluative Behavior: Influences of Praise and Blame on Perceptions of Ability in the Sport Setting". This study is being conducted under the supervision of Dr. Harty Ewing, on the staff of the Youth Sports Institute. This study involves written responses, of children (9-13), to ‘demographic questions, questions measuring perceived competence and responding to four short stories based on a hypothetical sport situation, to assess children’s self-perception of ability. are not intended to trick or deceive in any way. approximately five to seven minutes (15 to 10 total minutes) to complete. regard. You should understand that there are questions should be answered as honestly as possible. “assure-ants will be taken by means of paper and pencil, and Each questionnaire will take In this no right or wrong answers and that the You should further understand that your child’s participation in this study will be treated with strict confidence and anonymity in any reports or publications dealing with this project. In order to obtain useful information I need fill out the survey questionnaire. Children the questionnaire. All information received the study’s objectives and would kindly give would help ay research efforts treaendously. to participate, that there is no penalty for the cooperation of children (9-13 yrs.) to will not be required to put their names on is strictly confidential. If you approve of permission for your child to participate, it Let me stress that the children do not have not participating or for discontinuing participation, and their verbal permission will be sought before questionnaires are handed out. please sign your name where it reads "Parent" questions. please feel free to call. I am grateful for your help! Thank you, Cary Uilliamson Youth Sports Institute l.H. Sports Circle, Room 105 flichigan State University East Lansing, NI 68826 (517) 355-3858 mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm I give permission for my child ( If you consent to have ae ask your child if they would like to participate. s signature". If you have any further ...................................... l..... l to participate in this study under the (first name) conditions stated in the above letter. Parent's Signature APPENDIX C Analysis 2: Means and Standard Deviations for Non-Significant Three and Two-way Interactions -93- so.— mm.n na.— no.m on.~ ou.n ~o.~ nn.o m— —o.p . mo.n nn.w mp.s us.~ ms.n ”5., no.0 op mm.— no.n ms." mm.o —o.~ no.n ms.— 00.0 n— co.w ms.n mm." cm.o mo.— ms.~ ~o.p o~.s ~F no.p mo.n ou.— no.0 «G.F n~.n on." mm.s wp ~o.w nm.¢ oo.— mo.m om.~ n~.e oo.~ mk.o op po.w ms.n 00.— no.5 no.- n¢.~ os.— no.s o aw m mu m aw x aw x ooeuaox ocaomms oceans: unouuaa on... oeaosms ens-ca aeouuam 00¢ oueoueea so eeouuao so ou< so sumosoe ue>_eueoa to. eco_ue_>eo oceuceam uce aces: _99- F—.~ mp.e oo.— o—.s mm.- oo.~ no." .mo.m mp ~¢.~ mo.e os.— n—.s n~.~ mo.n oo.. no.0 up o~.~ n..o oo.F m~.s so.— ne.n —m.- mm.m n— ~—.~ on.n ~o.— no.0 ~o.- op.n Fo.— no.0 ~. 00., mo.n sm.P mo.s «0.. mm.n on.p m—.s .— oe.~ me.e o~.~ om.o ~—.~ nn.e no.9 nv.o a. so.— mo.~ mo.— o~.s n—.~ nn.n oc.~ ms.o o mu m mm m aw x mm x eta..es uoausss_o eeeuuaa u.:u_ssso ecoooes seem eeauuam seen one suoausssso see» so ulouuso so oo< so su_._n< uo>_oucua cos ecosue_>oo nuanceum uce aces: me.n ms.~ om.m ss.p mo.n oo.~ oo.e m— nw on.~ os.m Fo.~ no.“ n~.~ oo.m oo.~ n~.m «— m . oe.~ ms.m oe.~ no.m oo.— on.e om.~ oo.e mp oe.~ ne.n os.~ oo.e so.~ om.e om.~ mo.m ~p oo.~ om.m po.~ no.“ o~.~ nn.m ps.~ nn.m .9 oc.~ oo.m oo.~ nm.m as.— oo.m «o.~ mu.n or ~n.~ mo.m sn.n oo.n ~o.. ns.e ~n.n mn.n 0 mm m an n mm m an x also. also. co . oecuaez uoau_es_o co semeca nonesse_a .ecuaez seem ens-ea seen one ooanuooa so a.au_t._o ea.» so ou< so auc_a-¢ so.._o< uo>_ootua Law gun-mac Dugoum 2. 2302 Means and Standard Qeviagions for -101- WW We Age Success Failure ii 5.12 Ii £12 9 6.98 1.97 .09 .02 10 6.31 2.10 .38 .67 11 7.10 1.47 .60 .91 12 6.89 1.51 .24 .97 13 6.56 1.76 .78 .15 14 7.05 1.57 .85 .33 15 6.08 1.95 .53 .50 APPENDIX D Data Coding 1-3 4-5 10 -102- RAW DATA DATA DIRECTORY VARI Subject ID Age Outcome Task Difficulty Feedback Perceived Ability 20.125 , 01-20 Age in Years 1 2 Success Failure Easy Difficult Praise or Blame Neutral No Ability Very Low Ability Low Ability Below Average Ability Average Ability Above Average Ability High Ability Very High Ability Exceptional ability APPENDIX E Raw Data 921 921 921 921 921 921 921 921 921 921 921 921 921 921 921 921 921 19 921 20 921 11021 21021 31021 41021 51021 61021 71021 81021 91021 101021 111021 121021 131021 141021 151021 161021 171021 181021 191021 201021 11121 21121 31121 41121 51121 61121 71121 81121 91121 101121 111121 121121 131121 141121 151121 '0 coauaussuroo— s-es-s-s-s-s-e “Mauser- mes-s use 921 ' 921 921 921 921 921 921 921 921 921 921 921 921 921 921 921 921 921 921 921 20 921 11021 21021 31021 41021 51021 61021 71021 81021 91021 101021 111021 121021 131021 141021 151021 161021 171021 181021 191021 201021 11121 21121 31121 41121 .31121 61121 71121 81121 91121 101121 111121 121121 131121 141121 151121 awe-s- ”POO-qaubuue— prepares-ens- Savant.“ ‘911 911 911 911 911 911 911 911 911 911 911 911 911 911 911 911 911 911 911 11011 21011 31011 41011 31011 61011 71011 81011 91011 101011 111011 121011 131011 141011 151011 161011 171011 181011 191011 201011 11111 21111 31111 41111 51111 61111 71111 81111 91111 101111 111111 121111 131111 141111 151111 .- Oo-uoubuun O‘FF‘F‘F‘PHF’ .NOMDHNM NO. 00 911 911 911 911 911 911 911 911 911 911 911 911 911 911 911 911 911 911 911 911 11011 21011 31011 41011 51011 61011 71011 81011 91011 101011 111011 121011 131011 141011 151011 161011 171011 181011 191011 201011 11111 21111 31111 41111 51111 61111 71111 81111 91111 101111 111111 121111 131111 141111 151111 ..NOM“UN~ -103- N N O'N'bMDUNs-n GDN‘MFUNI‘ O N N 922 922 922 11022 21022 31022 41022 31022 61022 71022 81022 91022 101022 111022 121022 131022 141022 151022 161022 171022 181022 191022 201022 11122 21122 31122 41122 51122 61122 71122 81122 91122 101122 111122 121122 131122 141122 151122 ODNOU‘DUNs-e G p N 912 912 912 912 912 912 912 912 912 912 912 912 912 912 912 912 912 912 912 20 912 11012 21012 31012 41012 51012 61012 71012 81012 91012 101012 111012 121012 131012 141012 131012 161012 171012 181012 191012 201012 11112 21112 31112 41112 51112 61112 71112 81112 91112 101112 111112 121112 131112 141112 151112 OOVOVFUN—e 161121 171121 181121 191121 201121 11221 21221 31221 41221 51221 61221 71221 81221 91221 101221 111221 121221 131221 141221 151221 161221 171221 181221 191221 201221 11321 21321 31321 41321 51321 61321 71321 81321 91321 101321 111321 121321 131321 141321 151321 161321 171321 181321 191321 201321 11421 21421 31421 41421 51421 61421 71421 81421 91421 101421 161121 171121 181121 191121 201121 11221 21221 31221 41221 51221 61221 71221 81221 91221 101221 111221 121221 131221 141221 151221 161221 171221 181221 191221 201221 11321 21321 31321 41321 51321 61321 71321 81321 91321 101321 111321 121321 131321 141321 151321 161321 171321 181321 191321 201321 11421 21421 31421 41421 51421 61421 71421 81421 91421 101421 161111 171111 181111 191111 201111 11211 21211 31211 41211 51211 61211 71211 81211 91211 101211 111211 121211 131211 141211 151211 161211 171211 181211 191211 201211 11311 21311 31311 41311 51311 61311 71311 81311 91311 101311 111311 121311 131311 141311 151311 161311 171311 181311 191311 201311 11411 21411 31411 41411 51411 61411 71411 81411 91411 101411 161111 171111 181111 191111 201111 11211 21211 31211 41211 51211 61211 71211 81211 91211 101211 111211 121211 131211 141211 151211 161211 171211 181211 191211 201211 11311 21311 31311 41311 51311 61311 71311 81311 91311 101311 111311 121311 131311 141311 151311 161311 171311 181311 191311 201311 11411 21411 31411 41411 51411 61411 71411 81411 91411 101411 -104- 161122 171122 181122 191122 201122 23 25 23 29 25 11222223 21222 31222 41222 51222 61222 71222 81222 91222 101222 111222 121222 131222 141222 151222 161222 171222 181222 191222 201222 11322 21322 31322 41322 51322 61322 71322 81322 91322 101322 111322 121322 131322 141322 151322 161322 171322 181322 191322 201322 11422 21422 31422 41422 51422 61422 71422 81422 91422 101422 21 21 28 2: 21 21 2: 21, as 21 21 21 21 as as 21 21 21 21 2: 161122 171122 181122 191122 201122 11 15 11 17 17 11222211 21222 31222 41222 51222 61222 71222 81222 91222 101222 111222 121222 131222 141222 151222 161222 171222 181222 191222 201222 11322 21322 31322 41322 51322 61322 71322 81322 91322 101322 111322 121322 131322 141322 151322 161322 171322 181322 191322 201322 11422 21422 31422 41422 51422 61422 71422 81422 91422 101422 11 11 13 13 12 11 13 11 15 12 11 11 14 17 14 11 12 17 15 17 11 13 15 161112 171112 181112 191112 201112 11212 21212 31212 41212 51212 61212 71212 81212 91212 101212 111212 121212 131212 141212 151212 161212 171212 181212 191212 201212 11312 21312 31312 41312 51312 61312 71312 81312 91312 101312 111312 121312 131312 141312 151312 161312 171312 181312 191312 201312 11412 21412 31412 41412 51412 81412 71412 81412 91412 101412 161112 171112 181112 191112 201112 11212 21212 31212 41212 51212 61212 71212 81212 91212 101212 111212 121212 131212 141212 151212 161212 171212 181212 191212 201212 11312 21312 31312 41312 51312 61312 71312 81312 91312 101312 111312 121312 131312 141312 151312 161312 171312 181312 191312 201312 11412 21412 31412 41412 51412 61412 71412 81412 91412 101412 19 17 19 17 17 17 111421 121421 131421 141421 151421 161421 171421 181421 191421 201421 11521 21521 31521 41521 51521 61521 71521 81521 91521 101521 111521 121521 131521 141521 151521 161521 171521 181521 191521 201521 111421 121421 131421 141421 151421 161421 171421 181421 191421 201421 11521 21521 31521 41521 51521 61521 71521 81521 91521 101521 111521 121521. 131521 141521 151521 161521 171521 181521 191521 201521 111411 121411 131411 141411 151411 161411 171411 181411 191411 201411 11511 21511 31511 41511 51511 61511 71511 81511 91511 101511 111511 121511 131511 141511 151511 161511 171511 181511 191511 201511 111411 121411 131411 141411 151411 161411 171411 181411 191411 201411 11511 21511 31511 41511 51511 61511 71511 81511 91511 101511 111511 121511 131511 141511 151511 161511 171511 181511 191511 201511 -105- 111422 121422 131422 141422 151422 161422 171422 181422 191422 201422 11522 21522 31522 41522 51522 61522 71522 81522 91522 101522 111522 121522 131522 141522 151522 161522 171522 181522 191522 201522 111422 121422 131422 141422 151422 161422 171422 181422 191422 201422 11522 21522 31522 41522 51522 61522 71522 81522 91522 101522 111522 121522 131522 141522 151522 161522 171522 181522 191522 201522 16 18 12 17 11 12 17 11 13 17 17 15 18 11 17 12 13 12 17 13 13 11 12 17 17 13 12 17 11 111412 121412 131412 141412 151412 161412 171412 181412 191412 201412 11512 21512 31512 41512 51512 61512 71512 81512 91512 101512 111512 121512 131512 141512 151512 161512 171512 181512 191512 201512 111412 121412 131412 141412 151412 161412 171412 181412 191412 201412 11512 21512 31512 41512 51512 61512 71512 81512 91512 101512 111512 121512 131512 141512 151512 161512 171512 181512 191512 201512