:sr'l“ "$.13 -.‘l. . v; I,,A,.:'.::;L. r ‘ 3”:- II. .{J ~§uv~| {fixatqwx‘é‘A— g «v‘ :Uuk‘. :fiv $1 "I?" y. “' '37.: 593%: ‘. ' 4‘ .4 _ 1 13%? v”“‘-‘;|'zu I ‘3” 3“. I.‘ . “$13.3"! 34/1;- ‘.‘ p I" - a . ~ 2. 13"”: a}, ,. ' - ‘ ‘f’-.:r.3:3 IV: ‘mxn‘ . ""‘ ‘ * 1r” » 16,3:1‘35 r. {mi-I: 3‘ Q "‘ f! .4 $32: 9%: : Ly“! ~ ' 341.! a 33:11“?! 1:“; up; - 4. v . «Cu-.Jvaflu' 2\‘_ 1.3, 'la-t '3: "if: N V ”XL" -33... a,” 3? m» uE’SiS NIVERSITY LIBRARIE Illlllil’illl‘ll mall a Hill!!! 3 1293 00896 3237 iii This is to certify that the dissertation entitled ATTRIBUTION STYLES OF HIGH SCHOOL STUDENTS AND HIGH SCHOOL DROPOUTS presented by Larry Balkema has been accepted towards fulfillment of the requirements for Ph.D. degree“, Educational Administration Date May 1991 MSU is an Affirmative Action /Equal Opportunity Institution 0- 12771 7, __ _.._fi I - ,ndficflflfi—f "“ ‘ —' W—h“ M “Mm—m-A—"- 44‘ M~_A__.—A_ ,. 4_..p.—-s #5— -“..‘_ W‘hh _._~..- V‘flaa... - LIBRARY Michigan State 5 University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. ll DATE DUE DATE DUE DATE DUE WV— MSU is An Affirmative Action/Equal Opportunity institution , _ 7—-— ATTRIBUTION STYLES OF HIGH SCHOOL STUDENTS AND HIGH SCHOOL DROPOUTS By Larry Balkema A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Educational Administration 1991 (3 ‘,-‘ '/"- 7'0 :5 cs a< an Whi att sch: com; edu< inma Mult‘ stati ABSTRACT ATTRIBUTION STYLES OF HIGH SCHOOL STUDENTS AND HIGH SCHOOL DROPOUTS By Larry Balkema This study was designed to investigate the relationship between achievement behavior and attributional style of high school students and high school dropouts. The research was aimed at determining whether a student’s level of performance is related to his/her attributional style. The population included students enrolled in grades 10 to 12 in school systems with community education programs with high school completion programs. High school dropouts not enrolled in community education programs were selected from a population of prison inmates. ‘The instrument used for data collection was the Multidimensional-Multiattributional Causality Scale. Descriptive statistics were used as the primary analytical tool. T-tests with a .05 significance level were applied in some cases. The central findings of the study were as follows: l. High school dropouts in community education programs and dropouts in prison reported :1 more positive attributional pattern than students who had not dropped out of school. Larry Balkema 2. There was no significant difference in the attributional patterns of male and female students. 3. There was no significant difference in the attributional patterns of lOth-, llth-, and thh-grade students. 4. Black students reported a significantly more positive attributional pattern than white students. 5. Students enrolled in high school reported a slightly more positive attributional pattern as academic performance improved, as measured by grade point average. mad thei frui wife edit faiti assis on th ACKNOWLEDGMENTS It is with deep appreciation that I acknowledge those who have made the completion of this dissertation possible. I have used their wisdom, expertise, and encouragement to bring this project to fruition. I will always be grateful for the support of Lee Balkema, my wife. Special thanks go to Louisa Randall for all of her typing and editing, and to Dave Ringnalda and Dick Wagemaker, who never lost faith. I would like to thank Dr. Louis Romano for all of his assistance and Dr. Philip Cusick for continually helping me reflect on the implications of quality work. iv LIE Cha III IV TABLE OF CONTENTS LIST OF TABLES OOOOOOOOOOOOOOOOOOOOOOO Chapter I. INTRODUCTION .................... Purpose of the Study ............... Definition of Terms ................ Assumptions and Limitations ............ Assumptions ................... Limitations ................... Research Hypotheses ................ Research Questions ................ Overview of the Study ............... II. THEORY AND SUPPORTIVE RESEARCH ........... Introduction ................... Cognitive Psychology ............... Learned Helplessness ............... Attribution Style in Achievement Situations . . . . Learned-Helplessness Achievement Patterns ..... Race, Gender, and Income Levels .......... Summary ...................... III. DESIGN OF THE STUDY ................. Introduction ................... Population and Sampling Methods .......... Instrumentation .................. Data-Gathering Procedures ............. Statistical Treatment ............... Summary ...................... IV. ANALYSIS OF THE DATA ................ Introduction ................... Instrument Analysis ................ Demographic Data ................ Findings Related to the Multidimensional- Multiattributional Causality Scale ...... Findings ..................... 48 Tests of Significance .............. 48 Hypothesis l .................. 48 Hypothesis 2 .................. 5l Hypothesis 3 .................. 53 Hypothesis 4 .................. 55 Hypothesis 5 .................. 58 Hypothesis 6 .................. 59 Hypothesis 7 .................. 62 Hypothesis 8 .................. 64 Hypothesis 9 .................. 66 Hypothesis l0 .................. 69 Hypothesis ll .................. 7l Hypothesis 12 .................. 73 Research Question l ............... 74 Research Question 2 ............... 77 Research Question 3 ............... 78 Research Question 4 ............... 80 Summary ...................... 80 V. SUMMARY AND CONCLUSIONS ............... 82 Summary ...................... 82 Findings ..................... 83 Hypothesis l .................. 83 Hypothesis 2 .................. 84 Hypothesis 3 .................. 85 Hypothesis 4 .................. 87 Hypothesis 5 .................. 88 Hypothesis 6 .................. 88 Hypothesis 7 .................. 90 Hypothesis 8 .................. 9l Hypothesis 9 .................. 92 Hypothesis lO .................. 93 Hypothesis ll .................. 94 Hypothesis l2 .................. 94 Research Question 1 ............... 96 Research Question 2 ............... 97 Research Question 3 ............... 98 Research Question 4 ............... 98 Conclusions .................... 103 Recommendations .................. l06 Suggestions for Further Research ......... l07 vi Page APPENDICES A. THE MULTIDIMENSIONAL-MULTIATTRIBUTIONAL CAUSALITY SCALE AND THE BIOGRAPHICAL DATA SHEET ........ 108 B. PERMISSION LETTERS ................. 114 BIBLIOGRAPHY ........................ 118 vii Table .1 1 .h h-h-h-D-D 01-th .2 .10 .11 .12 .13 LIST OF TABLES Primary Reasons High School Dropouts 14 to 21 Years Old Left School, by Race and Gender ........ High School Dropout Rates for Youths 14 to 21 Years of Age, by Age, Race, and Gender--l979 ....... Distribution of Subjects by Race ........... Distribution of Subjects by Gender .......... Distribution of Subjects by Income Level ....... Distribution of Subjects by Class Standing ...... Distribution of Subjects by Class Standing, Race, and Gender ..................... Frequencies of Responses to MMCS Items ........ Results of 12 Hypothesis Tests on the MMCS ...... T-Ratios of Students in the Top One-Third and Middle One-Third of Their Class on the MMCS T-Ratios of Students in the Top One-Third and Bottom One-Third of Their Class on the MMCS T-Ratios of Students in the Top One-Third of Their Class and Dropouts Enrolled in Community Educa- tion on the MMCS .................. T-Ratios of Students in the Top One-Third of Their Class and High School Dropouts on the MMCS ..... T-Ratios of Students in the Middle One-Third and Bottom One-Third of Their Class on the MMCS T-Ratios of Students in the Middle One-Third of Their Class and High School Dropouts Enrolled in Community Education on the MMCS ......... viii 45 45 46 46 46 47 49 50 52 54 56 58 60 .14 .15 .16 .17 .18 .19 .20 .21 .22 .23 .24 .25 .26 T-Ratios of Students in the Middle One-Third of Their Class and High School Dropouts on the MMCS ........................ T-Ratios of Students in the Bottom One-Third of Their Class and High School Dropouts Enrolled in Community Education on the MMCS ......... T-Ratios of Students in the Bottom One-Third of Their Class and High School Dropouts on the MMCS . . T-Ratios of Students in Community Education and High School Dropouts on the MMCS .......... T-Ratios of Male and Female Students on the MMCS . . . T-Ratios of White and Black Students on the MMCS . . . T-Ratios of White and Black Students on the Attributional Scales of the MMCS .......... Means and Standard Deviations of Students’ Attribu- tions on the Effort, Ability, Context, and Luck Scales, by Race .................. Means and Standard Deviations of Students’ Attribu- tions on the Effort, Ability, Context, and Luck Scales, by Gender ................. Means and Standard Deviations of Students’ Attribu- tions on the Effort, Ability, Context, and Luck Scales, by Class Standing ............. Means and Standard Deviations of Students’ Attribu- tions on the Effort, Ability, Context, and Luck Scales, by Grade Level ............... Effect of Attributions on Educational Achievement: Analysis of Ability, Effort, Context, and Luck Scale Scores by Class Standing and Income Level Effect of Attributions on Educational Achievement: Analysis of Ability, Effort, Context, and Luck Scale Scores by Class Standing ........... ix Page 62 65 67 69 71 73 75 76 76 77 78 79 81 alah one schoc City (”Liv other youths recent studen increa. 1980). increas Year ( drOpout 1958. Year l< showed School. and ID! CHAPTER I INTRODUCTION The dropout rate of students from American high schools is an alarming 26%, 50% if the schools are located in urban areas. Almost one out of every two high school students from big inner-city schools drops out before graduation (Maurer, 1982). In the New York City school system, that amounts to about 40,000 children a year (New York Times Educational Supplement, 1980). In Michigan and other states, there is a concern about the increasing number of youths dropping out of high school. California illustrates the recent trend of increasing dropout rates. In 1967, only 12% of the students left school before graduation. By 1970, that rate had increased to 17%, then 20% in 1972, and finally 22% in 1976 (Camp, 1980). Enrollment data for Ohio show that the number of dropouts increased 15%“from the 1975-76 school year to the 1978-79 school year (Kaser, 1980). Peng (1982) reported that the high school dropout rate for pupils entering the fifth grade has been 25% since 1958. The Michigan Department of Education studies for the academic year 1983-84, the latest year for which statistics are available, showed that 22% of the freshman class studied never did finish school. The consistency of the dropout rates as reported by Peng and the increasing rate reported by Camp bring into question the st 91 an of dr edt hig horr Wer effectiveness of current interventions being used in) decrease the dropout rate. When discussing the problems of the high school dropout, we must take a much closer look at whom we are talking about. We need to focus our attention on the complexity of the problem. There is the voluntary dropout who has decided not to continue school for reasons that are under his control, and there is the involuntary dropout who is forced to leave by the school system or forces outside his control. Those students who have left the traditional school system can be further divided into those who are working and out of school, those who are not working and out of school, and those who are either working or not and have returned to some form of schooling other than the traditional high school. Some would also include in the dropout population those students who are still in school in body but in reality have academically dropped out. Many research studies have been conducted to identify the characteristics of dropouts and those factors contributing to the students’ voluntary leaving or being forced to leave. The reasons given by students for dropping out can be seen in Table 1.1. Data analysis by Ekstrom, Goetz, Pollack, and Rock (1986) revealed some of the major underlying home factors found to be associated with dropouts. Dropouts tended to come from homes with a weaker educational support system when compared with students who complete high school. Dropouts (a) had fewer study aids present in their homes, (b) had less opportunity for non-school-related learning, (c) were less likely to have both natural parents living at home, (d) had mothers with lower levels of formal education, (e) had mothers who were more 'likely to be working, and if) had parents who were less likely to be interested in (M‘ to monitor both in-school and out-of—school activities (Ekstrom et a1., 1986). Table 1.1.--Primary reasons high school dropouts 14 to 21 years old left school, by race and gender.a Females Males Reasons for Total Leaving School B H W T B H W T School Related 29 21 36 32 56 36 55 53 44 Poor performance 5 4 5 5 9 4 9 9 7 Dislike school 10 15 27 24 29 26 36 33 29 Expelled/suspended 5 1 2 2 18 6 9 10 7 School too dangerous 1 1 2 1 O 0 l l 0 Economic 15 24 14 15 23 38 22 24 20 Desired to work 4 7 5 5 12 16 15 14 10 Financial diffi- 3 9 3 4 7 9 3 5 4 culties Home responsi- 8 8 6 6 4 l3 4 5 6 bilities Personal 45 30 31 33 0 3 3 2 17 Pregnancy 41 30 31 33 0 0 0 0 9 Marriage 4 15 l7 l4 0 3 3 2 8 Other 11 25 19 20 21 23 20 21 19 Total % 100 ----------------- 100 Source: National Longitudinal Survey of Youth Labor Market Expe- riences. aDistributions are percentages. tc be en mar res bac the cre tra fac cre uner the the Increased attention is being focused on the many reasons children leave school. There is an equity issue. Those individuals who leave school before graduation do not receive the maximum benefit of the educational opportunity that is theoretically available. ‘The number of students who leave school early are disproportionately from low-socioeconomic—status (SES) families and racial/ethnic minority groups. As can be seen in Table 1.2, the proportion of black/Hispanic students who drop out of school is significantly higher than that of white students. Blacks are suspended from high school three times as often as whites (Howe & Edelman, 1985). These data suggest a question of equity. Are all of the subgroups of American society receiving an equal opportunity to participate in the American educational system? As society becomes. more complicated lll structure, the social problems encountered are becoming more complex. To develop solutions to the many critical social problems facing society, there must be a resource of well-educated, well-trained citizens of all ethnic backgrounds who are willing to commit themselves 1K) helping solve these problems. The technological orientation of society also creates an increasing demand for individuals having received better training before entering the labor market. A further complicating factor is the population shift from rural to urban areas, which has created metropolitan centers of population density, poverty, unemployment, and social unrest. In addition, structural changes in the composition of the work force have made it more difficult for the uneducated to find employment because a shorter work week and more leisure time have stimulated the growth of the service industries iri which workers need both social and verbal ease. Females and minorities suffered the most with respect to unrealized achievement gains if they dropped out of school. These unrealized achievement gains for women and minorities were largest in the language-development areas of vocabulary, reading, and math (Ekstrom et a1., 1986). Table 1.2.--High school dropout ratesa for youths 14 to 21 years of age, by age, race, and gender--1979. Age Gender/Race Total 14-15 16-17 18-19 20-21 Black 2% 10% 24% 25% 15% Female 2 8 22 20 14 Male 1 12 25 30 17 Hispanic 2 17 36 35 23 Female 2 17 39 33 24 Male 3 18 32 38 22 White 2 8 16 12 10 Female 1 9 14 11 9 Male 2 8 17 13 10 Total % 2 9 18 15 11 Source: National Longitudinal Survey of Youth Labor Market Experi- ence. aPercentage of the civilian population who are high school dropouts. 5( mt BS 5)’ (In M dyn There is evidence showing that the problems of school dropouts (including repeated failure, lack of basic educational skills, and personality problems) contribute to the development of an individual who finds it difficult, if not impossible, to compete adequately in society, both on a personal level and in the labor market where he must compete.for the means to sustain himself. Without the essential skill for coping in society and burdened with the failure syndrome that often becomes a part of the personality of the school dropout, youngsters too often become alienated, disillusioned, and frustrated, contributing to what Conant in the 19605 termed "social dynamite." Public policy has moved in a direction that stresses that equal options shall be open for all citizens in order that they may become equipped to enter the mainstream of society and reap the maximum benefits (H’ that society. Civil rights legislation intended to provide these options to minority groups is one example of this policy. Public Law 94-142 (education of the handicapped) is intended ix> extend the benefit of“ education to the handicapped population. Federally supported manpower and education programs to meet the needs of the socially and economically disadvantaged exhibit additional concern about high school dropouts. Further federal support is exemplified by the federal allocation of funds designed to retrain those who have left school and to promote dropout-prevention programs. The Manpower Development and Training Act of 1963, as amended, and the Economic Opportunity Act of 1964, as amended, are examples of this effort. Such allocations are based h a1 pr f0 Inc the res env Stu did or unde may Upor CaUS on ‘the assumption that education, including vocational training, will improve the individual’s employability so that he may compete more realistically in the labor market. Follow-up- studies have shown that "dropping out" can prove costly to both the individual dropout and society as a whole. A 1978 Department of Labor report noted that the unemployment rate for dropouts is more than double that of high school graduates. According to a Missouri Division of Corrections report, 95% of the individuals who are incarcerated in Missouri correctional institutions are high school dropouts (Long, 1981). These two facts alone emphasize the need to deal more effectively with the dropout problem and to provide workable solutions and viable alternatives for these young adults. High school students live 'hi 3. complex social environment. Individual students within the large social and academic setting of the school seek to understand their position and the salient factors responsible. Students seek to gain an understanding of their environment with the constant pursuit of the question "Why." Students seek to understand why they did not get the date, why they did not make the team, why they got in trouble, and why they passed or failed a particular class or assignment. Students want to understand the environment in which they find themselves, or they may wish to gain an understanding of how the environment impinges upon them. The individual student is looking for the cause or causes of important life events. Once~ a cause or causes are assigned as an explanation for an event, effective management of that type of event may then be possible, and a prediction of future outcomes in similar situations may be made. The process of assigning causation begins to give the individual predictive power concerning future occurrences of similar events. This predictive power enables one to take a more active part in controlling one’s significant life events. At this point it is possible to control specific events in order to secure a more predictable future. Attribution theory (Kelly, 1967) is concerned with how the individual ascribes causation. Individuals experience either a positive or negative event in their environment (the student fails an important examination); the student now is faced with explaining to himself, and possibly' others, why' this event occurred. The attribution the student uses to explahn why this failure occurred can have a significant effect on his gaining control over the future occurrence of similar events. High school students who are doing poorly in school or those who have already dropped out have been looked at in many different contexts by researchers. Often the researcher has focused on their home environment and community support systems, concluding that environmental and social factors led in) poor performance in educational settings. The problem with this type of research is that it does not provide the student with a workable plan or realistic process by which he can begin to make effective changes to reverse the poor academic performance. Attribution theory looks at the manner in which the student ascribes success and failure. If W ir hi prc the stL ext to mam Char his made orf the student is making self-defeating attributions and is made aware of this, he then may be able to take corrective action. The student may begin to attribute success and failure in a manner that would provide him with a better ability to control future events. An attribution for failure of "I lack the ability" may be changed to "I did not work hard." In the future, the student can choose to put forth more effort toward the learning task. High-school-age individuals make attributions concerning their success or failure in achievement settings. The accuracy of their attribution is run; the critical factor. The critical issue is whether the attribution gives the individual the hope of influencing future events. If a student makes a failure attribution to internal, stable causes (lack of ability), there is little hope for him to see a different future outcome. The reason for his past and present failure is not perceived as changeable. The student lacks the means by which to alter future events. In contrast, if the same student makes a failure attribution to unstable, internal or external factors (lack of effort), he could well expect the future to be quite different. He may alter his behavior in a positive manner. He may begin attending classes or studying; he may also change environmental factors that have had a negative influence on his achievement. The critical factor is the type of attribution made and the effect it may have on future expectations for success or failure. Ar ac at co fa di ac. Ed: he' lO Purpose of the Study The purpose of this study was to investigate the relationship between achievement behavior and attributional style of high school students and high school dropouts. Do students attribute their success and failure differently as their level of success in school changes? This research was aimed at determining whether a student’s level of academic performance is related to his attributional style. Does a student who is more successful in achievement situations exhibit a different attributional style? A strong interest in the study of cognitive motivational variables has continued since the Coleman report on Equality of Educational Opportunity identified a student’s sense of control over his environment as the best single predictor of school achievement (Coleman et a1., 1966). The present study dealt with attributions given by an individual for performance in an achievement situation. An attributional model of achievement motivation assumes that achievement behavior* is cognitively mediated by three pairs of attribution factors: internal/external, stable/unstable, and controllable/uncontrollable. After any experience with a success or failure, the attributions given to explain performance will differentially affect expectancies for outcomes in future achievement settings. Attributions have a powerful effect on a student’s total educational and life experiences. Studies in the area of learned helplessness in children have linked deterioration iri performance be ct st in fa th ex, Fa 9X1 ex; inc to lik 11 following failure to learned helplessness, the perceived inability to surmount failure (Dweck, 1978; Dweck. & Bush, 1976; Dweck & Reppucci, 1973). This perception is associated with attributions to uncontrollable, stable, internal factors such as lack of ability, rather than to controllable factors such as effort. In 'these children, success is attributed ix) external, uncontrollable, unstable factors such as "the teacher likes me," "the test was easy," "everyone did well," or "I was lucky," never to their own effort or ability. Failure, on the other hand, is attributed to internal, stable, uncontrollable factors such as "I’m dumb" or "I never do well," and not the attributes of lack of effort, "I was ill and not performing at my best," or "I just had a bad day; I’ll do better on the next test." It appears that this differential pattern of attributional choice could impair or improve the long-term achievement of students. Individuals attributing failure 1K) their own ability instead of their effort will likely continue to experience future failure in similar situations. Individuals who attribute failure to their lack of. effort instead of their ability may be able to experience improved performance in similar situations in the future. Failure attributions to internal stable factors (ability) or external stable factors (material is too difficult) give rise to expectations that the future will remain constant. The reasons the individual has failed are out of his control. Success attributions to unstable internal factors (luck) or external factors ("teacher likes me") do not prompt a positive causal link between success and a 1 ii 1‘6 12 individual ability or action; therefore, the expectancy for fUture success is not enhanced. The reason for the present success is out of the individual’s control. Conversely, the student making failure attributions to unstable internal factors (lack of effort) or external factors (poor test- taking conditions) may expect the situation to change in the future. The reasons for the present failure can be modified by the individual. He can study harder or work to change the poor environmental conditions. This ability to affect the causal reason for present failure provides the grounds for the individual to hope for future success. This same individual making success attributions to internal stable factors (ability) or external factors (easy material) can expect the same conditions that produced his present success to repeat themselves and eXperience continued future success. If students divided by class ranking and those who have dropped out do differentially attribute success and failure as predicted, it would indicate that educational programs that are being implemented to help the low achiever and school dropout would need to consider adding an attributional retraining component to their program. Definition of Terms The following definitions of terms are presented to aid in the interpretation and clarification of this study and to aid any future replications of this study that may be conducted. 13 Attributional style. Refers to an individual’s tendencies in ascribing causality for positive and negative events to some factors differentially. Within the attribution framework, individuals will vary in their ascription to measures of causality, stability, and internality (Abramson, Seligman, & Teasdale, 1978; Ickes & Layden, 1978; Weiner, 1979, 1980). Dropout. According to Good (1973), the term . designates an elementary or secondary school pupil who has been in membership during the regular school term and who withdraws or is dropped from membership for any reason except death or transfer to another school, before graduation from secondary school (grade 12), or before completing an equivalent program of studies; such an individual is considered a dropout whether his/her dropping out occurs before (N‘ between regular school terms, whether it occurs before or after he/she has passed the compulsory school attendance age, and where applicable, whether or not he/she has completed a minimum required amount of school work. (p. 32) Learned helplessness. Exhibited by individuals who believe they have no control over the outcome of certain events. In a school setting, students who curtail effort after experiencing failure are exhibiting a learned-helplessness response. Stable. This dimension defines causes that are invariant. An attribution to a stable cause would be ability, which remains relatively consistent over time. Unstable. This dimension defines causes that are variant. An. attribution to an unstable cause would be luck or effort. These factors can change greatly over short periods of time. Ir— lc W8 ab pa: dif fai W81“ I4 Assumptions and Limitations The validity of this study is affected by the following: Assumptions 1. Grade point average (GPA) is an: indication of academic success or failure. 2. Income levels reflect the SES of the responding students. Limitations 1. This study was limited to the three school districts located in the Grand Rapids area of western Michigan. 1h) attempt was made to generalize the results to other populations. 2. The collection of data regarding SES was dependent on the ability of participants to objectively determine their own or their parents’ income. Research Hypotheses To determine the relationship that might have existed among the different groups in regard to the attribution for success and failure irI achievement situations, the following null hypotheses were formulated: Hypothesis 1: There is no statistically significant difference in Multidimensional-Multiattributional Causality Scale (MMCS) scores for ability, effort, context, and luck between students in the top one-third of their class and students in the middle one-third of their class. Hypothesis 2: There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the top one-third of their class and students in the bottom one-third of their class. 15 Hypothesis 3: There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the top one-third of their class and students who have dropped out of high school and are enrolled in a community education program. Hypothesis 4: There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the top one-third of their class and students who have dropped out of high school. Hypothesis 5: There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the middle one-third of their class and students in the bottom one-third of their class. Hypothesis 6: There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the middle one-third of their class and students who have dropped out of high school and are enrolled in a community education program. Hypothesis 7: There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the middle one-third of their class and students who have dropped out of high school. Hypothesis 8: There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the bottom one-third of their class and students who have dropped out of high school and are enrolled in a community education program. Hypothesis 9: There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the bottom one-third of their class and students who have dropped out of high school. Hypothesis 10: There is no statistically significant differ- ence in MMCS scores for ability, effort, context, and luck between students who have dropped out of high school and enrolled in community education programs, and students who have dropped out of high school. Hypothesis 11: There is no statistically significant differ- ence in MMCS scores for ability, effort, context, and luck between male and female students. Hypothesis 12: There is no statistically significant differ- ence in MMCS scores for ability, effort, context, and luck between white and black students. 16 Research Questions This study was designed to answer the following research questions: 1. Which of the attributional categories of ability, effort, context, and luck is viewed as the most important determinant of academic success and failure in achievement settings, as reported by race, gender, and class standings? 2. Do students as they progress through grades 10, 11, and 12 change their attributional pattern? 3. What effect does the income level of the student’s family have on the student’s attributional pattern? 4. Do students who experience greater academic success, as measured by GPA, attribute success and failure in a more productive pattern than students who are not as successful in achievement settings? Overview of the Study This dissertation consists of five chapters. Chapter I included the purpose of the study, significance of the problem, hypotheses and research questions, assumptions and limitations, and definitions of terms. In Chapter II, 'literature related ix> high school dropouts, attribution theory, and learned helplessness is reviewed. Methods and procedures of the study are described and explained in Chapter III. An analysis of the data is presented in Chapter IV, and 17 Chapter V contains the findings and conclusions of the study, with recommendations for further research. CHAPTER II THEORY AND SUPPORTIVE RESEARCH Introduction Attribution theory postulates that behavior is strongly influenced by the perceived causes of past events. An attribution is the inference that an individual makes about the causes of his own or another person’s behavior (Bar-Tal, 1978). The investigation into how people "attribute this consequence to that cause" is rooted in Heider’s (1958) "common sense theory of action." "Common sense psychology" is the phrase used by Heider to explain the means by which individuals talk about and understand their daily experience. Individuals do this by developing cause-effect relationships between events and outcomes. This cause-effect relationship is then integrated into the view an individual has of himself as a person, which in turn influences the expectations for future behavior and outcomes. Attribution theory, as defined by Kelly (1967), is concerned with the study of the process by which an individual interprets events as being caused by particular parts of the relatively stable environment. Kelly also postulated that the attributions one makes following success or failure are mediated by one’s past experience and the social context, and that the expectancy for future success 18 cc so wh f0 pen l9} eve tho Ext 19 or failure is a result of the type of attribution one makes. In this instance, the validity or accuracy of the attributions made is not the issue. This study was concerned with what types of attributions one makes and how these influence future behavior. Coqnitive Psychology Cognitive psychology is attempting to understand how incoming information is acted upon and translated into a belief system that allows one to make sense of and give meaning to the external environment. Attribution theorists (Heider, 1958; Kelly, 1971; Weiner, 1974, 1976, 1980) have argued that the perceived causes of success and failure have an important influence on future behavior. Weiner (1979, 1980) further classified Heider’s (1958) four causal elements into a tridimensional model: (a) locus of causality (internal-external), (b) degree of stability (stable-unstable), and (c) controllability (controllable-uncontrollable). In the present context, locus is conceived as a backward-looking belief as to the source of the event under question. Locus of causality is a term which implies that thte attributions individuals propose as reasons for events stress factors that are internal or originate within the person or are the result of environmental sources. Weiner (1972, 1978, 1980) proposed that individuals who attribute the cause of an event to ability (we effort are nmking internal attributions, and those who attribute the cause to luck or task difficulty are making external attributions. Individuals who are internally oriented are 20 seen as self-directing, whereas externally oriented individuals believe they have little control over events in their lives. The stability dimension (stable-unstable) reflects the magnitude or expectancy of change following success or failure. When an individual attributes the cause of an event to ability or task difficulty, which are seen as relatively enduring characteristics, one would expect little future change in behavior. However, if the individual attributes the cause to either effort or luck, which are seen as relatively unstable characteristics, one would realistically expect a wide variation from present to future performance. Weiner’s (1979, 1980) third dimension, controllability (controllable-uncontrollable), refers to whether individuals believe they have control over a particular outcome. This dimension is not completely independent of the locus of causality and stability dimensions. We find causes that are internal and stable, as well as uncontrollable (e.g., mood and fatigue), and other causes that are internal and unstable but are subject to volitional control (e.g., effort) (Weiner & Litman-Adizes, 1980). The perceived centrollabil- ity of an outcome for a present event could have a significant effect on future expectancy of similar events. Attribution theory (Weiner, 1979, 1980) suggests that the combination of attributions made upon the three dimensions will have a profound effect on future events. In an educational setting, the attributions one constructs could well dictate success or failure in an academic setting. In achievement situations, ability, effort, fi de do ES “DC 01" 21 task difficulty, and luck have been identified as the factors most influential in explaining behavior (Andrews & Debus, 1978; Dweck & Reppucci, 1973; Frieze, 1973; Weiner, 1971, 1979). An attributional analysis of achievement motivation assumes that behavior is cognitively mediated by attribution to one or more of these factors. Learned Helplessness Learned helplessness is the theoretical foundation that accounts for the consequences of experiencing unControllable events in both animals and humans. It is the aspect of controllability that links Weiner’s attributional theory and Abramson et al.’s (1978) reformulation of learned-helplessness theory. Overmier and Seligman (1967) and Seligman and Maier (1967) first used the term "learned helplessness" to explain the debilitating consequences of uncontrollable events experienced by dogs. During their research, naive dogs efficiently learned to escape shock by jumping over a barrier in a shuttle box, whereas dogs who received shocks they could not escape or avoid subsequently showed a marked deficit in acquiring the avoidance response when the barrier was removed. Their research proposed that when animals were presented with uncontrollable events, they developed and incorporated into their behavioral pattern a response tendency that suggested that all future situations would also be uncontrollable. This deficit in behavior was the direct result of the uncontrollability of the shock rather than the frequency, intensity, or duration because only the subjects that could not control the 22 shocks subsequently failed to learn to escape or avoid responses (Seligman, Maier, & Solomon, 1971). Learned-helpless students, as identified by Dweck and Reppucci (1973), exhibit the same sense of lack of control over future events. These students have failed in the past, and they expect to fail again in the future. They have developed a behavior pattern that frequently leads to failure. Their response is one of helplessness. They cannot control the sequence of events leading to success or failure. They expect to fail, and they do. These students’ peers were able to persist in the face of repeated noncontingent failure, whereas the helpless students were not. Dweck and Reppucci hypothesized that the negative feedback received by the learned-helpless students over the years has been so consistent that they no longer attempt to alter the failure patterns they experience. The data analysis of Dweck’s study showed that the helpless students generally ignored the role of motivation or effort and made attributions for failure to low ability (an internal-stable factor), those factors over which the student had no control, whereas students who persisted in spite of the failure feedback they received tended to attribute success to high effort and failure to low effort (failure to internal-unstable factors), those over which they had control. The data suggest that attributions to unstable factors, such as effort, allow for the continuation of instrumental responding in the face of failure, whereas attribution to stable 23 factors, such as low ability, produces a sense of noncontrol and a belief in an inability to bring about change. In a follow-up study, Diener and Dweck (1978) gathered information about the comments children made to themselves when they were confronted with an insoluble task. More than half of the children described as "helpless" made statements about not being smart enough to solve the problem, whereas not one "persevering" child made such a comment. Almost half of the mastery-oriented, persevering children made self-instructional statements that guided them in their solution of the problem. It is clear that the helpless students expected to fail and believed that they were not smart enough to change this. In contrast, the persevering students, who associated failure with a lack of effort, acted on their belief that changes could be made that would lead to a more successful outcome. The belief that outcomes are uncontrollable results in deficits of motivation, cognition, and emotion (Abramson, Garber, & Seligman, 1980). Hiroto and Seligman’s (1974) studies suggested that learned- helpless individuals develop a unique response style that is characterized by a decrease in motivation to respond to a given situation in which they perceive they have control over the outcome. These students are also more likely to experience difficulty in learning a novel task. Seligman et a1. (1978) found inadequacies in the original learned-helplessness model when applied to humans. They discovered that the original model could not explain or account for human 24 differences. Persons could experience the same uncontrollable situation but would attribute the cause of the outcome differentially. Second, the original model failed to address the issues of the generality and chronicity of helplessness. Specifically, people could generalize the cause of an event to a new situation. Also, Seligman (1975) found that learned helplessness could not dissipate over time. Therefore, Seligman presented a reformulated model within the attribution framework. Individuals may attribute their performance in achievement situations to a variety of factors, including ability, effort, task difficulty, and luck (Weiner, 1972, 1974). In terms (H’ the learned-helplessness model, a person may make an attribution for the perceived control or lack of control in a: learning situation. For example, people may believe that they were unable to control the outcome of an event because they did not try hard enough (effort) or were not smart enough (lack of ability). The type of attribution will determine whether control or lack of control is to be expected in future situations. The learned-helplessness model is cognitive in that it suggests that the mere exposure to uncontrollable events is not sufficient in and of itself to produce the helplessness deficits in humans. The individual must also view and come to expect that the outcomes are uncontrollable in order to exhibit helplessness behavior (Abramson et a1., 1980). Abramson proposed a three-dimensional framework with which to study individual attributions. 'Hue three dimensions are In we be st su su 911 sh fut stu don he. "1% recc has math Verb Stud on t 25 internality, stability, and globality. These are the same as Weiner’s (1980), except Abramson replaced controllability with ' globality. The dimension of globality implies that factors that have an influence over a great number or range of events are global, whereas factors that are limited in scope or affect only a particular event are specific. This dimension can affect the student in two major ways. For example, suppose a student is doing poorly in physics but has previously been successful in biology and chemistry. The student is likely to view his poor performance in physics as subject-specific. 0n the other hand, if the student has never been successful in science, he is more apt to view his difficulty as global because he is more likely to do poorly in science in any situation. A second implication this dimension has is on the scope of future expectations for academic performance. For example, if’ a student has just taken the math part of the SAT and believes he has done poorly, he can make one of two globality attributions. First, he may attribute his poor performance to a specific cause, such as "I’m not very good at math problems." Here the student has recognized that his performance was not as good as he wished, and he has found a cause for his poor performance that reflects only on his math ability. This attribution should exert little influence on the verbal section to be taken later. The second attribution the student could make is to a global cause, such as "I always do poorly on tests; this test is no different." In this case, he has again n( W tc to fo as pe in in in) "Tl at‘ ste fut den abi Ung 26 recognized his poor performance and has found a cause for his poor performance that reflects on his ability to take all tests. This attribution is likely to have a negative effect on future test taking. .An attribution to global factors predicts that the expectation will recur even when the situation changes, whereas an attribution to specific factors predicts that the expectation need not recur when the situation changes (Abramson et a1., 1978). The internality dimension is based on the self-other dichotomy. When people believe that outcomes are more or less likely to happen to themselves than to relevant others, they attribute these outcomes to internal factors. Conversely, persons make external attributions for outcomes that they believe are as likely to happen to themselves as to relevant others. For example, if an individual is unable to perform a task that his/her peers can successfully accomplish, the individual is likely to attribute the outcome to personal inadequacy. Alternatively, if no one can perform the task, the individual is likely to attribute the outcome to an external cause: "The teacher gives dumb tests." The third dimension, stability, explains the consistency of the attribution and its influence over time. The effect of the stability dimension is reflected in individual expectancies for future performance following success or failure outcomes. It was demonstrated that stable attributions made to factors such as ability led to expectancies of continued success or failure, whereas unstable attributions such as effort can lead to fluctuations in 0L st mi at Th ac me un Th st en be: sul th. SUI po: ex; to per "101 He- 27 outcomes (McMahon, 1973; Valle & Frieze, 1976). For example, if a student is taking a test in math and receives a failing grade, he might make one of two stability attributions. First, he may make an attribution to a stable cause, such as ability: "I’m just dumb." This attribution will probably lead to fUture poor performance in academic situations. There is not much one can do for lack of mental ability. 0r second, he may attribute the failure to an unstable cause, such as effort: "I didn’t study for this test." This attribution leaves open the future possibility that he may study for and do well on the next test. By examining the types of attributions students make when they encounter a success or failure in an academic situation, we can begin to predict the possibility of future performance. Continued success or failure is likely when an individual makes attributions that are internal, stable, and global. Alternatively, attributing success or failure to specific, unstable factors allows for the possibility of the individual making changes and having different expectancies for the future. Attribution Style in Achievement Situations Achievement standards and expectancies for success are related to students’ achievement goals and anticipated future school performance. Weiner’s (1971, 1974) early research on achievement motivation and achievement behavior set the stage for the examination of students’ school performance. As previously stated, Weiner (1979) suggested that, 'hi an achievement situation, ti fc ab be 00 ON anc 28 individuals attribute outcomes to causes that fall along the dimensions of locus of causality, stability, and controllability. Weiner also realized that there are a number of different subordinate or equally useful causal dimensions, including, perhaps, intentionality and globality as proposed by Abramson et a1. (1980). Weiner (1979) linked these dimensions with expectancy change, self- esteem, and interpersonal judgments within an: achievement context. There are also linkages between the three dimensions and psychological affects; stability relates to depression-type affects, and control is associated with particular feeling states and behaviors. The dimension-consequence linkages influence motivated behaviors, such as persistence and choice. In school situations, students’ locus orientations toward achievement events permeate their daily classroom performance. Rotter (1966) is credited with the original studies on locus of control. Rotter postulated that an individual’s perception of the cause of an event could be attributed to dimension of locus of control (internal-external). Lefcourt (1976, 1981) suggested that the perceived internal control of achievement events was essential for the continued effort and continued confidence in one’s abilities. A review of Lefcourt’s work showed that children must believe that their own behaviors can effectively influence important outcomes and that they are competent to perform these activities in order for them to approach achievement situations with confidence and motivation for success. ir tt at li re wh Kai col out out cau mor med fai' eff: out< adec futL Stat or f Witt COnd 29 Weiner’s (1972) and Kukla’s (1972) work indicated that individuals who attribute causation to internal causes view themselves as mostly self-determining, whereas individuals who attribute causation to external factors believe that events in their lives are typically outside of their ability to control. Other researchers have clarified that individuals differ in the degree to which they believe they are able to influence the outcome of events (Crandall, Katkowsky, & Crandall, 1965; Gilmore, 1979; Johnson & Kanoy, 1980). Gilmore and Reed (1979) identified internal and external college students. and had them respond to positive and negative outcomes on a university examination. The internal and positive outcome students attributed responsibility of outcome to internal causal factors, whereas external and negative outcome students were more external in their causal attributions. The most important function of the stability dimension is to mediate expectancies for future performance following success or failure. Specifically, unstable attributions to such factors as effort or luck lead to the expectancies for future change in outcomes. 0n the other hand, attributions to stable causes, such as adequate or inadequate ability, lead to expectancies for consistent future success or failure (McMahon, 1973; Valle & Frieze, 1976). Stated somewhat differently, if the conditions that lead to success or failure are perceived as remaining the same, one would anticipate with increased certainty similar future events. But if the conditions or reasons for success or failure are perceived as 1’. cc cc sa su at Si re; Bill the ex; stL 9re pun per abi Stu the 3O temporary and likely to change, one would question whether the future outcomes will repeat the past. ' Weiner (1979) and Fontaine (1974) implied that success or failure outcomes are predictive of attribution. Fontaine’s data showed that, under success conditions, individuals have a high expectation and confidence for future success, whereas under a failure condition, individuals have a lower expectancy and confidence for success. Simon and Feather (1973), working with college students, found that individuals who passed the given exam saw ability as a more important cause of their outcome than did subjects who failed the exam. However, luck and task-difficulty attributions were greater following exam failure than success. Similar results for ability and task-difficulty attributions were reported by Fitch (1970), Frieze and Weiner (1971), McMahan (1971), and Weiner et a1. (1971). Students in an achievement setting must deal not only with their own expectations toward future outcomes, but also with the expectations of their teachers. Weiner (1972, 1976) found that students who are perceived by their teachers as having expended a great deal of effort are rewarded more when they succeed and punished less when they fail than are those students who are perceived as putting forth little effort. The interaction between ability and effort will affect students’ achievement in school. Students with high ability who put forth little effort will receive the greatest amount of teacher punishment, whereas the greatest 31 amount of teacher praise and reward goes in) the student with low ability who does well because of the great effort put forth. Prawat, Byers, and Anderson (1983) found that teachers believed that hard work should be rewarded and lack of effort should be punished. Teachers were also found to take the greatest pride in low-ability students who became motivated and became angry with the consistently unmotivated child who did well. Teachers want students to be motivated and to work hard. They attribute success in their students to effort. This, however, could come into direct conflict with students’ need to protect their self-esteem. Covington and Omelich (1979) found that many students attempt to avoid the implication that they lack ability by not trying. Learned-Helplessness Achievement Patterns Recently, researchers have begun to explore approaches to modify poor achievement behavior. They have been interested in understanding why children react differentially to achievement outcomes. Dweck and her colleagues (Dweck & Bush, 1976; Dweck & Dweck, 1978; Dweck & Goetz, 1978; Dweck & Reppucci, 1973) looked at the learned-helplessness phenomenon as a way to understand children’s cognitions. Dweck and Reppucci (1973) found that some children’s achievement performance deteriorated when they were exposed to noncontingent failure (unsolvable tasks), whereas other children became more persistent in their efforts (Roth & Bootzin, 1974; Thornton 8c Jacobs, 1972), even though all students were highly motivated and capable of academic success. as tf bl an pe re ta the mo: var res has fol 197: comr difi suc: comp blac 1963 aChi View 32 Dweck and Reppucci (1973) used fifth-grade students. Each was asked to perform insoluble block-design problems. They were told they would be rewarded for correct solutions with prizes. Insoluble block-design problems were interspersed with soluble ones given by another experimenter. They found that children who showed the most performance decrements were those who took less personal responsibility for the outcomes of their actions. When they did take responsibility, they attributed the outcome to ability rather than effort. Students whose efforts escalated after failure used more effort attributions. Race,,Gender, and Income Levels Race, gender, and income levels have been considered important variables in previous attributional research. "hi date, most research on minority motivation from an attributional perspective has focused on the particular attributions of blacks and whites following success or failure at achievement tasks (Friend & Neale, 1972; Willing, Harnisch, lel, &. Maeher, 1983). These studies commonly suggested that blacks tend to rate external factors of task difficulty (ease) and luck as the most important determinants of success and failure. Such reported findings are certainly compatible with the dominant view in the motivation literature of blacks as more externally oriented than whites (Battle & Rotter, 1963) and less sensitive to the value of effort as a cause of achievement (Katz, 1969). More recent studies have challenged this view by documenting no differences between blacks and whites in the sug rep ext l97 ach Fri att ext ext tenl fai deg: fat was con: low. eng this Succ dif1 33 their causal preferences (Willing et a1., 1983) or differences suggesting a more adaptive attributional pattern among blacks (Graham, 1984). The literature on locus of control of Hispanic populations has reported mixed findings. Hispanics have been reported to be more external (Kagan, l976; Pedhazur & Wheeler, 1971), less external (Cole & Cole, 1974, 1977), and equally external (Alvarez & Pader, 1978; Garza & Ames, 1974) in comparison with other groups. Research on gender differences in attributional style in achievement situation has had mixed results. In a recent article by Frieze, Whitley, Hanusa, and McHugh (1982), three basic models of attributional gender differences were reviewed. These were general- externality, self-degradation, and low expectancy. The general- externality model posited that, in an achievement situation, women tend to be generally external in their attributions for success and failure (Feather, 1969; Simon & Feather, 1973). The self- degradation model posited that women attribute success to external factors and failure to internal factors (Nichols, 1975). This model was based on the assumption that people attempt to maintain a set of consistent beliefs about themselves (Aronson & Mater, 1968). The low-expectancy' model suggested that women have a generally low expectation for success in achievement situations. According to this model, low expectations lead to unstable attributions for success and stable attributions for failure. Frieze et a1. (1982) analyzed 21 studies examining gender differences in success-failure attributions. The purpose of the re em Fr wo 19 in the shc l9E inf Par int ext may ext thec Ind‘ of 1 lder con- Uncf OVEY 34 research was to determine which of these three models had the most empirical support. Although none of the models was well-supported, Frieze et al. found that men make stronger ability attributions than women, regardless of the outcome (Dweck & Licht, 1980; Nichols, 1975). Furthermore, Frieze et a1. pointed out the inconsistencies in gender-related attribution research. Specifically, it was found that men attribute success and failure less to luck, whereas women show a greater tendency to use the attribution of luck (Sohn, 1982). Conversely, other researchers (McHugh, 1975; Wong 8 Weiner, 1981) have indicated that, in 21 free-response situation, luck is infrequently cited by both males and females (Elig & Frieze, 1979; Parsons, Meice, Adler, & Kaczala, 1982). Moreover, Wong (1982) interpreted his findings to conclude that males view internality and externality as opposing poles of a single dimension, whereas females may have a two-dimensional view as they see both internal and external causes operating jointly to determine an outcome. Swim—ant The preceding section outlined the components (H: attribution theory and its relationship to the learned-helplessness phenomenon. Individuals tend to attribute the cause of events to the dimensions of internality, stability, and controllability. Individuals who are identified as learned helpless tend to perceive that they have no control over a failure outcome. They perceive failure as unchangeable, and they are unable to see the role effort can play in overcoming failure. Learned-helpless students tend to attribute 35 failure to lack-of-ability causes (internal, stable) and success to luck (external, unstable). Whereas mastery-oriented students believe they can influence the outcome (H: events, these students attribute failure to external, unstable causes, such as lack of effort, and success to internal, stable factors, such as ability. CHAPTER III DESIGN OF THE STUDY Introduction This research was designed to examine the effect of school standing on the attributional patterns of high school students and high school dropouts. The sites were representative of schools in urban, suburban, and rural areas of western Michigan. The study was intended to serve as preliminary research in unraveling the effects that. class standing, 'gender, and race have on the attributional patterns of high-school-age students. The literature reviewed for this study delineated the process by which individuals make attributions for success and failure experienced in achievement settings. The effects these attributions may have on the future expectation of repeated success or failure was explored. The phenomenon of learned helplessness and its effect on student performance was also discussed. Presented in this chapter are the processes used to conduct this study. The population, sample, pertinent geographical data, and sampling techniques are presented first. ‘The development and selection of instruments and an overview of the data-gathering procedures follow. Finally, the statistical treatment is discussed. 36 37 Pooulation and Sampling Methods The sample was drawn from a population of school systems with community education programs in the western Michigan area. Schools were judged to be urban, suburban, or rural in composition. One site was selected from each of the categories. High school dropouts not enrolled in community education programs were selected from a population of prison inmates housed in western Michigan correctional facilities. The experimental design for this study included school class standing (top, middle, or bottom third), high school dropouts, high school dropouts who had returned to a community education program, gender, race, and level of income. The experiment was constructed so as to enable an examination of the interaction effect that each of the variables may have on the attributions of groups of individuals. The high school students were ranked in order of class by present grade point average (GPA), as reported 'hi school records. Students attending school 'were broken into three groups: top, middle, and bottom one-third of their class. Dropouts were selected from community education programs with an upper age limit of 25 years. High school dropouts were selected from inmates in western Michigan prisons. All students were randomly selected from each stratum. 38 Instrumentation The Multidimensional-Multiattributional Causality Scale (MMCS) was developed to assess achievement attribution for achievement situations (Lefcourt, VonBaeyer, Ware, & Cox, 1979). The MMCS consists of 24 causal attribution statements (see Appendix A). The MMCS is made up of eight three-item subscales designed to measure (a) the attribution of school success to effort, ability, the favorable context of the event, and good luck; and (b) the attribution of school failure to lack of effort, lack of ability, an unfavorable context, and bad luck. Examples of the context of an event are the teacher’s method of grading and the course materials. Responses were measured ("1 a five-point Likert-type scale ranging from 4 (Agree) to 0 (Disagree). The five response alternatives presented for each item are "I agree," “I mildly agree," "I agree and disagree equally," "I mildly disagree," and "I disagree." The categories are listed below by question. Each subject was also requested to complete a biographical information sheet (see Appendix A). l. Effort_£ajlure. Poor grades are due to lack of study. 2. Ability Failure. Low marks will cause you to question academic ability. 3. Context Success. Good grades are due to teacher’s easy grading system. 4. Luck Success. Doing well on exam depends on some luck. 5. Effort Success. Good grades are the direct result of effort. 6. Ability Success. Getting goodgrades requires academic ability. ll. 12. 13. 14. 15. l6. 17. 18. 19. 20. 21. 22. 23. 24. 39 Context Failure. Teacher’s perception of student as a less able student will cause you to receive lower grades than if someone else handed in the same work. Luck Failure. Low grades are due partially to bad luck. Effort Failure. Low achievement is due to lack of effort. Ability Failure. Failing a course indicates a lack of skill. Context Success. Good grades may reflect an easier course. Luck Success. Good grades depend on chance factors. Effort Success. Good grades are always the result of hard study. Ability Success. Good grades reflect directly on my aca- demic ability. Context Failure. Poor grades are due to the fact that the teacher makes the course too boring. tuck Failure. My academic low points sometimes make me think I was just unlucky. Effort Failure. Poor grades show that II haven”t worked hard enough. Ability Failure. Poor grades show that I’m unable to suc- ceed in that course. Context Success. Good grades are due to the fact that the material is easy to learn. Luck Success. When I get good grades I consider myself lucky. Effort Success. Hard work will overcome all obstacles. Ability Success. Good grades are due to academic compe- tence. Context Failure. Low grades are due to teachers being stingy with marks. Luck Failure. Bad grades may be a function of bad luck. 40 Data-Gathering Procedures The researcher contacted each data-collection site and obtained permission to conduct the survey from each administrator in charge. Students’ class rankings were determined for students by GPA. Students were randomly selected to fill the three categories of top, middle, and bottom one-third of their class. Upon selection to participate in the study, each subject’s legal guardians were contacted and permission was obtained for each student’s participation (see Appendix B). All subjects completed the MMCS (Lefcourt et a1., 1979). The development of the MMCS has been described elsewhere (Lefcourt, 1981). Its reliability and validity have appeared adequate in other studies (Powers, Douglas, & Choroszy, 1983; Powers & Rossman, 1983). Students attending high school or the community education programs were asked to fill out the questionnaire during class time. The high school students were dismissed from their study hall period and met in a central area where the questionnaire was completed, or the questionnaire was completed during their English class, depending on the high school. Community education attenders completed the questionnaire in their regular classroom. Prison inmates completed their questionnaire during scheduled activity time. ll brief statement was read to all high school students and high school dropouts: You have been asked to participate in a research study on high school students. Your participation is entirely voluntary and anonymous, and you are free to withdraw from the study at any time. 41 It is not possible to give you any additional information at this time as it may affect the outcome of the study. However, as soon as all questionnaires are completed, a meeting will be held and any questions you ask will be answered. When you are finished with the questionnaire, please leave them on the table and return to your next class. Again, thank you for your participation in the research project. Several minutes were allowed for each subject to complete the biographical data sheet. The students were then directed to turn to page three of their packet, and the following directions were read to them: The statements in this booklet express opinions on a number of issues. You may find yourself agreeing strongly with some of the statements, disagreeing just as strongly with others, and perhaps feeling uncertain about others. Your reactions reflect your own opinions and obviously in a belief survey there are no right or wrong answers. Whether you agree or disagree with any statement you can be sure that there are many people who feel the same way you do. Indicate on the test booklet how much you agree or disagree with each statement by circling one of the letters in the following way. I Agree I Mildly Agree I Agree and Disagree Equally I Mildly Disagree I Disagree monw> II II II II II If you have mixed feelings about an item, decide whether you lean slightly more to one side than the other. Try to avoid the middle response. Remember that the best answer is whatever your personal opinion is, and this is usually best reflected in your first reaction to an item, rather than after long debates with yourself over particular experiences. All students were given as much time as they needed to complete the questionnaire. 42 Statistical Treatment This research was designed to assess the effects of a student’s school performance (as measured by GPA) or dropping out on the attributions one makes in achievement settings for success and failure. The respondents indicated their degree of agreement with each item in the MMCS, using a five-point Likert-type scale. The response choices for the questionnaire were weighted in the following manner: Response Choice Scorinq Weight I Agree 4 I Mildly Agree 3 I Agree and Disagree Equally 2 I Mildly Disagree 1 I Disagree 0 Responses were scored from 4 through 0, proceeding from left to right in the answer spaces. Scale scores were determined by summing the weights for the responses chosen for the items on each scale. The MMCS used 24 items (three from each of the eight scales), yielding a score ranging from 0 to 12 for the scales Ability Success, Ability Failure, Effort Success, Effort Failure, Context Success, Context Failure, Luck Success, and Luck Failure. Each of the eight scales was further paired to form the four scales of Ability, Effort, Context, and Luck, yielding a score ranging from 0 to 24. The Ability and Effort scales were combined to form an Internality scale, and the Context and Luck scales were combined to form an Externality scale, yielding a score ranging from 0 to 48. 43 The Internality and Externality scales were combined to yield a total score ranging from 0 to 48. A t-test was used to establish whether a significant difference existed between the group means on the MMCS for students in the top, middle, and bottom one-third of their class and high school dropouts. A t-test was also used to establish whether a significant difference existed between the group means of subjects by gender or race. The level of rejection of the null hypotheses for this study was set at p = .05. Summary The samples for this study were drawn from school systems with community education programs in western Michigan. Three sites were selected, one from each of the urban, suburban, and rural areas. Inmates were selected from correctional facilities in western Michigan. The instrument used in this study was the MMCS, as well as a biographical information sheet. The data were collected on-site by the researcher and subjected to calculations of descriptive statistics and t-tests. The scores for the MMCS were reported as weighted scores. CHAPTER IV ANALYSIS OF THE DATA Introduction Descriptive information about the means and standard deviations of the 17 dependent variables for the five groups of respondents is discussed in Chapter IV. These five groups are: 1. High school students in the top one-third of their class. 2. High school students in the middle one-third of their class. 3. High school students in the bottom one-third of their class. 4. High school dropouts enrolled in community education pro- grams. 5. High school dropouts. The results of testing the 12 hypotheses on each of the dependent variables are presented. Instrument Analysis Demographic Data The demographic data are reported as relative frequencies and percentages in Tables 4.1 through 4.5. The data indicated that the majority of respondents were white (63.1%), followed by blacks (28.3%) (Table 4.1). There was a nearly equal distribution of males 44 45 (46.8%) and females (53.2%) for the sample (Table 4.2). The subjects’ income was grouped into six levels, ranging from under $10,000 to $50,000 and above. The largest number of reports was for the $30,001 to $40,000 income level, with 26.3% of the responses. However, 110 subjects did not respond to this item (Table 4.3). The number of respondents for class level varied widely, with a high of 278 (43.4%) in the top one-third of their class to a low of 84 (13.1%) in the bottom one-third of their class (Table 4.4). The number of respondents by class standing, race, and gender is reported in Table 4.5. Table 4.1.--Distribution of subjects by race. Race Frequency Percent White 404 63.1 Black 181 28.3 Hispanic 14 2.2 Native American 13 2.0 Asian 5 0.8 Table 4.2.--Distribution of subjects by gender. Gender Frequency Percent Female 340 53.2 Male 299 46.2 46 Table 4.3.--Distribution of subjects by income level. Income Level Frequency Percent Under $10,000 61 11.4 $10,001 to $20,000 56 10.4 $20,001 to $30,000 96 17.9 $30,001 to $40,000 141 26.3 $40,001 to $50,000 83 15.5 Above $50,001 100 18.6 Missing 110 Table 4.4.--Distribution of subjects by class standing. Class Standing Frequency Percent Top one-third 278 43.4 Middle one-third 134 20.9 Bottom one-third 84 13.1 Dropout, community education 96 15.0 Table 4.5.--Distribution of subjects by class standing, race, and gender. Males Females Class Standing Total W B 0 W B 0 Top one-third 95 18 15 105 37 7 277 Middle one-third 36 32 5 37 19 5 134 Bottom one-third 24 7 7 29 15 2 84 Dropout, community education 22 12 4 40 10 7 95 Total 177 69 31 211 81 21 590 Note: Total white males = 184, total black males = 82, total white females = 219, total black females = 99. 47 Findings Related to the Multidimensional- Multiattributional Causality,Scale The MMCS indicated the attributional choices for successes and failures in achievement situations. The MMCS consists of 24 questions. The frequencies of responses to each item are presented in Table 4.6. Table 4.6.--Frequencies of responses to MMCS items. Mildly Mildly Question/Attribute Agree Agree Neutral Disagree Disagree 1. Effort failure 281 230 90 21 19 2. Ability failure 151 157 119 89 125 3. Context success 75 115 99 147 205 4. Luck success 93 155 144 121 128 5. Effort success 285 221 89 29 17 6. Ability success 227 194 124 56 40 7. Context failure 156 132 130 95 128 8. Luck failure 66 128 156 171 120 9. Effort failure 291 201 79 44 26 10. Ability failure 111 121 147 118 144 11. Context success 123 146 132 101 129 12. Luck success 80 134 173 135 118 13. Effort success 218 214 99 62 48 14. Ability success 227 207 135 36 36 15. Context failure 209 185 124 66 57 16. Luck failure 22 67 138 188 225 17. Effort failure 317 164 94 39 27 18. Ability failure 89 112 143 148 149 19. Context success 205 192 105 68 71 20. Luck success 112 126 124 124 155 21. Effort success 403 129 85 14 10 22. Ability success 234 225 132 32 18 23. Context failure 78 102 149 151 161 24. Luck failure 64 84 126 148 218 48 Findings Tests of Significance The group means were subjected to t-tests to determine whether there were any significant differences between the different levels of academic success, as measured by GPA. These t-tests and the resulting probability values (p-values) were used to indicate rejection or nonrejection of Hypotheses 1 through 8. A p-value of less than .05 resulted in rejecting the null hypothesis. A summary of the results for each of the hypotheses is presented in Table 4.7. Hypothesis 1 There is no statistically significant difference in Multidimensional-Mu1tiattributiona1 Causality Scale (MMCS) scores for ability, effort, context, and luck between students in the top one-third of their class and students in the middle one-third of their class. Ability. The mean score on the Ability scale of the MMCS for students in the top one-third of their class was 14.5072, whereas the mean score for students in the middle one-third of their class was 14.2463. In the test of significance, the t-value was .56, which was significant at the .574 level (Table 4.8). Effort. The mean score on the Effort scale for the top one- third students was 18.5000, whereas for students in the middle one- third it was 18.4627. In the test of significance, the t-value was .09, which was significant at the .93 level (Table 4.8). Qgfltgxt. The mean score on the Context scale for the top one- third students was 12.941, whereas for students in the middle one- third it was 13.3582. In the test of significance, the t-value was -.86, which was significant at the .391 level (Table 4.8). 49 .m>wpmmm: mm; m=Fm>-g mzh « «mmo.uq x634 a_oo.na uxmpcou a_oo.ua pcoccm emoo.na »H_F_n< capumhmm xuan .> agwgz N_ umuumwmg uoz mpmsmm .> mpmz FF vmpumhmg “oz mpaoaogv .> .um xuwczseou op .Noo.ua spw_wn< umyuahmm mpzoaocu .> m\_ Eoppom m somo.ua »p___n< umuumham .ua xywczssou .> m\_ soupom m meo.ua mepcou amoo.uq spw_wn< cmpuaham muzoaocv .> m\_ a_vuwz N omo.na uxmucou uapumflam .ua xpwcseeou .> m\F apnuwz m umpumhac “oz m\F 508303 .> m\P m_nuwz m «N_o.ua apw__n< cawuahmm mpzoaocu .> m\F no» a aoao.ua “20cc“ capuawmm .nm suwcsssou .> m\F ace m mmo.ua x034 naoamflam m\_ sogpoa .> m\P ace N aagumhmc “oz m\F a_uuwe .> m\_ ace P m=F~>-a “Fammm masocu comwcmaeou mwmmcuoqaz .muzz as“ so momma mwmmspoa»; N_ co mp_=mmz--.N.¢ anmp 50 ngk. The mean score on the Luck scale for the top one-third students was 9.5018, whereas for students in the middle one-third it was 10.3609. In the test of significance, the t-value was -1.84, which was significant at the .067 level (Table 4.8). These scores did not prove to be significant at the .05 level. Therefore, Hypothesis 1 could not be rejected. Table 4.8.--T-ratios of students in the top one-third and middle one-third of their class on the MMCS. Scale Group Mean S.D. T-Value p - - Top 1/3 14.5072 4.177 Ab‘l'ty Middle 1/3 14.2463 4.521 0'56 '574 Top 1/3 18.5000 3.439 Eff°rt Middle 1/3 18.4627 4.290 0'09 '930 Top 1/3 12.9410 4.290 _ C°"teXt Middle 1/3 13.3582 4.685 0'86 '39] Top 1/3 9.5018 4.700 _ LUCK Middle 1/3 10.3609 4.296 1‘84 ‘067 Further analysis of the data revealed that the top one-third of high school students differed from the middle one-third only on the Ability Success scale. Ability Success. The mean score for the top one-third students on the Ability Success scale was 9.0180, whereas for students in the middle one-third the score was 8.2910. In the test of significance, the t-value was 2.52, which was significant at the .013 level. 51 Luck Success. The difference among the two groups approached significance on the Luck Success scale. The mean score of the top one-third students was 5.3538, whereas for the middle one-third students it was 5.8657. In the test of significance, the t-value was -1.77, which was significant at the .07 level. Hypothesis 2 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the top one-third of their class and students in the bottom one- third of their class. agility. The mean score on the Ability scale of the MMCS for students in the top one-third of their class was 14.5072, whereas for students in the bottom one-third of their class it was 13.7143. In the test of significance, the t-value was 1.44, which was significant at the .152 level (Table 4.9). fiftgtt. The mean score on the Effort scale for the top one- third students was 18.5000, whereas for students in the bottom one- third the score was 18.1190. In the test of significance, the t-value was 0.76, which was significant at the .45 level (Table 4.9). Context. The mean score on the Context scale for the top one- third students was 12.9460, whereas for students in the bottom one- third it was 12.4167. In the test of significance, the t-value was 0.82, which was significant at the .41 level (Table 4.9). tgtk. The mean score on the Luck scale for the top one-third students was 9.5018, whereas for students in the bottom one-third it was 10.7143. In the test of significance, the t-value was -2.08, which was significant at the .039 level (Table 4.9). The 52 Luck scale was the only one on which a significant difference was found. Therefore, Hypothesis 2 was rejected. Table 4.9.—-T-ratios of students in the top one-third and bottom one-third of their class on the MMCS. Scale Group Mean S.D. T-Value p - . Top 1/3 14.5072 4.177 Ab111ty Bottom 1/3 13.7143 4.487 1'44 '152 Top 1/3 18.5000 3.439 Effort Bottom 1/3 18.1190 4.190 0'76 '450 Top 1/3 12.9410 4.290 C°"tEXt Bottom 1/3 12.4167 5.422 0’82 '410 , Top 1/3 9.5018 4.700 _ LUCk Bottom 1/3 10.7143 4.661 2'08 ’039 Further analysis of the data revealed that the top one-third of high school students differed significantly from the bottom one- third on two different scales. Ability Success. The mean score on the Ability Success scale of the MMCS for students in the top one-third of their class was 9.0180, whereas for students in the bottom one-third the score was 7.6786. In the test of significance, the t-value was 3.84, which was significant at the .001 level. Luck Failure. The mean score on the Luck Failure scale for the top one-third students was 4.1480, whereas for students in the bottom one-third it was 4.9286. In the test of significance, the t—value was -2.32, which was significant at the .022 level. 53 Hypothesis 3 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the top one-third of their class and students who have dropped out of high school and are enrolled in a community education program. Atiiity. The mean score on the Ability scale of the MMCS for students in the top one-third of their class was 14.5072, whereas for students who attended community education the score was 15.2707. In the test of significance, the t-value was -1.33, which was significant at the .186 level (Table 4.10). fiiigtt. The mean score on the Effort scale for the top one- third students was 18.500, whereas for students who attended community education the score was 19.4375. In the test of significance, the t-value was -2.07, which was significant at the .04 level (Table 4.10). tgntgxt. The mean score on the Context scale for the top one- third students was 12.946, whereas for students who attended community education it was 11.9583. hi the test of significance, the t-value was 1.55, which was significant at the .124 level (Table 4.10). tgtt. The mean score on the Luck scale for the top one-third students was 9.5018, whereas for students who attended community education the mean score was 11.0313. In the test of significance, the t-value was -2.39, which was significant at the .018 level (Table 4.10). 54 Significant differences were found on the Effort and Luck scales. Therefore, Hypothesis 3 was rejected. Table 4.10.--T-ratios of students in the top one-third of their class and dropouts enrolled in community education on the MMCS. Scale Group Mean S.D. T-Value p - . Top 1/3 14.5072 4 177 _ Ab‘I‘ty Comm. ed. 15.2708 5.065 1'33 '186 Top 1/3 18.5000 3.439 _ Eff°rt Comm. ed. 19.4375 3.936 2‘07 '040 Top 1/3 12.9410 4.290 C°"teXt Comm. ed. 11.9583 5.723 1'55 '124 Top 1/3 9.5018 4.700 _ LUCK Comm. ed. 11.0313 5.626 2'39 '018 Further analysis of the data revealed that the top one-third of high school students differed significantly from dropouts attending community education on 6 of the 18 scales. Ability Failure. The mean score on the Ability Failure scale of the MMCS for students in the top one-third of their class was 5.4892, whereas for students who attended community education the mean score was 6.5208. In the test of significance, the t—value was -2.66, which was significant at the .009 level. Effort Success. The mean score on the Effort Success scale for the top one-third students was 9.2482, whereas for students who attended community education it was 10.0521. In the test of 55 significance, the t-value was -3.30, which was significant at the .001 level. Context Egjlure. The mean score on the Context Failure scale for the top one-third students was 6.5683, whereas for students who attended community education the mean score was 5.6667. In the test of significance, the t-value was 2.29, which was significant at the .023 level. Luck Success. The mean score on the Luck Success scale for the top one-third students was 5.3538, whereas for students who attended community education it was 6.4583. In the test of significance, the t-value was -2.97, which was significant at the .004 level. Failure. The mean score on the Failure scale for the top one- third students was 19.9856, whereas for students who attended community education it was 18.3333. In the test of significance, the t-value was 2.41, which was significant at the .017 level. Internal. The mean score on the Internal scale for the top one-third students was 33.0072, whereas for students who attended community education it was 34.0783. In the test of significance, the t-value was -1.99, which was significant at the .048 level. Hypothesis 4 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the top one-third of their class and students who have dropped out of high school. Ability. The mean score on the Ability scale of the MMCS for students in the top one-third of their class was 14.5072, whereas 56 for students who dropped out the mean score was 16.8276. In the test of significance, the t-value was -2.64, which was significant at the .012 level (Table 4.11). Table 4.ll.--T-ratios of students in the top one-third of their class and high school dropouts on the MMCS. Scale Group Mean S.D. T-Value p - . Top 1/3 14.5072 4.177 _ Ab‘I'ty Dropouts 16.8276 4.528 2'64 '012 Top 1/3 18.5000 3.439 _ Effort Dropouts 19.7241 3.972 1'60 '120 Top 1/3 12.9410 4.290 Context Dropouts 11.3448 4.790 1'73 '093 Top 1/3 9.5018 4.700 , LUCK Dropouts 10.1742 4.676 0'73 '468 ttfgtt. The mean score on the Effort scale for the top one- third students was 18.500, whereas for students who dropped out it was 19.7241. In the test of significance, the t-value was -1.60, which was significant at the .120 level (Table 4.11). Context. The mean score on the Context scale for the top one- third students was 12.9460, whereas for students who dropped out it was 11.3448. In the test of significance, the t-value was 1.73, which was significant at the .093 level (Table 4.11). tgtt. The mean score on the Luck scale for the top one-third students was 9.5018, whereas for students who dropped out the mean 57 score was 10.1724. hi the test of significance, the t-value was -0.73, which was significant at the .468 level (Table 4.11). There was a significant difference on the Ability scale. Therefore, Hypothesis 4 was rejected. Further analysis of the data revealed that the top one-third of high school students differed significantly from high school dropouts on four other scales. Ability Failure. The mean score for students in the top one- third on the Ability Failure scale was 5.4892, whereas for students who dropped out it was 7.2414. In the test of significance, the t-value was -2.69, which was significant at the .01 level. Effort Success. The mean score on the Effort Success scale for the top one-third students was 9.2482, whereas for students who dropped out the mean score was 10.7241. In the test of significance, the t-value was -3.62, which was significant at the .001 level. Context Failure. The mean score on the Context Failure scale for the top one-third students was 6.5683, whereas for students who dropped out the mean score was 4.0000. In the test of significance, the t-value was 3.90, which was significant at the .001 level. Failure. The mean score on the Failure scale for the top one- third students was 19.9856, whereas for students who dropped out it was 15.3793. In the test of significance, the t-value was 2.72, which was significant at the .011 level. 58 Hypothesis 5 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the middle one-third of their class and students in the bottom one- third of their class. Atiiity. The mean score on the Ability scale of the MMCS for high school students in the middle one-third of their class was 14.2464, whereas for students in the bottom one-third the mean score was 13.7143. In the test of significance, the t-value was 0.85, which was significant at the .397 level (Table 4.12). Table 4.12.--T-ratios of students in the middle one-third and bottom one-third of their class on the MMCS. Scale Group Mean S.D. T-Value p - . Middle 1/3 14.2463 4.521 Ab‘l‘ty Bottom 1/3 13.7143 4.487 0'85 '397 Middle 1/3 18.4627 4.290 Eff°rt Bottom 1/3 18.1190 4.190 0'58 '560 Middle 1/3 13.3582 4.685 contEXt Bottom 1/3 12.4167 5.422 1'3] '19] Middle 1/3 10.3609 4.296 _ LUCk Bottom 1/3 10.7143 4.661 0'56 '576 gifgtt. The mean score on the Effort scale for the middle one- third students was 18.4627, whereas for the bottom one-third it was 18.1190. In the test of significance, the t-value was 0.56, which was significant at the .56 level (Table 4.12). 59 tgfltgxt. The mean score on the Context scale for the middle one-third students was 13.3582, whereas for the bottom one-third it was 12.4167. In the test of significance, the t-value was 1.31, which was significant at the .191 level (Table 4.12). tugs. The mean score on the Luck scale for students in the middle one-third was 10.3609, whereas for those in the bottom one- third it was 10.7143. In the test of significance, the t-value was -0.56, which was significant at the .576 level (Table 4.12). These scores did not prove to be significant at the .05 level. Therefore, Hypothesis 5 could not be rejected. Further analysis of the data revealed that the students who were in the middle one-third of their high school class did not differ significantly on any of “the scales from students in the bottom one-third of their class. Hypothesis 6 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the middle one-third of their class and students who have dropped out of high school and are enrolled in a community education program. Agility. The mean score on the Ability scale of the MMCS for students in the middle one-third of their class was 14.2463, whereas for students who attended community education it was 15.2708. In the test of significance, the t-value was -1.58, which was significant at the .115 level (Table 4.13). tfiggt. The mean score on the Effort scale for the middle one- third students was 18.4627, whereas for students who attended 60 community education it was 19.4375. In the test of significance, the t-value was -1.78, which was significant at the .076 level (Table 4.13). Table 4.13.--T-ratios of students in the middle one-third of their class and high school dropouts enrolled in community education on the MMCS. Scale Group Mean S.D. T-Value p - - Middle 1/3 14.2463 4.521 _ Ab‘I‘ty Comm. ed. 15.2708 5.065 1'58 '115 Middle 1/3 18.4627 4.290 _ Effort Comm. ed. 19.4375 3.936 1'78 '076 Middle 1/3 13.3582 4.685 C°”tQXt Comm. ed. 11.9583 5.723 1‘97 '050 Middle 1/3 10.3609 4.296 _ LUCk Comm. ed. 11.0313 5.626 0'98 '329 Context. The mean score on the Context scale for the middle one-third students was 13.3582, whereas for students who attended community education the mean score was 11.9583. In the test of significance, the t-value was 1.97, which was significant at the .05 level (Table 4.13). L195. The mean score on the Luck scale for the middle one- third students was 10.3609, whereas for students who attended community education it was 11.0313. In the test of significance, the t-value was -0.98, which was significant at the .329 level (Table 4.13). 61 The Context scale was significant at the .05 level. This indicated a significant difference; therefore, Hypothesis 6 was rejected. Further analysis of the data revealed that the middle one-third of high school students differed significantly from the dropouts enrolled in community education on four other scales. Effort Success. The mean score on the Effort Success scale of the MMCS for students in the middle one-third of their class was 9.0373, whereas for students who attended community education it was 10.0521. In the test of significance, the t—value was -3.37, which was significant at the .001 level. Context Failure. The mean score on the Context Failure scale for the middle one—third students was 7.0224, whereas for students who attended community education it was 5.6667. In the test of significance, the t-value was 3.13, which was significant at the .002 level. Failure. The mean score on the Failure scale for the middle one-third students was 20.1654, whereas for students who attended community education the mean score was 18.333. In the test of significance, the t-value was 2.46, which was significant at the .015 level. Internal. The mean score on the Internal scale for the middle one-third students was 32.7090, whereas for students who attended community education it was 34.7083. In the test of significance, the t-value was -2.03, which was significant at the .043 level. 62 Hypothesis 7 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the middle one-third of their class and students who have dropped out of high school. Ability. The mean score on the Ability scale of the MMCS for students in the middle one-third of their class was 14.2463, whereas for students who dropped out it was 16.8276. In the test of significance, the t-value was -2.78, which was significant at the .008 level (Table 4.14). Table 4.14.--T-ratios of students in the middle one-third of their class and high school dropouts on the MMCS. Scale Group Mean S.D. T-Value p - . Middle 1/3 14.2463 4.521 , ‘ Ab‘I‘ty Dropouts 16.8276 4.528 2‘79 ‘008 Middle 1/3 18 4627 4.290 - _ Effort Dropouts 19.7241 3.972 1'45 '134 Middle 1/3 13.3582 4.685 Context Dropouts 11.3448 5.723 2'09 '046 Middle 1/3 10.3609 4.296 LUCK Dropouts 10.1724 4.676 0‘21 ‘843 Effort. The mean score on the Effort scale for the middle one- third students was 18.4627, whereas for students who dropped out the mean score was 19.7241. In the test of significance, the t-value was -l.53, which was significant at the .134 level (Table 4.14). 63 tgfltext. The mean score on the Context scale for the middle one-third students was 13.3582, whereas for students who dropped out it was 11.3448. In the test of significance, the t-value was 2.09, which was significant at the .046 level (Table 4.14). 1453. The mean score on the Luck scale for the middle one- third students was 10.3609, whereas for students who dropped out it was 10.1724. In the test of significance, the t-value was 0.21, which was significant at the .843 level (Table 4.14). The Ability and Context scales were both below the .05 level of significance. This indicated a significant difference; therefore, Hypothesis 7 was rejected. Further analysis of the data revealed that the middle one-third of high school students differed significantly from the dropouts on five other scales. Ability Success. The mean score on the Ability Success scale of the MMCS for students in the middle one-third of their class was 8.2910, whereas for students who dropped out the mean score was 9.5862. In the test of significance, the t-value was -2.15, which was significant at the .042 level. Effort Success. The mean score on the Effort Success scale for the middle one-third students was 9.0373, whereas for students who dropped out it was 10.7241. In the test of significance, the t-value was -3.36, which was significant at the .001 level. Context Failure. The mean score on the Context Failure scale for the middle one-third students was 7.0224, whereas for dropouts 64 it was 4.000. In the test of significance, the t-value was 5.20, which was significant at the .001 level. Internal. The mean score on the Internal scale for the middle one-third students was 32.7090, whereas for students who dropped out it was 36.5517. In the test of significance, the t-value was -2.68, which was significant at the .007 level. Failure. The mean score on the Failure scale for the nfiddle one-third students was 20.1654, whereas for students who dropped out it was 15.3793. In the test of significance, the t-value was 2.72, which was significant at the .013 level. Hypothesis 8 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the bottom one-third of their class and students who have dropped out of high school and are enrolled in a community education program. Ability. The mean score on the Ability scale of the MMCS for students in the bottom one-third of their class was 13.7143, whereas for students who attended community education the mean score was 15.2700. In the test of significance, the t-value was -2.l9, which was significant at the .03 level (Table 4.15). Effort. The mean score on the Effort scale for the bottom one- third students was 18.1190, whereas for students who attended community education it was 19.4375. In the test of significance, the t-value was -2.l7, which was significant at the .032 level (Table 4.15). 65 Table 4.15.--T-ratios of students in the bottom one-third of their class and high school dropouts enrolled in community education on the MMCS. . Scale Group Mean S.D. T-Value p - - Bottom 1/3 13.7143 4.870 _ Ab‘I‘ty Comm. ed. 15.2708 5.065 2'19 ‘030 Bottom 1/3 18.1190 4.190 _ EffOrt Comm. ed. 19.4375 3.936 2'17 '032 Bottom 1/3 12.4167 5.422 6°"tQXt Comm. ed. 11.9583 5.723 0'55 '582 Bottom 1/3 10.7143 4.661 _ LUCK Comm. ed. 11.0313 5.626 0‘4] ‘680 Context. The mean score on the Context scale for the bottom one-third students was 12.4167, whereas for students who attended community education it was 11.9583. In the test of significance, the t-value was 0.55, which was significant at the .582 level (Table 4.15). L_u<_:_t. The mean score on the Luck scale for the bottom one- third students was 10.7143, whereas for students who attended community education it was 11.0313. In the test of significance, the t-value was -0.41, which was significant at the .68 level (Table 4.15). The Ability and Effort scales were both below the .05 level of significance. This indicated a significant difference; therefore, Hypothesis 8 was rejected. 66 Further analysis of the data revealed that students in the bottom one-third of their class differed from high school dropouts- enrolled in community education on three scales. Ability Success. The mean score on the Ability Success scale of the MMCS for students in the bottom one-third of their class was 7.6786, whereas for high school dropouts who attended community education it was 8.7500. In the test of significance, the t-value was -2.54, which was significant at the .012 level. Internal. The mean score on the Internal scale for students in the bottom one-third of their class was 31.8333, whereas for high school dropouts who attended community education the mean score was 34.7093. In the test of significance, the t-value was -2.65, which was significant at the .009 level. Egiigte. The mean score on the Failure scale for students in the bottom one-third of their class was 20.1786, whereas for high school dropouts who attended community education it was 18.3330. In the test of significance, the t-value was -2.00, which was significant at the .047 level. Hypothesis 9 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the bottom one-third of their class and students who have dropped out of high school. Ability. The mean score on the Ability scale of the MMCS for students in the bottom one-third of their class was 13.7143, whereas for students who dropped out it was 16.8276. In the test of 67 significance, the t-value was -3.21, which was significant at the .002 level (Table 4.16). Table 4.16.--T-ratios of students in the bottom one-third of their class and high school dropouts on the MMCS. Scale Group Mean S.D. T-Value p - - Bottom 1/3 13.7143 4.870 _ Ab111ty Dropouts 16.8271 4.528 3'2] “002 Bottom 1/3 18.1190 4.190 _ Eff°rt Dropouts 19.7241 3.972 1'80 “070 Bottom 1/3 12.4167 5.422 Context Dropouts 11.3448 4.972 0'94 '320 Bottom 1/3 10.7143 4.661 LUCk Dropouts 10.1724 4.676 0'54 '590 gttgtt. The mean score on the Effort scale for the bottom one- third students was 18.1190, whereas for students who dropped out it was 19.7241. In the test of significance, the t-value was -l.80, which was significant at the .07 level (Table 4.16). QQDLBLL- The mean score on the Context scale for the bottom one—third students was 12.4167, whereas for students who dropped out it was 11.3448. In the test of significance, the t-value was 0.94, which was significant at the .32 level (Table 4.16). Lu. The mean score on the Luck scale for the bottom one- third students was 10.7143, whereas for students who dropped out it was 10.1724. In the test of significance, the t-value was 0.54, which was significant at the .59 level (Table 4.16). 68 The Ability scale was below the .05 level of significance. This indicated a significant difference; therefore, Hypothesis 9 was rejected. Further analysis of the data revealed that students in the bottom one-third of their class differed from high school dropouts on five scales. Ability Success. The mean score on the Ability Success scale of the MMCS for students in the bottom one-third of their class was 7.6786, whereas for students who dropped out it was 9.5862. In the test of significance, the t-value was -3.00, which was significant at the .005 level. Effort Success. The mean score on the Effort Success scale for students in the bottom one-third of their class was 9.0595, whereas for students whor dropped out it was 10.7241. In the test of significance, the t-value was -3.25, which was significant at the .001 level. Context Failure. The mean score on the Context Failure scale for students in the bottom one-third of their Class was 6.3452, whereas for students who dropped out it was 4.000. In the test of significance, the t-value was 3.27, which was significant at the .002 level. Internal. The mean score on the Internal scale for the bottom one-third students was 31.8333, whereas for students who dropped out it was 36.5517. In the test of significance, the t-value was -3.18, which was significant at the .002 level. 69 Failure. The mean score on the Failure scale for the bottom one-third students was 20.1876, whereas for students who dropped out it was 15.3793. In the test of significance, the t-value was 3.07, which was significant at the .01 level. Hypothesis 10 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students who have dropped out of high school and enrolled in community education programs, and students who have dropped out of high school. Ability. The mean score on the Ability scale of the MMCS for students in community education was 15.2708, whereas for students who dropped out it was 16.8276. In the test of significance, the t-value was -1.48, which was significant at the .12 level (Table 4.17). Table 4.17.--T-ratios of students in community education and high school dropouts on the MMCS. Scale Group Mean S.D. T-Value p . - Comm. ed. 15.2708 5.065 _ Ab‘I‘ty Dropouts 16.8276 4.528 1'48 '120 Comm. ed. 19.4375 3.936 _ Effort Dropouts 19.7241 3.972 0'34 '730 Comm. ed. 11.9583 5.723 C°"tEXt Dropouts 11.3448 4.790 0'52 '560 Comm. ed. 11.0313 5.626 LUCK Dropouts 10.1724 4.676 0'75 '410 70 Effort. The mean score on the Effort scale for community education students was 19.4375, whereas for students who dropped out the mean score was 19.7241. In the test of significance, the t-value was -0.34, which was significant at the .734 level (Table 4.17). Context. The mean score on the Context scale for community education students was 11.9583, whereas for students who dropped out it was 11.3448. In the test of significance, the t-value was 0.52, which was significant at the .567 level (Table 4.17). LQQK. The mean score on the Luck scale for community education students was 11.0313, whereas for students who dropped out it was 10.1724. In the test of significance, the t-value was 0.75, which was significant at the .413 level (Table 4.17). There were no significant differences on any (H: the scales. Therefore, Hypothesis 10 was not rejected. Further analysis of the data revealed that students in community education programs differed significantly from high school dropouts on one scale. Context Failure. The mean score on the Context Failure scale for community education students was 4.677, whereas for students who dropped out it was 4.0000. In the test of significance, the t-value was 2.26, which was significant at the .027 level. 71 Hypothesis 11 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between male and female students. Ability. The mean score on the Ability scale of the MMCS for male students was 14.3244; for female students the mean score was 14.6118. In the test of significance, the t-value was -0.82, which was significant at the .412 level (Table 4.18). Table 4.18.--T-ratios of male and female students on the MMCS. Scale Group Mean S.D. T-Value p - - Males 14.3224 4.472 _ Ab‘I‘ty Females 14.6118 4.352 0'82 ‘412 Males 18.4849 3.849 _ Effort Females 18.7588 3.827 0'90 '369 Males 13.0100 4.661 C°"tEXt Females 12.4029 4.916 1'60 '110 Males 10.0671 4.721 LUCK Females 9.9912 4.804 0'20 '84] titott. The mean score on the Effort scale for male students was 18.4849, whereas female students’ mean score was 18.7588. In the test of significance, the t-value was -0.90, which was significant at the .369 level (Table 4.18). tgfltgtt. The mean score on the Context scale for male students was 13.0100, whereas the score for female students was 12.4029. In 72 the test of significance, the t—value was 1.60, which was significant at the .11 level (Table 4.18). tgtt. The mean score on the Luck scale for males was 10.0671, whereas the score for females was 9.9912. hi the test of significance, the t-value was 0.20, which was significant at the .841 level (Table 4.18). No significant. difference *was found on any of the scales. Therefore, Hypothesis 11 could not be rejected. Further analysis of the data revealed that male and female students differed significantly on three of the attributional scales. Ability Failure. The mean score on the Ability Failure scale of the MMCS for male students was 5.5619, whereas the score for female students was 6.0853. In the test of significance, the t-value was -2.15, which was significant at the .032 level. Context Failure. The mean score on the Context Failure scale for males was 6.8060, whereas the score for females was 6.1558. In the test of significance, the t-value was -2.78, which was significant at the .006 level. Egiigtg. The mean score on the Failure scale for males was 20.5336, whereas the score for females was 18.8909. In the test of significance, the t-value was 3.438, which was significant at the .001 level. 73 Hypothesis 12 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between white and black students. Ability. The mean score on the Ability scale of the MMCS for white students was 14.1683, whereas the score for black students was 15.4033. In the test of significance, the t-value was -3.12, which was significant at the .002 level (Table 4.19). Table 4.19.--T-ratios of white and black students on the MMCS. Scale Group Mean S.D. T-Value p . - Whites 14.1683 4.235 _ Ab‘I‘ty Blacks 15.4033 4.417 3'17 '002 Whites 18.2946 3.869 _ Effort Blacks 19.5525 3.618 3'80 '00] Whites 13.5421 4.590 C°"tQXt Blacks 10.8785 4.918 6’18 '00] Whites 10.3391 4.607 LUCk Blacks 9.4190 5.086 2'07 '039 fiftgtt. The mean score on the Effort scale for white students was 18.2946, whereas the score for black students was 19.5525. In the test of significance, the t-value was -3.80, which was significant at the .001 level (Table 4.19). tgfltgyt. The mean score on the Context scale for white students ‘was 13.5421, whereas the score for black students was 74 10.8785. In the test of significance, the t-value was 6.18, which was significant at the .001 level (Table 4.19). LEQK- The mean score on the Luck scale for white students was 10.3391, whereas the score for black students was 9.4190. In the test of significance, the t-value was 2.07, which was significant at the .039 level (Table 4.19). A significant difference was found for each scale. Therefore, Hypothesis 12 was rejected. Further data analysis revealed that white and black students differed significantly on 13 other scales. The mean scores, standard deviations, t-values, and probability levels are reported in Table 4.20. The black students differed significantly from the white students in the basic underlying attributional pattern. Black students were much more likely to attribute their success to internal causes than were white students. Research Question 1 Which of the attributional categories of ability, effort, context, and luck is viewed as the most important determinant of academic success and failure in achievement settings, as reported by race, gender, and class standings? The rank orderings of the mean scores for the four attributional categories (ability, effort, context, and luck), as shown in Tables 4.21 through 4.23, indicate the order of causal ascription by race, gender, and class standing, respectively. 75 Table 4.20.—-T-ratios of white and black students on the attribu- tional scales of the MMCS. Scale Group Mean S.D. T-Value p 13:22:: 11:2: 21:12 22:: 1:11:22 11:2: 22:1: :12: 22:22:. 11:2: 211:: 2:22: 121135: 11.11%: 3:193? 35?? ’3-40 '0‘“ 1:22:22 1122: 2:221: 22:: 12:11:: 11:2: 21:2: 221: 12:1... 11:2: 221:: 2:11: 11:2: 22:21:: 22:2 11:2: 22:22:: 21:: 11:2: 12:12: 22:1 11:2: 11:22:: 2:22: 11:2: 22:11:: 12:22: 76 Table 4.21.-—Means and standard deviations of students’ attributions on the Effort, Ability, Context, and Luck scales, by race. Scale Blacks Whites Others Mean = 19.5525 18.2946 18.1455 Eff°rt S.D. = 3.618 3.869 3.865 - - Mean = 15.4033 14.1683 13.8634 Ab‘I‘t’ S.D. = 4.417 4.235 5.263 Mean = 10.8785 13.5421 12.5273 C°"tEXt S.D. = 4.918 4.590 4 472 Luck Mean = 9.4190 10.3391 9.9091 S.D. = 5.086 4.607 4.862 Table 4.22.--Means and standard deviations of students’ attributions on the Effort, Ability, Context, and Luck scales, by gender. Scale Males Females 1:2: : 22:29 ‘21:: 2:1: : 2:214 121:8 1:: : 22210 12:21: 1:1: : 2121‘ 2:211: 77 Table 4.23.--Means and standard deviations of students’ attributions on the Effort, Ability, Context, and Luck scales, by class standing. Top Middle Bottom Comm. Drop- Scale 1/3 1/3 1/3 Educ. outs Mean = 18.50 18.46 18.12 19.44 19.70 Eff°rt S.D. = 3.49 4.29 4.19 3.94 3.97 - . Mean = 14.51 14.25 13.71 15.27 16.80 Ab‘l‘ty 5.0. = 4.18 4.52 4.49 5.07 4.52 Mean = 12.95 13.36 12.42 11.96 11.30 C°”teXt S.D. = 4.29 4.69 5.42 5.72 4.79 Luck Mean = 9.50 10.36 10.71 11.03 10.20 5.0. = 4.70 4.30 4.66 5.63 4.67 These findings corroborate those of Lefcourt et a1. (1979) and Douglas and Powers (1983), who reported the same rank ordering among university undergraduates and precollege students. High school students and high school dropouts in community education programs clearly considered effort as the most important determinant of academic success and failure. Further, the smaller standard deviation of the effort scores indicated a greater consensus among students. Research Question 2 00 students as they progress through grades 10, 11, and 12 change their attributional pattern? The mean scores of high school students in the three grade levels--10th, 11th, or 12th grade—-showed no significant differences. Students in grade 10 made similar attributional 78 responses to those of students in grades 11 and 12. Table 24 shows the mean scores and standard deviations for each of the grades. Table 4.24.--Means and standard deviations of students’ attributions on the Effort, Ability, Context, and Luck scales, by grade level. Grade Ability Effort Context Luck 10 Mean = 14.3282 18.3795 12.4718 9.6995 S.D. = 4.352 4.174 4.702 4.782 1] Mean = 14.4525 18.2123 13.2849 10.1844 S.D. = 4.275 3.600 4.296 4.302 12 Mean = 14.0410 18.8115 13.2951 9.9590 S.D. = 4.383 3.524 4.854 4.733 Research Question 3 What effect does the income level of the student’s family have on the student’s attributional pattern? Socioeconomic status (SES) plays an important role in shaping a student’s developmental process. This researcher was concerned with the students’ SES level and its effect on their attributional pattern. For the purpose of this study, parents’ income or students’ own income if they were living independently was used as a measure of SES. Analysis of variance was used to determine whether income played a role in explaining the variance on the Ability, Effort, Context, and Luck scales for the four class standings: top one- third, middle one-third, bottonl one-third, and high school 79 dropouts. Students’ income level did not play a significant role in explaining the variance on the Ability scale (F = .508, significant at the .770 level) (Table 4.25). Income did account for some of the variance on the Effort scale (F = 2.422, significant at the .035 level). The Context and Luck scales were not significantly affected by students’ income level (Context: F = .649, significant at the .662 level; Luck: F = 1.245, significant at the .287 level). Table 4.25.--Effect of attributions on educational achievement: analysis of Ability, Effort, Context, and Luck scale scores by class standing and income level. Source of Sum of Mean Signif. Variation Squares df Square F of F Ability Main effects 217.457 8 27.182 1.331 .226 Class standing 122.552 3 40.851 2.000 .113 Income 51.906 5 10.381 0.508 .770 Effort Main effects 268.670 8 33.584 2.238 .024 Class standing 179.585 3 59.862 3.990 .008 Income 181.699 5 36.340 2.422 .035 Context Main effects 128.072 8 16.009 0.663 .724 Class standing 10.837 3 3.612 0.150 .930 Income 78.365 5 15.673 0.649 .662 Luck Main effects 432.904 8 54.113 2.244 .023 Class standing 157.489 3 52.496 2.177 .090 Income 150.098 5 30.020 1.245 .287 8O Resegggh Question 4 00 students who experience greater academic success, as measured by GPA, attribute success and failure in a more productive pattern than students who are not as successful in achievement settings? Significant differences were found among the five class- standing groups on seven of the attributional scales. Analysis of variance indicated a difference on the Ability Success, Ability Failure, Effort Success, Context Failure, Luck Success, Luck, and Internal scales. Results of the analysis of variance for the significant scales are shown in Table 4.26. Summary The statistical and descriptive analyses of the data were presented in this chapter. The data that were generated by the MMCS were presented. The tests of significance and their resulting probability values were discussed for each research hypothesis. Data concerning important research questions were reported, along with tests of significance. Eight of the 12 stated hypotheses were rejected. Significant differences in attributional patterns were found between the following groups: students in the top one-third of' their' class and students in the bottom one—third, community education, and dropouts; students in the middle one-third of their class and those in community education and dropouts; students in the bottom one-third of their class and those in community education and dropouts; and black and white students. 81 Table 4.26.--Effect of attributions on educational achievement: analysis of Ability, Effort, Context, and Luck scale scores by class standing. Source of Sum of Mean Signif. Variation Squares df Square F of F Ability Success Between groups 134.6040 3 44.8680 Within groups 4072.8808 588 6.9267 -4775 ~0003 Ability Failure Between groups 83.7049 3 27.9016 Within groups 5560.0502 588 9.4559 '9507 '0322 Effort Success Between groups 69.1408 3 23.0469 Within groups 2954.1295 588 5.0240 '5873 '0035 Context Failure Between groups 106.5173 3 35.5058 Within groups 5137.4557 588 8.7372 '0638 '007] Luck Success Between groups 92.6360 3 30.8787 Within groups 4988.8868 588 8.4990 '6332 '0128 Luck Between groups 222.9556 3 74.3185 .2637 .0211 Within groups 13343.9749 586 22.7713 Internal Between groups 404.4445 3 134.8148 Within groups 26811.1349 588 45.5972 '9566 '0319 CHAPTER V SUMMARY AND CONCLUSIONS This chapter includes a summary of the study, a discussion of the findings as they apply to the research hypotheses and the research questions, conclusions, and recommendations for future research. Summary The purpose of this study was to investigate the different attributional styles of high school students and high school dropouts in achievement situations. In preparation for the study, an extensive literature search was undertaken. Attribution research has been conducted extensively for the past ten years. The search was narrowed to focus on how students attribute success or failure in academic settings. Research was supplemented by pertinent literature from related fields: the psychology of motivation and reward, learned helplessness, and cognitive psychology. The Multidimensional-Multiattributional Causality Scale (MMCS) was used. This indicates the type of attribution one makes for success and failure in achievement settings. The subjects came from 'three school systems with community education programs in the western Michigan area, and western 82 83 Michigan prison inmates. An inner-city, a suburban, and a rural setting were selected. The data were gathered on-site by the researcher and submitted to statistical analysis. The computer program used to tabulate the scores was the Statistical Package for the Social Sciences (SPSS-X) on the Michigan State University IBM mainframe computer. Findings The findings of the data analyses that were presented in Chapter IV are discussed here in two parts. Part one is a discussion of the results of the t-tests that were applied to the research hypotheses. In part two, the research questions are restated, and the findings from the data pertinent to the questions are discussed. Hypothesis 1 There is no statistically significant difference in Multidimensional-Multiattributional Causality Scale (MMCS) scores for ability, effort, context, and luck between students in the top one-third of their class and students in the middle one-third of their class. Although the two groups did not differ on the test of significance, the students in the top one-third of their class did rate both ability and effort as more important in determining success and failure, whereas students in the middle one—third of their class rated luck and context as more important in determining success and failure than the students in the top one-third of their class. These differences indicate a more positive trend in the 84 attributional pattern for students in the top one-third of their class over those in the middle one-third of their class. Hypothesis 2 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the top one-third of their class and students in the bottom one- third of their class. A comparison of means from the composite scores on the MMCS between students in the top one-third of their class and those in the bottom one-third of their class revealed a statistically significant difference of less than .05 on the Ability Success, Luck Failure, and Luck scales. Students who were in the bottom one-third of their class believed that luck was a more important component in determining success and failure than did students in the top one- third of their class. This attributional pattern of attributing success and failure to luck will hinder the students who perform more poorly. Since luck was the reason for past performance, the students saw little reason for making the changes necessary to become more successful in academic situations. The reasons they failed were outside of their control. Students in the top one-third of their class attributed their success more to their ability than did students in the bottom one-third of their class. The top one- third students rated effort as a more important factor in determining success and failure than did students in the bottom one- third of their class. The attributional pattern exhibited by the top one-third students tended to be more positive and productive for academic success than that exhibited by those in the bottom 85 one-third of their class. The top one-third students believed that their ability caused the success, that their effort was the reason for their success, and that lack of effort caused their failure. Hypothesis 3 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the top one-third of their class and students who have dropped out of high school and are enrolled in a community education program. The data showed that high school dropouts attending community education programs were more likely to attribute their failure to lack of effort and their success to their effort than were students in the top one-third of their class. Along with this, dropouts in community education were more likely to attribute both their successes and failures to internal causes” ' Community education students reported that the locus of control was more internal than did high school students. Their performance was controlled more by those things occurring within themselves than by the environment. High school students in the top one-third of their class were more likely to attribute failure to external factors. The context of the event was often seen as the reason they failed; it was not something within themselves that caused the failure. These differences might show that community education students were more realistic in assessing their actual strengths and weaknesses than were high school students. Dropouts 'hi community education have been out of school and have now returned of their own accord, bringing with them new insights about themselves and the 86 learning situation. The attributions for failure to ability may be a more accurate assessment of specific weaknesses they perceive in themselves. The attribution of success to effort may reflect their greater understanding that they must work harder to achieve in a learning situation than a student who has had a fairly successful school experience up to this point. If students perceive they lack ability to be successful, it follows that they would view effort as being very important for success. These individuals know they must work harder than others to be successful. If community education students transfer their effort attributions to actual practice, they will enhance their chances of being successful. Community education students also rated luck as more important iri determining success and failure. This attribution does not enhance the expectations for future success. When one attributes success or failure to luck, one is placing the control of events outside of oneself. The student cannot expect to change the cause of failure or to repeat the cause of success. The community education students present a mixed picture. Their attributional pattern in some areas predicts that they can make the changes necessary to be more successful, whereas other patterns predict that altering their response style is outside of their control. 87 Hypothesis 4 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the top one-third of their class and students who have dropped out of high school. A comparison of means from the composite scores on the MMCS between students in the top one-third of their class and those who had dropped out of high school revealed a statistically significant difference of less than .05 on one scale. Students who had dropped out of high school and were in prison viewed lack of ability as more important in determining failure than did students in the top one- third of their class. ‘This greater emphasis on ability' might reflect a more realistic assessment of their own strengths and weaknesses. Prison inmates also viewed that their effort was more important in determining success. These two attributes complement each other; if one believes he lacks the ability to do well in school, then in order to be successful one must work harder. The prison inmates were more internal looking than the top one-third students. Prisoners reported that those attributes which came from or were controlled by themselves were important in determining the outcomes for themselves in academic situations. These individuals had dropped out of school and were now housed in prison. The greater number of life experiences available to them might have effectively shown them the things in which they did well and those that were problem areas for them. "n: this point, prisoners had failed at school and had been removed from society for some criminal behavior. While housed in prison, these individuals had had time to 88 reflect on those life events that had affected them in positive and negative ways. Prison inmates presented EH1 attributional pattern that could be predictive of possible future success in academic settings. If these individuals transferred this increased awareness to their decision-making process, they could expect to perform better in the future. Hypothesis 5 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the middle one-third of their class and students in the bottom one- third of their class. Students in the middle one-third of their class rated effort, ability, and context as slightly more important than did students in the bottom one-third of their class, who rated luck as slightly more important in contributing to success and failure. The direction of the attributional pattern was slightly more positive for students in the middle one-third of their class. Both groups viewed effort as the most important attribute for success and failure. If they transferred these attributions into action, they could expect to make some of the Changes necessary to continue success or remediate the causes of failure. Hypothesis 6 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the middle one-third of their class and students who have dropped out of high school and are enrolled in a community education program. 89 Students in the middle one-third of their class ranked the context of the situation as a more important contributor to failure than did community education students. This difference indicated that the middle one-third students were more likely to look at the environment as a cause for their failure, thus enabling them to make some changes in their environment that might lead to future success. This attribution would allow students to begin to exercise some control over future outcomes. The community education students rated effort as more important in determining success than did students in the middle one-third of their class. This might be because students in community education had experienced previous failure in school and realized that their failure might have been due to lack of effort. The present attribution for effort-success implies that community education students understood that effort played an important part in determining whether they would succeed in an academic setting. Community education students were more internal in their attributions than students in the middle one-third of their class. The community education students reported that the locus of control was within themselves and that the outcome of events was more dependent on factors within themselves than the middle one-third students. 'The greater emphasis (n1 effort might have been a result of the self-selecting process of who voluntarily returns to a community education program. 9O Hypothesis 7 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the middle one-third of their class and students who have dropped out of high school. Students in the middle one-third of their class did not see ability as being as important an attribution of success or failure as did the dropouts. The dropouts reported that their performance was influenced by their ability or lack of ability in determining both success and failure. The dropouts reported that their effort was important in determining success. This attribution dovetails with the ability failure attribution; if one lacks the ability, he then must work harder to achieve success. This difference may have been caused by the dropouts being older and having had more experience by which to judge their ability. The dropouts were removed from the school setting and might have been taking a more realistic look at their abilities, realizing that they had abilities in specific areas and not in others. The students in the middle one-third of their class differed significantly from the dropouts on the Context Failure and Failure scales. The high school students saw their environment as playing a more important part in their failure than did the dropouts. If the students attribute this failure to events in the environment that they can control, they might then take corrective action to change the environmental conditions that have contributed to their failure. However, if they attribute the cause of failure to uncontrollable events, they will be unable to change the environmental conditions 91 and will continue to repeat their past performance. In this study, the high school dropouts were more internal than students in the middle one-third of their class. The dropouts reported that the locus of control was more within themselves and that the outcome of events was more dependent on factors from within themselves than did students in the middle one-third of their class. Hypothesis 8 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the bottom one-third of their class and students who have dropped out of high school and are enrolled in a community education program. Students in the bottom one-third of their class reported that the context of the situation played a larger role in determining failure than did community education students. The high school students looked more to the environment to explain why they failed at a given task. If the students attribute their failure to events in the environment that they can control, they may then take corrective action to change the environmental conditions that have contributed to their failure. If, however, they attribute the cause of failure to uncontrollable events, they will be unable to change the cause of failure and will continue to repeat their past performance. Students in community education attributed their success and failure more to their own ability and effort than did students in the bottom one-third of their class. Community education students attributed the cause of their success and failure more internally 92 than did students in the bottom one-third of their class. These students reported that they were responsible for both their- successes and failures; the causal factors came from within themselves. The community education students reported a more positive attributional pattern than did the students in the bottom one-third of their class. The community education student is in the position to make some of the necessary changes that are needed to become or continue\to be successful in academic settings. These data for the community education student may be shifted in the positive direction on account of the self-selecting process of who has returned to school on their own. Hypothesis 9 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students in the bottom one-third of their class and students who have dropped out of high school. High school dropouts who were housed in prison attributed their success and failure more to effort and ability. They saw that their success was more dependent on themselves than did students in the bottom one-third of their class. High school dropouts were more internal looking than bottom one-third students. The high school dropouts reported that the locus of control was more internal, and that the outcome of events was more dependent on factors from within themselves. This may be because the dropouts were older and had experienced some significant problems in the past. This increased opportunity to assess one’s strengths and weaknesses may have 93 accounted for the greater emphasis on personal responsibility for their success and failure attributions. Students in the bottom one-third of their class looked more to the context of the situation and environmental factors to explain failure. This difference might have been because the prison inmates lived in a: very controlled environment and could exert little control over it; thus, they reported that the environment was not as important a factor as high school students said it was. Hypothesis 10 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between students who have dropped out of high school and enrolled in community education programs, and students who have dropped out of high school. Community education students were more likely to report that the context of an event was more responsible for failure than were prison inmates. This may have been due, in large part, to the prisoners living in a very controlled environment on which they could have little effect. This view of not being able to affect the environment and its not playing a role in attributing to either success or failure was reflected in their neutral score of 4.0. Both groups showed a positive attributional style. Effort was the most important factor in determining success and failure, followed by ability, context, and luck. Both groups reported an attributional pattern that should allow them to continue those activities that lead to success in academic settings and to make those changes that would improve their chances for future success. 94 The piece of information that needs to be explored is whether these individuals were putting into action and effect the responses that would be expected from their attributional pattern. Hypothesis 11 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between male and female students. Males were more likely to attribute their failure to the environment and context of the situation than were females. They tended to look outside of themselves to explain why they did not perform well. Females were more likely to view lack of ability as a causal factor for failure. The reason they did not perform well was because they lacked ability. These two responses are consistent with earlier research, which has reported that males are more likely to attribute success to ability and failure to external factors, whereas females are more likely to view failure as due to internal factors. Hypothesis 12 There is no statistically significant difference in MMCS scores for ability, effort, context, and luck between white and black students. The reported differences between black and white students were significant. The attributional patterns reported by the black students emphasized effort and ability as more important than those of white students. These results are contrary to earlier attributional research, which found blacks to be more external in their attributions for success and failure. The pattern of 95 attributional choices shown here predicts that black students would expect to continue to experience success in the future. They attributed their present success to the internal stable factor of ability, to the internal unstable factor of effort--they can work hard again and do well, and their lack of success to effort-~they didn’t try hard enough but that can change in the future. The attribution of failure to ability is not consistent with the other attributions made by typical standards. However, it may be viewed that the attribution of failure to ability is a more realistic assessment of one’s failure in a particular subject area. Then the pattern would remain consistent. White students in this study showed an attributional pattern that was not as positive as that reported by black students. White students still showed an attributional pattern that emphasized ability and effort as being the most important components in their academic success or failure. For both groups of students, hard work was the most important factor in success, and lack of effort was the largest contributor to their failure. Effort was followed by ability, context, and luck as reasons given for academic success or failure. This pattern would be predictive of a student who would expect to experience success in the future. Success was attributed to an internal stable cause--ability, and to an internal unstable cause--effort, whereas failure was attributed to internal unstable causes such as lack of effort and external causes such as poor environment. 96 Black and white students differed to a significant extent in this study. Still, both reported a fairly positive attributional pattern. The overall positive pattern, however, must be viewed in light of the larger number of academically successful students completing the survey. This situation might have shifted the results in a more positive direction. Research Question 1 Which of the attributional categories of ability, effort, context, and luck is viewed as the most important determinant of academic success and failure in achievement settings, as reported by race, gender, and class standings? Each of the subject groups (male, female, black, white, and class standing) reported that effort was the most important factor ir1 determining success or failure. High school dropouts rated effort more important than did any other school group. This greater emphasis on effort may have been a more realistic evaluation of their own strengths and weaknesses. Dropouts may have viewed themselves as less able on specific school-related tasks and hence thought they must work harder to be successful. The students who were still in school showed a natural progression from the top to the bottom in their ratings of the importance of effort; the more successful students rated effort as more important than did those who were not performing as well. This decreased emphasis on effort as students performed less well followed the predicted pattern, with students taking less and less responsibility for their own success or failure as their academic success decreased. Students who did not view effort as important in success or failure gave up an aspect 97 of control over the future. If they did not believe that hard work would influence future outcomes, they would not be likely to change poor study or work habits that had led to the present failure, or they might not repeat the effort that had led to success. Females rated effort higher than males, but the differences were not significant. Blacks rated effort more important than did whites and other minorities. There was a significant difference between black and white respondents, and between black and other minority respondents. Research Question 2 00 students as they progress through grades 10, 11, and 12 change their attributional pattern? The data indicated that the attributional patterns of students were fixed by the time they entered the tenth grade and did not change during the next three years. There were no significant differences on any of the attributional scales among students in any of the grades. The consistency of the responses among grades 10, 11, and 12 reveals that, by high school, a student’s attributional pattern is well established. This indicates that, without intervention, students will continue to make the same attributions for success and failure in the future as they presently make. For students making positive attributional responses, this enhances their chance for continued success and provides the framework within which to mitigate the effects of failure; their future looks bright. Students with an attributional pattern that hinders them from being 98 successful in school have much to worry about. The data suggest that their attributional patterns will remain relatively constant over time, and they can expect to continue to fail and be unable to make the changes necessary to become more successful students. Students in each grade viewed effort as the most important component in determining success or failure, followed in order by ability, context, and luck. The data suggest that, if students have an attributional pattern that inhibits them from taking action to become more successful, an intervention must take place at an earlier date. 'The students must be taught how their negative attributional pattern is affecting their performance and how to change their attributions to be more productive. Research Question 3 What effect does the income level of the student’s family have on the student’s attributional pattern? Examining the four attributional scales for the effects of income revealed that the only scale affected was the Effort scale. Looking more closely at the data for the Effort scale’s two component parts showed that income did not affect the attributions for effort success but did contribute to the differences on the Effort Failure scale. Research Question 4 00 students who experience greater academic success, as measured by GPA, attribute success and failure in a more productive pattern than students who are not as successful in achievement settings? 99 Effort was reported as the most important determinant of success or failure. High school dropouts did not differ on the Effort Failure scale from any of the in-school groups, and there were no significant differences between any of the groups on the Effort Failure scale. Effort success was highly significant for the high school dropouts when compared to the other four groups. This finding indicates that, for the dropouts, effort played an important role 'hi attaining success. Students in school rated effort important, with community education students giving it the highest rating of the four, followed by the top one-third students, then the low one-third and the middle one-third, with virtually no differences between them. The high school dropouts had come to view effort as an important component to be successful in academic settings. The question, however, is whether the conception that effort is important for success had led to the individual taking action in that direction. Ability was the second most important determinant of academic success or failure. Again, the high school dropouts rated ability the highest, with the community education students second, followed by the top one—third, the middle one-third, and the bottom one-third students. This pattern of the high school dropouts rating ability more important than in-school students may be explained if one finds that the dropouts were making a much more realistic judgment of their actual abilities. After spending time out of school, these individuals had the opportunity to more objectively reflect on and draw conclusions about why they did well or poorly in school. 100 Closer examination of the data showed that high school dropouts viewed ability as a key determinant in contributing to both their success and their failure. The attributions for ability success were ordered as follows: dropouts, top one-third students, community education, middle one-third, and bottom one-third. The Ability Failure scale revealed a different pattern, with the scores arranged as follows: dropouts, community education, bottom one- third, middle one-third, and top one-third. The Ability Failure scale followed the predicted pattern, with students who had experienced more success in school believing that ability played less of a role in their failure than did students who had experienced little success in school. The dropouts identified that their ability in specific situations would help them be successful or contribute to the learning task being more difficult and failure more likely. The pattern of the more successful students viewing ability as more important than the poorer students also followed the predicted pattern of responses. Students who have experienced success are more likely to believe that they can do the task required of them and will expect to do well in the future. Students who have experienced failure or poor results tend to view themselves as less able and expect to see the same poor performance in the future. Ability is an internal stable factor that is unlikely to change over time. The attribution of their success or failure to ability predicts future performance to repeat that of the past. This pattern of attribution bodes well 101 for the successful student. However, for the poorer student, the future in school is viewed as holding more of the same frustration and experiencing continued failure because their ability is unlikely to change. The context of the academic experience decreased in importance for determining success or failure. The sequence of importance follows: middle one-third students, top one-third, bottom one- third, community education students, and high school dropouts. The context variable is complex and needs to be analyzed in two parts: context for success and context for failure. The data indicated virtually no difference among the four groups on the Context Success scale. The groups believed that the context of an event contributed little to their success. The attribution of context for failure showed that students in the middle and top one-third of their class viewed this as a more important determinant of failure than did students in the bottom one-third of their class, community education students, or high school dropouts. High school dropouts viewed the context of the situation as having the least effect on success. The score indicates that, the more difficulty a student had in an academic setting, the less likely he was to view the setting as important in determining failure. This attribution does not encourage these students to look at their environment to see whether it may be contributing to their failure. This precludes them from making the structural changes that may be necessary to become successful. Students probably will not change a poor work or study environment to a more positive one if their failure is not seen as 102 attributable to the environment. By attributing failure to the context of the event, individuals can hope for a different set of circumstances the next time around, which will enable them 'to be successful. This attribution also allows students to protect their self-image as good students. They were not responsible for the failure. It was the environment, and it is possible to change that in the future. Least important in determining success or failure in this study was the attribution of luck. The use of luck as an attribution plays an important part in future expectations for success or failure to be repeated. Luck is an external unstable factor; if students attribute success to luck, they cannot count on being successful in the future. If 'failure is attributed to luck, students will be able to view the future as changeable; they may be able to influence future events. The sequence of attribution for luck as important was as follows: community education students, bottom one-third, middle one-third, dropouts, and top one-third. These findings indicate that, the more successful students were in school, the less likely they were to use luck to explain success or failure (except for the dropouts, which may have been the result of a smaller sample size). This pattern indicates that successful students believed they had a more active role in academic performance and could alter future outcomes. Students who had not been successful in school believed that luck was the cause of an 103 outcome and gave up the idea of changes affecting the event in the future. Conclusions In this section, the conclusions drawn from the findings regarding the hypotheses and research questions are presented. Hypotheses 1 through 10 and Research Question 4 explored the relationship of high school students, by class standing, and high school dropouts. The data showed that, for students in school, the more successful the student was in achievement situations, the more productive were his attributional patterns. Conversely, the less successful the student was, the more counterproductive was his attributional pattern. High school dropouts who had' enrolled in community education had begun to make a more positive attributional pattern for school success. Dropouts presently in prison demonstrated an even further shift in making attributions that would help them become more successful in academic settings. Both groups of dropouts appeared to have made a realistic assessment of their strengths and weaknesses and reflected this in their patterns of attributions. Students in each group still need to alter their attributions for success and failure if they wish to improve their chances for being successful in academic settings. The attributional pattern that one expresses does not, however, predict success or failure. 0ne’s attributional pattern gives an indication of whether one will take the steps necessary to enhance 104 the~ possibility for future success and limit the potential for failure. The individual must still act. Regarding Hypothesis 11, 1") difference 'hi attributional patterns was found between males and females. However, examining the data from the subscales indicated that males were more likely to attribute failure to the context of the situation, whereas females attributed failure to lability. ‘This attribution of failure by females to ability could be a detriment to future success. Why should one try if she is not able? The attribution of failure by males to the context of the event could have a positive effect on future expectations for success; after all, the academic environment can be altered, removing the causes of past failures. The findings regarding Hypothesis 12 revealed that there was a significant difference between white and black students. In this study, the black students were found to make much more constructive and positive attributional causal links than the white students. These attributional patterns should be helpful for the black students in future academic settings. They viewed effort as an important component of success and lack of effort as a contributor to failure. Even though the white students were significantly lower than their black counterparts, the attributional pattern for them also exhibited the necessary components for future success. High-school-age students reported that effort was the most important determinant of academic success or failure, followed by ability. The generality of this belief among the population is significant. It indicates that the students had developed a 105 positive work attitude. They believed that the harder one works the better one will do. The majority of the students were taking a proactive role in determining their academic success. The causal links between effort and success had been formed. Ability was rated as the second most important determinant of success. Students were making realistic assessments of the components that go into producing a successful outcome. If students can and do make realistic assessments of specific strengths and weaknesses, one would then expect the individuals to develop realistic goals for their performance and allocate their effort to those areas that require the greatest amount of work. The consistency of the attributional patterns from the tenth grade to the twelfth grade indicates that students had developed a consistent attributional pattern by that time. Individuals who have developed a positive productive pattern by this time will continue to benefit from it. However, students with a negative, counterproductive pattern will continue to be hindered by it in the future. The long-term outlook is for individuals to maintain their present attributional style unless some significant intervention or experience takes place. The income level of the students themselves or their family had little effect on attributional style. ‘The only area affected was the attribution to effort/failure. Here it was found that income caused part of the variation in responses between low-income and high-income students by class standing; it did not affect 106 students in the nfiddle income levels. The tenuous nature (H: the data provided and the slight differences reported cause one to view these findings with reservation. Recommendgtions Overall, it was found that students did attribute success and failure differently, depending on their success in school. These differences can be used to identify students who have the mental ability to perform well in school but may be experiencing academic difficulty. If the student has a counterproductive attributional style, one that attributes success to luck. and the context and failure to ability, he may be a candidate for attributional retraining, where one learns to attribute success to effort and ability and failure to lack. of 'effort and the context of the learning situation. In school, students often experience success and failure. By using attributional theory, educators can develop programs to maximize the effect of the success by helping the student realistically identify the causes behind such success: ability and effort. Failure can also be used to help the student identify its causal link, such as lack of effort or poor context. However, not all students have the ability to do well in all areas of school. If lack of ability is the reason for failure, attribution theory would enable the student to make specific failure causal links that are subject or situation specific, allowing for success in all other areas . 107 Suggestions for Further Research In this study the researcher analyzed the attributional styles of high school students and high school dropouts. The study was intended to serve as preliminary research in unraveling the relationship between academic success and attributional patterns. The following suggestions are offered as areas in which further research could be conducted. 1. Because attributional patterns did not change between grades 10 and 12, a longitudinal study could provide a more detailed account of the acquisition and maintenance of attributional patterns. 2. Western Michigan is a small socioeconomic environment. A replication of this study in an area with a more diverse population might provide different but comparable conclusions. 3. In this study, a significant difference was found between black and white students, contrary to past findings. This set of students warrants further study to find those factors that have set them apart. 4. A study might be conducted in which individuals with coun- terproductive attributional styles undergo attributional retraining to ascertain the effects retraining has on school performance. APPENDICES APPENDIX A THE MULTIDIMENSIONAL-MULTIATTRIBUTIONAL CAUSALITY SCALE AND THE BIOGRAPHICAL DATA SHEET 108 The statements in this booklet express opinions on a number of issues. You may find yourself agreeing strongly with some of the statements, disagreeing just as strongly with others and perhaps feeling uncertain about others. Your reactions reflect your own opinions and obviously in a belief survey there are no right or wrong answers. Whether you agree or disagree with any statement you can be sure that there are many people who feel the same way you do. Indicate on the test booklet how much you agree or disagree with each statement by circling one of the letters in the following way. A = I AGREE B = I MILDLY AGREE C = I AGREE AND DISAGREE EQUALLY D = I MILDLY DISAGREE E = I DISAGREE If you have mixed feelings about an item, decide whether you lean slightly more to one side than the other. Try to avoid the middle response. Remember that the best answer is whatever your personal opinion is and this is usually best reflected in your first reaction to an item, rather than after long debates with yourself over a particular experience. Before you begin, please fill in the information requested on the top of the separate data sheet. 109 BIOGRAPHICAL DATA SHEET Name (optional) Age: Years month(s) Last grade completed: Sex: Male_______ Female Race: Black_______ Asian Hispanic Native American White (Anglo)_______ 0ther_______ Parental Income (if living at home); own income if living independently: Under $10,000_______ $10,001 to $20,000_______ $20,001 to $30,000_______ $30,001 to $40,000_______ $40,001 to $50,000_______ Above $50,000_______ Grade point average, if known: 110 MULTIDIMENSIONAL-MULTIATTRIBUTIONAL CAUSALITY SCALE When I receive a poor grade, I usually feel that the main reason is that I haven’t studied enough for that course. A B C D E AGREE DISAGREE If I were to receive low marks it would cause me to question my academic ability. A B C D E AGREE DISAGREE Some of the times that I have gotten a good grade in a course, it was due to the teacher’s easy grading scheme. A 8 C D E AGREE DISAGREE Sometimes my success on exams depends on some luck. A B C D E AGREE DISAGREE In my case, the good grades I receive are always the direct result of my efforts. A B C D E AGREE DISAGREE The most important ingredient in getting good grades is my academic ability. A B C D E AGREE DISAGREE In my experience, once a teacher gets the idea you’re a poor student, your work is much more likely to receive poor grades than if someone else handed it in. A B C D E AGREE DISAGREE 8. 9. 10. ll. 12. 13. 14. 111 Some of my lower grades have seemed to be partially due to bad breaks. A B C D E AGREE DISAGREE When I fail to do as well as expected in school, it is often due to a lack of effort on my part. A B C D E AGREE DISAGREE If I were to fail a course it would probably be because I lacked skill in that area. A B C D E AGREE DISAGREE Some of my good grades may simply reflect that these were easier courses than most. A B C D E AGREE DISAGREE I feel that some of my good grades depend to a considerable extent on chance factors, such as having the right questions show up on the exam. A B C D E AGREE DISAGREE Whenever I receive good grades it is always because I have studied hard for that course. A B C D E AGREE DISAGREE I feel that my good grades reflect directly on my academic ability. A B C D E AGREE DISAGREE 15. 16. 17. 18. 19. 20. 21. 112 Often my poorer grades are obtained in courses that the teacher has failed to make interesting. A B C D E AGREE DISAGREE My academic low points sometimes make me think I was just unlucky. A B C D E AGREE DISAGREE Poor grades inform me that I haven’t worked hard enough. A B C D E AGREE DISAGREE If I were to get poor grades I would assume that I lacked ability to succeed in those courses. A B C D E AGREE DISAGREE Sometimes I get good grades only because the course material was easy to learn. A B C D E AGREE DISAGREE Sometimes I feel that I have to consider myself lucky for the good grades I get. A B C D E AGREE DISAGREE I can overcome all obstacles in the path of academic success if I work hard enough. A B C D E AGREE DISAGREE 22. 23. 24. 113 When I get good grades, it is because of my academic competence. A B C D E AGREE DISAGREE Some low grades I’ve received seem to me to reflect the fact that some teachers are just stingy with marks. A B C D E AGREE DISAGREE Some of my bad grades may have been a function of bad luck, being in the wrong course at the wrong time. A B C D E AGREE DISAGREE APPENDIX B PERMISSION LETTERS 114 Dear Parent: I am writing to ask permission for your child to participate in a research project directed toward finding out more about how children see themselves as learners. The purpose of the study, being con- ducted to fulfill the research requirement for the completion of my doctoral degree, is to learn more about the attributions children make for their performance, and how this relates to their successes and failures in achievement settings. All students’ participation is entirely voluntary and anonymous, and students will be able to withdraw at any time if they feel it is necessary. The administra- tion has reviewed and approved this research project. Previous research has indicated that a child’s interpretation of the reason for past performance, in a large part, determines his or her reaction to a future task. For example, if a student believes he did well because he studied, it is probable that he will study before his next test. If he believes he did well because it was an easy test, he is less likely to put forth good effort next time. It is my hope that the findings of this study will contribute to our understanding of how to motivate students, and how to help them realize that they can influence their level of achievement. Upon its completion, the results of this study will be made available to you. I plan to hold several meetings to share the results and discuss the implications of the findings with you. In order to increase the validity of the results, the teachers have very little information about the study at this time and probably will not be able to answer your questions. Please call me and leave a message at 457-1402 ex. 250 if you have any questions or concerns. Thank you. Larry Balkema My son/daughter, , has permission to participate in the research project being conducted by Mr. Balkema. 115 To, LARRY BALKEMA one. 5/12/89 MICHIGAN DEPARTMENT OF FROM, warden Martin Makel, cognac-nous Dunes Correctional Facility SUBJECT: RESEARCH REQUEST I have received approval from the Research Division of the Corrections Department Program Bureau of your proposal to gather information from state prisoners. The approval, which is attached, outlines the guidelines to which we must adhere. I have also contacted warden H. Gary wells at the Muskegon Correctional Facility; and he has given approval for you to meet with their School Principal, Bruce Corim, for gathering further data. You are to contact Mr. Corim at 616/773-3201 to establish a date and time. The Department of Corrections and this facility are always interested in assisting professional research. Good luck in your project. cc: warden H. Gary wells, MCF file gaaapwogag 116 MICHIGAN DEPARTMENT OF CORRECTIONS RESEARCH DIVISION NIB MIDItiLN'D U M TO: Warden Martin Makel DATE: May 5, 1989 Dunes Correctional Facility . _: . r- FROM: Terry Murphgb Chief Research Division SUBJECT: Mr. Balkema’s Proposal We have reviewed Mr. Balkema’s proposal to conduct research at your institution. We further understand that the proposed research is being conducted for purposes of completing an academic degree, and hereby grant approval for him to proceed, subject to the following conditions: 1. It is our understanding that no analysis of the unique characteristics of prisoners or the effectiveness of correctional programs will be made. If it is Mr. Balkema's intention to perform such analyses, he must advise the Department of that fact, at which time a more extensive review of the proposed research will be undertaken prior to approval. 2. All study subjects must be advised that participation is voluntary, and must Sign a consent form. 3. Mr. Balkema must submit a statement releasing the Department of Corrections from any liability occurring during face to face contact with prisoners or resulting from the nature of the questionnaire. 4. Mr. Balkema must submit a statement guaranteeing the anonymity of all study subjects. The statement should indicate what measures will be taken to ensure that there is no reasonable possibility of responses and results being attributed to any individual or specific group of study subjects. 5. The warden of each institution at which Mr. Balkema wishes to conduct research must submit a letter giving their consent, and specifying that any 117 staff involvement, e.g. providing security during interviews, will not interfere with normal job duties and will not jeopardize the routine and safe operations of the institution. Please note that all of the above conditions before actual research begins. Finally, the Corrections, through the Bureau of Programs, right to review the final report for factual to submit written responses or corrections, must be met Department of reserves the accuracy, and as appropriate. If you or Mr. Balkema have any further questions, please feel free to contact me. I wish Mr. Balkema well in his future academic endeavors. THM:sh BIBLIOGRAPHY BIBLIOGRAPHY Abramson, L. Y., Garber, J., & Seligman, M. E. P. (1980). Learned helplessness in humans: An attributional analysis. In J. Gamber & M. E. P. Seligman (Eds.), Human helplessness: Theory and applications. New York: Academic Press. Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psycholoqv, a1, 49-74. Alvarez, C. M., & Pader, 0. F. (l978). Locus of control among Anglo-Americans and Cuban-Americans. Journal of Social Psychology, lOS, 195-198. Andrews, G. R., & Debus, R. L. (l978). Persistence and the causal perception of failure: Modifying cognitive attributions. Journal of Educational Psychology. 19(2), 154-166. Avonson, E., & Mater, D. (l968). Dishonest behavior as a function of differential levels of induced self-esteem. Journal of Personalityaand Social Psycholoqx, 2 (2, pt. l), lZl-l27. Bat-Tal, D. (1978, Spring). Attributional analysis of achievement- related behavior. Review of Educational Research, 3a, 259- 27l. Camp, C. (l980). School dropouts. National lonqitudinal studies handbook. Columbus: The Ohio State University, Center for Human Resource Research. Cole, 0., & Cole, S. (l974). Locus and cultural conformity: On going against the norm. Personality and Social Psychology Bulletin, 1, 351-353. Cole, 0., & Cole, S. (l977). Counternormative behavior and locus of control. Journal of Social Psychology, 101, 21-28. Coleman, J. S., Campbell, E. R., Hoabson, C. J., McPartland, J., Weinfield, F. D., & York, R. L. (l966). Egualitv of educa- tional opportunity. Washington, DC: U.S. Government Printing Office. 118 119 Covington, M. V., & Omelich, C. L. (l979). It’s better to be able and virtuous too: Student and teacher evaluative responses to successful effort. Journal of Educational Psychology, 11(5), 688-700. Crandall, V., Katkowsky, M., & Crandall, V. J. (1965). Children’s beliefs in their own control of reinforcement in intellectual academic achievement situations. Child Devel0pment, aa, 91- l09. Diener, C. I., & Dweck, C. K. (1978). An analysis of learned help- lessness: Continuous changes in performance, strategy and achievement cognitions following failure. Journal of Person- ality and Social Psychology, aa, 45l-462. Dweck, C. S. (l978). Sex differences in learned helplessness: II. The contingencies and evaluative feedback in the classroom. Developmental Psychology, 15, 268-276. Dweck, C. S. (l978). Sex differences in learned helplessness: III. An experimental analysis. Developmental PsychologY, 15, 268-276. Dweck, C. S., & Bush, E. S. (1976). Sex differences in learned helplessness: 1. Differential debilitation with peer and adult evaluators. Developmental Psychology, 121, l47-l56. Dweck, C. 5., & Goetz, T. (1978). Attributions and learned helplessness. In H. Harvey, M. Ickes, & R. Lidd (Eds.), flaw directions in attribution research (vol. 2). Hillsdale, NJ: Laurence Erlbaum. Dweck, C., & Lichft, B. (l980). Learned helplessness and intel- lectual achievement. In J. Garber & M. E. P. Seligman (Eds.), Human helplessness: Theory_and applications. New York: Aca- demic Press. Dweck, C. S., & Reppucci, D. (l973). Learned helplessness and reinforcement responsibility in children. Journal of Person- ality aad Social Psychologv, 25(1), 109-116. Ekstrom, R. B., Goetz, M. E., Pollack, J. M., & Rock, 0. A. (l986). Who drops out of high school and why? Findings from a national study. Teachers College Record, a1. Elig, T., & Frieze, I. (1979). Measuring causal attributions for success and failure. Journal of Personality and Social Psychology. §Z(4), 621-634. 120 Feather, N. (1969). Attribution of responsibility and valence of success and failure in relation to initial confidence and task performance. Journal of Social Psychology, 13, 129-144. Fitch, G. (1970). Effects of self-esteem, perceived performance, and choice on causal attributions. Journal of Personality and Social Psychology, lé, 311-315. Fontaine, G. K. (1974). Social comparison and some determinants of expected personal control and expected performance in a novel task situation. Journal of Personality and Social Psychology, ag(4), 487-497. Friend, R., & Nefale, J. (1972). Children’s perceptions of success and failure: An attributional analysis of the effects of race and class. Develonmental Psychology, 1, 124-128. Frieze, 1., & Weiner, B. (1971). The utilization and attributional judgments for success and failure. Journal of Personality, 39, 591-606. Frieze, I. H., Whitley, B. E., Hanusa, B. H., & McHugh, M. C. (1982). Assessing the theoretical models of sex differences in causal attributions for success and failure. Sex Roles: A Journal of Research, 8(4), 333-345. Garza, R. T., & Ames, R. E. (1974). A comparison of Anglo and Mexican American college students on locus of control. Journal of Consulting and Clinical Psychology, AZ, 919. Gilmore, T., & Reid, D. (1979). Locus of control and causal attribution for positive and negative outcomes on university examinations. Journal of Resaarch in Personality, 13, 154-160. Good, C. V. (1973). Dictionary of education (3rd ed.). New York: McGraw-Hill. Graham, S. (1984). Communicating sympathy and anger to black and white children: The cognitive (attributional) consequences of affective cues. Journalaof Personality and SocialaPsychologv, 11, 40-54. Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley. Hiroto, D. S., & Seligman, M. E. P. (1975). Generality of learned helplessness in man. Journallof Personality and Social Psy- chology, 31, 311-327. 121 Ickes, W., & Layden, M. (1978). Attributional styles. In J. Harvey & R. Kidd (Eds.), New directions in attribution research (vol. 2). Hillsdale, NJ: Erlbaum Press. Johnson, 8., & Kanoy, K. (1980). Locus of control and self-concept in achieving and underachieving bright elementary students. Psychology in the Schools, 1(13), 395-399. Kagan, S. (1976). Preference for control in rural Mexican and urban Anglo-American children. Revista Interamericana de Psicologia, 19, 51-59. Kelley, H. H. (1971). Attribution in social interaction. New York: General Learning Press. Kelley, H. H. (1976). Attribution theory in social psychology. In D. Levine (Ed.), Nebraska Symposium of Motivation. Lincoln: University of Nebraska Press. Kukla, A. (1972). Attributional determinants of achievement related behavior. Journal of Personalityaand Social Psychology, 21(2), 166-174. Lefcourt, H. M. (1976). Locus of control: Current trends in theory and research. Hillsdale, NJ: Erlbaum Press. Lefcourt, H., VonBaeyer, C., Ware, E., & Cox, D. (1979). The Multidimensional-Multiattributional Causality Scale: The development of a goal specific locus of control scale. Canadian Journal of Behavior Science, 11(4), 286-303. Maurer, R. E. (1982). Dropout prevention: An intervention model for today’s high schools. Phi Delta Kappan, aa, 470-471. McMahan, J. D. (1971). Sex differences in causal attributions following,success and failure. Paper presented at the meeting of the Eastern Psychological Association, New York, NY. McMahan, J. D. (1973). Relationship between causal attributions and expectancy of success. Journal of Personality and Social Psychology, 28(1), 108-114. Nichols, J. G. (1975). Causal attributions and other achievement related cognitions: Effects of task outcome, attainment value and sex. Journal of Personality and Social Psychology, §l(3), 379-389. Parsons, J. E., Meice, J. L., Adler, T. F., & Kaczala, C. M. (1982). Sex differences in attributions and learned help- lessness. Sex Roles: A Journal of Research, 8(4), 421-432. 122 Pedhazur, I., & Wheeler, L. (1971). Locus of perceived control and need achievement. Perceptual and Motor Skills, 33, 1281-1282. Powers, 5., Douglas, P., & Choroszy, M. (1983). The factorial validity of the Multidimensional-Multiattributional Causality Scale. Educational and Psychological Measurement, 13, 611-615. Powers, 5., & Rossman, M. H. (1984). Attributions for success and failure among Anglo, Black, Hispanic, and Native American community college students. Journal of Psychology, 117, 27-31. Prawat, R. S., Byers, J. L., & Anderson, A. H. (1983, Spring). An attributional analysis of teachers’ affective reactions to student success and failure. American Educational Research Journal, 39, 137-152. Roth, 5., & Bootzin, R. (1974). Effects of experimentally-induced expectations of external control: An investigation of learned helplessness. Journal of Personality and Social Psychology, 23, 253-264. Seligman, M. E. P., & Maier, S. F. (1967). Failure to escape traumatic shock. Journal of Experimental Psychology, 14, 1-9. Seligman, M. E. P., Maier, S. F., & Solomon, R. L. (1971). Unpredictable and uncontrollable aversive events. In F. R. Brash (Ed.), Aversive conditioning and laarning. New York: Academic Press. Seligman, M. E. P., & Overmier, J. B. (1976). Effects of inescap- able shock upon subsequent escape and avoidance learning. Journal of Comparative and Physiological Psychology, 63, 28-33. Silmon, J. G., & Feather, N. T. (1973). Causal attributions for success and failure at university examinations. Journa1,of Educational Psychology, §fl(1), 46-56. Sohn, D. (1982). Sex differences in achievement, self- attributions: An effect-size analysis. Sex Roles: A Journal of Research, 3(4), 345-350. Thornton, J., & Jacobs, P. (1971). Learned helplessness in human subjects. Journal of Experimental Psychology, 3, 367-372. Valle, V. A., & Frieze, I. H. (1976). The stability of causal attributions as a mediator in changing expectations for success. Journal of Personalityaand Social PsychologY, 33, 579-587. 123 Weiner, 8. (1972). Attribution theory, achievement motivation and the educational process. Review of Educational Research, 1;, 203-215. Weiner, B. (1972). Theories of motivation: Fron:mechanisn to cognition. Chicago: Rand-McNally. Weiner, B. (1974). Achievement motivation and attribution theory. Morristown, NJ: General Learning Press. Weiner, 8. (1979). A theory of motivation for some classroom experiences. Journal of Educational Psychology, 11(1), 3-25. Weiner, 8. (1980). Human motivation. New York: Holt, Rinehart and Winston. Weiner, 8., Frieze, I., Kukla, A., Reed, L., Rest, S., & Rosenbaum, R. M. (1971). Perceiving the causes of successgand failure. New York: General Learning Press. Weiner, 8., Russell, D., & Lerman, D. (1979). The cognition emotion process in achievement-related contexts. Journal of Personality and Social Psychology, 31(7), 1211-1220. Willing, A. C., Harnisch, D. L., Hill, K. T., & Maeher, M. L. (1983, Fall). Sociocultural and educational correlates of success-failure attributions and evaluation anxiety in the school setting for Black, Hispanic, and Anglo children. American Educational Research Journal, an, 385-410. Wong, P. T. (1982). Sex differences in performance attributions and contingency judgment. Sex Roles: A Journal of Research, 9(8), 381-388. Wong, P. T., & Weiner, 8. (1981). When people ask questions and the heuristics of the attributional search. Journalgpf Personalitynand Social Psychology, 19, 650-663. ”11111111111111.1111