THE EFFECT OF SYSTEMATIC STORY CHANGES 0N INTENTIONAUTY IN CHTLDREN’S MORAL JUDGMENT S Thesis for the Degree of M. A. MICHIGAN STATE UNIVERSITY DANIEL. CHDDA GUTKIN 1970 ' {vi-Whflflm J .::-= :2» WW ”W WWW/[fl T 151131541; Y 2093 001 . vegan Eitata Umvermty This is to certify that the thesis entitled An Inquiry Into The Development Of Moral Intentionality presented by Daniel Choda Gutkin has been accepted towards fulfillment of the requirements for _Eh*_11.__degree in Mining; T 14/“; ”’6? Major professor/ -‘_._._._., amomc IV "053 & SUNS l 300K BINDERY INC LIBHIRY BIND .3: MM 3W" ”CF-.7- ~- ABSTRACT THE EFFECT OF SYSTEMATIC STORY CHANGES ON INTENTIONALITY IN CHILDREN'S MORAL JUDGMENTS BY Daniel Choda Gutkin Many investigators starting with Piaget have found that children's moral judgments progress from an early reliance on amount of damage to a later reliance on inten- tions (Piaget, 1965; Boehm and Nass, 1962; Bandura and McDonald, 1963; Cowan et al., 1969). The present study is an investigation of systematic changes in stories on children's use of intentionality in moral judgments. Story-type A is the type used by Piaget in his original investigation, in which good intentions go with more serious damage and bad intentions go with little damage. Five additional types of stories were used. In Story-Type B damage was equal (and low) while intentions varied. In Type C damage was equal (and high) while in- tentions varied. In Type D intentions were equal (and l Daniel Choda Gutkin good) while damage varied. In Type E intentions were equal (and bad) while damage varied. In Type F, good intentions and little damage were united and opposed to bad intentions and high damage. It was predicted that: 1) Types A and F would produce an equal number of inten- tional (mature) judgments; 2) Types B and C would produce more intentional judgments than A or F; 3) Type B would produce more intentional judgments than Type C; 4) Types D and E would produce fewer intentional judgments than A or F; 5) Type D would produce fewer intentional judgments than E. Sixty fourth-graders (30 boys) were group tested, writing their judgments after reading two stories of each type. The results showed an overall effect for story- type. Types D and E produced significantly fewer inten- tional judgments than all other types, as predicted. No other comparisons were significant. These results are interpreted in terms of Breznitz and Kugelmass' (1967) crude—refined story dimension. To make an intentional judgment in Types D and E a more re- fined use of the principle of intentionality is needed 2 Daniel Choda Gutkin than in the other types. The failure to find significant differences between B and C on the one hand and A and F on the other may have been due to a ceiling effect. Fi- nally, the results pose some interesting questions for a social learning and imitation theory of moral judgments. (Bandura and McDonald, 1963; Cowen et al., 1969). xi/ 5 I ..- 63‘. [QC-(.1. LL, ”' ‘ ‘r . ‘1» ..... . A.» ..,. 1'1" ,..},1( . - I -( 4,\_‘ . Approved: Ellen Strommen l7 I‘e»' xvegx Date : / 4/ fur/.7 . THE EFFECT OF SYSTEMATIC STORY CHANGES ON INTENTIONALITY IN CHILDREN'S \ MORAL JUDGMENTS BY Daniel Choda Gutkin A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1970 To the memory of my father. To the memory of my father. ACKNOWLEDGEMENTS. I would like to thank the members of my committee, Dr. Ellen Strommen, Dr. Ralph Levine, and Dr. Lucy R. Ferguson for their assistance. Dr. Strommen in particular deserves my gratitude for her help in planning the re- search, and for offering me advice and encouragement throughout the project. I would also like to thank the children, teachers, and staff of the Dimondale and Wilcox elementary schools of the Holt, Michigan school system for their help in data collection, and M. Christine Falvey who helped score the data. iii LIST OF TABLES . . . INTRODUCTION . . . . METHOD RESULTS. . . . . . . DISCUSSION . . . . . TABLE OF CONTENTS TEST ITEMS TO GROUPS. QUESTION ORDERS TO TEST COMPLETE DESIGN FOR A GIVEN GROUP . REFERENCES . . . . . APPENDIX A TEST ITEMS. . B ASSIGNMENT OF C ASSIGNMENT OF D E SCORING RULES iv Page 17 33 38 45 47 58 59 60 62 LIST OF TABLES Table Page 1. MEANS AND STANDARD DEVIATIONS FOR STORY- TYPES. . . . . . . . . . . . . . . . . . . . 35 2. PREDICTED VERSUS OBTAINED ORDER OF STORY- TYPE MANS O O O O O O O C O O 0 C O O O O O 3 5 3. SUMMARY OF ANALYSIS OF VARIANCE. . . . . . . . 36 4. NEWMAN-KEULS COMPARISONS SIGNIFICANT AT .05 LEVEL. . . . . . . . . . . . . . . . . . . . 36 5. MEANS AND STANDARD DEVIATIONS FOR EACH STORY‘TYPE BY GROUP CELL . . . . . . . . . . 37 INTRODUCTION Piaget noted that the young child "seems to go through a stage when rules constitute an obligatory and untouchable reality" (Piaget, 1965, p. 121). One manifes- tation of this moral realism, as Piaget calls it, is the sort of judgments children make concerning the relative guilt of two hypothetical children, one of whom causes a large amount of damage by accident, while another causes a small amount of damage while acting from malice or greed. Piaget wrote a number of stories that illustrate this con- flict situation and tested school age children on them. He found a tendency for the younger children to make judg- ments in terms of the amount of damage done. The older children tended to use the intentions of the children in the stories as the basis for their judgments. Piaget found that at age seven most children made judgments on the basis of amount of damage (objective judgments), while by age ten most children made judgments in terms of intention (subjective judgments). Piaget writes that "The objective concept of re- sponsibility arises, without any doubt, as a result of the constraint exercised by the adult . .l." Lihié., p. 133). He goes on to say that "The rules imposed by the adult . . . constitute categorical obligations for the child before his mind has preperly assimilated them . . ." (gg;g., p. 135). Thus the younger child in the objective stage of moral judgment is led to believe that it is breaking or hurting things as such which is reprehensible, quite apart from intentions. Piaget believes that the child is eventually led from moral realism, which is based on adult restraint and the unilateral respect of children for adults, to a "mor- ality of reciprocity and not of obedience" (gpgg,, p. 138). Here the child will judge an action in terms of the moti- vation behind it instead of in terms of its material con- sequences. Piaget believes this development takes place as a result of peer interaction which encourages a desire for c00peration, and fosters mutual sympathy and the ability to examine the intentions of others. Many subsequent investigators have attempted to replicate, extend, criticize, or further clarify Piaget's pioneering study of the development of moral judgment as it relates to the problem of damage versus intentions. Boehm did a series of studies in which she exam- ined the relation between the use of intentionality and variables such as age, sex, social class, religion, and intelligence. She also investigated Piaget's idea that the development of intentionality is related to the over- coming of unilateral respect for adults and is based on the attainment of peer cooperation and reciprocity. In the first study (Boehm and Nass, 1962) it was predicted that: l) middle-class children would be more concerned with intentional factors, while working class children would be more concerned with amount of damage factors; 2) girls would show mature responses earlier than boys; 3) the child's responses would be story specific rather than reflecting one common level in all situations. The only subject variable that was statistically significant was age. For the three stories in which in- tention and damage were placed in conflict, children below 9 years of age gave more immature responses (re- sponses based on damage) while children above 9 years gave a greater number of mature responses (responses based on intention). Sex and social class were not re- lated to maturity of moral judgment in this study. The prediction of story specificity was sustained. The average number of mature responses was not the same for each of the stories designed to tap intentionality. Boehm and Nass included one story in which the subjects had to choose between peer cooperation and adult authority. The investigators found that mature responses on the stories dealing with intentionality were not re— lated to preference for peer c00peration on this story. Another study by Boehm (1962a) compared children from two different intelligence levels (gifted and aver— age) and two different socioeconomic classes (middle class and working class) on essentially the same task used in the paper discussed above. She found that the academically gifted children developed judgments based on intentionality more quickly than children of average intelligence. The middle-class children developed judgments based on inten- tionality more quickly than the working class children. There was an interaction between intelligence level and social class such that the gifted middle class children were more different from the average middle class children than were the gifted working class children from the average working class children. For the stories on peer cooperation and indepen- dence from adults, the working class children at both intelligence levels showed more preference for peer co- operation and independence than their respective counter- parts among the middle-class children. Thus, contrary to Piaget's hypothesis, the use of intentions as a cri- terion of moral judgment did not seem related to inde- pendence from adult authority and the attainment of peer cooperation. A related idea suggested by Piaget's theory is that a democratic or permissive home should result in the earlier attainment of intentionality, or in general that there should be a relation between parental authority and the development of intentionality. Both MacRae (1954) and Johnson (1962) report no relation between this factor and moral intentionality. Kugelmass and Breznitz (1967) compared the devel- opment of intentionality in kibuttz and city adolescents. One might expect that the communal style of life in the kibuttz would result in greater peer solidarity and hence in earlier moral intentionality. The kibbutz uses comparatively little adult restraint and this also should produce earlier intentionality. However, in this study no differences were found in the development of intentionality between kibuttz and city subjects. In a study comparing children in Catholic parochial schools with children in public schools, the finding of most relevance here was that the Catholic parochial group made more intentional judgments (Boehm, 1962b). This find- ing is not too surprising since Catholic education stresses the distinction between a sinful act which must be con— fessed, and an accidental mishap which need not be con- fessed. A further study by Boehm (1963a) compared the Catholic group of the previous study with a small group of children attending a Jewish parochial school. On the two stories concerned with intentionality the Jewish group was found to be in between the Catholic and the public school group. On one of the authority-versus—peer stories the Jewish group was superior to both the Catholic and the public school group. Thus once again Piaget‘s expec- tation of a relation between peer reciprocity and inten- tionality is not confirmed. Boehm concluded that "These apparent differences [between the Catholic and Jewish groups] suggest definite variations in content of conscience created by particular in-group training by religious insti- tutions" (Boehm, 1963a, p. 110). Piaget (1965) had found that very young children could not understand stories relating to intentionality. Boehm (1963b) attempted to find whether perhaps young children could be tested on this kind of problem after all. She tested a group of 51 children ranging in age from 3 years 9 months to 5 years 2 months on one Piaget-type item. She found that children as young as these could indeed give meaningful answers. Thirty-nine percent of the total group made subjective judgments, although none of the subjective answers were given by any of the three- year-old children. There was again a difference between middle-class children and working class children such that the middle-class group responded more maturely. The evidence which shows that religious training and social class membership are related to type of moral judgment made suggests that such judgments may be acquired through a process of social learning and reinforcement. With an eye to testing this idea, Bandura and McDonald (1963) attempted to see whether reinforcement and modeling procedures could be used to change a child's judgments from intentional (subjective) to damage-based (objective) and vice versa. Bandura and McDonald determined the operant level of their subjects by administering 12 stories to them. Here the usual relation between age and intention- ality documented by Boehm and Nass (1962) was found. After the operant level phase of the study was completed those children who showed the most consistent response patterns were subjected to one of three experimental conditions. In the "model and child reinforced condition" an adult model and the child alternated in responding to a set of stories similar to those used in determining the operant level. The model consistently made judgments and gave explanations Opposite to the sort originally made by the child. The experimenter gave social reinforcement following each response of the model. The experimenter also gave social reinforcement to the child each time he' made a moral judgment that ran counter to his dominant tendency as determined during the operant phase. In the "model reinforced, child not reinforced condition" the same procedure as that described above was used except that the child did not receive reinforcement for matching the judgments of the model. In the "model absent, child reinforced condition" there was no model and the child simply was socially rein- forced for making a judgment that ran counter to his pre- vious dominant tendency. After the children had been exposed to one of these three experimental conditions, they were immediately tested again (posttest phase) on a third set of stories, some of which were identical with the stories used during the operant phase, some of which were new. Bandura and McDonald found that the children in the two modeling conditions altered their moral judgments in the direction of the conditioning during the condition- ing phase of the experiment. This change carried over to the posttest phase of the experiment. The model absent condition did not produce statistically significant changes in moral judgments. Bandura and McDonald conclude that the operant phase of their experiment "fail(s) to substantiate Piaget's theory of demarcated sequential stages of moral deve10p- ment“ (Bandura and McDonald, 1963, p. 280) since they found that all children show both objective and subjective re- sponses at least some of the time. A careful reading of Piaget shows that he does not advocate such a theory. He 10 states that "It may even happen that one and the same child judges sometimes one way, sometimes the other. Besides, some stories point more definitely to objective responsi- bility than others . . . therefore, the material cannot be said to embody stages properly so called. Broadly speaking, however, it cannot be denied that the notion of objective responsibility diminishes as the child grows older" (Piaget, 1965, p. 124). Bandura and McDonald's Operant phase results substantiate the foregoing quotation. Moral objectivity is negatively related to age, and the same child judges sometimes one way, sometimes the other way. Considerable light is shed on the finding of a lack of perfect consistency in any one child by an examination of the actual stories used by Bandura and McDonald. Some are much clearer examples of a conflict between motivation and damage than are others. Bandura and McDonald also conclude that "the utility of Piaget's stage theory of morality is further limited by the finding that children's judgmental responses are readily modifiable, particularly through the utiliza— tion of adult modeling cues" (Bandura and McDonald, 1963, p. 280). This conclusion also seems based on a misunder- standing of Piaget's position. In the introduction to 11 their experiment Bandura and MbDonald speak of Piaget as a 'stage theorist.' As they contrast their social learning, reinforcement position with Piaget's position, they imply that Piaget is oblivious to environmental determinants of behavior and that he believes behavior to be "an unfolding of genetically programmed re5ponse predispositions" LEELQ., p. 274). To be sure, Piaget is not a learning theorist in the sense that Bandura is, but he assigns experience a key role in the development of behavior and explicitly rejects nativism (Piaget, 1952). Bandura and McDonald's experiment is of great in- terest. It demonstrates a mechanism through which moral judgments can be acquired and changed. It dOes not prove that this mechanism is in fact how such judgments are ac- quired in the real world, nor does it contradict Piaget's rather mild claim that younger children tend to make ob- jective judgments while older ones make subjective judg- ments. Cowan et a1. (1969) attempted a replication with some modifications of the Bandura and McDonald experiment. Essentially they corroborated the original results. In addition they showed that: l) the changes due to modeling were reflected in the actual explanations given as well as 12 in the simple choices made. That is, not only did the subjects pick the child which the model's behavior implied was the right one, they also gave the explanation the model had provided. 2) The changes show up on a posttest given two weeks after the experimental manipulation as well as on a posttest given immediately after the condi— tioning phase. 3) The changes generalized to stories of a somewhat different type (called here switch items) as well as to stories of the same type. Cowan et al. also found that the learning curves for the children who were conditioned up (conditioned from objective to subjective) were different from the curves of those who were conditioned down (conditioned from sub- jective to objective). And as far as the explanations (in contrast to the simple choices) are concerned, the conditioned down group tended to revert to their original (more advanced) level when tested after two weeks, while the conditioned up group continued to increase in the number of mature explanations given two weeks later. Subjects conditioned down also showed less generalization to the switch items than the subjects conditioned up. These findings were interpreted by Cowan et a1. as showing that conditioning up follows different laws 13 from conditioning down, and hence imitation alone cannot wholly account for what is going on. Further, Cowan points out that these results are in accord with the ideas of developmental theorists like Werner (1948) and Turiel (1969) who suggest that changes that are in line with the natural progress of develOpment ought to be more easily maintained than changes that run counter to the natural direction. In discussing Bandura and McDonald's interpretation of their results, Cowan makes the interesting point that since objective judgments predominate at lower ages, then imitation of models cannot solely account for moral judg— ment behavior since adults presumably do not provide models of or reinforcement for lower level responses. Bandura (1969) replied by saying, "it would . . . come as no sur- prise to find that parents are more inclined to take in- tentions into account in judging their children's behavior as they advance in age" (p. 278). Research into this con- troversy would be extremely interesting. As noted above, Boehm has found that moral judg— ments of intentionality are story specific. The finding that very few of the children tested were completely ob- jective or subjective across a series of stories, which is reported both by Bandura and McDonald and by Cowan 14 et al., is further evidence in support of the story spe- cificity of such judgments. None of these investigators looked into the characteristics of different stories that might cause one story to produce different results from another. Breznitz and Kugelmass (1967) have attempted to refine our knowledge of intentionality in moral judgments by looking at some of the within—item variables that might increase or decrease the likelihood of occurrence of a mature judgment. They constructed items that represented different combinations of three two-valued variables: usage/verbalization, recognition/recall, and crude/refined. Although eight items are theoretically possible by combin- ing all levels of 3 dichotomized variables, Breznitz and Kugelmass found that only 5 items were psychologically possible. The reader is referred to their report for a complete description of the 5 items used. The order in which intentionality would be achieved was predicted for 4 of the 5 items. If an item at a cer- tain level was passed (that is if the judgment made on it was in terms of intentions) then it was expected that all items below that level would be passed. Further, once an item was failed it was expected that no item at a higher level would be passed. Thus a Guttman scale was predicted. 15 As Breznitz and Kugelmass point out, authors such as Fla— vell (1963), Hunt (1961), and Wohlhill (1960) have sug- gested the applicability of a Guttman scale for Piaget's theory, since his theory demands that stage A must be passed through before stage B can be attained. The authors administered their items to a large sample of Israeli school children between the ages of 11 and 17. Very high coefficients of reproducibility for the 4 items for which predictions had been made were ob— tained. The remaining item was fitted into the scale post hoc and the coefficient of reproducibility remained at a high level for the full five item scale. Thus the prediction of a Guttman scale was confirmed. Breznitz and Kugelmass concluded that their data showed that the cognitive principle underlying the use of intentionality goes through four successive stages. "Starting with pre-verbalized usage of the principle, the development moves through the stage of verbalization, then to the ability to recall the principle spontaneously until, finally, very refined use of the principle is possible" (Breznitz and Kugelmass, 1967, p. 476). In a later analysis of the same data the investi— gators found that deve10pment in moral intentionality 16 continued throughout adolescence (Kugelmass and Breznitz, 1968). This is in contrast to their finding that, on the simple stories of intentionality usually used, no progress occurs beyond the age of eleven. The present investigation is concerned with the effect of systematic story variations on moral judgment. These variations can be considered in the light of the crude/refined dimension suggested by Breznitz and Kugel- mass. Although the content of the items here is very different from that of the items used by Breznitz and Kugelmass, these investigators point out that this dimen- sion is content free. Thus the present study can be seen as a further exploration of how moral judgments are af- fected by changes in the amount of refinement required to produce a mature response. METHOD The situation that Piaget and others have used in assessing intentionality pits the variable of intention (good or bad) against the variable of damage (high or low) so that they conflict. The child who acts from good mo- tives cause more damage, while the one who acts from bad motives causes less. Logically these two variables and their two respective values could be arranged in other combinations. These combinations may be psychologically interesting. There are five other combinations in addition to the classic form which can be generated. Piaget's classic case will be called Type A. Here good intentions are united with high damage while bad intentions are coupled with little damage. Here is an example of a Type A story: A little boy called John decides to help his mother by setting the table. But as he is reaching for the cups he slips and breaks 15 of them. Once there was a little boy whose name was Henry. One day when his mother was out he tried to get some candy l7 18 out of the cupboard. But the candy was too high up and he couldn't reach it. While he was trying, he knocked over a cup. The cup fell down and broke. In a Type B story the amount of damage is held constant (and low) while the intentions vary. In terms of the example story, imagine that both John and Henry break one cup each. John breaks his cup while trying to help his mother, while Henry breaks his in an attempt to snitch some candy. A Type C story is the same as a Type B except that damage is held constant at a high level. Thus our Type B example becomes a Type C story if both boys break 15 cups. In Type D the intentions are held constant (at the "good" value) while the amount of damage varies. To pro- duce a Type D transformation of the example story, the amount of damage must vary (high or low) while the inten- tion of each character must be the same and good. Imagine that both John and Henry are trying to help their mothers. However, John breaks 15 cups and Henry breaks only one. Type E stories are identical with those of Type D except that in Type E stories intention is held constant at the "bad" value. Thus our Type D story becomes a Type E story if both Jehn and Henry are snitching candy. 19 Finally, a non-conflict situation from the point of view of moral judgments will be called Type F. If we imagine that John breaks one cup while helping his mother, and that Henry breaks 15 cups while taking candy we would have an example of a Type F story. Here regardless of whether the child is judging on the basis of intentions or on the basis of damage he would be expected to make the same choice. The only difference would be the different reasons the children would give to justify their choices. Thus Type F stories drive a wedge between simple choice and the explanation for the choice. In their replication of Bandura and McDonald's study Cowan et al. used stories (called by them "switch items") of this type with a sim- ilar rationale (Cowen et al., 1969). Type B stories put pressure on the child who makes an immature judgment in a Type A situation to judge ma— turely in the new situation. "Objectively" the situations are identical; each child in a Type B story causes the same amount of damage. The only difference is in terms of intention. If we imagine that the child who judges a Type A story on the basis of amount of damage is incapable of considering intention at all, then we would predict that he will also judge a Type B story on the basis of 20 amount of damage. In this case he will call the two children in a Type B situation equally guilty since they caused an equal amount of damage. If, however, we imagine that the immature judger may not ignore intentional factors altogether but only overlooks them when they are placed into conflict with amount of damage factors, we can pre- dict that an immature judgment on a Type A story could easily co-exist with a mature judgment on a Type B story. Hypothesis one follows from this reasoning: Stories of Type B will elicit a greater number of mature judgments than will stories of Type A. The same chain of reasoning leads us to predict that stories of Type C will also put pressure on the child who made the immature judgment on stories of Type A to judge maturely in the new situation. Thus hypothesis two states that: Stories of Type C will elicit a greater number of mature judgments than will stories of Type A. In Type B stories only a small amount of damage is caused. In Type C stories a large amount of damage occurs. This difference might influence children who make immature judgments to Type A stories. Perhaps some of these children will judge the characters in the stories as equally guilty (the immature response) only in the 21 the cases in which damage was high (Type C). If only a small amount of damage results (Type B) then perhaps these children will be more sensitive to the different intentions involved and make the mature judgment. The idea here is that the child who makes an im- mature judgment is overconcerned with physical damage as a basis for judgment. If the damage is great (Type C) he will be so impressed by the amount of damage that he will be unable to pay attention to a relatively subtle thing like intention. However, if the damage is not so over- whelming (Type B) perhaps then the child will be able to attend to the different intentions that led to the damage. Hypothesis three follows from this argument: Stories of Type B will elicit a greater number of mature judgments than will stories of Type C. Type D stories put pressure on the child who judged maturely when confronted with a Type A situation to judge immaturely.l (A mature judgment to a Type D story is to declare that neither character is morally reprehen- sible since both acted from good intentions.) "Subjec- tively" the situations for Type D stories are identical; each child in a Type D situation acts from equally good motives. The only difference is in terms of amount of 22 damage. If we imagine that the child who judges a Type A story on the basis of intention is completely indifferent to amount of damage factors then we could predict that he will judge maturely on a Type D story. If, however, we imagine that the mature judger on a Type A story may not ignore amount of damage factors altogether but only con- siders them less important than intentional factors, we can predict that a mature judgment on a Type A story could easily co-exist with an immature judgment on a Type D story. In order to make a mature judgment on a Type D story the child must ignore the amount of damage factor and declare that this difference is irrelevant, morally speaking. This might be a more difficult task than to make a mature judgment on a Type A story. Thus hypothesis four states: Stories of Type A will elicit a greater number of mature judgments than will stories of Type D. A similar argument leads to the conclusion that stories of Type E will put pressure on the child who judged maturely on stories of Type A to now judge immaturely. (For Type E stories the mature judgment is to declare both children equally guilty since they both acted from the same bad motives.) 23 Thus hypothesis five predicts: Stories of Type A will elicit a greater number of mature judgments than will stories of Type E. In order to make the mature judgment on Type D stories the child has to give the children in the stories credit for having good intentions. In order to make the mature judgment on Type E stories the judger has to note that both children in the story had bad intentions. Let us assume that young children are more likely to be pun- ished for doing something from bad intentions than they are to be rewarded for having had good intentions. Or to put it another way, assume that a child will first incur displeasure for his bad intentions and only later be re- warded for his good intentions. Given these assumptions hypothesis six follows: Stories of Type E will elicit a greater number of mature judgments than will stories of Type D. Finally, hypothesis seven predicts that stories of Type F will elicit about the same number of mature judg- ments as will stories of Type A. The children who judge maturely on Type A stories are expected to also make mature judgments on Type F stories. The converse holds for 24 children who make immature judgments on Type A stories. Hypothesis seven follows from these expectations. Subjects Sixty fourth—graders from two local public schools served as subjects. Previous research (Bandura and McDonald, 1963; Cowan et al., 1969) shows that a group of children of this age makes about an equal number of mature and immature judgments. Thus fourth graders are appropriate to this study in which attention will be paid to both types of judgment. Test Items Twelve stories designed to test moral maturity, in the sense discussed above, were used. Ten stories were taken from a set developed by Bandura and McDonald (1963). Most of the stories taken from Bandura's set have been substantially changed. In some cases the Bandura stories did not lend themselves to being transformed into the different types used in the present study. In other cases the Bandura items were changed since they did not seem to provide a clear test of moral maturity as originally 25 written. The additional two stories were written by the present investigator. The stories used appear in Ap- pendix A. The children were group tested on these 12 stories. The following instructions were read to them: "I want to find out what kind of things boys and girls think are good, and what kind of things they think are bad. You have each been given a booklet with 12 stories in it. Read each story and then answer the questions on the page facing the story. You should make a circle around the answer you think is true. Read all 4 answers carefully before decid- ing. Remember, you can only circle one of them. When you have decided which answer is true, write your reasons in the blank space on the answer sheet. Try to explain as well as you can why you think the answer you circled is true. This might be hard to do for some of the stories but take your time and try to do your best. This is not a test and there are no right or wrong answers. I want to know what YOU think. So try hard to do your best work. Each time write your reasons for picking the answer you did. If you have trouble reading any of the stories, or if you have any questions, raise your hand and I'll try to help you. Are there any questions?" 26 Experimental Design The 60 83 were divided into 6 groups of 10 each. Each group of 85 received two of the 12 stories as Type A, two as Type B, two as Type C, and so on through Type F. A latin square design was used to assign stories to type in which presented so that each story appeared equally often as all of the types. (See Appendix B.) The 6 groups therefore represent 6 replications of the same basic ex- periment. An equal number of girls and boys were placed into each of the 6 groups. Except for this restriction placement of Ss into groups was random. Each of the Ss within a given story-by-type group received the 12 stories in a different order. The assign— ment for each S corresponded to the sequence from 1 row of a 12 by 12 latin square. Since there are only 10 subjects in each group, only 10 of the 12 rows were used. Each of the 12 stories consists of two sub-stories, one about a "bad" child and the other about a "good” child. Since the order in which the sub-stories appears could affect performance, this factor was counterbalanced. Each S received half of the stories so that the "good" child (according to the mature judgment of the item as a Type A 27 story) was described first. The other half of the stories were presented so that the "bad“ child was described first. Recall that each S received two stories in each of the 6 types (A through F). For a given 3, one story of each type appeared in good-bad order while the other story appeared in bad-good order. Within a given story—by-type group each S was alternated with the following S so that if S received a 1 particular story as a good-bad story then 82 received that story in bad-good order, S received it in good-bad order, 3 and so on. Thus within a given story-by—type group a given story appeared to half of the Ss in good-bad order and to half of the Ss in bad-good order. At the end of each story the child was asked to pick which of the following four statements he thought was true and to give his reasons: 1. X is worse than Y. 2. Y is worse than X. 3. X is as bad as Y. 4. Neither X nor Y is bad. The order in which these questions ap- peared was randomized so as to control for a response set. There are 24 orders in which 4 questions can be arranged. Five of these orders which make up a Luden square were selected. They will be designated as orders a, b, c, d, and e. 28 It is desirable for each story to be followed by the four questions in all of the five orders. It is also desirable for each of the question orders to appear an equal number of times over all. To insure these condi- tions the first two 55 within each story-by-type group had the questions that followed story #1 appear in order a, the questions that followed story #2 appear in order b, and so on until for story #5 question order e appeared. Then the sequence started over again with question order a following story #6. The second pair of Ss had question order b follow story #1, question order c follow story #2, and so on. This plan was followed for each subsequent pair of 83 so that the fifth pair had story #1 followed by question order e. (See Appendix C.) Note that the same story-by-question order condi- tions were given to pgi£§_of $3. In this way there was no confounding of internal order of story (does the "good" child appear first or second) with question order (is the "good" child asked about first or second). Recall that within a given story-by-type group each story that S re- 1 J ceived in good-bad order, S received in bad-good order, 2 and vice versa. S1 and S2 were also paired for the story- by-question order condition. Assume that Sl received a 29 particular story in good-bad order and that he received the questions to that story in order a. This means that the first question concerned the "good" child. 82 received this same story in bad-good order. Since he too received the questions in order a, S 's first question concerned 2 the "bad" child. Thus for this story, S1 was asked about the good child first, while 52 was asked about the bad child first. In general, for any particular story, and for any particular question order, one S was asked about the "good" child first while another S was asked about the "bad" child first. Thus the internal order of the story was not confounded with the question order. Appendix D shows the entire design for a given group. Scoring Each judgment was scored either as mature (2) or as immature (1). Although theoretically these categories are clear and non-overlapping, pilot work showed that scoring judgments is not as straightforward as could be hoped. Some subjects gave unintelligible or otherwise unscorable responses. Therefore the following steps were 30 taken in scoring the data: 1) A detailed set of rules explaining how to score the data was written (see Appendix E). The rules were based on an examination of the pilot data and illustrated with examples taken from that data. The data from the main study was then scored according to the rules generated from the pilot data. The rules are largely concerned with the scoring of difficult or anoma- lous cases and are not easy to summarize. In general, both the Ss choice and explanation were taken into account in scoring the data. 2) The investigator scored all the data using the rules described above. 3) An independent rater trained to use these rules also scored the data. During the first administration of the experiment, data were collected from 53 subjects. The raw data were scored by the investigator according to the criteria de- scribed above. A second trained rater independently scored 15 protocols. Per cent agreement between the two raters was 92.3%“ This percentage of agreement seemed to justify the conclusion that the data had been both objectively and reliably scored. Where the raters disagreed, the ratings of the first rater were used. 31 On the basis of the pilot work results it had been decided in advance to eliminate any subject who made 4 or more unscorable judgments. Inspection of the results from the first 53 subjects of the main experiment indicated that this criterion was too stringent. Seventeen of the first 53 subjects (almost one-third of the sample) failed to meet this stringent criterion; their records contained 116 un- scorable stories but also 88 scorable stories. A less stringent criterion of 6 or more unscorable responses was used instead; only 7 of the first 53 subjects failed to meet this criterion. The records for these 7 contained 72 unscorable stories but only 12 scorable stories. At a later date the 14 additional subjects needed to complete the sample were tested. By dropping 7 subjects for giving too many unscorable responses the distribution of scores was probably skewed toward mature responses. The subjects who had to be dropped were ones who apparently had diffi- culty in reading the stories or in writing their answers. Hence, it seems reasonable to expect that these dropped subjects probably would have given immature responses. However, this bias should not have seriously affected the hypotheses under investigation. 32 Each S made 12 judgments, 2 in each of the 6 story- type categories. The two judgments of each S within each story-type category were averaged. Thus each 8 contributed 6 scores, and there were a total of 360 data points. Ss who had unscorable responses were assigned either the score for the second story within that story-type, or if both stories were unscorable, the mean for that group by story- type cell. RESULTS The means and standard deviations for the six treatments (story-types) appear in Table l. The relative magnitudes of these means are roughly in accord with the predictions. The relative positions of Type D and Type E are reversed, and Type F is just greater than Type C. Otherwise the order is as predicted (see Table 2). The data were analyzed by a three-way sex x group x treatment (story-type) analysis of variance with repeated measures over story—types (see Table 3). According to this analysis there was a significant main effect of treatments (F e 8.977, df = 5/25, p < .01), and a significant interac- tion between treatments and groups (F = 2.542, df = 25/240, p < .01). None of the other main effects or interactions was significant. The significant effect of treatments contributed 15.5% of the total variance, while the signif- icant interaction contributed 8.6%. The Newman—Keuls test showed that Types D and E differed significantly at the .05 level from all other types (see Table 4). None of the other differences were 33 34 significant. Thus story types D and E did elicit fewer mature judgments than were made in the other types. This result is in accordance with the predictions. The means for Types B and C were not significantly higher than those of Types A and F. However, this finding may reflect a ceiling effect. The means for Types A and F were already so high that Types B and C could scarcely have been signif- icantly higher. The distribution may already have been skewed toward mature reSponses by the elimination of 7 pre- sumably immature 85. Hence the ceiling effect may have been heightened. The significant interaction between treatments and groups is difficult to explain. It was unpredicted and un- wanted. Most of the interaction seems to have been due to Group Two which didn't show the effect of treatments at all (see Table 5). All other groups showed effect of treat- ments in roughly comparable ways. Each group was distinguished by a particular combi— nation of which stories appeared in which story—types. The groups were used simply as a counterbalancing device, to insure that each story appeared equally often in all types. It was not considered, in advance of the data analysis, that the variable of story by story-type had any 35 psychological significance. At present we are at a loss to explain this result, which is after all reliable beyond the .01 level, and can only suggest a replication to see whether it shows up again. Even if this interaction is reliable it is a weak effect which accounts for little variance (eta2 = .086). TABLE 1 MEANS AND STANDARD DEVIATIONS FOR STORYJTYPES —— Story-Type A B C D E F Mean 1.706 1.825 1.767 1.471 1.372 1.772 SD .389 .351 .356 .475 .426 .355 TABLE 2 PREDICTED VERSUS OBTAINED ORDER OF STORY-TYPE MEANS Predicted Obtained D E g C B E D A C F B Few Many Few Many Mean # of Mature Judgments 1 Mean # of Mature Judgments 36 TABLE 3 SUMMARY OF ANALYSIS OF VARIANCE Source of Sums of DF F P Etaz Variance Squares A Sex 0.066 1 .287 NS error 1.141 5 B Groups 1.725 5 .705 NS AB 1.141 5 .467 NS error 23.496 48 C Treatments 10.279 5 8.977 <.01 .155 error 5.725 25 AC 0.509 5 1.461 NS error 1.742 25 BC 5.725 25 2.542 <.01 .086 ABC 1.742 25 .773 NS error 21.619 240 Total 66.300 359 TABLE 4 NEWMAN-KEULS COMPARISONS SIGNIFICANT AT .05 LEVEL Treatments B F C A D E ** ** ** ** g* ** ** ** Treatments tub>0fijw 37 TABLE 5 MEANS AND STANDARD DEVIATIONS FOR EACH STORYJTYPE BY GROUP CELL I) StoryJType B C D E F M 1.70 1.90 1.80 1.50 1.61 1.72 Group I SD .40 .30 .40 .45 .37 .39 M 1.83 1.70 1.70 1.78 1.70 1.80 Group II SD .32 .40 .33 .40 .46 .40 M 1.45 1.65 1.72 1.50 1.25 1.83 Group III SD .47 .45 .39 .45 .40 .32 M 1.70 1.95 1.85 1.15 1.22 1.80 Group IV SD .24 .15 .32 .32 .32 .33 M 1.75 1.90 1.78 1.30 1.05 1.70 Group V SD .336 .30 .33 .46 .15 .24 M 1.80 1.85 1.75 1.60 1.40 1.78 Group VI SD .40 .32 .34 .49 .37 .39 DISCUSSION The results show that story-types has a signifi- cant effect on intentionality in moral judgments, as does the interaction of story-type with group. The story-types effect reflected smaller mean numbers of mature judgments made in response to stories of Types D and E than to stories of other types. Thus the results indicate that children judged less maturely when confronted with a Type D or Type E story than when confronted with a story from one of the other four types. The significant difference between story-types D and E and the other story—types can be looked at in terms of Breznitz and Kugelmass' (1967) crude-refined dimension. On a Type A story the characters are differentiated both by their intentions and by the amount of damage they do. On a Type D or E story the characters are identical in terms of intentions and only differentiable according to the amount of damage each produces. A Type D or Type E story thus requires a more refined application of the principle of intentionality than a story of the other types. 38 39 In order to make a mature response in Type D or E one must be freer from the notion that if something bad happens to someone, they must have deserved it. The idea of immanent justice as described by Piaget is the belief that if someone does something bad, he will be punished automatically and naturally. Here we seem to have the converse of this idea. If something bad has happened, then the person must have done something bad to deserve it. In a Type D or E story a child who judges immaturely seems to say, "Well, the character who caused more damage must have done something worse, because something worse happened to him." In a Type A story such reasoning could be checked by remembering that the child who caused more damage also had better intentions. Thus the child can reason that intentions, relative to amount of damage, are more important. In Type D or E stories this check cannot operate. Both characters in these stories operated from the same motives, so in order to make a mature judgment the subject must commit himself essentially to the idea that amount of damage factors are irrelevant, not merely less important relative to intentions. Mature judgments on stories of Type B and C may result from a cruder use of the principle of intentionality 40 than is needed for stories of Type A. Here the child is not presented with a conflict between amount of damage and intentions. Both characters produce the same amount of damage. The only thing that distinguishes them is their different intentions. So in order to judge maturely the child need only believe that intentions have some relevance to moral worth. He could judge maturely on Types B and C and still believe that intentions are less important rela- tive to damage in the assessment of moral worth. As stated in the results section, the failure to obtain significant differences between story types B and C on the one hand and A on the other is probably due to a ceiling effect. Research is currently underway with a younger age group in order to see whether reliable differences among these types will emerge. Thus going from Types B and C through Types A and F and ending up at Types D and E we go through 3 stages of progressive refinement in moral judgment. In B and C a mature answer entails only the belief that intentions play some part in such judgments. In A and F the mature answer depends on the belief that intentions should be given greater weight relative to damage. In D and E the mature 41 answer rests on the assumption that amount of damage factors are not to be considered in making moral judgments. The failure to find the predicted differences be- tween D and E and between B and C is not too surprising. The reasoning on which these predictions was based was conjectural and tentative. Another way to conceptualize the different story- types is in terms of discriminability. In stories of Types B and C the only difference is in terms of inten- tions. Here to take note of this difference leads to the mature judgment. In Types D and E the only difference is in terms of amount of damage. Here to take note of this difference leads to the immature judgment.) Thus we can predict that a greater number of mature judgments will occur to B and C while a smaller number will occur to D and E (the results actually obtained) simply by virtue of the expectation that any possible discriminatory cue will be used. In Types B and C using the one discriminatory cue leads to a mature judgment. In Types D and E using the one discriminatory cue leads to an immature judgment. The discrimination explanation seems to have some problems. For one thing, one of the cues given to the children on all stories and on all story-types was that 42 an acceptable choice could be the decision that both char- acters were equally good or equally bad. That is, the task was not presented necessarily as a discrimination problem where the best answer was the one that capitalized on whatever differences existed between the characters. In fact, to make a mature judgment on B and C the child had to note the one difference, whereas to make the mature judgment on D and E he had to ignore it. Discrimination of when to discriminate was thus required. According to the discrimination hypothesis the number of mature responses in B and C ought to be equal to the number of immature responses in D and E. This is because a mature judgment in B and C is based on making a discrimination, as is an immature response in D and E. However, in B and C 83%.and 78% respectively of the scor- able responses were mature, while in D and E only 52% and 60% respectively of the scorable responses were immature. Thus simple discrimination does not seem to provide an adequate account of the results obtained. The results provide further support for Boehm's contention that intentionality is not a simple global concept but is responsive to the effect of different stories. This study shows that the likelihood that 43 intentionality is used depends on the variable of story-type. A story analysis of the data would provide further informa- tion on the story specificity of such judgments. The data seem to support the idea of Breznitz and Kugelmass that "As the child mature and broadens his range of experience, it is possible for him to respond to a wider range of situations with the more developed type of response based on intentionality.” (Breznitz and Kugelmass, 1967, p. 470.) Strictly speaking the present data, based on only one age group, does not and cannot tell us anything about the develogment of intentionality. Another study now under- way will aim at finding out whether older children do re- 3pond to a wider range of situations (Types D and E as well as A, B, and C) with intentional judgments than do younger children. The results also raise an interesting question for the studies by Bandura and McDonald and by Cowan et al. which show that these judgments can be modified by reinforcement and modeling. Did the children in our sample make fewer mature judgments on Types D and E because they had been so conditioned? Or does it make more sense to think of the re- sults in terms of the application of a rule which, given a sufficiently difficult problem, may break down? Most likely 44 before serving in the experiment the children had never considered the problem posed in Types D and E stories. (The investigator certainly never explicitly considered such problems before undertaking this study.) If this is true, then a direct conditioning or modeling explanation seems unlikely. What the children probably had learned, quite likely from reinforcement and imitating adult models, was the importance of considering intentions in making moral judgments. How they applied this rule probably de- pended less on Specific training and specific experiences than on a general developmental process which is probably age and IQ related. This general deve10pmenta1 process has to do with the increasing ability of the growing child to make more and more refined and more and more abstract use of the general principles and rules be has learned. REFERENCES REFERENCES Bandura, A. Social learning of moral judgments. Journal of Personality and Social Psychology, 1969, 11, 275-279. Bandura, A. and McDonald, F. Influence of social rein— forcement and the behavior of models in shaping children's moral judgments. Journal of Abnormal and Social Psychology, 1963, 61, 274-281. Boehm, L. The development of conscience: A comparison of American children of different mental and socioeconomic levels. Child Development, 1962a, 33, 575—590. Boehm, L. The deve10pment of conscience: A comparison of students in Catholic parochial schools and in public schools. Child Development, 1962b, 3;, 591-602. Boehm, L. The development of conscience: A comparison of upper-middle class academically gifted children attending Catholic and Jewish Parochial schools. Journal of:Social Psychology, 1963a, 59, 101-110. Boehm, L. The development of conscience of preschool children: a cultural and subcultural comparison. Journal of Social Psychology, 1963b, 52, 355-360. Boehm, L. and Nass, M. L. Social class differences in conscience deve10pment. Child Development, 1962, éég 565-574. Breznitz, S. and Kugelmass, S. Intentionality in moral judgment: DevelOpmental stages. Child Develop: 45 46 Cowan, P., Langer, J., Heavenrich, J., Nathanson, M. Social learning and Piaget's cognitive theory of moral development. Journal of_Personality and Social Psychology, 1969, ll, 261-274. Flavell, J. The developmental Psychology of Jean Piaget. Princeton, N.J., Van Nostrand, 1963. Hunt, J. McV. Intelligence and Experience. New York: Ronald Press, 1961. Johnson, R. A study of children's moral judgments. Child Development, 1962, gs, 327-354. Kugelmass, S. and Breznitz, S. The development of inten- tionality in moral judgment in city and kibbutz adolescents. Journal of Genetic Psychology, 1967, .lll, 103-111. Kugelmass, S. and Breznitz, S. Intentionality in moral judgment: Adolescent development. Child Develop- ment, 1968, §2_(first half), 249-256. MacRae, R. A test of Piaget's theories of moral deve10p- ment. Joupnal of Abnormal and Social Psychology, 1954, 42, 14-18. Piaget, J. The Origins of Intelligence in Children. New York: Int. Univer. Press, 1952. Piaget, J. The moral judgment of the child. New York: Free Press, 1965. Turiel, E. DevelOpmental processes in the child's moral thinking. In P. H. Mussen, J. Langer, and M. Covington (Eds.) New Directions in Developmental Psychology. New York: Holt, Rienhart and Winston, 1969. Werner, H. Comparative Psychology of Mental Development. Chicago: Follet, 1948. Wohlwill, J. A study of the deve10pment of the number concept by scalogram analysis. Journal of Genetic APPENDIX.A TEST ITEMS APPENDIX A ‘ TEST ITEMS 11 Carol's mother asked her to help sweep up in the playroom. While Carol was sweeping she accidentally hit a table with the broom. Five records that were on the table fell off and broke. Laura's mother asked her to help sweep up in the living room. Laura didn't feel like it very much. She was pretty mad at her mother for asking her to sweep. So Laura took the broom and hit a chair with it. A record that was on the chair fell off and broke. 12 Carol's mother asked her to help sweep up in the playroom. While Carol was sweeping she accidentally hit a table with the broom. A record that was on the table fell off and broke. 47 48 Laura's mother asked her to help sweep up in the living room. Laura didn't feel like it very much. She was pretty mad at her mother for asking her to sweep. So Laura took the broom and hit a chair with it. A record that was on the chair fell off and broke. 13 Carol's mother asked her to help sweep up in the playroom. While Carol was sweeping she accidentally hit a table with the broom. Five records that were on the table fell off and broke. Laura's mother asked her to help sweep up in the living room. Laura didn't feel like it very much. She was pretty mad at her mother for asking her to sweep. So Laura took the broom and hit a chair with it. Five records that were on the chair fell off and broke. l4 Carol's mother asked her to help sweep up in the playroom. While Carol was sweeping she accidentally hit a table with the broom. Five records that were on the table fell off and broke. 49 Almost the same thing happened to Laura. Laura's mother asked her to help sweep up in the living room. While Laura was sweeping she accidentally hit a chair with the broom. A record that was on the chair fell off and broke. 15 Carol's mother asked her to help sweep up in the playroom. Carol didn't feel like it very much. She was pretty mad at her mother for asking her to sweep. So Carol took the broom and hit a table with it. Five records that were on the table fell off and broke. Almost the same thing happened to Laura. Laura's mother asked her to help sweep up in the living room. Laura didn't feel like it very much. She was pretty mad at her mother for asking her to sweep. So Laura took the broom and hit a chair with it. A record that was on the chair fell off and broke. l6 Carol's mother asked her to help sweep up in the playroom. While Carol was sweeping she accidentally hit a table with the broom. A record that was on the table fell off and broke. 50 Laura's mother asked her to help sweep up in the living room. Laura didn't feel like it very much. She was pretty mad at her mother for asking her to sweep. So Laura took the broom and hit a chair with it. Five records that were on the chair fell off and broke. 21 Jerry's father is building a wooden table. Jerry sees that it is not finished. He decides to help his father by fin— ishing the table. While he is working Jerry breaks one of the table's legs by accident. Tony's father is building a workbench. Tony decides to see if he can break it. He takes a hammar and hits it as hard as he can. He makes a little dent but the workbench is still okay. 31 Ed didn't know the names of streets very well, and was not sure where River Street was. One day a man came to Ed and asked him where River Street is. So Ed said, "I Ehlpk it's that way," but it was not that way. The man really got lost, and could not find the place he was looking for. 51 Joe knows the names of streets very well. One day a man came to Joe and asked him where Bayshore is. But Joe wanted to play a trick on him, and said, “It's that way," and he pointed the wrong way. But the man didn't really get lost, because he found his way again. 41 John was in his room when his mother called him to dinner. John went down, and Opened the door to the dining room. But behind the door was a chair, and on the chair was a tray with fifteen cups on it. John did not know the cups were behind the door. He opened the door, the door hit the tray, bang went the fifteen cups, and they all got broken. One day when Henry's mother was out, Henry tried to get some cookies out of the cupboard. He climbed up on a chair, but the cookie jar was still too high, and he couldn't reach it. But while he was trying to get the cookie jar, he knocked over a cup. The cup fell down and broke. 52 51 Barbara decides she will clean her room and put her toys away so that her mother will not have so much work to do. Barbara put her big doll in the toy box, and then she put the wooden blocks in too. Barbara did not think the blocks would hurt the doll, but when she put the blocks in the toy box, they fell on the pretty doll and broke it all to pieces. Amy's mother asked her to clean up her room and put her toys away. Amy didn't want to do this. When her mother asked again, Amy got mad, picked up a doll and threw it on the floor. When the doll hit the floor, the doll's finger broke off. 61 Pam went grocery-shopping with her mother. Her mother has a lot of ShOpping to do, and is getting tired from.walking around the store. They are almost finished sh0pping, and her mother remembers that she needs a bottle of mustard which is at the back of the store. Pam says, "Ill get it for you, Mother," and runs to get the bottle of mustard. 53 As she is taking the bottle off the shelf, she drops it. The bottle of mustard falls to the floor and breaks. Kate is getting tired of sitting and waiting while her mother is ShOpping in the grocery store. So Kate runs up and down the aisles in the grocery store. The clerk tells her to slow down and be more careful. But Kate doesn't pay much attention to him. She starts to run again when he isn't looking. As she turns the corner, her hand hits a bottle of ketchup. The bottle of ketchup falls to the floor without breaking. 71 Chuck's friend is building a house with blocks. When Chuck sees his friend building the house, he helps his friend by bringing the blocks to him. The house is just about fin- ished when Chuck starts to get some more blocks, but as he gets up he trips and falls against the house, and it crashes down. Harry asks one of the boys to play ball with him. But the boy says, "No, I'm going to build a tower out of blocks." This makes Harry mad at the boy. He decides he will try to knock down the boy's tower. Harry waits until the tower 54 is just about finished and then he throws his ball at it. The ball hits the tower and knocks off one block. 81 Jane comes in from playing outside. She feels real tired so her mother tells her to take a rest. Jane didn't know her mother left her hat on the sofa. She walks over to the sofa, and plops down. When Jane lies down she squashes her mother's hat all out of shape. One day Doris' mother and father go visiting and Doris is all alone at home. She wants to see the things in the t0p of her parents' closet even though she has been told not to go up there. Doris knows her parents willnot be home till suppertime, so she climbs up on a chair but as she is reaching for the things in the closet the mother's hat box falls down and the hat gets bent a little bit. 91 Kathie's mother was late coming home from shopping. Kathie thought she would like to help her mother, so she began to set the table for supper. As she was carrying the dishes 55 to the table, two large dishes slipped out of her hand, fell, and broke. One morning Alice came in from the yard and found that her mother was not back from the store yet. Alice thought she would get some ice cream even though she knew her mother was saving it for company. She got a little dish from the cupboard. As she was going to the refrigerator, the dish drOpped and broke. 101 Peter and his parents are sitting down having supper. The mother needs some sugar which is by Peter's plate so Peter tells his mother that he will pass it to her. As Peter reaches quickly for the sugar bowl his hand hits the bottle of milk and the milk spills all over the table. Gary is sitting at the supper table, and he has eaten all his food except the potatoes. Gary tells his mother that he does not want to eat the potatoes. The mother says he has to eat his potatoes, or he will not get any dessert. This makes Gary mad. So Gary pushes his plate and it knocks over the salt shaker, and a little salt spills. 56 111 Jim's father had a clock which was kept on t0p of a tall bookcase. One day Jim noticed that the clock had st0pped running. He decided to help his father by fixing the clock. But as Jim was trying to get the clock down he knocked it off the bookcase by accident. It smashed on the floor in a thousand pieces. Sam's father had a clock which was kept on top of a cabinet. One day Sam decided it would be fun to try to break the clock. So he climbed up and threw the clock down from the cabinet. The clock was not hurt except for a small scratch. 121 Two first grade classes are playing a game to see which class can kick the kickball the farthest. All the children in both classes have had their turn except Ross. So far his class is losing, but if Ross can kick the ball real far, his class will win. His class is cheering for Ross. Ross takes careful aim and kicks the ball with all his might. The ball goes sailing across the playground, and smack into a window and breaks it. 57 Harvey and his class are playing kickball when the bell rings. The teacher says, “Recess is over, let's go back to the classroom." Harvey is mad because his team lost. So when the teacher is lining up the children, Harvey kicks the ball across the playground and someone has to go and get it. APPENDIX B ASSIGNMENT OF TEST ITEMS TO GROUPS APPENDIX B. ASSIGNMENT OF TEST ITEMS TO GROUPS II III IV V VI 1-2 11-12 9-10 7-8 5-6 3-4 3-4 1-2 11-12 9-10 7-8 5-6 5-6 3-4 1-2 11-12 9-10 7-8 7-8 5-6 3-4 1-2 11-12 9-10 9-10 7-8 5-6 3-4 1-2 11-12 11-12 9-10 7-8 5-6 3-4 1-2 58 APPENDIX C ASSIGNMENT OF QUESTION ORDERS TO TEST ITEMS ASSIGNMENT OF QUESTION ORDERS TO TEST ITEMS APPENDIX C 2 3 4 5 6 7 10 11 12 Question order 1, 2 a b c d e a b e a b 3, 4 b c d e a b c a b c 5, 6 c d e a b c d b c d 7, 8 d e a b c d e c d e 9, 10 e a b c d e a d e a 59 APPENDIX D COMPLETE DESIGN FOR A GIVEN GROUP 6O G‘LIJ'U Find! (DO U .m @ .mu Q. U a m m c o m \m e .m (e m (He 0 a a m u a m 4 m m a o \NH \H HA .oe e m 0 p w 0 Q m a m o m a m .e .m m «H H .m e a a m a m m o m o a a .e e .me e m (a m m m p U Q o a m m m o m HH OH 4 m e mbomw stHU 4 mom szmmn MBEQAOU Q NHQmemd (DD amp NH KOO mo muomnesm 61 [\Q Aggro mQrQ Nec'o LOU .Hmpuo coaummso lelm .mmmu snoum atria .Hmeuo Ooomuemn .# snoum \mHuuu.H .Hmpuo omnipoom .# huoum NHIIIH ems Q. U m 0 c m a m OH \O \OH e HH m O a O m o m m O .HH m (O OH m m U Q m a m m O .NH O \OH O U 0 U M m m o a s OH NH m m o m m p APPENDIX.E SCORING RULES APPENDIX.E SCORING RULES You will be rating the judgments of school children on the stories of Appendix A. Take a look at story 31. In this story Joe intended to do harm (he wanted to play a trick) but in fact he did no harm (the man didn't get lost). Ed did not intend to do harm (he tried to tell the man where the street was) but in fact he was mistaken and the man really got lost. After reading this story (or a comparable one) the children were asked to choose one of the following four answers as true: Ed is worse than Joe; Joe is worse than Ed; Ed is as bad as Joe; Neither Ed nor Joe is bad. Then they were asked to explain the reasons for their answer. You are to use three categories in rating the childrens' judgments: mature (l), immature (0), and un- scorable (?). A mature judgment is one in which the child uses the intentions or motivations of the characters in the 62 63 story as the basis for deciding. An immature judgment is one in which the child uses the amount of damage caused and disregards intentions or motivations in making a decision. Thus in story 31, a mature judgment would state that Joe is worse than Ed because Joe intended to deceive the man while Ed wanted to help. An immature judgment would state that Ed is worse than Joe because the man Ed talked to really got lost while the one Joe talked to did not. The meaning of the unscorable category will be explained in detail later. As you will note by looking at Appendix A, each of the 12 stories appears in six different forms. The presen- tation of each of the 12 stories in each of the 6 forms is followed by the four questions given above, sometimes in different orders, and with appropriate modifications for name changes. Story 31 is an example of a type 1 story. The last digit of a story tells you its type. For story 32 (a type 2 story) the mature response is to say that Joe is worse than Ed because Joe meant to trick the man and Ed did not. The immature response is to say that Ed is as bad as Joe since their actions resulted 64 in the same amount of damage (the man was slightly incon- venienced). For story 33 (type 3) the mature response is to say that Joe is worse than Ed because Joe meant to trick the man and Ed did not. The immature response is to say that Ed is as bad as Joe because both caused the same amount of damage. For story 34 (type 4) the mature response is to say that neither Ed nor Joe is bad since neither one intended any harm. The immature response is to say that Ed is worse than Joe since the man Ed talked to really got lost but the man Joe talked to did not. For story 35 (type 5) the mature response is to say that Ed is as bad as Joe since both of them tried to trick the man. The immature response is to say that Ed is worse than Joe since the man Ed talked to really got lost, but the man Joe talked to did not. For story 36 (type 6) both the mature and immature judgments would state that Joe is worse than Ed. The judg- ment is scored as mature if Joe is considered worse because he tried to trick the man. The judgment is scored as im— mature if Joe is considered worse because the man he talked to got lost. 65 The general principles of what is meant by mature and immature have been explained above and should be well understood now. The next most baSic principle is as fol- lows: Put more weight on the child's explanation of his answer than on the answer itself. Although for any par- ticular story-type, one answer represents a mature judgment, one represents an immature judgment [these two are called the "desisuated" answers], and two are irrelevant and ought not to be picked at all, it has been found that children will often pick from the two "irrelevant" answers. They then justify their choices in ways that are scorable as mature or immature. Therefore do not use the child's choice of one of the four statements as the main basis for rating his judgment. Your rating will be based on both factors. The details will be spelled out below. Here is an example of a clearly mature rating for story 31. Answer--"Joe is worse than Ed." Explanation-- "because Ed did not no what the name of the streets where. And Joe did now what the name of the streets." Despite the garbled grammar and spelling it is clear that the judgment is based on the idea that Ed made an innocent mistake while Joe was being deceitful. 66 Here is an example of an immature response to a type 1 story: Story 41, Answer--"John is worse than Henry." Explanation-—"John borke fifhteen cups, and Henry broke a cup." Here the child is clearly concerned with the amount of damage and not with the motivations that led to the damage. Here is a mature response for a type 2 story: Story 42, Answer——"Henry is worse than Jehn." Explana— tion--"Henry is worse than John because Henry souldent of have got in the cokie jar any way." The different reasons that led to the breaking of a cup by each boy is empha- sized. Here is an immature response for a type 2 story: Story 62, Answer--"Neither Pam nor Kate is bad." Explan- ation--"because the bottles did not break." Amount of damage is the criterion for choice here. The differing motivations of Pam and Kate are ignored. Here is a mature response to a type 3 story. Story 13: Answer--"Laura is worse than Carol." Explana- tion—-"Laura did not do it on acat." "Acat" is interpreted as "accident." It is often necessary to make guesses about the meaning of words on this rating task. This is a mature 67 response since the child is noting the different motiva- tions that led up to the broken records. Here is an immature response to a type 3 story. Story 33: Answer--“Ed is as bad as Joe." Explanation-- "Ed did the same as Joe in the story he told the the wrong street and the got lost." The child has focused on the fact that the men got lost. He ignores the different reasons why. Here is a mature response to a type 4 story. Story 14: Answer--"Neither Laura nor Carol is bad." Ex- planation--"becs thae both accidentally broke somthing." This is mature since the child notes the identical and innocent motivations and ignores the differing amounts of damage. Here is an immature reSponse to a type 4 story. Story 114: Answer--"Jim is worse than Sam." Explanation-- "because Jims smashed into a thousand pieces But sams was only scratches." Here the child is clearly using the dif- ference in amount of damage as a basis for judgment. Here is a mature judgment to a type 5 story. Story 95: Answer--"Alice is as bad as Kathie." Explana- tion--"Alice new her mother was saving the ice and so did 68 Kathei." The judgment is based on the intentions of the two girls and ignores the differences in amount of damage. Here is an immature judgment for a type 5 story. Story 25: Answer--"Jerry is worse than Tony." Explana- tion—-"Because Jerry broke a leg and Tony put a dent in it." Here the child is judging on the basis of amount of damage and ignoring the equality of intention. Here is a mature judgment of a type 6 story. Story 116: Answer--"Sam is worse than Jim." Explanation-- "Jim dint try to breck the clock but Sam did." Sam is con- sidered worse because of his bad intentions. Here is an immature judgment of a type 6 story. Story 86: Answer--"Doris is worse than Jane." Explana— tion——"When dorises mother went away she got up and her mothers hat got squashed out of shape when Jane got up on the sofa the hat bend a little bit." Here the explanation of why Doris is worse is based on Doris having caused more damage than Jane. The preceding examples are all relatively clear-cut cases of mature or of immature judgments. In many cases the answers that children give are much more difficult to rate. Here are a list of rules and examples to be used in rating difficult cases. 69 It is permissible to be generous. The child is not required to spell out with logical precision his reasons as long as he somehow indicates the basis for his choice. Example-—Story 73: Answer--"Harry is worse than Chuck." Explanation--"Chuck trip and shmash the house Hary shmash the tower whit a ball." This counts as a mature response. The child implies that Harry meant to knock down the tower while Chuck did it by accident. Example--Story 23: Answer--"Tony is worse than Jerry." Explanation--"Because Tony went ahead and broke it and Jerry just broke a leg." Here the child seems to be saying that Tony purposely broke it while Jerry broke it by accident. Thus the judgment is rated as mature. As noted above, sometimes a child will pick an "irrelevant" answer but then give a ratable explanation. In such cases give a rating of mature or immature if at all possible. Example--Story 24: Answer-—"Tony is as bad as Jerry." Explanation--"I think they are both had because they both broke something." Here the answer is one of the "irrelevant" ones although it most closely resembles the mature answer, "Neither Tony nor Jerry is bad." The ex- planation is clearly immature, however, for motivation is not mentioned at all and it is the fact of damage that is 7O determining the child's judgment. Therefore the item is rated as immature. It should be noted that in this example the choice of the answer and the explanation given are logically com- patible. However, if the choice of answer and the explan- ation are logically incompatible, then the item should be rated as unscorable. Example-—Story 94: Answer--"Neither Kathie nor Alice is bad." Explanation--"Kathie borke two dishes and Alice borke one dish." The explanation refers to an inequality, while the answer asserts an equality. This is uninterpretable, and so the response is rated as unscorable. Example--Story 45: Answer--"thn is as bad as Henry." Explanation--"because John knocked 15 cups over." Again the answer affirms equality, while the ex- planation takes note of a difference between the two char— acters. So the judgment is rated unscorable. Example-- Story 91: Answer--"A1ice is as bad as Kathie." Explan- ation-~"because when katheys mother is late she wants to help her mother so she starts setting the table 2 plates brake. When Alices mother gone away she gets into the ice creme when she knows her mom was saving it for com- pany while shes bringing the dish over it brakes." There 71 is no apparent connection between the eXplanation and the answer so the item is rated as unscorable. There is an exception to the rule that the answer and the exPlanation must be logically compatible in order for an item to be rated mature or immature. If it is clear that the child has got the names mixed up, the response is still rated mature or immature, though strictly speaking this represents a "logical incompatibility." Example-- Story 81: Answer-—"Doris is worse than Jane." Explana- tion--"Doris didn't do it on purpose." Here what must have been intended was "Jane did't do it on purpose." Here the child is given the benefit of the doubt and cred- ited with a mature judgment. Example--Story 92: Answer-- "Alice is worse than Kathie." Explanation--"Because Alice was trying to help." Here again it seems safe to infer that the child meant to write "Kathie was trying to help." So the judgment is scored as mature. If the child chooses one of the four statements for an answer but offers no explanation rate the judgment as scorable if the circled answer is one of the two "de- sisuated" ones. If the child gives a comprehensible exPlanation but does not choose any of the four alternatives for an 72 answer, rate the judgment on the basis of the explanation if possible. If the child's explanatiOn merely reiterates his answer from the four alternatives, rate the judgment as scorable if the circled answer is one of the desisuated ones. If the child makes a choice of answer that is scor- able as mature or immature, but then rather than explaining his choice, states that the child who is worse ought to be punished, the item should be rated as scorable. If the eXplanation seems to indicate that the child has missed the basic point of the story, rate the judgment as unscorable. Example: Story 95: Answer--"Kathie is worse than Alice." Explanation--"Kathie is worse than Alice because Kathie got the ice crem.when she knew it was for company any way." This explanation indicates a lack of comprehension of the story, for both Alice and Kathie took the ice cream even though they knew they were not supposed to. Hence this item.must be rated as unscorable. Example--Story 55: Answer—-"Barbara is worse than Amy." Explanation--"Bardar did not mind." Again the child must have misunderstood the story for both children were dis- obedient. 73 Example--Story 112: Answer--"Jim is worse than Sam." Explanation--"because when Jim got up on the book shelf the books fell down and the clock was hurt when Sam got on the cabinet he throught it down and.was hurt." Both the choice of answer and the explanation indicate a lack of comprehension of the story. So this item is rated unscorable. Example—-Story 53: Answer--"Barbara is worse than Amy." Explanation--"Barbara is worse than Amy because Barbara broke her new doll." It is tempting to score this item as immature since damage rather than motivation is clearly the basis of the judgment. However, the child has obviously not understood the story, for.Amy also broke her doll. Therefore the item must be rated as unscorable. Do not be too stringent. If the child shows some awareness of motivation or of damage, rate his judgment accordingly. Example: Story 43: Answer-—"Henry is worse than John." Explanation--"Henry knocked them off the shelf but Jehn did not know that the cups were on the tray." This is a little vague, but the child indicates an aware- ness that Jo n was not intending any harm. So the item is rated mature. 74 Example--Story 35: Answer--"Joe is as bad as Ed." Explanation—-"because they both were bad." This is not a high level answer. It borders on mere restatement, which as specified above is to be rated unscorable. However, the child has indicated an awareness that both Ed and Joe did something bad and he is not basing his judgment on the amount of damage, Which varies. Therefore, this item is rated as mature. If the choice of answer indicates a mature or an immature judgment but the explanation is unintelligible, then rate the item as unscorable. Example--Story 81: Answer-—"Doris is worse than Jane." Explanation--"Jane men to rod (illegible) mother (illegible) but Doris did." The answer indicates a mature judgment but the explana- tion is not comprehensible so the item is rated as un- scorable. If the explanation can be deciphered at all score it. Example——Story 76: Answer--"Harry is worse than Chuck." Explanation--"Harry went mod away." 'This makes no sense, strictly speaking, but it seems that the child is referring to the fact that Harry acts out of anger. Therefore the item is rated as mature. This rating might not have been possible if the child's choice of answer 75 had not been the officially mature one. Thus an ambiguous explanation in conjunction with a clear answer can yield a scorable judgment. Some of the stories are Open to several interpre— tations concerning motivation. If the child's judgment is in terms of motivation, but in a somewhat unusual way, rate the item as mature. Example—-Story 126: Answer--"Neither Ross nor Harvey is bad." EXplanation--"Ross did not get angry. Harvey did not try to break the window." This is not one of the anticipated responses. However, it indi— cates a concern with motivation more than with damage and so is rated as mature. Example: Story 42: Answer-—"Neither John nor Henry is bad." ExPlanation--"They both did not no the cup was there." Again this is not the way that anyone was expected to respond. However, the response shows a concern with motivation and so is rated mature. If the child bases his judgment on an irrelevant aspect of the story, rate the item as unscorable. Example-- Story 63: Answer--"Kate is as bad as Pam." Explanatione- "Kate should not been running neither should have Pam." This misses the point altogether but is neither mature 76 nor immature as the terms are here used. Therefore this item is rated as unscorable. If the child indicates in his explanation that he has mixed up some non-essential aspect of the story, rate the item anyway. Example--Story 76: Answer--"Harry is worse than Chuck." Explanation--"because when harry was building the blocks chuck brung the blocks to harry so he wouldn't have to get up all the time." Here the child has combined the two parts of the story into one whole, but his response is clearly scorable as mature. If the exPlanation is ambiguous so that there is no way to tell whether the child is referring to motiva- tion, or damage, rate the item as unscorable. Example-- Story 83: Answer-—"Neither Jane nor Doris is bad." Ex- planation--"Because they both did the same thing." It is just not clear here what the child had in mind. Does he think that Jane and Doris both acted from the same motives? Or does he think that they both caused the same amount of damage? There is no way to decide, the choice of answer does not provide any clues, and so the item is rated as unscorable. 77 If the child in his explanation seems equally con- cerned with both amount of damage and motivation rate the item as unscorable. NOV :3 1370 ICHIGQN STATE UNIV. LIBRARIES llll Ill llllllll Ill 9 4 312 30107 2041