Students' Perception of Moral Atmosphere in ...

Students’ Perception of Moral Atmosphere in Secondary Schools, their Moral Reasoning Competence, and their Practical Judgement in School1

Daniel Brugman, Louis W. C. Tavecchio, Bart Jan van Os & Karin Høst

Parts of this paper were presented at the 21st Conference of the Association for Moral Education, New York, November 1995

About the authors

Daniel Brugman is teaching social and moral development at Utrecht University, the Netherlands. His research interest focuses on conditions that influence individuals’ moral development. Address: Dr D. Brugman, Department of Developmental Psychology, Utrecht University, Heidelberglaan 2, 3584 CS Utrecht, The Netherlands.

Louis W.C. Tavecchio is Associate Professor at the Center for Child and Family Studies of Leiden University, the Netherlands. Besides 'moral development and moral education' his research areas and activities include 'sex-specific socialization and children's sex-role attitudes' and 'effects of daycare on children's socio-emotional development'.

Bart Jan van Os graduated in quantitative Psychology at Leiden University and is participating in research at the Center for Education and Instruction. Currently he is working at the Data Theory Group, Department of Education, Leiden University.

Karin Høst has been working as a junior researcher at the Center for the Study of Education and Instruction, Leiden University, The Netherlands.

2

Content Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 The Just Community approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 The relationship between moral atmosphere, practical judgement, and moral competence . . . . . . . . . 4 Investigating moral atmosphere in regular schools from the viewpoint of a Just Community . . . . . . . 5 Four conditions of secondary schools in the Netherlands that may have an impact on students’ perception of the school moral atmosphere . . . . . . . . . . . . . . . . . . . . . . 6 Research questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Moral atmosphere, practical judgement and moral competence . . . . . . . . . . . . . . . . . 8 Control variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Hypothesis 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Hypothesis 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Hypothesis 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Hypothesis 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Summary and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Appendix 1: Validity of the measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Appendix 2: Confirmatory Factor Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

1

Abstract The aim of the present study is to investigate whether the moral atmosphere in secondary school, as perceived by the students, exerts a negative influence on students’ practical judgement in schoolrelated dilemmas. Hypotheses based upon the Just Community approach are tested in research carried out in regular secondary schools in the Netherlands. The Just Community approach describes the quality of the school moral atmosphere as it is perceived by the students in comparison with students’ developing moral competence. The approach especially tries to create a moral atmosphere in school which stimulates students’ moral development. In our research, which covers a period of three years, we used several student samples of widely differing secondary schools. To measure the quality of the moral atmosphere in school as perceived by the students, we used the Just Community constructs which characterize the development of a school with a moral atmosphere of a low quality into a school with a moral atmosphere of a high quality (a ”Just Community”). First, we will investigate whether there are differences between regular secondary schools in their moral atmosphere as perceived by the students. We hypothesize that such differences can be expected. Second, we will investigate whether there exist gaps between students’ perception of the school moral atmosphere, their practical judgement and their moral competence. Two hypotheses are tested: • On average, the stage of students’ practical moral reasoning in school-related dilemmas is less advanced than their stage of moral reasoning competence, but is higher than the perceived stage of reasoning in school moral atmosphere. • Students’ predicted action in practical judgement is more prosocially oriented than as perceived in their school moral atmosphere. Third, we will investigate the strength of the relationships between the constructs moral atmosphere, moral competence and practical judgement. The following hypothesis is tested: • Students’ perception of the school moral atmosphere influences their practical judgement more strongly than does students’ moral reasoning competence. When these hypotheses are confirmed we may conclude that the moral atmosphere in school as perceived by the students exerts a negative influence on students’ practical judgement in schoolrelated dilemmas, and that the influence of moral atmosphere differs between schools.

2

Introduction Our research on moral atmosphere started in the fall of 1991 and consisted of ethnographic interviews with students to find out how they perceived the moral atmosphere in their school. The opening question went like this: Question: Suppose I came to live round here. I am a new boy, same age as you, and I am thinking of going to your school. So I ask you what your school is like. What answer would you give? Answer student (girl, aged 15): It's nice. At first I didn't fancy it much. A lot of people think it's a school for swots, students who work extremely hard and who take little interest in other things apart from their studies. Because it's exclusively a grammar school they think it's for real clever kids, but in fact it's a really tight community because it's also a small school, it's very social, you know a lot of people. It's like, well when you come into first grade, people don't tease you or anything like that. That's what you see in lots of other schools, I think. It's all very close-knit, just like your own class is. Question: (...) Do you think all the students at the school feel the same way about it as you do? Answer: No, I don't think so, but most of them do, I suppose. Well, there are a few kids..who are outsiders ... and a few ... who don't like it at all, but the vast majority really do, they do have this feeling honestly. From the perspective of this student her school is a community. She also makes clear this is not her private view only, but a viewpoint shared by most students of her school. She acts as a representative of the students of her school. According to the Just Community approach of Power, Higgins and Kohlberg (1989), judgements concerning situations in school that reflect the perspective of the majority of the students characterize the culture or atmosphere in school. In contrast, judgements reflecting individual views are referred to as practical judgements. In this example, the perspective of the majority as the individual perceives it and the perspective of the individual student converge.

The Just Community approach The Just Community approach is well-known as an educational approach which stimulates the process that members of an institution go through when they develop the moral atmosphere within their institution (Power et al., 1989). The moral atmosphere is characterized by 5 constructs, each of them divided into steps, phases or ‘soft’ stages to represent the process of development. The constructs are: (1) Valuing of the School as an Institution, which refers to the extent to which students value the school intrinsically. (2) Stage of Community, which refers to the shared understanding of the community as a terminal value. (3) Degree of Collectiveness, which refers to the degree in which a norm is shared by the students. (4) Phase of the Norm, which refers to students’ commitment to seeing that the norms are upheld. (5) Stage of Reasoning on the Norm, which refers to the way the meaning of the norm is shared. Not all of these constructs, however, are used in all our instruments. Finally, the following, non-developmental construct is distinguished, the (6) Content of the Norm, which refers to the norms within a community: caring, trust, integration, participation, publicity, collective responsibility, substantive and procedural fairness, equity, and order. Just Community studies have been carried out as educational intervention projects mostly with volunteer students within secondary schools. Just Community schools have a small population of about one hundred students and a participatory democracy. The moral atmosphere in these schools is of a high quality: students feel they are responsible for the well-being of all members of the community and for the community as a whole; students are committed to the maintenance of shared moral values; interpersonal and collective relationships are valued in themselves and given such priority that members are willing to make personal sacrifices for the sake of their common life. The effects of these educational interventions have been investigated on the development of students’ practical reasoning in school-related dilemmas, their prosocial behaviour within and outside the 3

school, and their moral competence. Students participating in the just communities grew in their practical reasoning and their moral competence and demonstrated more prosocial behaviour (less agression, racism, etc., and higher educational goals). The gap between moral reasoning competence and practical reasoning which is observed under regular conditions disappeared in these schools (Power et al., 1989). Proponents of the Just Community approach have claimed that the moral culture or the perceived atmosphere in regular high schools in the USA exerts a negative pressure on practical judgement in school. For example, according to Power (1988, p. 202): "Our research indicates that the cultures of conventional, large, public high schools have a ... depressing effect on students' judgements about school-related problems". The negative influence on students’ practical judgement attributed to the school’s moral atmosphere is thought to show itself in the following: •

In regular schools, students judge more prosocially and reason at a more advanced stage regarding school-related dilemmas when answered from the individual perspective on themselves (practical judgement) than when answered from the perspective of the majority of most other students (moral atmosphere). For example, Higgins, Power and Kohlberg (1984, Table 5.3) report that 75% of the students from a regular high school predicted a prosocial choice for themselves in a school-related moral dilemma, while only 34% of the students predicted a prosocial choice for the other students. The choice for themselves in a moral dilemma in a practical judgement is evoked by asking “What would you do?” The choice which reflects the moral atmosphere is evoked by asking: “What, do you think, would most other students do?

•

In regular schools, students score higher on hypothetical dilemmas measuring the moral reasoning competence than on school-related dilemmas measuring practical moral reasoning. For example, Higgins et al. (1984, Table 5.3) report that students from regular high schools have a mean score for practical moral reasoning (deontic judgements) on school-related dilemmas which is lower by about half a stage than for their moral reasoning competence on hypothetical dilemmas. When one wants to know students’ the reasoning on the norm or their perception of the reasoning on the norm concerning the moral atmosphere ‘why’ questions are asked (“Why would (should) you do that?”, “Why would they do that?”).

•

In Just Community schools, however, the gap between practical moral reasoning and moral reasoning competence becomes nonexistent or changes to a stimulating one, i.e. students score even higher on school-related dilemmas than on hypothetical dilemmas (Higgins et al., 1984; Power et al., 1989).

The Just Community studies only obtained results with respect to a few regular schools in as far as these served as controls for the Just Community schools. We investigated whether these claims were supported when using a great number of widely differing regular secondary schools.

The relationship between moral atmosphere, practical judgement, and moral competence The Just Community approach is viewed here from the broader perspective of the relation between institutional moral atmosphere and the development of a sense of responsibility, practical judgement and the moral reasoning competence of individuals participating in institutions. As in the Just Community approach, we investigated whether the moral atmosphere in schools as perceived by the students exerts a negative influence on students’ practical judgement. 4

Moral competence underlies the reasoning of subjects when confronted with hypothetical moral dilemmas, dilemmas in a fictitious and remote context. The stagewise development of the moral reasoning competence is claimed to satisfy the ‘hard’ stage criteria: qualitatively distinct cognitive structures, an invariant developmental sequence, structural wholeness, and hierarchical integration (Colby & Kohlberg, 1987; Kohlberg, 1984). A dilemma is a situation in which different parties or persons have competing rights. Hypothetical dilemmas aim at eliciting the highest stage of moral reasoning subjects can attain at that point in his or her development. In general, empirical research has confirmed this finding (Krebs, Vermeulen, Carpendale, & Denton, 1991; Walker, DeVries & Trevethan, 1987). Practical reasoning is elicited by real-life dilemmas embedded in a familiar, institutional context, for example in a school context. Higgins et al. (1984) and Power et al. (1989) have called these institutional dilemmas practical dilemmas. These authors (Higgins et al., 1984, p.76; Power et al. 1989, p. 273) proposed that standard dilemmas embedded in a familiar context may be used to measure moral performance or morality-in-action. The moral performance contains prescriptive as well as descriptive aspects: "Real moral decision-making is not only a prescription but also a description of facts of the situation and of the self's needs and motives and those of others in an integrated deliberative practical judgement" (Higgins et al., 1984, p. 77). On the other hand, Power et al. (p. 134) also made clear that actual moral performance is rarely studied; they refer to one study only. In some studies, like those of Gilligan and Belenky (1982), real-life situations were used in which persons faced a difficult decision such as whether or not to have an abortion, which is of course closer to morality-in-action than the more widely used standard dilemmas placed in an institutional context. Because we use these standard dilemmas, we prefer to speak of practical judgement when referring to the predicted action of individuals themselves, and of practical reasoning when referring to the reasoning justifying this judgement. The development of practical (moral) reasoning is thought to satisfy soft stage criteria: (a) while stages are still accepted as qualitatively distinct cognitive structures, a sharp distinction is no longer made between content and structure; instead, content and style are part of the definition of the stages; (b) while development is still progressive, regression is not excluded; (c) the hierarchically integrative, unitarian, simple stage conception of morality gives way to a complex stage model or a contextual constraint model, which permits a subject to use widely different stages dependent on the circumstances (Power et al., 1989, p. 136-142; cf. Krebs & Van Hesteren, 1994; Teo, Becker & Edelstein, 1995). Nevertheless, Colby and Kohlberg (1987, p. 8), while acknowledging some fluctuation in moral performance as well as in competence, believe that subjects always use the best reasoning they are capable of, and only use lower stage reasoning in situations characterized by a significant downward pressure. These situations are characterized by a moral atmosphere of a low quality.

Investigating moral atmosphere in regular schools from the viewpoint of a Just Community One may wonder whether it makes sense to study moral atmosphere in regular schools through categories that characterize the development of a school into a Just Community. Studying moral atmosphere in regular schools through these categories entails the risk that no differences will be found, since becoming a Just Community is not a goal of these schools. Indeed, according to the advocates of the Just Community approach, probably no differences can be expected. The moral atmosphere in USA public high schools is regarded as having an unvariably low quality: “We think that despite real differences, these (...) schools share common characteristics that press on students and teachers and create the same hidden curriculum of authoritarian, individualistic and selfprotectionistic, and instrumental norms and institutional valuing.” (Higgins, 1991, pp. 131, 132), or Power (1985, p. 223): “The need for democratic moral education is perhaps most evident when one 5

considers the high rates of violence, crime and vandalism in contemporary high schools. (...) The problems of crime and disruption in our schools have their origin in a declining school culture or more specifically, in a low level of moral atmosphere”. This picture of the moral atmosphere has been built upon the results in schools that served as control schools in the Just Community studies. However, we cannot generalize these results to all regular public secondary schools. One important goal of our research is to find out whether differences exist between schools in students’ perception the school moral atmosphere and what conditions may influence this perception. Four conditions of secondary schools in the Netherlands that may have an impact on students’ perception of the school moral atmosphere It is hard to believe that no differences exist between secondary schools as regards moral atmosphere. •

In the Netherlands, within-school democracy --through the establishment and functioning of a student council or student parliament-- varies widely across secondary schools.

•

The following secondary educational levels can be distinguished: Junior Vocational Secondary Education (LBO), Intermediate Secondary Education (MAVO), Higher Secondary Education (HAVO) and Preparatory College Education (VWO). A child leaving primary school at about 12 years of age can make a choice among 15 types of schools from these four educational levels or combinations thereof. The educational programme of the Junior Vocational Secondary Education and Intermediate Secondary Education takes four years, the Higher Secondary Education a five year programme and the Preparatory College Education a six year programme. The educational level of a school is rather strongly related to moral reasoning competence (Stams, 1994, using the SROM-SF in a sample of 160 students, reports a correlation as high as .49, and a correlation of .31 with intelligence and social economic status partialed out), and when one is interested in assessing differences between schools with regards to moral atmosphere it seems necessary to control for student characteristics such as moral competence. Theoretically, moral competence at least sets a ceiling to the perception of the stagelike constructs of moral atmosphere.

•

In addition to state public schools, there are a large number of specifically denominational schools in the Netherlands such as Protestant, Catholic, and anthroposophic. Also, the educational practice in these schools, including state public schools, may be based upon different educational theories -- such as Montessori, Dalton and Jenaplan. Such characteristics may be accompanied by a certain community feeling.

•

Some findings based upon traditional questionnaires for measuring the social atmosphere in schools do suggest differences in moral atmosphere between schools. Correlations have been found between the educational level of secondary schools and social atmosphere, and between educational level and delinquency (Junger-Tas & Kruissink, 1990). Baerveldt (1990) found a negative correlation between integration in school and petty crime by students. Students tended to integrate better into schools that paid more attention to the students, and also into schools that emphasized improving their educational system. This finding is in agreement with the results presented by Battistich, Solomon, Kim, Watson, & Schaps (1995). They found a positive relationship between a student’s sense of school community and student outcomes - this relationship was stronger for students with the most disadvantaged background.

6

Research questions Our research covers different studies undertaken in the course of a research project to develop a reliable and valid written instrument for the measurement of the moral atmosphere in secondary school as perceived by the students. The hypotheses are: (1) There are differences between secondary schools in their moral atmosphere as perceived by the students, using the constructs and categories that characterize the development of a school with a moral atmosphere of a low quality into a school with a moral atmosphere of a very high quality (a ”Just Community”). (2) On average, the stage of individual practical reasoning in school-related dilemmas is less advanced than the stage of individual moral reasoning competence, but is higher than the stage of the norm perceived in moral atmosphere; (3) Practical judgement is more prosocially oriented than the atmosphere judgement. (4) The perception of the moral atmosphere in school affects practical judgement more than does students’ moral reasoning competence. Were hypotheses (2), (3) and (4) confirmed, it would lead to the conclusion that the moral atmosphere in schools as perceived by the students exerts a negative effect on students’ practical judgement. These hypotheses were derived from the Just Community approach but will be investigated here with research carried out in regular schools (the validity of the scales is elaborated in Appendix 1).

Method Design To develop a questionnaire for the measurement of the perception of the moral atmosphere in secondary school we started with open ethnographic interviews in Study 1. Subsequently, in the second study, structured interviews containing standard schooldilemmas were used, and finally, in the third study, we used the structured interviews as well as the questionnaires, containing both the same standard schooldilemmas. -------------------------------------------------Table 1: Instruments and Constructs -------------------------------------------------All six scales/instruments have been used together with a scale for verbal intelligence and social desirability. The first hypothesis was tested using all samples, hypotheses 2, 3 and 4 were tested using the third sample only.

Samples Three samples participated in this study. These samples reflect the different stages in our research to construct a valid and reliable written moral atmosphere questionnaire. We started with an open, ethnographic School Moral Atmosphere Interview and ended with the written School Moral Atmosphere Questionnaire. In the third study (a multitrait-multimethod study, MTMM), standardized interviews and written questionnaires for moral atmosphere, practical judgement and reasoning, and moral competence were used. 7

Sample 1: 96 students from the second, third and fourth grades of 12 secondary schools representing all educational levels2 . From each school 8 students participated. We were especially eager to interview the prominent students ) they were supposed to be better informants than average students, as they are presumably better acquainted with various social networks in school (cf. Johnson, 1990, p. 10) and, therefore, may contribute more substantially to the moral atmosphere themselves. In this study we used an open interview, which started with the question quoted in the opening paragraph. Sample 2: Another 144 students from the second, third and fourth grades of the same 12 secondary schools as in sample 1. From each school this time twelve randomly selected students participated. We wanted to find out whether these students could answer the questions of a structured interview that was constructed on the basis of the open interviews with the prominent students (sample 1). Students were interviewed about four schooldilemmas (helping, theft, school participation, and bullying). Sample 3: This study is a so called multitrait-multimethod (MTMM) study (Appendix 1). The students came from sixteen secondary schools reflecting systematically the various educational levels in the Netherlands. From each of the following educational levels four schools were selected: (a) Junior Vocational Secondary Education, (b) Intermediate Secondary Education, (c) Higher Secondary Education and Preparatory College Education. Moreover, four schools were selected which offered Intermediate as well as Preparatory level. In each school two classes from the third grade were randomly chosen. From the schools that offered both educational levels (b as well as c) four of these classes were chosen. From each class three students were randomly selected. One hundred and twenty students, half of them male, half female, participated in the study. Age varied between 13.8 years and 17.3 years. Mean age was 15.0 years. Students received the structured interview and the School Moral Atmosphere Questionnaire (SMAQ), which was constructed on the basis of students’ answers on the structured interviews in sample 2. The written instrument was administered in groups, the interview instrument was administered individually. ---------------------------Table 2: Samples ----------------------------Five of the twelve schools in sample 1 are particularly interesting because they kept participating in the research from the beginning and were part of sample 1 to sample 4. In a separate large scale study, reported elsewhere, only the questionnaires were used (Høst, Brugman, Tavecchio & Beem, 1998). This study was carried out with a sample of 32 secondary schools, containing 192 classes and 1556 students. In this sample the Secondary School Moral Atmosphere Questionnaire (SMAQ) was administered to the students. Of each school two or three grade levels participated, depending on the duration of the course (4 or 5 years). From each grade level, two classes participated and from each class, eight students were randomly selected. This study has been reported separately elsewhere (Høst et al., 1998). However, we will give also some figures of schools that also participated in the earlier studies mentioned above. The MTMM study was carried out as a part of this larger study, using half of the schools of the larger study.

Instruments In addition to measures for moral competence, practical judgement and moral atmosphere, instruments for verbal intelligence and social desirability were used as controls. We wanted to investigate whether the instrument for e.g. moral competence measured something other than the measure for verbal intelligence or social desirability. In this section the instruments are described in detail. All instruments are described in this section. Moral atmosphere, practical judgement and moral competence 8

The School Ethnographic Moral Atmosphere Interview (SEMA I) is an open interview that consists of series of open questions to explore students’ perception of the moral atmosphere. The opening questions of this interview were given in the intr oduction to this article. Questions concerning practical (moral) judgement and the moral atmosphere were asked as follow-ups or in c lose conne ction with each other. This p rocedur e can be fo und in all instruments on practical judgement and moral atmo sphere (interv iew, question naire, obse rvation). Instrum ents make use of the same moral institutional dilemmas to measure both constructs (Higgins et al., 1984; Kohlberg, Scharf & H ickey, 197 2; Powe r et al., 1989 ). The School Standardized Moral Atmosphere Interview (SMAI) is a standardized interview constructed analogou sly to the ‘Practical School Dilemmas Interview’ of Power et al. (1989). The interview guide contains four school-relate d dilemm as (helping an unpopu lar classmate, p articipation in a school happening, stealing, bullying), two of which are selected for an interview, one prosocial and one ‘antisocial’. These dilemmas were constructed because they addresse d norms frequently mentio ned acro ss schools d uring the ethno graphic interviews. The interview contains open-ended questions and probes in order to stimulate students to communicate their own perceptions and the pe rceptions of the majority of students in their class or school. The interview contains questions referred to as ‘Content of the N orm’. These include q uestions about predictive behaviour: ‘What are you going to do?’, ‘What do you expect your classmates to do?’, and prescriptive behaviour: ‘What should your classmates do?’ Because the prescriptive statement was not included in the practical judgement it was not included in the score for the Atmosphere Stage of the Norm either. The interview also contains questions enabling students to justify the behaviours, which refer to Stage of Reasoning on the norm. Again from two perspectives are distinguished: the perspective of oneself or practical judgement (“What would you do? Why?”), and the perspective of most other students or moral atmosphere (“What, do you think , would mo st other studen ts do? W hy?”). Answ ers on the ‘wha t’ questions are referred to as ‘Content of the Norm’ re spectively, and answers on the ‘why’ question s as ‘Stage of R easoning o n the Norm ’. The questions ar e raised for v arying situations in e ach dilemm a. In the mult itrait-multimethod study, only the Helping dilemma w as used in the sta ndardize d interviews o n moral atm osphere b ecause of p ractical limitations. All interviews were transcribed in full. Statements b y the students wer e scored o n Content o f the Norm and Stage of Reasoning on the Norm. Inter-rater reliabilities (Cohen’s kappa) for coding the interviews on Content of the Norm varied between .75 and .92, with a mean of .83. The reasoning statements were sco red separ ately (Cohe n’s kappa .90). The highest Stage of Community could not be reliably distinguished from Stage of Reasoning on the Norm a nd was there fore dismisse d. The School Moral Atmosphere Questionnaire (SMAQ) is a multiple-choice instrument. It covers the same two dilemmas mentioned above, one about helping an unpopular classmate and one about stealing. The helping dilemma goes like this: “In some classes there are students who don't belong to the group, who are unpopular or not very well liked. John is one of these students, and you don't really like him either. He's not doing well in a particular subject, so he might not be promoted to the next grade this year. Most of his classmates are achieving well on this subject. After the common part of the lesson, students are allowed to do something for themselves. John asks if so meone w ill help him with his ho mework.” After each dilemma subjects are requested to give their opinion about what happened, what they think they themselves would do and what m ost of their classmates would do. Next, sets of four questions are asked, for instance concerning the reasons for helping, from the perspective of practical judgement as well as from the perspective of the atmosphere. For example, after presenting a reasoning “If you do not help John, the teacher might get angry”, students are asked “Is this a reason that you would give?” respectively “Is this a reason that most of your classmates would give? ”. Then, subjects are asked to choose the reasoning closest to the one they themselves would pro bably give in this situ ation, and the reason c losest to the on e the majo rity of the students would probably give in this situation. Such sets of questions are also asked for the reverse situation, for instance not helping. As in the SROM-SF (see below), reasonings are keyed to specific stages. Other questions concern what the subject and their classmates will do when the situation changes, for instance you will be bullied by your classmates, you miss your favourite tv programme, or the teacher is asking you to help. This instrument was used in the multitrait-multimethod study reported below as well as in a larger study in wh ich 32 seco ndary scho ols participated. The instrument is described in detail in Høst et al. (1998). With the School Moral Atmosphere Questionnaire, only the scores on Stage of the N orm (Cronba ch’s " =.67), Content of the Norm (" = .68) and Stage of C ommunity (P ersonal gain s, stage 1 and 2: " = .61; Soc ial Relations, stag e 3 and 4 : " = .78) wer e used.

9

The Sociomo ral Reflection Mea sure - Short Form (SRM-SF, Gibbs, Basinger & Fuller, 1992) is a simplified and shortened oral version of Kohlberg's Moral Judgement Interview. It measures production, i.e. students are asked to produce moral reasonings. The instrument is administered individually. The interview is typed out and the written text is scored. At an earlier stage of the research, our interviewers and coders trained themselves with the Dutch version of this instrument (Zwart-Woudstra, Meijer, Fintelman & Van IJzendoorn, 1993) and amply met all inter-rater criteria with Gibbs as a standard. Inter-rater reliability between both observers was good compared with the criteria given by Gibb s et al. (1992 ) with a SRM S correlatio n of .94 (no rm .80), a me an a bs olute SRMS discrepancy of .09 (nor m .20), a glo bal stage agreement of 100% (norm 80%) and an exact stage agreement of 80% (norm 50%). The average SRM-SF score (reported above) is in agreement with the average score reported by Gibb s et al. (1992 , Table 1 , p. 40) for gr ade 8 stud ents with a mean age of 14.1 years. The Sociomo ral Reflection Objec tive Measure - Sh ort Form (SROM -SF, Basin ger & Gibb s, 1987; cf. also Gibbs, Arnold, M organ, Sch wartz, Gav aghan & T appan, 1 984) is a wr itten instrument and contains two dilemmas, one of them being the Heinz dilemma. After the dilemma has been presented a subject is asked what Heinz should do and why (open question). Next, some aspects of the stories are changed: it is no lo nge r He inz's wife but Hei nz's best friend who suffers from the illness. Subjects are asked which reasons they would use if the person involved would be their best frien d. For eac h reason sub jects are aske d to indicate whether it is close to the reason they the mselves wo uld give or not. Subjects are supp osed to recognize the re asons they themselves use or would use. Each of these responses gives a close score. Finally, subjects are asked which of the four reasonings given most closely reflects their own. This series of questions is repeated with a stranger replacing the best friend. The SROM-SF score combines the close and closest score. The range of the total score is between 100 and 400. Two respondents were excluded be cause they met one of the exclusion rules set b y Gibbs. Cronbach’s " was .71.

Control variables T o measure verbal intelligence a part of the Groninger Intelligence Test - Short Form was used (Kooreman & Luteijn, 1987) w hich contains 20 multiple -choice items asking the subject to find a logical connection between pairs of wor ds. The item s gradually incr ease in difficulty. Gu ttman’s Split-half co ëfficient was .67 . The Social D esirability Sca le contains 11 items, for instance “I am honest ... sometimes/always”. Cronb ach’s " was .54 (n=116), which was well below the .73 in the pilot study (n=198, Stams, 1994).

Results Hypothesis 1 The first hypothesis questions, whether differences in moral atmosphere as perceived by the students exist between various regular secondary schools. Results of all four studies will be presented with regard to this hypothesis. In the first study we used open interviews with pr ominent students. From the beginning of our research on moral atmosphere, differences in students' perceptions of the moral atmosphere were found within all schools. For example, while some students perceived a rather strict discipline at school, others perceived an easy companionship between students and teachers. In the resulting description of the school's moral atmosphere, events that students agreed upon were stressed. We also found meaningful differences between some of the 12 schools that participated in this first study, although these differences were not as big as the differences between regular schools and Just Community schools (Power et al., 1989). In Table 3, examples from the open interviews with eight randomly selected prominent students are presented for five of the twelve schools toghether with an overall score on ‘Valuation of the school’. These particular five schools were selected because they kept participating in the research from the beginning. ---------------------------------------Table 3: Valuation of the school ----------------------------------------10

Statements of the students were scored according to the levels, stages or categories which are distinguished for each moral atmosphere construct (see Power et al., 1989). As regards the construct Valuation of the school as an institution, in all schools we found statements by students reflecting extrinsic motivation (level 1), in which the school is valued as an institution that helps the individual to meet their academic needs. School A is characterized predominantly by this category. Some typical statements by students are given that reflect this score. Students of school D, on the other hand, showed a strong sense of belonging (level 3), scored as ‘spontaneous community’. Here, the school is valued as the kind of place in which the students feel a sense of closeness to others and an inner motivation to help them The students from school H reflected this orientation also, although the teacher-student relationship in this school differed from that in school D. Students of school J showed not only extrinsic motivation. Many statements reflected a rejection of the school (level 0). Students of school L finally gave a wide variety of statements. Only in this school did we find statements by students reflecting a normative appreciation of the school (level 4), in which the school is valued for its own sake and the school can obligate its members in special ways. Statements of the students were scored according to the levels, stages or categories which are distinguished for each construct (see Power et al., 1989). Using these scores the 5 schools were ordered on each construct from a relatively low quality of moral atmosphere (5) to a relatively high quality (1). The overall rank order, taking into account the scores available on the different moral atmosphere constructs, is given in Table 3. The second study using structured interviews was carried out in the same school year. Now we interviewed students of these schools using structured interviews. A new group of 12 randomly selected students of the same grade levels (2, 3 and 4) from each school participated in this study (Brugman, Høst, Van Roosmalen & Tavecchio, 1994). The interviews contained standard dilemmas to measure the moral atmosphere in school (SMAI). Statements of the students were scored, which resulted in an overall score for each school. Our 5 schools showed nearly the same rank order as regards the quality in moral atmosphere when compared with the results of the etnographic interviews. Rank order of the schools on moral atmosphere is given in Table 3. In the third and fourth study written instruments with randomly selected students were used. Here, we only will pay attention to our 5 schools presented in Table 3. It was amazing to find that the overall measure for moral atmosphere (mean z-score on atmosphere-perspective for School as a Community, Valuation of the School, Content of the Norm, Stage of the Norm, Cronbach’s " =.91 at school level) in the five schools mentioned above, after three years showed the same rank order in quality as in the former study (Table 3). Moreover, schools D and L, characterized in the ethnographic study by an atmosphere of a relatively high quality, did show up again among the schools with the highest quality of moral atmosphere out of these 32 schools, while school J appeared among the schools with the lowest quality in moral atmosphere. Thus the range found between these 5 secondary schools in quality of moral atmosphere using the etnographic interviews is also the range found in a much larger set of schools using the moral atmosphere questionnaire. In the studies mentioned above, different student samples were investigated using different methods for assessing moral atmosphere. Because the study reported by Høst et al. (1998) was carried out three years later than the first study described here in which the ethnographic interview was used, it is likely that rather stable differences in moral atmosphere exist between schools. The differences between schools in atmosphere seem rather constant while the student population (but presumably not its characteristics) changes. Deliberately we selected widely differing schools. Therefore, one might object that differences were found in moral atmosphere due to the differences in student populations. Indeed, these five schools did show the same rank order as regards mean score on students’ moral competence as measured with the SROM-SF, as on moral atmosphere. At school level, a correlation of .61 was found between Moral Atmosphere and Moral Competence. At the individual level, however, the correlation was much lower (.16). In this study, significant and possibly substantial differences in moral atmosphere beween schools were shown within all educational levels, even when controlling for students’ moral competence (Høst et al., 1998). 11

Hypothesis 2 According to the second hypothesis, on average, the stage of students’ practical reasoning in school dilemmas is less advanced than the stage of their moral reasoning competence, but is higher than the perceived stage of reasoning of the norm of the moral atmosphere in school. Thus the stage of practical moral reasoning is a kind of compromise between the relatively high stage of moral reasoning competence and the relatively low stage of the perceived stage of reasoning on the norm of the moral atmosphere. One has to prove the existence of the gaps mentioned in this hypothesis before one can attribute a negative influence to the moral atmosphere. We will restrict ourselves to the results from the MTMM study, because they demonstrate with different measures the phenomenon mentioned in the hypothesis most clearly. These results are restricted for practical moral reasoning and moral atmosphere to the norm ‘Helping an unpopular classmate’. Results of the studies using the etnographic and structured interviews, however, point into the same direction (Brugman et al., 1994). In the MTMM study, interviews as well as questionnaires were used for moral competence (SRM-sf and SROM-sf respectively, see Table 4 first column), practical moral judgement (SMAI and SMAQ, column 2 in Table 4), and moral atmosphere (a different set of questions in the SMAI and SMAQ, column 3 in Table 4). First we compare the scores on the interview versus the questionnaire within each column. The scores can vary between 100 (stage 1) and 400 (stage 4). The moral competence mean score on the interview measure is lower than the questionnaire mean score. This reflects, as expected on theoretical grounds and on the basis of empirical results in other studies, that on a production measure (the interview, SRM-sf) the reasoning is of a lower level than on a recognition measure (questionnaire, SROM-sf). This phenomenon can be observed also on practical moral reasoning: the mean score on the interview measure is lower than on the questionnaire. However, on the stage of reasoning on the norm of the moral atmosphere the mean scores for the interview and the questionnaire are equal. Here, on moral atmosphere, we find on both measures the same relatively low mean score. ------------------------------------------------------------------------Table 4: Mean score etc. about here -------------------------------------------------------------------------The hypothesized gap can be observed when the mean scores are compared between the columns, while using the same type of instrument. Using the interview measure the moral competence score is 260, the practical moral reasoning 245 and the moral atmosphere 231. Using the questionnaire the moral reasoning competence stage is 298, the practical moral reasoning stage is 255 and the stage of reasoning on the norm in the moral atmosphere 229. Thus the observed stage of the moral atmosphere is about a third stage lower than the stage of moral competence when using a production measure (interview) and two-third of a stage lower when using a recognition measure (questionnaire). One also can observe that practical moral reasoning is on average nearly a mathematical compromise between moral reasoning competence and moral reasoning on the norm in moral atmosphere..

Hypothesis 3 According to the third hypothesis the predicted action or the content of the norm of the practical judgement is more prosocially oriented than that in the perceived moral atmosphere. Students think more prosocially about themselves than about their schoolmates. Here the score ranged from 0 (not prosocial) to 100 (prosocial). On the questionnaire the mean for practical judgement was 62 and for moral atmosphere 29; on the interview the mean for practical judgement was 74 and for moral atmosphere 55. Again, the difference between practical judgement and moral atmosphere is higher on the questionnaire than on the interview. 12

For Stage of Community, another construct that can be measured from the individual perspective (practical judgement) and the perspective of most other students (moral atmosphere) the difference, although significant, was much smaller, M=292 on the prosocial stages 3 and 4 for practical judgement, and M=286 for moral atmosphere (questionnaire). As could be expected, on the combined lower stages 1 and 2 of Stage of Community, the score for practical judgement (M=245) is lower than the score for moral atmosphere (M=268), which refers, however, to the same phenomenon as mentioned above. These stages are typically associated with a relatively large number of anti-social decisions.

Hypothesis 4 Given the considerations mentioned above, we present the results of a Confirmatory Factor Analysis (CFA) with respect to the fourth hypothesis, i.e. whether students’ predicted action about oneself in this situation (the Content of the Norm in practical judgement) is more strongly influenced by their perception of the moral atmosphere than by their moral reasoning competence. Again we used the scores of the MTMM study as described under hypothesis 2 and 3. We use two measurements for all constructs: moral reasoning competence (verbal: SRM-sf and written: SROM-sf), practical moral reasoning (verbal: SMAI and written: SMAQ) and stage of moral reasoning on the norm in moral atmosphere (also SMAI and SMAQ). The observed scores on these instruments are called the manifest variables. The theoretical constructs which they aim to measure are called the latent variables. In figure 1 the manifest variables are enclosed in square boxes, while the latent variables are enclosed in circles. Measures on the moral atmosphere variables and practical judgement were available for the Helping dilemma. To assess the strength of the relationships between the constructs CFA was used (the interested reader is referred to Appendix 2 for more information). We fitted a structural model for these latent variables using the practical judgement on Content of the Norm as a dependent variable (see Figure 1, Content of Norm Self). -----------------Figure 1 -----------------According to CFA the structure gives an acceptable description of the data (Normed Fit=0.98, Nonnormed Fit=1.05, Comparative Fit=1.00, df=6, p=.66). The effect size between moral atmosphere on Content of the Norm and practical judgement on Content of the Norm (.45) is much greater than that between Moral Competence and practical judgement on Content of the Norm (.11). The model suggests a socialization effect on Content of the Norm of the moral atmosphere variable on practical judgement. When we combine this result with the results of hypothesis 2 and hypothesis 3, which confirm that average score on Content of the Norm in moral atmosphere is lower than for Content of the Norm in practical judgement, we conclude that this socialization effect has to be interpreted as negative.

Summary and discussion In our research on students’ perception of moral atmosphere, their practical judgement and moral reasoning competence in regular secondary schools two claims of the Just Community approach were corroborated: (1) The stage of students’ practical moral reasoning in school-related dilemmas is less advanced than the stage of their moral reasoning competence, but it is higher than the stage of the norm in moral atmosphere. (2) The moral atmosphere in school is more influential with respect to the predicted action of practical judgement than the individual’s moral reasoning competence. Combining these results shows a negative socialization effect of the moral atmosphere in secondary schools on 13

students’ practical judgement. However, contrary to the opinion that the moral atmosphere in secondary schools is of a uniformly low quality, we found that even when controlling for students’ moral competence, differences exist between regular secondary schools with respect to moral atmosphere as perceived by their students. Probably, these differences are rather stable over time. This result raises questions about the institutional factors that may contribute to these differences between schools and about the mechanisms that transmit the shared ways of thinking about the meaning of school and the norms regulating social relationships within schools across student populations (Anderson, 1982). Students portray a more prosocial picture of their own content of the norm than of their classmates content of the norm, and a higher stage of their own reasoning than of their classmates reasoning. One has to be aware of the possibility that the difference between the content of the norm in practical judgement and in the moral atmosphere can be partly attributed to social desirability effects (Appendix 1). Besides social desirablity effects, we also found strong method effects for Stage of Reasoning on the Norm in practical judgement and in moral atmosphere. Probably this phenomenon is related to the observation of Power et al. (1989) that students in regular schools showed a lack of knowledge of the reasoning of other students. Students are better acquainted with the content of the norm of other students than with the reasoning justifying this content. Students know better that one would or would not help in a specific situation, than that they know for what reason one would help or not in that situation. Contents of the norm, like to help or not to help an unpopular classmate, may be deduced from observing the behaviour of other students in specific situations. Knowledge of the reasoning of other students on the other hand needs a kind of discourse which may not be customary in everyday interaction between the students of most schools. Two conclusions can be deduced from these observations, one methodological and one practical. The methodological implication is that statements about the content of the norm and reasoning on the norm in practical judgement should be gathered separately from these statements concerning the moral atmosphere. Until now a format was used in which they were gathered in close connection with each other, which enhances difficulties in disentangling them. Because of the methodological problems with the measurement of the constructs, the negative socialization effect of school moral atmosphere on students’ practical judgement needs to be studied further. Taking these methodological problems in perspective, we think the phenomenon observed has also practical consequences for the learning and developmental process: while students can derive their orientation for what to do by observing the content of the norm of other students, such an orientation is lacking for the moral reasoning process. The results stress the importance of particular intervention programs to get students better acquainted with each other’s moral reasoning. If they would put their own prosocial choice and higher stage reasoning to the front and would take eachother’s choice and reasoning seriously, students will perceive a higher stage moral atmosphere in school. When students perceive a higher stage school moral atmosphere, this could be less used as a self-serving bias to legimitize their own antisocial behavior. In this study the strength and the direction of the relationships between moral competence, moral atmosphere and students’ own practical judgement and reasoning were investigated in one particular dilemma. Elsewhere we have investigated the relationship between school moral competence, moral atmosphere and antisocial behavior and the importance of developing the moral atmosphere in school from the lowest to the highest quality found in our sample was assessed (Høst et al., 1998). From the results of these studies we may conclude that in order to stimulate prosocial behavior in a normal adolescent schoolpopulation it would be much more gratifying to stimulate the moral atmosphere in school than to stimulate individuals’ moral competence. Given the situation that the moral atmosphere in most of the schools from our samples corresponds more closely to the quality of the moral atmosphere in the USA sample of large public high schools than to the moral atmosphere of Just Community schools, a plea for the improvement of moral atmosphere in secondary schools is easy to make. Research on Just Community schools with volunteer students suggests that an improvement of the moral atmosphere results in long-term effects on prosocial and moral behavior, including students’ educational career (cf. also Battistich et al., 1995). 14

In the Netherlands, public concern on the moral atmosphere in secondary schools is growing. We have sent an evaluation report concerning the quality of the moral atmosphere in comparison with that in other schools to the staff of each school. Possibly such a report will create some pressure and enthusiasm in teachers about improving the moral atmosphere. To achieve this goal they may try interventions that are in accordance with their own educational philosophy, and/or which may reflect the Just Community approach.

References Anderson, C.S. (1982). The search for schoolclimate: A review of the search. Review of Educational Research, 52, 368-420. Baerveldt, C. (1990). De school: Broedplaats of broeinest? [The school: breeding ground or hotbed?] Dissertatie RUU. Arnhem: Gouda Quint. Basinger, K.S. & Gibbs, J.C. (1987). Validation of the Sociomoral Reflection Objective Measure Short Form, Psychological Reports, 61, 139-146. Battistich, V., Solomon, D., Kim, D, Watson, M., & Schaps, E. (1995). Schools as communities, poverty levels of student populations, and performance: A multilevel analysis. American Educational Research Journal, 32, 627-658. Bentler, P.M. (1989). EQS structural equations program manual. Los Angeles: BMDP Statistical Software. Brugman, D., Høst, K., Van Roosmalen, M., & Tavecchio, L.W.C. (1994). Moreel klimaat van scholen voor voortgezet onderwijs: de perceptie van de leerlingen. [Moral atmosphere in secondary schools: students’ perceptions] Comenius, 14, 64-82. Brugman, D., Tavecchio, L.W.C., Van Os, B.J., Høst, K., Meijer, T., & Roosmalen, M. van (1995). Moreel schoolklimaat en morele competentie bij scholieren van het voortgezet onderwijs: convergerende en discriminerende validiteit [Moral Atmosphere in Secondary Schools and Students' Moral Competence: Convergent and Discriminant Validity]. Leiden University: Department of Educational Studies / Utrecht University: Department of Developmental Psychology. Campbell, D.T. & Fiske, D.W. (1959). Convergent and discriminant validation by the multitraitmultimethod matrix. Psychological Bulletin, 9, 13-27. Colby & Kohlberg (1987). The Measurement of Moral Judgment. Vol. 1: Theoretical Foundations and Research Validation. Cambridge: Cambridge University Press. Edelstein, W. (1985). Moral interventions: A sceptical note. In M. W. Berkowitz & F. Oser (Eds.), Moral education: Theory and application (pp. 387-403). Hillsdale, New Jersey: Lawrence Erlbaum. Gibbs, J.C., Arnold, K.D., Morgan, R.L., Schwartz, E.S., Gavaghan, M.P. & Tappan, M.B. (1984). Construction and Validation of a Multiple-Choice Measure of Moral Reasoning. Child Development, 55, 527-536. Gibbs, J. C., Basinger, K. S., & Fuller, D. (1992). Moral Maturity. Measuring the Development of Sociomoral Reflection. Hillsdale, N.J.: Lawrence Erlbaum. Gilligan, G. & Belenky, M. F. (1980). A naturalistic study of abortion decisions. In R. L. Selman & R. Yando (Eds.), Clinical-Developmental Psychology. New Directions for Child Psychology 7 (pp. 69-90). San Francisco: Jossey-Bass Inc. Higgins, A. (1991). The Just Community approach to moral education: Evolution of the idea and recent findings. In W.L. Kurtines & J.L. Gewirtz (Eds.), Handbook of Moral Behavior and Development. Vol. 3: Application (pp. 111-141). Hillsdale, N.J.: Lawrence Erlbaum. Higgins, A., Power, C., & Kohlberg, L. (1984). The relationship of moral atmosphere to judgments of responsibility. In W.L. Kurtines & J.L. Gewirtz (Eds.), Morality, Moral Behavior and Moral Development (pp. 74-106). New York: Wiley. Høst, K., Brugman, D., Tavecchio, L.W.C., & Schouten, J. (1995). Moral Atmosphere in High 15

Schools and Students’ Moral Competence: Some Aspects of Objective Measurement. Paper presented at the 21st Conference of the Association for Moral Education. New York, November. Johnson, J.C. (1990). Selecting Ethnographic Informants. Newbury Park CA: Sage. Junger-Tas, J. & Kruissink, M. (1990). Ontwikkeling van de jeugdcriminaliteit. [Development of juvenile delinquency]. Arnhem. Kohlberg, L. (1984). Essays in Moral Development. Vol 2. The Psychology of Moral Development. San Francisco: Harper & Row. Kohlberg, L., Scharf, P. & Hickey, J. (1972).The justice structure of the prison: a theory and an intervention. The Prison Journal, 51(2), 3-14. Kooreman, A. & Luteijn, F. (1987). Groninger Intelligentie Test - schriftelijke verkorte vorm. Lisse: Swets & Zeitlinger. Krebs, D.L. & Van Hesteren, F. (1994). The development of altruism: Toward an integrative model. Developmental Review, 14, 103-158. Krebs, D. L., S. C. A. Vermeulen, J. I. Carpendale and K. Denton: 1991, ‘Structural and situational influences on moral judgment: The interaction between stage and dilemma’. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook of Moral Behavior and Development. Vol. 2: Research (pp. 139-169). Lawrence Erlbaum: Hillsdale, N.J. Power, C. (1985). Democratic moral education in the large public high school. In M.W. Berkowitz & F. Oser (Eds.), Moral education: Theory and application (pp. 219-238). Hillsdale, New Jersey: Lawrence Erlbaum. Power, C. (1988). The Just Community approach to moral education. Journal of Moral Education, 17, 195-208. Power, C., Higgins, A., & Kohlberg, L. (1989). Lawrence Kohlberg's Approach to Moral Education. New York: Columbia University Press. Stams, G.J. (1994). De sociomorele competentie van scholieren. [Secondary school students’ sociomoral competence.] Unpublished master thesis. Leiden: Vakgroep Algemene Pedagogiek, RUL. Teo, Th., Becker, G., & Edelstein, W. (1995). Variability in structured wholeness: Context factors in L.Kohlberg’s data on the development of moral judgment. Merrill-Palmer Quarterly, 41, 381393. Walker, L.J., De Vries, B.J., & Trevethan, S.D. (1987). Moral stages and moral orientations in reallife and hypothetical dilemmas. Child Development, 58, 842-858. Zwart-Woudstra, H.A., Meijer, T., Fintelman, M., & Van IJzendoorn, M.H. (1993). Vragenlijst Sociale Relaties. (Dutch translation of the Sociomoral Reflection Measure - Short Form) Leiden University, Center for Child and Family Studies.

16

Table 1: Verbal and written nstruments for moral competence, practical judgement and moral atmosphere Constructs

Interview

Questionnaire

Moral competence

Sociomoral Reflection Measure - Short Form (SRM-SF)

Sociomoral Reflection Objective Measure - Short Form (SROM- SF)

Practical Judgement Content of the Norm - Self

School Moral Atmophere Interview School Moral Atmosphere (SMAI) Questionnaire (SMAQ)

Stage of the Norm - Self

SMAI

SMAQ

Content of the Norm

SMAI

SMAQ

Stage of the Norm

SMAI

SMAQ

Stage of Community

SMAI

SMAQ

Moral Atmosphere

17

Table 2: Hypotheses investigated and instruments used in four student samples. Study

Number of Selection of students Number of students Schools

Hypothesis

Instruments

Study 1

144

12

Prominent students

Hypothesis 1

SEMAI:open interview

Study 2

144

12

Randomly selected students per class

Hypothesis 1

SMAI:standardized interview

Study 3 MTMM -study

120

16

Randomly selected students by Hypotheses randomly selected class within 1,2,3, 4 two grade levels

18

SMAI:standardized interview SMAQ:written questionnaire SRM-SF:interview SROM-SF:questionnaire Control scales (written): Verbal Intelligence Social desirability

Table 3: Moral atmosphere concerning ‘Valuation of the school’ in five secondary schools as assessed by different instruments (etnographic interviews, structured interviews and questionnaires) using different student samples and administered at different times. school A

school D

school H

school J

school L

1/0: Instrumental extrinsic / rejection - It's not boring i n this school, an d it's not nice either. It's just a school. - You get a good education at this school. - If you have a free period you can hear everybody shout fo r joy. - If this school caught fire, everyone would cheer. - About 80% of the students hate the school. - If a teacher was ill, the students would never ever send him or her a card.

1/2/3 (4): Instrumental extrinsic/ enthusiastic identification / spontaneous and normative community - Then [this teacher] is telling a great story and everybody is fed up when the bell rings. - [As a student] You should do something for your school. - The school is made by the students. - It was a beautiful sight, this great unity. Everybody sat down. That meant that we were all behind it. - If this school was closed, all students would go on strike. - [As a student you should not play truant] because you have a contract with the teacher, so to speak.

Construct: Valuation of the school Level 1 (2): Instrumental extrinsic/enhusiastic identification - At this school you get a good education. - Teachers keep good records of students’ achievements. - You have to work hard and you get a lot of homework. -

Level 3 (1): Spontaneous community - Teachers and students have a good contact with each other. - It’s very social; everyb ody knows everybody; contacts are much more personal than in other schools. - When a teacher was ill, everyone sent him a card. - Teachers think it’s important to know students’ opinions. - If somebod y is bullied , some students stand up for this person.

3 (1): Spontaneous community - Because it’s a small school everyone knows everyone. - Newcomers are immediately accepted. - Students stick up for each other. - Extracurricular activities are great; there’s always something going on. - Most teachers are nice and explain their lessons well, but some cannot keep order and are bullied. - It’s a very good school, I’m proud of it.

Rank order of the schools as to the overall quality of moral atmosphere, including the other constructs, based upon ethnographic interviews 4

2

3

5

1

Rank order of the schools as to the overall quality of moral atmosphere, including the other constructs, based upon the structured interviews 3

2

4

5

1

Rank order of the schools as to the overall quality of moral atmosphere, mean z-score on this quality of moral atmosphere based upon written questionna ire. Mean score on moral competence (based upon the SROM-SF) is added. 3 M moral atmo sphere = 0 ,7 Mmoral competence = 302

2 M = 1,5 M = 313

4 M = -0,5 M = 302

19

5 M = -1,2 M = 288

1 M = 2,1 M = 320

Table 4: Mean scores for students’ stage of moral reasoning on three constructs and for content of the norm on two constructs, using both interview and questionnaire measures (N=120). Moral Competence

Practical Moral Judgement

Stage of Stage of reasoning reasoning M =298 M =260 sd=26 sd=26

Moral Atmosphere

SMAQ questionnaire

SMAI interview

SMAQ questionnaire

Stage of reasoning on the norm: M= 245

Stage of reasoning on the norm : M= 255



Content o f norm: M= 74

Content of norm: M= 62



SRM-SF SROM-SF SMAI interview questionnaire interview

Comm unity: M=256

Comm unity: M= 261 Note: Scores are multiplied by 100

20

Table 5: Correlations between Moral Competence, Content of Practical Judgement and Atmosphere, and Moral Stage Atmosphere in School (corrected below diagonal, uncorrected above), N=120 MC Moral Competence

C - PJ

C-A

Stage

.23

.35

.22

.68

.43

Content-Practical Judgement

.20

Content-Atmosphere

.29

.57

Stage Atmosphere

.20

.42

21

.63 .57

Appendix Figure 1: Observed variables are enclosed in square boxes: MCV and MCW (Moral Competence Verbal and Moral Competence Written); ASV and ASW (Moral Atmosphere Stage of the Norm Verbal and Written); ACV and ACW (Atmosphere Content of the Norm Verbal and Written); CSV and CSW (Practical Judgement Content of the Norm-Self Verbal and Written). The latent variables are enclosed in circles. The measurement error is not enclosed. Causal links are represented by single-headed arrows. Normalized equations.

22

Figure 1: Observed variables are enclosed in square boxes: MCV and MCW (Moral Competence Verbal and Moral Competence Written); ASV and ASW (Moral Atmosphere Stage of the Norm Verbal and Written); ACV and ACW (Atmosphere Content of the Norm Verbal and Written); CSV and CSW (Practical Judgement Content of the Norm-Self Verbal and Written). The latent variables are enclosed in circles. The measurement error is not enclosed. Causal links are represented by singleheaded arrows. Normalized equations.

23

Appendix 1: Validity of the measurements Before discussing the results with respect to testing our hypotheses, it is important to note that validity of measurements on the constructs involved is a necessary precondition for testing a theory on the relation between these constructs. In this case, the validity of the measurements deserves special attention since the operationalization of moral atmosphere and practical judgement is troubled by their conceptual similarity. For measuring practical moral judgement and moral atmosphere the same dilemmas were used, the difference being that in the first case the perspective from the individual him or herself is asked for, while in the latter the perspective from the majority of the institutional members is being investigated. Statements on these perspectives were sampled in close connection with each other in the studies carried out so far. In our study, practical moral judgement and moral atmosphere statements were also sampled in this traditional way. Until now, no validity study has been reported on this distinction between practical judgement and moral atmosphere statements. Therefore, part of our research was designed to establish explicitly the convergent and discriminant validity of our constructs, resulting in a more complicated (multitraitmultimethod) overall design for testing hypotheses two and three than otherwise would have been necessary. Here, we will not fully exploit this design but only summarize the convergent and discriminant validity of the instruments for ‘practical judgement’ and ‘practical moral reasoning’, ‘moral atmosphere’ and ‘moral reasoning competence’. Full detail on the design and results of the validity study can be found in Brugman, Tavecchio, Van Os, Høst, Meijer, & Roosmalen (1995). To assess the validity of our constructs, a multitrait-multimethod study was carried out on one sample. Some instruments were then used to further test the theory. The central idea of a multitraitmultimethod study is to measure all constructs using more than one method. In this particular example we measured the constructs ‘moral competence’, ‘practical moral judgement’ and ‘moral atmosphere in school’ using both a structured interview and a questionnaire (see Table 1). Then, convergent validity can be established by verifying that measurements of the same construct measured by two different methods are similar. Discriminant validity can be established by verifying that two different constructs measured by the same method are distinct. The convergent and discriminant validity of the constructs moral competence, practical moral judgement and institutional moral atmosphere have been investigated in the MTMM study (Brugman et al., 1995). Both a qualitative approach (using the Campbell & Fiske criteria, 1959) and a quantitative approach (using a Confirmatory Factor Analysis, CFA, see Bentler, 1989) were used to study the validity. In general, the distinction in the constructs for practical moral judgement and institutional moral atmosphere did not meet the criteria of discriminant validity, with the exception of the construct ‘Content of the Norm’. Correlations between practical judgement and atmosphere proved too high to maintain a distinction in separate constructs; especially the practical judgement and atmosphere perspectives in Stage of Community correlated highly (Pearson r=.81 for Stage 1 and 2, and .80 for Stage 3 and 4). However, they did reflect a systematic difference in average score, which reflects some discriminant validity. For those constructs not meeting the validity criteria for the distinction between practical judgement and perceived atmosphere, the discriminant and convergent validity was studied by combining the practical judgements and atmosphere measurements. Using these combined measurements, the overall discriminant and convergent validity for Moral Competence and Content of the Norm was moderate and the discriminant and convergent validity of the constructs Stage of the Norm and Stage of Community was weak. The Stage of Community scales showed a low but significant correlation with the Social Desirability scale (.26). As regards Moral Competence (SROM-SF), the convergent validity was comparable to those reported by Basinger and Gibbs (1987) and Gibbs et al. (1984) for homogeneous age groups. For the moral atmosphere and practical judgement constructs no comparable studies have been carried out. However, the lack of discriminant validity suggests that research in this domain should not only score the statements reflecting the perspective on self (practical judgement) independently from 24

statements reflecting the moral atmosphere perspective of the majority, but should also gather these kinds of statements separately. To enhance differentiation, questions may be asked about differences between actions and reasonings of oneself and those of others. The gap between moral atmosphere and practical judgement can at least be partially explained by social desirability effects on practical judgements. However, a relatively substantial positive relationship between Social Desirability and Stage of Community was found. These findings raise questions about the interpretation of the effect of moral atmosphere on practical judgement, and may raise questions also about the educational success of the Just Community approach on Stage of Community. These measures were not corrected for the effects of social desirability. In the results presented in this paper, constructs lacking discriminant validity for this distinction will only be used for exploiting either the practical judgement or the atmosphere measurement.

25

Appendix 2: Confirmatory Factor Analysis We fitted a CFA-model to estimate the variance and covariance of the combined measurements for each construct. Then the correlations between the latent variables (constructs) were as shown in Figure 1. ----------------------Appendix Figure 1 ---------------------The fit of this model proved only to be satisfactory (Normed Fit,=.95, Nonnormed Fit=1.04, Comparative Fit=1.00, df=13, p=.67) if we allowed for covariance between the measurement error of the written instruments for the practical judgement on Content of the Norm and the perceived moral atmosphere on Content of the Norm (shown in the figure by the arc between E6 and E5). This covariance indicates a substantial relation on the written instrument (SMAQ) between practical judgement and perceived atmosphere on Content of the Norm that is not fully explained by the relationship between the two latent concepts. We suspect this to be the result of the high visual integration of the two measurements in the instrument. In Table 5 we show the estimated correlations between the latent concepts, both uncorrected and corrected for this anomaly. ----------------------------------------------------------------Table 5: Correlations corrected and uncorrected ----------------------------------------------------------------Both the uncorrected and corrected correlations show that the practical judgement on Content of the Norm is most strongly related to the perceived moral atmosphere on Content of the Norm. The structural model that was fitted, was corrected for the MTMM measurement model by including two latent variables for the methods, one for the interview instruments and one for the written questionnaires. The corrected model is shown in the text Figure 1 (the measurement components are excluded).

26

Notes 1. Ackno wledge ments This study was supported by grants from the Dutch Foundation for Scientific Research (NWO) and the Dutch Foundation for Educational Research (SVO). The authors wish to thank Marinka Fintelman and Tineke Meijer for holding the interviews and coding the tra nscripts of the S ociomo ral Reflection Measur e - Short Fo rm, and to Drs. Mechteld van Roosmalen for coding the transc ripts of the M oral Atmo sphere Inte rview. W ithout their assistance this study could not have been carried out in its present form. Inquiries and requests for reprints should be sent to the first author.

2. A total of 144 students, 12 of each school was interviewed. However, when we structured the materials of the interviews it became clear that using a random selection of 8 interviews by school resulted in an information basis that could be hardly improved upon by using the last four interviews. We restricted ourselves to the information of 8 interviews.

27