Testing and Measuring the Impact of Character Education on the Learning Environment and its Outcomes

The Project to Test and Measure the Impact of Values Education on Student Effects and School Ambience (T&M) came toward the tail end of a series of practice and research-based projects that ran under the umbrella of the Australian Values Education Program (AVEP) (2003-2010). AVEP was an Australian Government initiative that developed as part of a revision of policy concerning the role of values education as a proper goal for public education. The National Framework for Values Education in Australian Schools (Australian

Department of Education Science and Training [ADEST], 2005a, p 8.) defined values education as “any explicit and/or implicit school-based activity which promotes student understanding and knowledge of values, and which develops the skills and dispositions of students so they can enact particular values as individuals and as members of the wider community.” Thus, while values education borrows conceptually and in research terms from related fields of citizenship education, character education and moral education (Arthur, 2003; Berkowitz & Bier, 2005; Watson, Bat-tistich, & Solomon, 1998), it is deemed by the authors that values education is a more comprehensive term with wider and more holistic implications for effective teaching and schooling, and student behavior. Unlike in many European and Asian countries, such forms of education have endured a patchy if not divisive history in Australia. This article will foreground the AVEP with some of the relevant educational history, outline AVEP’s main components and results and conclude with a detailed summary of the T&M project and its significance.

The history of moral education in Australian schools is one of contention and division. The belief or suspicion that it was overly tied to a religious agenda rendered it an area of school education that was never satisfactorily dealt with. The combined influence of 19th century secularist assumptions and the politics between six colony states preparing for nationhood (Atkinson, 2014), and furthermore the politics between the disparate colonies and divided religious forces, made for difficult educational policymaking around anything but the most incontestable learning areas. Hence, we find a divided education lobby (Barcan, 1976, 1982) and the further tendency to associate anything smacking too much of morality as being tied to the influence of the churches and so to be regarded by public educators as inimical to good educational practice (Lovat, 2018; Lovat & Clement, 2008). As an example, the New South Wales (NSW) Public Instruction Act of 1880 (NSW, 1912) appears to juxtapose sentiments toward normal citizenry formation within a secular education, including even forms of acceptable religious education, with an oppositional “dogmatical or polemical theology” (p. 15).

Partly because of the strength of public (government) schools positioning so avowedly as bastions of secularist education, the Catholic Church absented itself from the act (and similar ones across the six colony states), with the Catholic Archbishop of Sydney of the time, Roger Vaughan, declaring the new public system to be “godless,” a secularist option that Catholic parents must avoid lest they contribute to the immorality and lawlessness that would be its inevitable legacy (cf. O’Farrell, 1985). Granted the rampant sectarianism of the times, such a view was bound to influence not only the Catholic audience for which it was principally intended but to harden the secularist view of public education’s advocates as well. By nailing Catholic education’s colors so firmly to the moral mast in the way Vaughan did, he effectively forced the public system to react. No doubt for reasons beyond Vaughan’s influence alone, public education moved further and further toward a philosophy of moral or values neutrality (ADEST, 2005a), at least in any sense that these forms of education might deviate public education from its essential purpose to be “free, compulsory and secular” (NSW, 1912), its own de facto values charter (Grellman, 2012). Subsequent attempts to change this, such as seen in the proposed General Religious and Moral Education syllabus (NSW, 1964), have been fiercely resisted to the point that, in this particular case, a government minister actually lost his job over the furore. Ironically, the resistance came most fiercely from the oddly combined forces of churches and teacher unions.

The rediscovery of the role of moral content and pedagogy came late in the twentieth century with, for example, the combined public education authorities declaring in the National Goals for Schooling in the Twenty-first Century (Ministerial Council on Education, Employment, Training and Youth Affairs, 1999), that: “Schooling provides a foundation for young Australians’ intellectual, physical, social, moral, spiritual and aesthetic development” (Preamble). The statement, along with the tenor of the entire document, struck a chord with many and was castigated by others who interpreted it as essentially the work of a conservative government looking to impose White, male values on others (cf. Ailwood et al., 2011). Certainly, the statement was itself contentious in the context of a public system that had been so avowedly secularist and so was bound to stir some enmity. At the same time, it did signal some shift in the assumptions that had been made at the founding of Australian education and this showed in subsequent documents and innovations, many of which survived the change of government to one more socialist in orientation. In this sense, it seemed the enlarged perspective on the needs of education was beyond mere ideological leanings. For instance, in 2008, under a very different government, the Declaration on Educational Goals for Young Australians (MCEETYA, 2008) stated:

Among the goals is one that states:

In a word, this declaration makes it plain that public schooling should be constituted of an environment that encourages, supports and nurtures the moral development of its students. One senses there is more than the mere expediency of changes in society from homogeneity to heterogeneity, including a heavily multicultural component, although this clearly would have been a motivating factor. Moreover, one sees the influence of a spate of recent research in neuroscience, neuropsychology and neuro-biology pertaining to learning and development that makes explicit reference to moral intelligence as an inherent feature of human development (Churchland, 2012; Sternberg, 2004) and moreover that moral exploration is an especially powerful way in which vital cognitive functions essential to educability are enlivened (Narvaez, 2013, 2014). One can see that it was far from a mere ideological or matter of political preference that the early part of the 21st century saw efforts toward recovering the gap created in Australian public education around moral education. The AVEP represented a high point in these efforts.

The AVEP began with a pilot Values Education Study in 2003 (Australian Department of Education, Science and Training, 2003), followed by the development of a National Framework for Values Education (NFVE) in 2005 (ADEST, 2005a). A number of research and practice projects took place from 2005 to 2010, the largest of which were the two stages of the Values Education Good Practice Schools Project (VEGPSP) (ADEST, 2006; Australian Department of Education, Employment and Workplace Relations, 2008) and the T&M (Lovat, Toomey, Dally, & Clement, 2009) followed on as an attempt to apply more rigorous research methodology to select components of VEGPSP. A summative project, Giving Voice to the Impacts of Values Education (Australian Department of Education, Employment and Workplace Relations, 2010) rounded off the Program and the funding.

The VEGPSP implementation was guided by the National Framework (NVFE, ADEST, 2005a) which aimed to ensure that Australian children are educated in an environment where essential values, such as respect, tolerance and compassion, are incorporated in the school’s policies and practices, taught explicitly across key learning areas and demonstrated implicitly by staff. The NFVE described a three-way implementation of values education comprising: process—learning values through interactions with others, notably through cooperative learning activities; content—explicit teaching and learning about the nature and meaning of values; and application—applying values to solve academic or interpersonal problems. The NFVE recognized that education is as much about building character as it is about academic progress. While the guiding principles of the NFVE provided a unified focus on transmitting and fostering values and strengthening students’ resilience, schools were also impelled to develop their own approach to values education in partnership with their local school communities, including students, families and teachers. All Australian schools were provided with a resource kit containing teaching resources and professional learning materials to support the implementation of values education (ADEST, 2005b).

Three hundred sixteen schools, involving some 100,000 students, 5,000 teachers and 51 university researchers were involved in the VEGPSP. Fifty-one projects, grouped in cross-sectoral clusters, followed the principles of the NFVE in constructing an intervention, in partnership with families and their school community, that focused on providing a safe and supportive learning environment where values were embedded in all school procedures, rules and curricula. Each project was evaluated progressively and summatively by an appointed researcher in conjunction with the VEGPSP research team. The findings illustrated the connection between values education and good practice pedagogy, with positive effects being demonstrated across a range of educational goals, including emotional, social, moral and academic goals. Many of the reports from the school-based projects identified a greater sense of calm and improved behavior and communication among students and between students and teachers. Reports spoke of enhanced reflectivity on the part of students, greater responsibility being demonstrated over local, national and international issues, enhanced student resilience and social skills, improved relationships of care and trust among students and between students and teachers, with students claiming a greater sense of belonging, connectedness and resilience. These features represent the kind of “sustaining classroom climate” that Narvaez (2010) claims is essential to support ethical behavior and academic excellence. In the T&M project, the term “school ambience” was used to describe the social, moral and learning climate that fostered values development and where learning was, in turn, strengthened by values acquisition. In closely related statements, teachers spoke of this improved school climate for learning leading to demonstrated advances in intellectual engagement, more focused work habits and what was termed strengthened academic diligence (Lovat, 2017). Academic diligence is defined as performing tasks with thoroughness and effort and persevering in the face of difficulties (Moore, Lipman, & Ryburg, 2015). Academic diligence, along with self-regulation, social competence, emotional well-being, and educational engagement have been described as essential “nonaca-demic” measures that mediate academic progress (Moore et al., 2015).

The training of teachers throughout the preparatory phases of implementation was informed by recent neuroscientific insights regarding the link between affective and cognitive development and the importance of interpersonal relationships and a supportive school and classroom climate to promote student engagement and learning (Damasio, 2003; Immordino-Yang, 2011). This link had been explicated in NFVE as potentially one between values education and good practice pedagogy (ADEST, 2005a). Consciousness of the link strengthened as VEGPSP rolled out, with increasing anecdotal testimony from participant teachers and researchers around the notion of values education and quality teaching constituting a veritable “double helix” (Lovat & Toomey, 2009). One item of teacher testimony from VEGPSP summarizes the findings most succinctly:

As suggested, the perceived link between values education, school ambience and student academic diligence was the subject of much testimonial evidence in the two stages of VEG-PSP. Across the 4 years in which the project rolled out, the nature of the evidence gradually developed from being largely anecdotal to the collecting of data with greater potential for quantitative and qualitative analysis. This included head teachers, classroom teachers and administrative staff, as well as parents, reporting on contrasts between previous and current levels around factors such as behavior management reports, classwork and homework diligence and quality, classroom order and school attendance/ attrition data for students and staff. Across the four years in which the VEGPSP rolled out, the school reports detailing the impact of values education were consistently and overwhelmingly positive. Although the school reports revealed converging results about the widespread transformational effects of values education, the diverse nature of the data collection and analysis meant that this evidence was largely “anecdotal.” The T&M project aimed to impose a more formalized and rigorous evaluation process involving the integration of quantitative and qualitative data.

A critical issue in evaluating the impact of values education is the lack of validated instruments to measure many of the psychological constructs associated with the implementation and outcomes of values education, as documented in the school reports from VEGPSP. Because schools were tasked with developing their own approach to implementing and appraising their values education initiatives, the outcomes of these projects and the nature of the data gathered as “evidence” of school effects were diverse and the interpretation somewhat subjective. The T&M project involved two experienced qualitative researchers and two experienced quantitative researchers in order to guide the data collection and conduct an objective secondary analysis of the school reports. This deeper investigation by the four researchers aimed to examine whether the results from disparate values education programs could be tested empirically, through a mixed-methods approach, to provide converging evidence to support the often-reported links between values education, quality teaching and improved student effects (Lovat et al., 2009, p. 3). In the T&M study, there was interest in all of the claims being made, with a dedicated focus on understanding how school and teacher factors coalesced in the context of values education to create a learning environment which appeared to promote positive student behavior and academic engagement. Key educational and neuroscientific literature regarding this cognitive-affective nexus is briefly described next.

Since the report by the Carnegie Corporation Taskforce on Student Achievement (Carnegie Corporation, 1996) in the mid-1990s, there has been much educational research showing that effective learning depends as much on the social context of schools and classrooms as it does on academic content and learning activities. Researchers such as Fred Newmann (Newmann & Associates, 1996) and Johnson and Johnson (2009) have identified that an optimal learning environment is built on trustful and supportive student-teacher relationships and respectful and cooperative student-student relationships. The mediating effect of the school and classroom ecology on student engagement and behavior has also been a focus of proponents of social and emotional learning programs. Emotional skills comprise qualities in the intrapersonal domain relating to self-awareness and self-management, while social skills fall within the interpersonal domain and encompass capacities, such as empathy, that are necessary for relating to and interacting effectively with others. The contribution of these skills to self-regulation, social relationships and academic achievement was confirmed in a meta-analysis of 180 longitudinal studies of school-based social and emotional learning programs in the United States by Pay-ton et al. (2007) who reported significant improvements in students’ socioemotional skills (e.g., self-control, problem-solving), attitudes about self (e.g. self-concept, self-esteem), attitudes about others (e.g. attitudes toward aggression), positive social behavior (e.g. cooperating with others), and connection to school, as well as improvements in academic performance as indicated by significant gains in test scores and grades (mean effect sizes of .28 and .34 respectively).

Neuroscientific research also supports the contention that effective learning is underpinned and accompanied by a socially and emotionally supportive learning context. Immordino-Yang and Damasio (2007) have provided neurobiological evidence that suggests that the cognitive processes required for learning, such as attention, memory and decision-making, are profoundly affected by emotional processes and states. These authors emphasize the important role that teachers play in scaffolding students’ social and emotional development and in providing enjoyable and engaging learning activities in a safe and respectful learning environment. According to Narvaez (2010), such environments are characterized by caring and a feeling of community, employing democratic practices and discourses to promote both moral and academic development. The neuroscientific research has also revealed how learning occurs through both explicit and implicit processes. Explicit learning refers to the conscious and controlled committal of new information to long-term memory, while implicit learning is largely unconscious and is established through observation of and interactions with others (Wilson, 2009). Implicit learning, such as teachers modeling care and respect, is a powerful mode of transmitting values because these are “absorbed” without conscious scrutiny or effort.

Both the educational and neuroscientific literature indicate that optimal learning contexts are created from a simultaneous focus on educating the child and instilling positive changes in teachers and the ecology of the school (Zins,

Weissberg, Wang, & Walberg, 2004). This dual approach, both targeting students’ and teachers’ social and emotional competencies as well as systematically intervening to create caring and responsive schools, defines the implementation of values education in the Australian context.

From this research background and the already amassed “anecdotal” evidence regarding the positive effects of values education, three research questions were formulated to guide the T&M project. These were:

1. How is values education being implemented in ways that elicit positive change in teacher practice and student response?

2. How do we find evidence for the many claims based on teacher testimony that values education has a positive effect on: (a) student academic diligence; (b) school ambience; (c) student-teacher relationships; (d) student and teacher well-being; and, (e) parental and family participation?

3. What added impact does the explicit teaching of values have on relevant student effects? (Lovat et al., 2009, p. 3)

A mixed-methods approach, employing a sequential explanatory design (Creswell, Cre-swell, Plano Clark, Gutmann, & Hanson, 2003), was used to integrate quantitative and qualitative data in a process of “indefinite triangulation” which both illuminated different aspects of the phenomenon under study as well as assisting in validating interpretations of the findings (Hammersley, 2008). In this study, quantitative and qualitative data were collected from one cluster of schools (Group A) over two time periods and analyzed. Case study data (Group B schools) were collected during the second time period and were analyzed separately to help explain and elaborate on the quantitative results. The qualitative and case study data helped to refine and explain the statistical results by incorporating more detailed information from the perspectives of a broader group of research participants (Gläser-Zikuda & Järvela, 2007).

The quantitative and qualitative data collected from the Group A schools included pre-and postintervention staff and student surveys to investigate any changes after 1 year of implementation of a values education program and to assess its impact on student behavior, school climate, classroom ambience, pedagogical practices, student academic diligence, and student-teacher relationships.

Qualitative data were obtained from 20 representative Stage 2 schools (Group B). The schools were chosen in order to provide examples from a range of values education projects and to ensure fair representation across states and territories, educational systems and educational levels. Extant case study data from the Group B schools were subjected to a secondary analysis to identify recurring school-reported effects of values education on outcomes such as student behavior, attendance, parental satisfaction with school, student well-being, teacher morale and student performance on state wide tests. A range of techniques was used by the school-based researchers who constructed the case studies. Some researchers used extensive periods in the school, working as “observer as participant’, whereby they immersed themselves in the school over a long period of time with a view to gaining understanding of what the school was attempting in general and in the values education area in particular. Many conducted interviews and focus groups to obtain data about the fit between what the school was undertaking by way of values education and student academic diligence, school ambience, student-teacher relationships, student and teacher well-being, and parental and family participation. Others supplemented such things with statewide testing data and other routinely collected data about school climate, parental views of the school, and the like.

The T&M project employed a secondary or meta-analysis (Timulak, 2009) of the Group B data in order to provide a more rigorous and substantive interpretation of the individual case study reports (Finfgeld, 2003). The analysis was guided by the conceptual framework established through the VEGPSP that suggested a strong link between values education, school ambience and student academic diligence. Following Timulak’s (2007) method of credibility checks, the primary qualitative researcher initially established the five thematic categories identified in the second research question, that is, student academic diligence; school ambience; student-teacher relationships; student and teacher well-being; and, parental and family participation. Two researchers then independently assigned units of text to each category. During this process, these three researchers also looked for emergent common themes in the case study descriptions of the various ways in which the schools implemented values education. The fourth researcher acted as an auditor and checked all steps of the procedure including the selection of studies, appraisal of studies, category descriptions and data analysis.

Participants. The empirical study consisted of the populations of one cluster of eight Australian primary schools from VEGPSP. The schools volunteered to work together through frequent cluster meetings to support each other in introducing an explicit approach to values education and embedding values across key learning areas in the curriculum. The schools were in close geographical proximity and ranged from small to large according to Australian standards. The smallest school had 180 pupils enrolled while the largest school had 550 students with an overall average class size of 25 students. The following analyses include 453 students from Grades 3-6 for whom complete data sets were available for both measurement occasions (about 50% of the population) as well as 112 teachers. The main reason for attrition in the student sample was that postsurvey data were not available for Year 6 students who completed the preinter-vention survey as these students had transi-tioned to high school by the follow-up year. Similarly, students in Year 3 in the follow-up year had not completed the preintervention survey. There were no significant differences in survey response means between the students with incomplete data and those with complete data. Of the total sample, 48% of the students were girls and 83% of the teachers were female.

The measures included student and staff questionnaires relating to student enactment of the targeted values. The surveys were administered at the beginning of the schools’ values initiative and then again 1 year after the program commenced. The survey items were developed by the researchers in consultation with the lead teachers involved in the project and were originally designed to measure student behavior relating to the six targeted values of respect, responsibility, care, compassion, integrity and honesty. However, the factor analysis of the student survey did not support the formation of six unique factors reflecting the targeted values and instead revealed two distinct components. One component described the student’s assessment of their own behavior and the other component described student assessment of their peers’ behavior. The statements describing a student’s or peers’ enactment of particular values were subsumed under these two main factors. The teacher surveys yielded three distinct factors, measuring student inclusive behavior, responsible behavior and student engagement. The details of the exploratory factor analyses which were employed to help define these constructs are provided in Lovat et al. (2009).

The student surveys captured student perceptions regarding two aspects of their own behavior How I Treat Others (a = .85) and Personal Responsibility (a = .80), and two aspects of their peers’ behavior—Peer impact on me (a = .75) and Peer impact on school (a = .75). The scale items and the respective factor loadings are reported in the Appendix. A 4-point Likert scale was employed, with higher scores indicating a more positive response. For questions beginning with “How often do you ...?” the response options were: none of the time; some of the time; most of the time; and all of the time. Teachers had familiarized their classes with the scales to ensure the students understood the scoring process. Students were given a copy of the questions and the teacher also read the questions aloud as students completed the survey. An open-ended question was included in the postimplementation survey, asking students to name a specific value and provide an example of when they had observed or demonstrated that value.

As described in Lovat et al. (2009), the exploratory factor analysis of the teacher surveys yielded three factors. One factor measured Student engagement and the other two factors reflected Inclusive behavior and Responsible behavior. Teachers were provided with a statement (e.g. “Students in my class cooperate with each other”) and rated the frequency of each behavior, using a 4-point Likert scale as follows: “Rarely/Never”; “Sometimes”; “Mostly”; and, “Always.” The alpha coefficients for these scales ranged from 0.73 to 0.79 and the items and factor loadings are reported in the Appendix. The postsurvey included an additional section containing open-ended questions seeking comments on whether teachers perceived any changes in relation to a range of aspects including student behavior, staff practice and attitudes, student-staff relationships, staff-parent relationships, relationships among staff, the ambience of the classroom or playground, and overall school climate.

The quantitative data reported here come from the Group A schools and relate to a pre-post paired samples t test comparison of student ratings of their own and their peers’ behavior and an independent samples t test comparing teacher ratings of student behavior before and after the values intervention. Only 50% of the teachers’ pre- and postsurveys could be matched and the results of a paired samples comparison would not adequately capture the range of teacher views. Because teachers were reporting on a different class of students in the follow-up year, it was important to retain as broad a sample as possible. The use of an independent samples t test for the teacher survey allowed all teacher data to be included and provided a more representative and conservative estimate of the effects. The paired samples comparison using 50% of the teacher sample revealed the same significant findings as those reported here.

The qualitative data from the Group A schools included the teacher and student responses to the open-ended questions in the postimplementation survey and provided more detailed information about the effects of values education from the perspectives of the research participants (Creswell & Plano Clark, 2007). The same common themes emerged from the student and staff surveys. These included student behavior, teacher practice and school and classroom ambience. Representative quotes from each of these themes are used to help explain the statistical results for the Group A schools. The meta-analysis of the Group B case study data was conducted using the five researcher defined themes as well as three emergent themes relating to the implementation of values education, that is, explicit and implicit teaching, schoolwide practices and effective leadership.

The findings from the empirical study conducted in the Group A schools are reported in the Results section. The integrated qualitative data from the Group A and Group B schools are interpreted to answer the three research questions in the Discussion section.

The pre- and postindependent samples t test comparison of teacher perceptions of student be0havior showed statistically significant improvements across two of the three aspects measured, that is, Student engagement (t = −3.89, p = 0.00) and Inclusive behavior (t = -2.313, p = 0.02). Because multiple comparisons were made, the Bonfer-roni criterion was applied in order to avoid Type 1 errors. As recommended by Kinnear and Gray (2000, p. 251) the p value was divided by the number of comparisons (three) and thus a significance level of 0.02 (0.05/3) was set. Although the postimplementation surveys indicated an increase in teacher ratings of Responsible behavior, this scale did not meet the adjusted criterion (t = -2.15, p = 0.03). The pre-post analysis of the student surveys revealed that there was no statistically significant change on the two scales measuring Peer behavior (Peer impact on me and Peer impact on school), though there was a trend for scores to increase. However, there was a statistically significant decrease in scores on the two scales measuring My behavior, that is, How I treat others (t = 6.06, p = .00) and Personal Responsibility (t = 3.13,p = .00).

Thus, the results of the quantitative analyses presented two apparently contradictory findings. While the teachers appeared to see positive improvements in student behavior after the intervention, the students rated themselves (though not their peers) more critically after the program. These results indicate that after 12 months of the values intervention, teachers were reporting that students were demonstrating values related behaviors more frequently. In contrast, students reported that they were demonstrating values related behaviors less often. While, on the surface, the decrease in students’ ratings of their own behavior could be interpreted as an actual decline, the fact that there was no significant change in the students’ ratings of their peers’ behavior suggests that students were not behaving “worse” but had become more self-critical in the postimplementation survey. If there had been an actual deterioration in student behavior (How I treat others), it should have been reflected in the ratings of Peer impact on me, since each student who rated their own behavior more negatively was also being evaluated by their peers. In addition, contradictory evidence to students’ self-reported decline in behavior was also provided by the teacher perceptions which indicated improvement on scales measuring similar aspects of behavior. The teacher rated Inclusive Behavior scale contained items similar to the student-rated scale How I treat others (do kind things for others, make friends with children who are different) while the teacher-rated Responsible Behavior scale also contained items similar to those in the student self-rated Personal Responsibility scale (e.g., put rubbish in the bin, take care of belongings).

When combined, these findings suggest that after the explicit teaching of values and through exposure to the self-reflective practices that teachers introduced, the students had become more self-aware and self-critical and were perhaps being more “honest” or more realistic in their self-judgments. The student and teacher comments reported below are representative of the student and teacher samples with almost all respondents describing positive effects of values education on student behavior, teacher practice and school and classroom ambience. The quotes were chosen to reflect a diversity of schools and grade levels. The quotes were initially selected by two authors and verified as representative of the sample by the other two authors.

The student responses to the open-ended questions in the postimplementation survey, which asked for an example of an occasion when the student had demonstrated or observed one of the values, indicated that learning what the values meant and how they could be enacted appeared to be associated with greater self-awareness and self-examination, leading to more desirable behavior. Following the quotes, the first digit in the ID code indicates the school and the second digit indicates the year level.

These student comments indicate evidence of self-regulation and show students grappling at a deeper level with the “moral choices” that confront them on a daily basis. It suggests that the students are reflecting on their first impulses, evaluating their behavior and then attempting to act in accordance with the values they have learned. Such “internal struggles” may not be visible to others, nor measurable, but represent a growing “wisdom” that ultimately leads to more self-regulated behavior. Values education appeared to raise students’ awareness that their first impulses may not always be the best course to follow and made them more conscious of evaluating the consequences of possible actions and responses.

Teacher comments also provided indications that students were taking greater responsibility for managing and controlling their own behavior. The teachers made frequent reference to the importance of establishing a shared “values language” across the school and of setting clear expectations about the kinds of behavior that the values represented. In most schools, the school rules were revised and rewritten to reflect a values focus. Rather than prescribing what should “not be done,” the revised student welfare policies explicated the school’s values and these were referred to when interpreting and addressing any instances of inappropriate behavior. Rather than admonishing students for “breaking the rules’, students were encouraged to reflect on “the value they were not showing” and to identify ways in which they could apply that value. Equipping students with clear and positive expectations as well as the words to express and reflect on the impact of their actions on themselves and others not only encouraged appropriate behavior but also helped students understand how they should treat each other. Teachers explained how these strategies not only reduced conflict in the playground, but also empowered students to attempt to resolve their own disputes:

As is evident in the latter comment, teachers and students remarked on the positive changes in student-student interactions, which were most evident in the playground but which also affected the learning climate of the classroom.

Both the student and teacher comments revealed that the school climate had improved in a number of aspects owing to the changes in social dynamics emanating from the values education program. There were numerous student comments to suggest that the playground was a calmer, more caring and cooperative environment than before the values program.

Student and teacher comments also suggested that improved relationships among students contributed to a more cooperative and productive learning environment in the classroom.

Although the results of the pre-post comparison did not reveal a statistically significant increase in the postintervention teacher ratings of students’ Responsible behavior, the teachers’ comments generally indicated that, overall, students appeared to be taking more responsibility for classroom routines and this appeared to be contributing to more independent learning and greater self-motivation:

The association between students’ improved behavior and greater application to their classwork was also evident in the moderate postintervention correlations of student engagement with inclusive behavior (r = 0.34) and responsible behavior (r = 0.45). These findings support the contention that classroom contexts which encourage and promote students’ social and emotional skills also optimize student learning (Brock et al., 2008).

Teachers often attributed the observed improvements in students’ self-regulation, social interactions and school and classroom ambience to the whole-school approach and to the greater consistency in their own and other teacher responses to both appropriate and inappropriate student behavior. The teacher comments on the open-ended questions in the postsurvey identified that their involvement in the project often altered the way teachers interacted with their students. For some, this meant being more receptive to the students’ points of view, being “more aware of listening to students” and “allowing time for two-way communication.” Teachers also mentioned that they were more conscious of being role models for children and made greater efforts to demonstrate the values in their teaching practice and in everyday interactions and to model the kinds of behavior they expected from their students:

A number of comments also indicated that values education had either reaffirmed long-held beliefs or helped teachers come to new realizations about the importance of their role in nurturing and supporting students, particularly those from disadvantaged backgrounds:

Values education appeared to help teachers reappraise the way they treated students and the way they viewed their role as a teacher. As can be seen in the comments above, these teachers are recognizing the importance of assuming the dual role of both educating and socializing children (Dally, 2010). The processes involved in implementing an explicit whole-school approach to the teaching of values seemed to divert teachers from applying directive or punitive responses to student misbehavior to the adoption of more proactive and positive strategies.

As part of the explicit teaching approach, each value was discussed at school assemblies and individual students were either publicly acknowledged or given an award for demonstrating a particular value. In class, students were encouraged to describe ways of enacting the values and to record or report instances when they had shown or seen a particular value. In addition, teachers looked for daily opportunities to “notice” when values were being demonstrated:

A noticeable change was also evident in teachers’ reactions toward students with recalcitrant problematic behaviors. Some teachers appeared to have adopted a more constructive and compassionate approach and reported that they now used the “language” of values in dialogue with students as a reflection point for students to consider how they could act in accordance with the school’s values. Rather than blaming students for poor behavior, teachers were using instances of inappropriate behavior or student-student conflict as “real-life” problem-solving opportunities to help students analyze their behavior and its consequences in terms of values:

Such statements reflect Noddings’ (2008) concept of “confirmation,” wherein effective teacher scaffolding involves the identification of positive aspects of behavior, even when actions are inappropriate or undesirable. Embedding values into the curriculum and into student welfare and discipline policies served to direct teacher attention away from judgment and punishment toward practices aimed at developing students’ capacities to demonstrate behavior consistent with the values that the school community espoused. Thus, the incorporation of values as part of the school’s “core business” legitimized and validated the time and attention that “educators” devoted to these “socializing” aspects of their role.

The discussion draws on the findings reported above from the empirical study integrated with the analysis of the Group B schools’ case-studies reported in Lovat et al. (2009). The discussion is organized around the three research questions.

1. How is values education being implemented in ways that elicit positive change in teacher practice and student response?

In the Group A schools, values education was implemented in different ways across the cluster but all schools adopted both implicit and explicit approaches. Implicit practices involved embedding values in student welfare policies and teachers modeling and scaffolding the schools’ values. Explicit teaching centered on clearly defining and discussing the values and creating authentic learning experiences where students could put the values into practice. Explicit teaching was also incorporated through songs and school displays about values, as well as presenting Values Awards at assembly, and using reflective practices with students to assist them in gaining greater insight into their own actions and the actions of others. The case-study schools identified a set of effective implementation practices that included the following features:

• Values Education must be seen as core business;

  Values must be scaffolded by supportive schoolwide practices;

  Values must be taught in an explicit way in the classroom and through other media (e.g., Assemblies, etc.);

  Values Education must be allied to “real world learning” (involving deep personal learning);

  Values Education must be reinforced through routine, visual positive media (e.g., through posters and/or other emblems); and,

  Values Education must be allied to expressed high standards for overall performance and achievement.

The importance of effective leadership was also noted, with successful implementation linked to the guidance provided by “a person with significant influence within the school” (Lovat et al. 2009, p.74).

1. 2. How do we find evidence for the many claims based on teacher testimony that values education has a positive effect on: (a) student academic diligence; (b) school ambience; (c) student-teacher relationships; (d) student and teacher well-being; and (e) parental and family participation?

There was substantial quantitative and qualitative evidence from the Group A schools suggesting that there were observable and measurable improvements in students’ academic diligence. Aspects of academic diligence included increased student attentiveness in class, a greater capacity to work independently as well as more cooperatively, greater care and effort being invested in schoolwork and students assuming more responsibility for their own learning as well as classroom “chores.” The Group B schools reported similar student and teacher testimony about students spending more time on task and applying greater effort in class with at least three case study schools confirming that this increased student engagement had translated into improved results on statewide tests (Lovat

et al. 2009, p.74).

The Group A schools provided evidence of improved student behavior, both in the classroom and in the playground, with teachers and students observing more incidents of caring, honest, respectful and responsible behavior leading to more cooperative student interactions and contributing to a safer and more harmonious school environment. The Group B case studies also provided clear and unequivocal reports about the fit between values education and the “tone” of the school. Since the schools’ involvement with values education, virtually all the case study schools reported the following improvements:

• conflict among students decreasing or being managed more constructively;

  students demonstrating greater empathy, honesty and integrity;

  more tolerant and cooperative student interactions;

  safer and more harmonious classrooms and playgrounds;

  greater kindness and tolerance among students;

  students actively seeking to include peers without friends;

  students taking greater responsibility with school equipment and routine tasks; and,

  students treating the school buildings and grounds “with respect.”

While there were no quantitative data collected on student-teacher relationships in the Group A schools, there was some evidence from the students’, and particularly the teachers’ responses to the open-ended survey questions, suggesting that both students and teachers were behaving more “respectfully” toward each other. Teachers described improved relationships with students as evidenced by a “rise in levels of politeness and courtesy,” “open friendliness,” “better manners,” “offers of help,” and students being “more kind and considerate” to teachers as well as their fellow students (Lovat et al., 2009, p. 10). Only four Group B case study schools presented data on student-teacher relationships and these schools reported improved and positive patterns in both student interpersonal relationships and student-teacher relationships. Teachers found that the values focus embedded in their curriculum units “developed students’ social skills by increasing cooperation, empathic character, self-management and self-knowledge which, in turn, led to more supportive and safer learning environments” (Lovat et al, 2009, p. 79).

In the Group A schools, while there was no single instrument employed to measure “well-being,” the pre-post comparisons of the student and teacher scales and the analysis of the open-ended questions on the student and teacher surveys provided converging evidence to imply that both student and teacher well-being were impacted on in a number of ways. Student well-being was positively impacted by the improved interpersonal relationships among students resulting in a safer and more cooperative playground where students were taking better care of each other and of the school grounds. Students’ intrapersonal skills were also developed through the explicit teaching of values and teacher recognition and reinforcement when values were demonstrated, thus providing greater incentive for student self-appraisal and self-regulation. With regard to teacher well-being, teachers noted that students were now more likely to smile at or greet them and offer help when it was needed. It appeared that the increased respect from students contributed to teachers believing the school was a “happier,” “friendlier,” and more enjoyable place to work (Lovat et al.,

2009, p. 12)

The Group B case studies also indicated a generally higher sense of student well-being but acknowledged that there were influences other than the values education program that may have contributed to this. Some schools had adopted additional programs specifically targeting student assertiveness, confidence, self-concept, and positive thinking skills. With regard to teacher well-being, the case study summaries suggested that the organizational structures and leadership approaches around the values project improved staff morale dramatically. This created a dynamic in which student well-being both supported and was supported by increased teacher well-being.

The effect of values education on parental and family participation in school life was scant and ambiguous. The teacher surveys in the Group A schools indicated that most teachers felt the parents supported the program and that the common focus and common language shared between home and school was mutually beneficial. However, some teachers expressed frustration when they felt parents were either disengaged or unsupportive and that the onus to teach values rested solely with the school. The Group B case studies were similarly lacking in evidence of any increased family involvement as a result of values education.

1. 3. What added impact does the explicit teaching of values have on relevant student effects?

The mode of teaching values in an explicit way appears to both require and create a caring, supportive and positive school culture. The data reported in the T&M project indicated two main findings. First, the Group A teacher ratings of student behavior indicated that there was a statistically significant improvement in teacher perceptions in two of the three aspects measured, namely, student engagement and inclusive behavior. Teacher comments from both groups of schools also indicated that the values education program resulted in most, but not all, students working more cooperatively, being more attentive in class, taking better care of their own belongings and being kinder to and more considerate of others. The improvements in “values-related” behavior appeared to create more productive classroom learning environments and more harmonious playgrounds. Similar findings are evident as a result of Social and Emotional Learning programs where improvements in student empathy and interpersonal management skills have contributed to more effective educational environments (Elias, Parker, Kash, & Dunkeblau, 2007). In addition, embedding and scaffolding values education into other curriculum areas and providing opportunities for students to put the values into practice in everyday interactions meant that students were supported to “know the good” as well as “do the good” (Berkowitz,2002). These changes are also consistent with an enhanced sense of a caring school community which supports both the academic and social progress and moral development of students (Noddings, 1988, 2003, 2008; Solomon, Battistich, Watson, Schaps, & Lewis, 2000). This enhanced sense of community would appear to have been facilitated through teacher pedagogical scaffolding of values which, in turn, created opportunities for peer-to-peer scaffolding of values (Horn, Daddis, Killen, 2008; Turner & Berkowitz, 2005).

The second major finding was that the teacher comments from both Group A and Group B schools suggested that what was “different” in the schools after the introduction of values education was that a “common language” had been established to describe expected behavior and that “everyone,” including students and parents, was aware of the school’s expectations for behavior and had acquired the language to describe and understand those expectations. Thus, the whole school approach and the explicit nature of the values education intervention appeared to be important components in the implementation process. The explicit teaching of values and the embedding of values into curriculum learning activities provided opportunities for teachers to clarify unclear or implicit messages about behavioral expectations, and engaged peer-to-peer scaffolding. The teacher comments also indicated that teachers were more conscious of modeling the values in their interactions with students, thus providing an “implicit” message about how values can be enacted. The combination of implicit and explicit instruction appears to be a more powerful learning paradigm than either approach in isolation. As Lickona (1992, cited in Berkow-itz, 2002, p. 60) emphasized, “we need not only to practice what we preach, but we also need to preach what we practice.”

The integrated data revealed teachers indicating that greater self-regulation of behavior and improved interactions between students had led to more harmonious and productive learning environments in which students were demonstrating greater kindness to each other and, at the same time, taking more care and pride in their work. As noted in a number of other studies (e.g., Benninga, Berkowitz, Kuehn, & Smith, 2006; Hawkes, 2009), it seems that the ambience, relationships, self-reflection, and discourse germane to values education have potential to impact positively on students’ academic work habits, without any other contaminating factor being obvious.

The study therefore provided confirming evidence around the many testimonial claims made in the earlier project reports about the impact of values education on overall school ambience. The whole-school approach that typified the implementation of values education in Australian schools engaged all staff, and ideally the parent community, in a collaborative endeavor to define and actualize a shared vision for the school, so resulting in changes in the broader school culture (Lovat, et al., 2009, p. 31).

Findings portrayed plausible reasons why this improved school ambience might well lead to enhanced academic diligence. Greater efforts toward self-regulation and improved student relationships meant that less teaching time was being diverted to behavior management and more time could therefore be devoted to the core business of education. The teachers observed that these new found skills and behaviors, together with the more positive ambience created by them resulted in students taking more control over routine tasks, so adding to their self-confidence and sense of competence and this appeared to lead to more independent learning and increased intrinsic motivation. In turn, teachers reported that students were putting greater effort into their work and trying harder, were more engaged in learning, taking greater responsibility for their learning and working together more cooperatively. In turn, changes in student attitude and application to school work led teachers to raise their own expectations for the students (Lovat, et al., 2009, p.100).

As noted by Holburn (2002), it is difficult to accurately measure interventions which are flexible and responsive to individual situations and contextual demands; however, it is important to deconstruct the elements of phenomena that appear to be associated with socially valid goals, even if such descriptions might not meet the precision generally required for “scientific analysis’. While the cumulative VEGPSP reports and the findings reported here from the T&M project may not meet the requisite level of scientific precision, they do provide compelling and remarkably consistent evidence of how values education impacts teacher and student well-being and serves to create the kind of learning environment most conducive to effective learning. The T&M project revealed the inherent difficulties in relying solely on quantitative data to measure a complex phenomenon such as values education. Future studies investigating the implementation and outcomes of values education in schools should employ both quantitative and qualitative methods in order to provide a clear picture of the complex elements in a changing school culture and to understand the dynamic interactions among teacher, student and school factors. A longitudinal design with more frequent assessment points will help disentangle the reciprocal associations between teacher practice, student behavior, school ambience and academic diligence.

As indicated above, the history of Australian education has been marked by division between public (government) and private (largely religious) schooling. This resulted early on in its history in anything related to moral, character or values education being seen as a natural part of the private school but inimical to the secular imperative of the public school, the latter often taking pride in its purported “values-neutrality.” In spite of late 20th and 21st century government reports that spoke of the need for issues of moral development to be dealt with as part of the charter of any sound education, and regardless of increasing evidence from the neurosciences of the integral importance of moral growth to the individual and to educability, there was nonetheless resistance to the implementation of the AVEP. Predictably, this came largely from the public school sector and its teacher unions but also, less predictably, from areas of the private sector that saw it as an unwelcome infiltration into one of its distinctive domains of influence. This reality rendered the early stages of AVEP with some obstacles and no doubt some of these persisted throughout the project. Nonetheless, the positive influence on so many crucial features of effective schooling, such as seen in the results of the Project to Test and Measure the Impact of Values Education on Student Effects and School Ambience above, has offered lasting testimony to the universal applicability of a values education approach, an approach to schooling that renders the priority of pedagogy to be directed to the value of the child and the values implicit in curriculum content. In a day and age plagued yet again by priorities concerned with academic outcomes, especially those that are seen as essential to international test results, values education stands as an oppositional force reminding policymakers that education is first and foremost a humane enterprise and, ironically, that it is through such an approach that we can achieve the best outcomes, including academic ones.