A comparison between the views of teachers in South Africa and six other countries on involvement in school change

Worldwide, and especially in South Africa, change and decentralised decisionmaking have been topical issues in the provision of education for the past years. It appears that teachers — the key agents in implementing the policies concerned — are largely ignored in the pre-implementation phases, and treated merely as implementers of these policies. The results from an empirical investigation revealed that the teachers in the South African sample expressed an exceptional degree of eagerness to be involved in decision-making and responsibility-taking concerning school change, even in aspects of management that could be considered as the principal’s ‘turf’. Although the views of a group of teachers in six other countries showed very similar result patterns, the sample of South African teachers was considerably more eager to be involved in initiatives of school change and related responsibilities than the teachers in the samples of the other countries. The results are illuminating, taking into consideration the increased workload of teachers, as well as certain other factors. Possible explanations for these observations are discussed.


Introduction
Worldwide educational change has been a topic of discussion for many years.The prominence and scope of interest of this topic are illustrated by the following remark: "… everywhere, it seems, educational change is not only a policy priority but also major public news" (Hargreaves, Lieberman, Fullan & Hopkins, 1998:1).Also in South Africa, change impacting on all aspects of life has been the core of discussion of the past decade or two and will probably continue that way for years to come.The entire education system in particular, which is often seen by politicians and governments as an instrument for social engineering and the creation of economic growth, has been affected rather radically.In this regard Berkhout (2007:405) stated that the restructuring of South African education poses continuous challenges for educational leaders to contribute towards constituting a just and equitable society.
Considerable research has been done on many aspects of educational change.However, research findings have also revealed that as far as implementation of educational change is concerned, change initiatives have frequently failed.The following are a few examples of publications that can be consulted in this regard, namely, research by Hargreaves and Bascia (2000), Ball (1999), Hargreaves et al. (1998), Chikoko (2007), Mortimore and Mortimore (1998), Frederics, Blumenfeld and Paris (2004), Prinsloo (2006), and Leithwood and Jantzi (2000).Various reasons have been indicated for the failure of change initiatives.An explanation often given deals with the tendency that the initiators of change, whether politicians or bureaucrats, do not bear in mind the crucial role that teachers play as implementers of change.It seems they fail to appreciate that involvement of teachers, in decisionmaking and responsibility-taking processes, relates very closely to successful implementation of educational change.It appears that Departments of Education, according to Carl (2005:227), are probably under the impression that they are aware of the teachers' needs.
One aspect of educational change deals with the effect of the change on the work-life experience of teachers, who are the key figures in the facilitation of such change.A Consortium for Cross-Cultural Research in Education, of which I am a member, commenced in 1994 with an international research project on educational change.The aim of this Michigan-based Consortium was, and still is, to gain insight into how different types of educational change alter teachers' work-lives, and how these alterations influence their dispositions in respect of the changes.A number of countrywide tendencies have crystallized from these results.It was ascertained, amongst other things, that the active involvement of teachers in school change proved to be the most prominent variable amongst those included in the study with regard to promoting positive work-life outcomes for teachers, as well as generating teachers' receptiveness and a positive disposition towards change itself (Poppleton & Williamson, 2004:289).It appears that the more teachers are involved in initiating and taking responsibility for school change, the more positive they feel about the change, and the more willing they are to engage in future change.

The teacher as educational change agent
It is evident that, as far as educational reform and restructuring are concerned, the decentralisation of decision-making powers to local and school level has become an international trend (Kruger, 2003:206).In fact, according to Botha (2007:29), for more than 30 years policy-makers, educators, and academics have considered school-based management a key ingredient in school improvement and reform efforts.Therefore, during the past 20 to 30 years, there has been a major shift towards greater self-management and selfgovernance in educational institutions throughout the world (Botha, 2006: 341).South African schools have also become part of this trend (Van Niekerk & Van Niekerk, 2006:85).The notion of decentralization of decision-making powers is apparently based on the assumption that participation of all stakeholders in leadership can play a major role in transforming education (Mosoge & Van der Westhuizen, 1998:73;Singh & Lokotsch, 2005:279), and that the significant changes and reforms demanded of schools can only be attained through devolution of power and through shared decision-making that encourages people to change and to address educational problems (Bradshaw & Buckner in Botha, 2004:239).As a result, more and more countries worldwide are implementing school-based management, and princi-pals are consequently empowered by giving them more authority over what happens in their schools.These principals increasingly exhibit the power to make on-site decisions, even decisions on what should be taught in the classrooms (Botha, 2006:341).
As far as South Africa is concerned, a new constitutional dispensation was established in 1994, and new education legislation was promulgated which stipulates that the government is committed to the development of a democratic system that provides for the participation of all stakeholders with a vested interest in education.According to legislation it is expected from educators, parents, learners, and non-teaching staff to actively participate in the governance and management of schools, with a view to providing a better teaching and learning environment (Van Wyk, 2004:49).The pivotal role to be played by the school principal in a school-based management dispensation is obvious, and the impact he/she has on the tone and ethos of a school is even more crucial than before in the process of building a culture conducive to change.School principals are responsible for creating a collegial environment, which offers educators opportunities to take on participatory leadership roles in a variety of portfolios, and this entails the devolution of power to educators.He/she influences an educator's sense of job satisfaction, morale, loyalty to the organisation, and level of motivation to do well (Singh & Manser, 2008:112).However, it is also true that with school-based management, the responsibilities and workload of school principals are assuming even larger proportions than previously (Botha, 2004:239).This state of affairs must surely impact on the principals' views of teacher involvement in shared decision-making, since the latter management style could be regarded as much more time-consuming than the more familiar autocratic style of management.Support for this notion is found, inter alia, in a relatively recent collection of writings on dilemmas that principals are faced with in their endeavours to engage teachers in the sharing of leadership in schools (Chrispeels, 2004).
Where do South African teachers find themselves in the scenario sketched above?Obviously, doubt exists about the establishment of school-based management in spite of the policies and legislation that have been promulgated in South Africa and the follow-up pronouncements of government, such as the intention of the Minister of Education to introduce legislation that will increase the powers and authority of school principals.According to the Minister, there is a need to re-assert the professional responsibility of principals, and such legislation will very likely enhance the leadership role of the principal and ensure that (s)he becomes more influential regarding matters such as the school budget, recommendation of teachers for vacant posts, and admission policies (Naidoo, 2005).According to Chikoko (2007), numerous studies have reported on the dilemmas of implementing decentralised school governance and educational reform, while Botha (2007:28) claims that, in spite of the widespread implementation of school-based management, it has received only moderate attention locally in terms of the participation of stake-holders, and the impact of stakeholder values on the school-based management process.In addition, Prinsloo (2006:355) asserted that the State, its functionaries, and the organs of the State have been trying to assert themselves to an increasing extent by limiting, or interfering in the authority that can be exercised by school-level governance structures.This is reflected in school governing bodies that are compelled to turn to the courts to stop officials from unlawfully interfering in the management and governance of their schools.It appears that often teachers are regarded as employees whose autonomy must be limited, and every attempt has to be made to ensure that any form of change is as teacher-proof as possible (Norlander-Case, Reagan & Case, 1999:11).
The teachers' desire to be involved is often mentioned in research publications such as Singh andManser (2008), Carl (2005;2007), Ngidi andQwabe (2006), andChikoko (2007) to name a few.For example, Carl (2005: 228) stated that teachers do not wish to be viewed as mere recipients who are to implement the changes, but that they expect to be included in the initial processes of meaningful decision-making where their voices will be heard.In addition, Poppleton and Williamson (2004:289) mention that the more teachers participated in the responsible and initiating roles in school change, the more positive they felt about the change, and the more willing they were to seriously engage in future change.Since teachers in many countries have been faced with the task of continuously facilitating and implementing education reform, that was designed without involving them (Hargreaves & Bascia, 2000), it appears that teachers have a perception that they are being excluded from decision and management processes.Carl (2005:223) found, for example, that teachers' perceptions were that, although they were subject and/or learning area specialists, little attention, if any, was given to their opinions.It seems that, as long as teachers perceive themselves to be the mere recipients who have to implement changes decided upon elsewhere, not only will their professional status but also actualisation of well-meant school changes be placed in jeopardy.

Research problem and aim of the investigation
Without fear of contradiction it can be stated that the teacher is the keystone in the arch of education.We concur with Lenyai (Mathibe, 2007:523) that you can eliminate the finest buildings and the most widely developed curriculum, but leave the learner with an intelligent, cultivated and humane educator and the educational process will continue satisfactorily.On the contrary, provide all the material necessities, but without the educator or with an incompetent one, and the results will be catastrophic.However, researchers such as George, Louw and Badenhorst (2008:140) have reported that results indicated significant levels of dissatisfaction among teachers pertaining to factors related to their workplace.A lack of job satisfaction resulted in frequent teacher absenteeism from school, aggressive behaviour towards colleagues and learners, psychological withdrawal from work, burn-out, and early exits from the teaching profession.In the same vein, Carl (2007:221) asked what was done to South African teachers that their motivation levels and their lack of trust were taking on such crisis proportions.
Resistance to change is viewed by many as a natural human phenomenon due to a 'concern for an expected loss' experienced by those affected by the change.However, Carl (2007:205) mentioned in this regard that the apparent negativity of teachers towards change is not necessarily that they resist change but rather that they are uncertain about what is expected from them.In addition, Chikoko (2007:175) argued that people desire involvement in the making of decisions on matters that affect them, even when they sometimes have no capacity to effectively make such decisions.Therefore, decentralisation should be accompanied by sustainable capacity-building among stakeholders.Linking up with this train of thought, theory and practice have shown that significant taking of responsibility by members within organisational settings increases the likelihood of a healthy institution, and the willingness to participate in future organisational improvements or change.
Another perspective on teacher involvement in shared decision-making is that, whether the change amounts to reform, restructuring, or innovation, it frequently introduces a new layer of responsibilities into the teacher's work (Poppleton & Williamson, 2004:310).Currently, teachers in South Africa have to cope with a workload that has increased significantly over the last couple of years.In research conducted by the Human Sciences Research Council it was found, for example, that more than 80% of teachers believed that their workload had increased considerably since the year 2000 (Rademeyer, 2005: 2).In addition, it must be borne in mind that educators have had very little experience of participatory decision-making seeing that, in the past, principals were generally considered to be the only persons with the knowledge and authority to make decisions (Heystek & Paquette, 1999:19).Complicating the issue of teacher involvement, even more, is the fact that a considerable proportion of South African teachers are not sufficiently qualified or trained, and lack the competencies to either implement the new policies capably or take part in decision-making processes in the school.To embrace transformation, when one is not properly qualified and probably overworked, is demanding too much.
The scenario sketched above leaves one with some doubt as to whether teachers want, unconditionally, to be involved in more responsibilities, even if they are change initiatives that could have a positive influence on their work environment.Therefore, the following research questions were formulated:

•
To what extent do teachers really wish to be involved in decision-making and responsibility-taking as regards processes of school change?• How do the views of a sample of South African teachers compare to those of teachers in some other countries?

Research design
The research commenced with a review of existing literature to establish the nature and limitations of teacher involvement in educational change.This was followed by a cross-country empirical investigation.The latter comprised quantitative surveys, which can be described as exploratory in nature.Structured questionnaires were developed in order to obtain the views of secondary school principals and teachers on the involvement of teachers in responsibility-taking during the implementation of school change initiatives.Samples of principals and teachers were drawn from seven countries.Although the research project was designed to obtain the views of both teachers and school principals, in this article I focus only on a cross-country comparison of the results obtained from teachers.

Data gathering
In order to obtain comparable data from the seven countries, the research team from the University of Michigan took the lead in developing a survey instrument (questionnaire).It comprised 20 tightly focused items, based on the roles and expectations derived from existing literature, as well as from their own experiences.The 20 items represented different possible responsibilities relating to school change in which teachers could be involved.Role expectation theory was important in this process.The primary role expectation of a teacher was defined to encompass any action that has a direct impact on day-to-day classroom activities.Back-translation procedures were employed in the precise construction of the items and country-specific fieldtesting was conducted.Any deviation from the questionnaire by a particular country, except native alterations, was not allowed because it would jeopardize cross-country comparisons.Therefore, an attempt was made to ensure that, as far as possible, the questionnaire made provision for each country's unique educational setting.
In the questionnaires teachers were asked to indicate, inter alia, to what extent they would like to be involved in each of the 20 responsibilities (items).A five-point Likert-type scale, namely, none/little/some/much/very much, was utilized for response purposes.

Sample
Research teams from seven countries conducted the research.These countries were Canada, China, Hungary, Israel, Singapore, South Africa, and the USA.In accordance with the collective cross-country design, it was agreed that each country's sample would consist of approximately 50 teachers or more, from not less than 10 secondary schools.The members of the research teams chose, by reason of their special responsibilities, to concentrate on secondary education as main focus of concern.Each sample reflected the following ratio: one recently appointed teacher, two mid-career teachers, and two senior teachers occupying different roles.Furthermore, the research teams were requested to ensure that the samples of teachers were representative in terms of age, gender, seniority, and socio-economic catchments.In accordance with these guidelines the South African sample of teachers consisted of 176 secondary school teachers of whom 84% were females and 85% were black.They were drawn from urban-suburban areas in Gauteng province by means of purposeful sampling (McMillan & Schumacher, 2006:126) to reflect the inequalities that exist in South African schools, due to the previous political and educational dispensation.Whilst 53% of the respondents were teachers at small schools (schools with an enrolment figure of less than 600 learners), 39% were employed at medium-sized schools and 8% at large schools (attended by more than 1,200 learners).To ensure that the views of experienced teachers were represented, 77% of the respondents included in the sample had more than 10 years teaching experience.The samples of the other countries varied from 48 to 141.

Data analysis
The data analysis commenced with the grouping of the 20 items of the questionnaires into four clusters.The role expectation theory (Hoy & Miskel, 1987) was applied to cluster the items according to their conceptual similarity.The four clusters were considered to be conceptual units and formed four new variables for analysis of the data.It was argued that this process would lead to a reasonable amount of data reduction and clusters would be obtained, each consisting of sufficient items to yield a reasonable reliability coefficient.Furthermore, this clustering made comparisons between teachers' views more meaningful and also reduced the complexity of cross-country analysis.The clusters were labelled Administration and Co-ordination, Human Relations, Teacher Support, and Classroom Activities.The Administration and Coordination cluster, which related to the primary role of principals, consisted of items that dealt with staffing, budgeting, assigning students to classes, and staff meetings.Items in the Human Relations cluster dealt with relations with external groups or teacher-staff-administration relations.The Teacher Support cluster comprised items that dealt with assistance or support of teachers.Items grouped in the Classroom Activities cluster related to curriculum matters, student behaviour, and class schedules.Also included in the last cluster were the issues of implementing and evaluating change since it was assumed that most changes affect the classroom environment and classroom activities.
It was, furthermore, hypothesised that the cluster means would increase in the order in which the clusters were listed in Table 1.In this regard the previously mentioned role expectations of teachers and principles were used to determine the ascending order of clusters.It was argued that, theoretically, it was expected that the positioning of the clusters would correspond to the following four involvement levels along a continuum of role responsibility: • primarily a principal's function; • mostly a principal's function with some teacher responsibility; • a principal's function with major teacher responsibility; and • mostly a teacher's function with some principal responsibility.The statistical analysis of single country data started with obtaining data response percentages, distributional percentages, the means for each item and each cluster, as well as the grand questionnaire means.Subsequently, the data for the six countries were combined so that this group's data could be compared to those of the South African sample.Cronbach alpha reliability coefficients were calculated for the cluster averages and the grand questionnaire means.In order to identify the systematic similarities and differences in result-patterns of the individual countries, a Newman-Keuls test of differences among ordered means (Pagano, 1990:361) was applied.

Discussion of results
A synopsis of the results obtained from the questionnaires, completed by the samples of teachers in South Africa and the six other countries, is presented in Table 1.The means of teachers' responses were calculated for each item, as well as the four item clusters and the questionnaire in the aggregate.In Table 1 the means of the South African sample are compared to the means of the group of six countries (South Africa excluded).For each of the 20 items (responsibilities), the percentage of South African teachers who indicated that they would 'much' or 'very much' like to take part in the specific responsibility is also included in the table.The five-point scale used for response purposes appears at the bottom of the table.
As mentioned in the section on data analysis, the analysis commenced with grouping the 20 items into four clusters according to their conceptual similarity.Each cluster consisted, therefore, of items with a similar theme.It was argued that indices consisting of items with common themes tend to be much more reliable than individual items.The international team also decided that a cluster score would be considered to have sufficient reliability to provide useful results if it obtained a Cronbach alpha reliability coefficient of at least 0.70.This expectation was fulfilled in the sense that reliability coefficients higher than this criterion were obtained for all the clusters (see bottom of Table 1).Likewise, the reliability coefficients of 0.90 and higher, which were obtained for the questionnaire totals, were acceptable.
As mentioned earlier (see section on data analysis) the four clusters were listed, in ascending order, according to the theoretical expectation of their size or teacher involvement level.In this way the theoretical premise allowed the researchers to predict the relative size of the teachers' ratings for the various groups of responsibilities.The results included in Table 1 indicate that the ascending order of the cluster means concurred with the theoretical expectation and listing.This tendency was confirmed in the results of the South African group, as well as the six-country group.The confirmation of the theoretical expectation by the statistical analysis suggested that both the underlying theory and the questionnaire complied with the requirements of construct validity.
The item percentages and item means listed in Table 1 indicate, at first glance, that the teachers in the South African sample expressed a high degree of eagerness to be involved in responsibilities concerning school change.The means of almost all the items and clusters are above 4, bearing in mind that the mean of the five point-scale used is 3.Although the teachers' means for the Administration and Co-ordination cluster are a little lower than the means for the other clusters, the percentage of teachers who indicated that they 'much' or 'very much' wished to be involved in the aspects within this cluster is rather high, except in one item.For the other three items the latter percentage is 76% or higher.This observation is revealing, when one keeps in mind that this cluster, as well as the Human Relations cluster, more than the other two clusters, is linked to the domain that could be considered as primarily a principal's function.It seems that teachers in the South African sample generally want to be involved in the processes of decision-making and responsibility-taking concerning all school aspects; more than one would expect.
A comparison between the results of the South African group and the six-country group revealed that the result patterns of these two groups are very similar.However, in spite of this similarity, it is noticeable that without exception the item, cluster, and questionnaire means of the South African group are considerably higher than those of the six-country group.In order to determine how much value could be attached to this observation, it had to be established whether the differences between the mentioned means are significant.Consequently, a Newman-Keuls test was applied to the cluster means and grand questionnaire means of the seven countries.The results of this application are included in Table 2.The country means as well as the significance of the differences between these means are indicated.
From Table 2 it is obvious that in the case of the South African sample the grand questionnaire mean and all the cluster means, except of the Classroom activities cluster, differ significantly from those of the other countries.The South African means are higher than those of the other countries.Although the South African mean of the Classroom activities cluster is also higher than those of all the other countries, the difference between the means of South Africa and of the USA is not significant.We can therefore, confidently conclude that the South African teachers who were included in the sample are, in general, keener to be involved in school change initiatives and related responsibilities than the teachers in the samples of all the other countries.
Looking at the country means listed in Table 2 one notices a few additional trends that can be added to those already highlighted: Israel, China, and Hungary yielded means that are quite similar, and also Singapore, Canada, and the USA.However, the latter groups' means are higher, which indicates that teachers in these countries seem more eager to be involved in school change responsibilities.As it is beyond the scope of this article, no attempt will be made to provide explanations for these trends.

Concluding remarks
Although the topic of educational change has been prominent during the past decades, research findings have revealed that the successful implementation thereof has frequently failed.Apparently, one of the reasons is that worldwide authorities do not make full use of teachers as a professional source, or they even deliberately exclude them from initiating, planning, and formulating school change.It appears that the role of teachers as key agents in the facilitation of such change is overlooked for the most part or plainly disregarded.However, from another perspective there is doubt whether it can be taken for granted that the majority of South African teachers want to be involved in change initiatives, even if this has a direct impact on their work lives.The workload of South African teachers has increased considerably in recent years and shared decision-making in the pre-implementation phases of educational change would introduce a new layer of responsibilities.Besides that, a considerable proportion of South African teachers are not sufficiently qualified or trained, and lack the competencies to capably implement new policies, all the more to take part in the initiation and decision-making processes in respect of these policies.
With the above opposing perspectives in mind, this research was undertaken to obtain more clarity on how enthusiastic South African teachers really are about being involved in the decision-making and responsibility-taking processes of school change, and how their desires compare to those of teachers in a number of other countries.A questionnaire was developed to gather data in the countries identified.Although such a generic instrument may often appear to be unsuited to the setting of each country, the results indicated that the questionnaire used was a valid and reliable instrument for the purposes of this study.The pattern of results supported the theoretical framework used to classify the school change activities into particular clusters.
The results of the research revealed that the teachers in the South African sample were exceptionally keen to be involved in responsibilities concerning school change, even concerning aspects of management that could be considered the principal's 'turf'.Although the group of teachers from the six other countries indicated very similar result patterns, the South African sample was considerably more eager to be involved in school change initiatives and related responsibilities than the teachers in the samples of all the other countries.A possible explanation for this observation is that teacher involvement has not yet been a reality in South Africa, in spite of comments received from many school principals in an earlier survey (Poppleton & Williamson, 2004), such as, "teachers are involved in any case"; "it is already happening"; "teachers are daily involved in issues that affect them"; and "in most cases structures are already in place to achieve this".The experience of exclusion faced by these teachers may explain their desire to be involved, despite a heavy workload.
Although this research was only exploratory in nature and one is hesitant to generalise, it appears that these results support Carl's (2007:203) indication of how negative teachers' experience of the top-down approach to change can be and Soga's (Botha, 2007:31) claim that many local schools are still overly bureaucratic, causing frustration to stakeholders meant to be served by the process.The conclusion, that the teachers in the South African sample expressed a seemingly unrealistic enthusiasm regarding involvement, could probably be ascribed to stakeholders in school-based management being given the responsibilities for less important issues, leaving the core school-based management decisions to the principal (Botha, 2007:35).It is imperative that teachers are more involved in the pre-implementation phases of educational change initiatives, despite their apparent lack of capacity and probable limited contribution in this regard.However, there is still a long way to go to bridge the gap between the teachers' desire to be part of the decision-making and responsibility-taking processes and the readiness of both the authorities and the school principal to venture in this regard.

Table 1
Teachers' preferences regarding their involvement in school change initiatives

Table 2
Teachers' involvement preferences: differences between country means (Newman-Keuls test)