Cluster supervision practices in primary school of Jimma Zone

The main objective of this study was to assess the supervisory practice of cluster resource centre (CRC) supervisors in Jimma Zone primary schools. To achieve this purpose, the descriptive survey design was employed. Data were collected from 238 randomly selected teachers, and 60 school principals with a response rate of 98.6%. Moreover, 12 CRC supervisors and 6 woreda supervision coordinators participated through purposive sampling. Questionnaires, interview and observation were used to collect data. Data were analyzed by using percentage, t-test, mean and standard deviation besides using descriptions for the qualitative data. Consequently, the obtained result confirmed that CRC supervisors are less successful in playing their administrative, linking, and pedagogic and community mobilization roles and responsibilities on continuous base. In order to solve the problems and alleviate the challenges, a collaborative work is recommended to the CRC supervisors of Jimma zone, Woreda Education Office, Regional Education Bureau and other stakeholders to capacitate the supervisors through providing training, allocating budget and improving their working conditions.


INTRODUCTION
The future of a nation is by large determined by the quality of education that it is able to provide to its citizens as education is more crucial to the society as compared to other public services (Ayalew, 1999).Gall (2009) in this aspect, stated that "from numerous considerations, it must be evident, that education claims the first rank among the sciences...........not perhaps in respect of its difficulties, but most certainly in respect of its importance".
Hence, the concern for quality education becomes the main agenda in almost all education systems across the globe, and consequently supervision has received substantial attention by national policy makers as a tool for monitoring the quality of education (De Grauwe, 2001).As a result, almost all countries installed school supervision services into their education system (IIEP).
Despite its importance in monitoring the quality of education provided in school, the traditional supervision practice was suffering from multitude of challenges relating to ineffeciency and ineffectiveness of its organization and structure.Consequently, many countries across the globe have hence attempted to reform their school supervision services to enhance educational quality (Adu et al., 2014).
Grauwe (2008) indicated that the reforms in the supervision structure and organization have been carried out by either of these three ways: 1.By decentralizing the service further 2.By building up a new structure (the resource centre) between the supervisors and the schools 3.By strengthening in-school supervision.
The major purpose of these reforms endeavors, among other things, is "to bring supervision closer to where the action is taking place, that is to say to the school site level." School clusters with all ther varying terminologies such as nuclei, pedagogical zones, complexes, school learning cells, core and satellite schools, were one of the mechansims by which countries tried to bridge the gap between the school and the supervisory services (Grauwe, 2008) According to Giordano (2008), the estabilishments of schools clusters dates back to the early 1940s in Great Britain and India so as to enable rural schools to pool educational resources together.Traditionally, clustering of schools involves bringing several schools around one core school which plays the leading role for the other schools (Giordano, 2008).
CRC "strategy has grown to be a common feature of educational reforms and improvement programmes throughout the developing world, particularly in Asia and Africa, after the formulation of the World Declaration on Education for All (EFA) at the Jomtien Conference as a manifestation of the renewed commitment of education ministries and donor organizations towards the quality of education (Giordano, 2008).Giordano (2008) indicated that school clusters facilitaite pedagogical supervision, and support among a number of administrative and pedagogical purposes through their linking of schools at the sub-district level.The cluster resource centre (CRC) helps not only to bring the supervison support near to the school but also makes the support very realistic as the "supervisor can have a more inside view of the issues faced by cluster teachers and head teachers" (De Grauwe andCarron, 1997, 2001).
Supervision at the CRC is usally carried out in two forms."The first is by creating them as an intermediate structure between the school, and the district level that is part of the administrative hierarchy" while a committe of teachers, represented from each of the schools in the cluster is in charge of providing a supervision support to the schools in the second form (Giordano, 2008).Supervisors including the cluster supervisors are supposed to top play three distinict but interrelated roles to control, support and liason (Rai and Singh, 2013;UNESCO, 2007).Each of these roles has two fields of application that are not always easy to disentangle, namely the pedagogical and the administrative.
CRC supervision practices have yield in some positive out comes in terms of improving school management (Raj Khaniya, 1997), and in terms of teacher attendance and accountability (Giordano, 2008).These out come are acheived due to the regular visits of schools made by the CRC supervisors coupled with the improvements in the monitoring and training of principals and teachers (Giordano, 2008).
"In spite of their advantages, school clusters and resource centres have shown somewhat disappointing results in terms of actually improving teaching, and at worst, can be counterproductive in their goals" (Giordano, 2008).Similarly, Grauwe (2001) emphasized that supervision practices across the globe are critisied for their lack of positive impact on quality of teaching and learning, of which CRC supervision is not exceptional.The pitfalls in the supervison practices in this reagrd are due to the imbalances between their mandates, and the resources availabile to discharge these mandates along with poor management and planning decisions.(2001) and Barro (2006) noted that the priority of all countries, particularly the developing ones, is to improve the quality of schools and the achievement of students since learning outcomes, economic growth and development expected by a country depend largely on the quality of education being offered.

Grauwe
The quality of primary education has of paramount importnace not only for its contribution towards poverty reduction by increasing the productivity of citizen trough education strategy (World Bank,1996) but also for its role in preparing students to the forthcoming levels of secondary education.The Ethiopian education and training policy in this regard states that "primary education will be of eight years duration, offering basic and general primary education to prepare students for further general education and training" (MoE, 1994).The quality of primary education hence by large determines the quality of students that the seccessive levels received as well as their overall quality.Govinda and Tapan (1999) indicated that, supervision is a key factor for ensuring the good functioning of the primary education.In line with this, Education Sector Development Program IV (ESDP) by the Ministry of Education (MoE) noted the importance of providing quality based supervision to improve the quality of education (MoE, 2010).In line with this, it is indicated that, school cluster is an important way to improve the quality of teaching and learning in the schools very closely (MoE, 2006).Prasertsri (1996) indicated that, school clusters are established to provide an administrative and pedagogic support and considered as "an effective, decentralized means of developing primary education with full community participation".
According to Tesfaye (2013), school clustering has been introduced in Ethiopia very recently with the intoroduction of school and cluster based teacher professional development as national policy, as a tool for improving teaching condition by responding to the local needs around the school.The form of clustering in Ethiopia resembles the later approach where by 3 to 5 primary schools are connected to form the cluster resource center (CRC).The major school around which the other schools networked is known as the cluster while the networked primary schools are called sattelite schools.The Woreda education office designated one supervisor for each cluster resource center, who is responsible to provide supervisory service as well as to closely support in built supervisions in schools, and should report to the Woreda level educational program and supervision (Million, 2010).MoE (2012) indicated that cluster supervisors are not part of the line managers but they play a role in monitoring, supporting, evaluating and linking schools vertically and horizontal.The school clusters according to De Grauwe (2001) are established to provide a closer and more regular supervision for schools.The cluster supervisor is responsible for many activities.These activities are summarized as support, control and linking (De Grauwe, 2001;MoE, 2012).In Ethiopia, CRC supervisors are supposed to carry out the following activities (Million, 2010): 1. Check, follow up, monitor and evaluate school teaching learning activities in order to maintain expected quality and standard.2. Ensure that educational programs inclined to local conditions and community needs.3. Organize and demonstrate appropriate teaching methods to teachers.4. Organize in-service training programs through seminars, workshops, conference etc. to school based supervisors and teachers.5. Conduct periodic planned visits to schools to render support at the spot.6. Prepare reports to Woreda education office on issues and problems for school which are beyond capacity of the schools.
In general, it is apparently clear that supervision is crucial to enhance the quality of education that schools provide.However, studies conducted in the area indicated that, supervisors are not able to play the expected role because of many problems (De Grauwe, 2001).Similarly, in Ethiopia supervisors are not playing the essential roles that they are supposed to do.The fifth education sector development program of ethiopia states that: The problem that principals and supervisors face in improving school quality is ..... little understanding of how to convert additional resources into improved learning outcomes for students through the achievement of school minimum standards.In addition, neither group currently has the capacity to conduct informed classroom observation, and provide appropriate and constructive feedback to improve teacher performance.
In Ethiopia, as far as the practice of supervision in primary and secondary schools is concerned some researchers such as Chanyalew (2005), Getachew (2001), Million (2010), and Desalegn (2012) conducted studies in different contexts, and found out that the instructional supervision is not effective and successful in developing teachers profession, and improving instruction.
Although the aforementioned studies were conducted at zonal and national contexts, their focuses were on the instructional supervision.Besides, as far as the knowledge of the authors are concerned, two researchers Gashaw (2008) and Keshaun (2014) studied CRS supervision practices in Ethiopia.The former focused on the role of CRC supervision in promoting teachers professional competence in West Gojjam Administrative Zone, while the latter was about the practice of primary school cluster supervision in Jimma town.Both studies identified that CRC supervisors are not effective in providing support to teachers to improve their profession and fail to contribute to the improvement of teaching learning in the schools with a close concern.The present research differs from the aforementioned studies in context, as in the case of the former, and in scope in the latter case.
For that reason, this study focused on assessing the current practices of CRC supervision in Jimma Zone primary schools.It particularly emphasized more on assessing the Supervision Practices such as teachers" professional development, and improving school management practices in Jimma zone primary schools in improving teaching learning.In order to achieve this purpose, the following basic research questions were answered in the study: 1. To what extent do cluster supervisors discharge their administrative role and responsibilities? 2. To what extent do cluster supervisors discharge their pedagogic roles and responsibilities? 3. To what extent do cluster supervisors discharge their liaison/linking role and responsibilities? 4. To what extent do cluster supervisors discharge their community mobilization roles and responsibilities?

METHODOLOGY
The descriptive survey method was employed to study the problem.A research design according to Bryman (2003), gives a framework for data collection and analysis of data in conducting a research.This design was chosen because it can provide sufficient information concerning the practices and challenges of primary school cluster supervisors of the study area.In addition, it helps to draw valid generalization and conclusions (Yalew, 2012).Consequently, the quantitative research approach was used by supplementing with the qualitative method in order to answer all the basic research questions.

Research method
Quantitative research method was employed in this study since the study involved quantifying the responses of the participants to the questionnaire items, and quantitative analysis of the data was carried out.

Sources of data
The study accounts for both primary and secondary data.The primary data was gathered from Woreda education supervision coordinator, primary school cluster supervisors, primary school principals, and primary school teachers since they can give firsthand information regarding cluster supervision.In addition to this, the secondary source of data was used to get important data or information from different documents such as the school report, feedback from woreda supervisors and supervision manuals.

Samples and sampling techniques
To select samples from the zone, multi-stage sampling was used in this study.First, by using cluster sampling, 6 Woredas (sub district) are taken from the total 18 Woredas of Jimma Zone which were classified into four groups/strata by zonal education office based on their geographical location and distance from the zonal office.Woreda Education Offices (WEOs) are educational units at sub district level which are in charge of the provision of education at the sub district.They usually assist and supervise schools within their vicinity.This is due to the fact that the Woredas in the zone are large and spread over a large area.In such instances, cluster sampling is preferable over the other techniques (Walliman, 2006) as it helps to control field costs, especially those related to travel.
Accordingly, six clusters were identified.The sampled woredas are selected from each stratum by simple random sampling considering their proportion in each division.The use of simple random sampling method is justified for its potential in providing all the Woreda Education offices (WEOs) an equal and independent chance to be included in the sample.Then, using stratified proportional sampling technique, 12 CRCs and 60 schools are taken from the sampled six Woredas, and each sampled CRC and school are randomly selected from the Woreda and CRCs respectively.Prportional sampling was used to ensure equal representations of all differing characterstics in each of the clusters.
At Woreda level 6, Woreda supervision coordinators are selected by availability sampling, and 12 cluster supervisors and 60 school principals were purposively selected as a sample since their knowledge and experience has contribution for the study (Figure 1).
Finally, 242 teachers are taken as sample size of the study based on Krejcie and Morgan (1970) table for determining sample size from a given population.In this case, the population is more than 600.Consequently, we calculate the size of the sample in each stratus through the following expression to keep the proportionality as recommended by Barreir and Albandoz (2001): Where ni is sample required from each strata n is the sample size Ni is the population in each strata/woreda N is the total population in six woredas The samples that were taken from each woreda are presented in Table 1.

Data collection ınstruments
In order to collect relevant data from the respondents, the researcher used questionnaire, semi-structured interview and observations.

Questionnaire
Questionnaire was chosen and considered appropriate because it can cover a large sample of respondents, thereby allowing a reasonable degree to generalize the findings.The questionnaire was designed with close and open ended questions, and distributed to the selected teacher and principal respondents to obtain their views concerning the practices and challenges of cluster supervisors.All the questions were prepared from literature.

Semi-structured interview
The researcher used semi-structured interview guide for Woreda education officers and cluster supervisors in order to get in depth information on the practice of cluster supervision.

Observations
The researcher used check list to observe the availability of office, materials and other facilities for supervisors besides the comments written in the schools log book by supervisors.

Validity and reliability
To ensure the quality of the tools in this study, a pilot test were be conducted on the validity and reliability of the instruments prior to the actual data collection.The objectives of the pilot test was to check out the clarity of the contents of each item, the consistency of items under each theme, relevance of the questions for the study area, and to see the difficulty of the language.According to Cohen et al. (2007), validity concerns the extent to which the test tests what it is supposed to test.Consequently, in this study the content validity of the questionnaire will be examined.So, the researcher first ensured that the questionnaire is to be based on related literature, and items reflects representative themes.Then the questionnaires were given to the research advisor, and two other teachers who had experienced on teaching supervision, and working as a CRC supervisor for their comments.
Reliability on the other hand, focuses on the degree of confidence that can be placed in the results and the data, which is often a matter of statistical calculation and subsequent test redesigning (Cohen et al., 2007).The pilot study was conducted on twenty teachers and four principals from Omo Nada woreda, Asendabo primary schools, which is not included in the sampled woredas.After the distributed questioners were returned, necessary modifications were made based on the comments given by the respondents and advisor"s suggestions.Most importantly, the reliability of different items of the questionnaires were checked by the help of Cronbach"s alpha (Coefficient alpha) calculated through statisitical package for social seinces (SPSS) version 20 windows.Consequently, the reliability statistics was 0.924 for 43 items used to address the basic research questions of the study (Table 2).

Methods of data analysis
The collected data from questionnaires and interview were analyzed, and presented by the combination of quantitative and qualitative data analysis method.Based on the nature of the basic questions that were developed, and the data collected from the respondents regarding the present practices and challenges of primary school cluster supervision of Jimma zone, Oromia Regional State, the following data analysis method were employed by using SPSS version 20.To analyze, the respondents" characteristics descriptive statistics like frequency and percentage were used while mean and the independent sample t-test were carried out to determine the significance level of differences in the responses of teachers and school principal respondents.Moreover, information and/or opinion reported by respondents through the open ended questions, semistructured interview and observation were presented thematically under the appropriate theme of the study or the quantitative data to triangulate it properly.Finally, possible summary, conclusions and recommendations were made (Table 3).

RESULTS AND DISCUSSION
The analysis and presentation of the data gathered from respondents on the practices and challenges of primary school CRC supervisors were presented based on questionnaires, interview and observation.The questionnaires were prepared having five point Likert scale range.Mean scores, standard deviations, overall mean, and t-test result from the responses were used to analyze quantitative data.Within the five point ranges, three trisecting scores were used to make the analysis clear as suggested by Anbessa (2012); these scores were 2.49, 3.49 and 4.49.Thus, the questionnaire items were analyzed based on the responses of the respondents with a mean value from ≤1.49 were ""Never", 1.5 to 2.49 were "rarely", from 2.5 to 3.49 were "sometimes", from 3.50 to 4.49 were "often", and from 4.50 to 5.00 were "always".Results from open-ended items and interview questions were also analyzed to support and validate the quantitative findings (Table 4).
In Table 4, item 1 respondents were asked about the follow up of CRC supervisors on the implementation of government education policy and regulations.Accordingly, the mean ratings of the teachers on this issue was 3.17, and that of the principals was 3.10 which show less frequent follow up since supervisors play this role sometimes.The overall mean 3.15 also shows that CRC supervisors sometimes practice this role.The p-value of 0.706 > 0.05 proves the two groups of respondents was not significantly differed in their response on the item.Note: P-value was calculated at α=0.05 levels, and df 296; Scales; ≤ 1.49 = never, 1.5-2.49= rarely, 2.5-3.49= sometimes, 3.5-4.49= often, ≥ 4.5 = Always.
Item 2 is about cluster supervisors" role on collecting statistical data on the number of students, teachers, sections etc.The mean scores of teachers and principals were 3.26 and 3.07 respectively showing that the supervisors perform this activity sometimes.The overall mean 3.16 has also the same verbal interpretation.The p-value of this item was 0.278 > 0.05 proves that there is no statistically significant differences between the two groups of respondents on the item.
Moreover, mean scores about the cluster supervisors monitoring and evaluation on the activities of various committee or clubs (item 3), 2.63 and 2.82 for teachers and school principals respectively show that this role of the CRC supervisor is sometimes practiced in their schools.The p-value of 0.299 > 0.05 proves that the two groups of respondents was not significantly difference in their response on the item.An overall mean 2.72 of the two groups also shows that similar verbal interpretation towards the item.
Never the less, the mean scores concerning reward provided by CRC supervisors for well performing and achieving team/staff members was 2.28 and 2.50 for teachers and principals respectively.These mean scores depicts that the supervisors were rarely motivating teachers and well performing team/staff members.The pvalue of 0.196 > 0.05 verifies that the two groups of respondents was not significantly difference in their agreement on the item.Also, the overall mean of 2.40 showed that the CRC supervisor is playing a little role in this regard.In the interview with CRC supervisors, one supervisor had this to say: I and perhaps other CRC supervisors are very busy with many administrative tasks such as collecting, organizing and reporting varies statistical data from many schools scattered in far distance from the cluster.It is also very difficult to reward well performing staff/team members where I do not have any financial source for this purpose.I always use verbal appreciation and encouragement which may not satisfy the well performers" need.In general, from the data presentation, t-test result and interview, it can be said that Jimma zone primary schools" CRC supervisors are less effective in playing their administrative roles, and responsibilities through frequent support to give the needed service, and to support the teaching learning process in the primary schools.

Pedagogic roles and responsibilities of CRC supervisors
The researcher also went ahead to assess the practice on the pedagogic roles and responsibilities in the next section, and Table 5 present the issue in Jimma zone primary schools context.
It can be seen from Table 5 item 1 that, teachers and principals were asked to give their response whether CRC supervisors conduct class observation for instructional improvement or not.The mean score of teachers" respondents is 2.66 and that of the principals" respondents is 2.58 with an average mean of 2.62.Hence, each mean showed that the CRC supervisors carried out class observation in their school sometimes.The p-value 0.683 > 0.05 indicates that there is no statistically significant difference between the two groups of respondents towards the item.The interview made with the curriculum officers and supervision coordinators give supportive evidence to confirm the responses of respondents, and revealed that the CRC supervisor less frequently conduct class observation since the large number of schools, sections, and number of teachers as well as other non pedagogic roles impede their performance.
Regarding item 2 of the same Table, CRC supervisors provide supportive feedback to teachers based on class observation sometimes as confirmed in the mean scores of the teachers 2.52 and principals 3.00.The t-test result with p-value 0.013 < 0.05 shows that there is statistically significant difference between the two groups of respondents towards the item as the mean score of the Note: P-value was calculated at α=0.05 levels, and df 296; Scales; ≤ 1.49 = never, 1.5-2.49= rarely, 2.5-3.49= sometimes, 3.5-4.49= often, ≥ 4.5 = Always.
principals is higher than the teachers mean.Hence, it is possible to say that the supportive feedback was not given in the study schools.
In the same way, the data obtained from open ended items reflect that supervisors did not give continuous feedback for teachers based on observation.They only provide general feedback about the school system that is not specific and valuable for individual teachers.
With regard to item 3 in Table 5, respondents mean score on facilitating professional growth of teachers through training, workshops and seminars was 2.29 for teachers and 2.20 for principals show that this role is practiced rarely.Also, overall mean 2.24 shows similar interpretation.The p-value of 0.616 > 0.05 verifies that the two groups of respondents was not significantly difference in their response on the item.
When respondents asked about the supervisors" role in introducing modern teaching methods to teachers to improve their skills with a main focus to improve students" performance, the mean score 2.75 for teachers and 2.60 for school principals show that the practice in this regard is less frequent/sometimes.The p-value of 0.437 > 0.05 proves that the two groups of respondents were not significantly differ in their views on the item.
Moreover, Table 5 item 6, about carrying out experience sharing by CRC supervisor, showed that the mean score 2.61 for teachers and 2.80 for school principals show that this pedagogical role is performed sometimes.The p-value of 0.301 > 0.05 proves that the two groups of respondents were not significantly differs in their response since their mean is very close with an average mean 2.70 shows similar verbal interpretation (sometimes practiced) towards this item.
With regard to, CRC supervisors" encouragement to teachers to conduct action research to solve instructional problems, the mean score 2.37 for teachers and 2.28 for school principals show that this pedagogical role is performed rarely.The p-value of 0.650 > 0.05 proves that the two groups of respondents were not significantly differs in their response since their mean is very close with an overall mean of 2.32 which shows similar verbal interpretation (rarely practiced) towards this item.
From the aforementioned data analysis, it is possible to deduce that the CRC supervisors were less successful in playing their pedagogical roles and responsibilities.They rarely support the school teachers and principals in the pedagogical aspects.The finding of this study is congruent with the finding of the study by De Grauwe (2001) in Eastern and Southern African countries that found out supervisors give high priority to non-pedagogic tasks, and urgent administrative issues than pedagogical area to bring quality education through the improvement of teaching.
As can be seen from Table 6 item 1, respondents were asked to give their response about the CRC supervisors work as a linking agent to create good relationship between schools and district office, the mean scores results of teachers and principal respondents were 2.98 and 3.00 respectively with an overall mean 2.99.The pvalue of 0.927 > 0.05 indicates that there is no Note: P-value was calculated at α=0.05 levels, and df 296; Scales; ≤ 1.49 = never, 1.5-2.49= rarely, 2.5-3.49= sometimes, 3.5-4.49= often, ≥ 4.5 = always.
statistically significant difference between the two groups of respondents towards the item.This showed that the CRC supervisors sometimes play this role in creating good relationship between the school and district office.
As to CRC supervisors facilitation of cooperation among member schools, and staffs, Table 6 item 2 depicts that the teachers and school principals" mean cores were 2.48 and 2.45 respectively with an overall mean of 2.46, which showed that supervisors rarely facilitating schools" and principals" cooperation.The Pvalue of 0.833 > 0.05 proves that there is no statistically significant difference between the teachers and principal groups.Similarly, the data obtained from principals" through open ended items reveals that less cooperation and collaboration is observed between the CRC schools and satellite schools as well as teachers working in the CRC.
In promoting smooth communication among stakeholders, Table 6 item 3 depicts that the teachers and principals mean scores were 2.64 and 2.53 respectively with an overall mean of 2.58 which was practiced sometimes.The p-value of 0.532 > 0.05 proves there is not statistically significant difference between the two groups" mean on the item.
One of the roles of CRC supervisor is linking schools and teachers with other schools and Woreda Education offices (WEO), but in our experience our supervisors fail to do this.May be this comes from the problem with the transportation facilities, the supervisors skills and other factors.Thus, it is difficult to be a good linking manager.In whole, the linking function of the supervisors presented from item 1 to 3 was practiced less frequently as the results of the t-test and interview revealed.In this regard, one of the school principals answers are presented in Table 7.
The results in Table 7 focus on the practice of community mobilization by CRC supervisors of Jimma zone, as reported by respondents.Accordingly, the mean scores of the teachers were 2.61, and that of the school principals was 2.83 in which the supervisors involve the community to participate in the planning process of the school activities sometimes.Also, overall mean 2.72 shows the practice is performed sometimes.The p-value of 0.217 < 0.05 proves that the two groups of respondents was not significantly different in their response on the item.
When respondents were asked how often the CRC supervisors promote community-school cooperation in solving resource problems, as shown in Table 6 of item 2, the teachers mean score was 2.57 and the school principals mean score was 2.80 which indicate that CRC supervisors sometimes perform this activity as compared with the expectation.Also, overall mean 2.68 shows that the supervisors play such role sometimes in promoting community-school cooperation in solving resource problems in the study area.The p-value of 0.208 > 0.05 shows that the two groups of respondents was not significantly different in their response on the item.
In the same Table 7 of item 3, the mean score of respondents concerning the cluster supervisors" practice in involving the community in decision making about their schools, were 2.63 for teachers and 2.95 for school principals with overall mean of 2.79 which showed sometimes practiced.The p-value of 0.79 > 0.05 proves that the two groups of respondents was not significantly difference in their view on the item.
Concerning, item 4 of in Table 7, the mean scores of teachers and school principals on the CRC supervisors" encouragement to the community to work with teachers on education of their children were 2.52 and 2.70, respectively, in which both group"s rating show the role with this regard which is been performed sometimes.Similarly, the overall mean 2.61 has similar interpretation towards the item.The p -value 0.354 > 0.05 showed that the two groups of respondents were not significantly different in their responses on the item.
From Table 7, the t-test analysis, interview with school principals, curriculum officers and CRC supervisors, it is possible to conclude that the primary school CRC supervisors were less successful in mobilizing the community as frequent as expected to bring changes in the schools.

DISCUSSION
1.The work experiences and the level of education of teacher and principal respondents of the study area were found to be satisfactory to get adequate data for this study and to contribute in the provision of quality education if managed and supported well.2. The results of the study on the practice of the CRC supervisors concerning their administrative, pedagogical, linking and community mobilization role, and responsibilities, most of the overall means of the two groups were found between 2.5 to 3.49 with a verbal interpretation "sometimes".Likewise, no statistically significant differences was found between the responses of the teachers and principals at α = 0.05 levels, with 296 degrees of freedom.Therefore, the CRC supervisors support in various educational aspects is not satisfactory.

Conclusion
Findings confirmed that the practice of the CRC supervisors on administrative, pedagogical, linking and community mobilization aspects were less successful and infrequent.Therefore, the CRC supervisors support is not satisfactory in terms of improving the management and academic aspects of the school system with a specific objective to improve the students" achievement.This implies that, the cluster supervision process in Jimma zone is not in a position to provide the necessary support for the improvement of teaching and learning, which it is supposed to.This might be due to various reasons that future research endeavors might take into account for investigations.

Table 1 .
Samples of the study.

Table 4 .
The mean scores of administrative role and responsibilities of CRC Supervisors.

Table 5 .
Mean scores of responses on the pedagogic roles and responsibilities of CRC supervisors.

Table 6 .
Mean scores of the liaison/linking roles CRC supervisors.

Table 7 .
Mean scores of community mobilization role of CRC supervisors.