Decentralization and the Autonomy of Heads of Devolved Secondary Schools in Tanzania. A Case Study

Table of Contents

1. Introduction

2. Literature review
2.1 Theoretical review
2.2 Empirical review

3 Methodology
3.1 Validity and piloting for reliability testing

4. Results and Discussion
4.1 How politicians and higher authorities’ professionals affect the autonomy of heads of devolved secondary schools in decision making
4.2 How secondary schools heads’ autonomy in decision making has changed since 2009
4.3 How to improve secondary school heads’ autonomy in school decision making

5. Conclusion
5.1 Recommendations for further researchers

6. References

Abstract

The purpose of this study was to find out how decentralization of the education system influenced the autonomy of the heads of public secondary schools. Quantitative research approach was adopted with a questionnaire survey design. Questionnaires with closed-ended and open-ended questions were administered to the heads of the government owned secondary schools. It was found that: the autonomy of the heads of secondary schools before decentralization was higher than their autonomy after decentralization; the political influences and the higher authorities’ professional interventions affected the autonomy of the heads of the devolved secondary schools in decision making with high magnitudes; the political influences had the relative negative effect while the professional interventions had the relative positive effects; and decentralization of the education system imposed more negative improvement than positive improvement on the autonomy of the heads of the devolved secondary schools in decision making. The heads of the schools suggested for: the government to grant them more power and respect in decision making; close monitoring of schools in decision making by education stakeholders in a meaningful way; and making clear legal limits of the obligational responsibilities of the school management. The study recommends that the heads of schools should be granted more powers to run the schools through clear lines of responsibilities, than the prevalent situation. Furthermore, it is important to establish best decision making approaches in secondary schools.

Key words: Education service decentralization, Decision making, School heads’ autonomy, Autonomy improvement.

1. Introduction

This research paper presents a study on how decentralization of the education system influence the autonomy of heads of devolved secondary schools in decision making. The need for research arose from alleged unpleasing emergence of interferences from politicians and higher authorities on delicate school management issues related to students and schools’ development programmes whereby politicians and higher authorities’ professionals were said to intervene in school based decisions in non-meaningful way. The paper is organized in the form of literature and methodology used in the study, study results and discussion, conclusion and recommendations for further researches.

2. Literature review

2.1 Theoretical review

This study was guided by three important decision making models including the rational comprehensive decision making model, the incremental decisions model and the mixed-scanning decision making model. According to the rational comprehensive decision making model a decision maker can diagnose the problem which: his or her goals, values and objectives are clear and ranked in accord with their importance; the high utility alternative ways of addressing the problem are considered; the cost and benefits or advantages and disadvantages of each alternative are investigated; alternatives and their consequences can be compared with other alternatives; and the decision maker will choose the alternative that maximizes the attainment of his or her goals, values and objectives (Levin & Milgron, 2004). Accordingly, individuals have preferences upon which make choices through rationalizing to figure out whether they are compatible with the optimization of the agent’s preferences. The application of the approach takes the form of exchange theory whereby all actions are fundamentally rational in character, and decision makers calculate the likely costs and benefits of actions before deciding what to do, while denying the existence of any kind of action other than the purely rational and calculative (Scott, 2000).

However, problems are not always clearly defined and much of the decision process may appear to be irrational. Amid the process, decision makers must have vast amounts of information in order to make use of the rational comprehensive decision making technique (Lunenburg, 2010). Consequently, the high utility alternative calls for regret or rejoicing of decision maker. By regret is meant a painful sensation of recognizing that ‘what is’ compares unfavorably with ‘what might have been’. The converse experience of a favorable comparison between the two is regarded as ‘rejoicing’. Indeed, there needs to be an ability to predict the future consequences of the decisions made. Particularly, the problems confronting decision makers often embody conflicting values since it is tough to ignore the invested and lost costs of former decisions that may foreclose many alternatives. This model assumes that there is one decision maker making decisions involving a great many people, interests and institution. It makes use of two-attribute utility function that incorporates two measures of satisfaction namely utility of outcome and quantity of regret (Loomes & Sugden, 1982 cited in Hasson, 2005).

Moreover, Hechter and Kanazawa (1997) criticize the rational choice that it lacks realism in its assumption of calculating the expected consequences of decision makers’ options and choosing the best of them. Accordingly, the paradigm would require much to invest at understanding how such resources as time, information, approval and prestige are involved. Moreover, individuals may act impulsively, emotionally or merely by force of habit thereby making biased options. “Rational decision makers choose the alternative that is likely to give them the greatest satisfaction” (Heath, 1976: 3 cited in Scott, 2000).

According to the incremental decisions model, the selection of goals and objectives is intertwined with the scientific analysis of the problem. Shafriz, Lyne and Borick (2005) present decision making through the model as the systematic process involving influx of information, probing for outcome, articulation of norms, establishing correspondences, execution of prescriptions, appraising and termination. Accordingly, decision makers only consider alternatives for dealing with a problem that differs marginally or incrementally from existing policies, suggesting that they do not completely remake policy every time they make a policy decision, but instead refashion existing policy. Problems confronting decision makers are continually redefined and for each alternative, only important consequences are prioritized (www.unc.edu). In particular, constant approach-to-solution adjustments are made to better manage policy but having ever single decisions or totally correct solution available to resolve a problem is a rare case. The means are conceived to be evaluated and chosen in the light of ends finally selected independently of and prior to choice of means whereby means-ends relationship is possible only to the extent of the agreed upon, reconcilable and stable values or policies (Lindblom, 1959). Consequently, a decision is that policy makers agree on, not one that may be most appropriate for agreed objective.

The mixed-methods scanning decision making model is accredited to Etzion (1967) who found fault with the rational comprehensive and incremental models. He argues on mixed-methods scanning approach that rationalistic approach focuses on the disparity between the requirements of the model and the capacities of decision makers, and that the need for the mixed-methods scanning model results from the fact that social decision making centers do not have specific agreed upon sets of values that could provide the criteria for evaluating alternatives. The approach considers both fundamental and incremental decisions. It incorporates a broad-based analysis sometimes and in depth analysis at the other times. It considers the differing capacities of decision makers. However, it is not specific about how mixed-methods scanning can be used in decision making (www.unc.edu) in that it doesn’t speculate how the rational being should reconcile with the set policy and premise. So, either rational decisions may outweigh the incremental decisions in certain situations or the reverse becomes true in other settings.

2.2 Empirical review

The term decentralization governance and management perspective refers to transfer of authority and power in decision making from the central government, which is “the locus of decision making” to local governments, and depends on political legacy for its establishment but operations require professionalism for efficiency (Smith, Barr & Burrke, 1976). Therefore in educational contexts, decentralization means transfer of authority from a higher to lower levels of decision making such as provincial departments or schools. Decentralization process differs from one country to another and some countries retain centralization. The driving forces for decentralization are drives for efficiency in management and governance, political democratization through consulting people and involving them in decision making, clarifying the lines of accountability, search for new resources, and in some other governments include the desire to reduce the powers of teachers’ unions. (McGinn & Welsh, 1999; Scottish Office, 1997). There are three important styles that define decentralization state in education system: deconcetration, devolution and delegation.

Decentralization by deconcentration involves reorganization of decision making within the ministry of education and the bureaucracy, whereby the central government retains full responsibility, but administration is handled by provincial department in regional or district offices, or schools (Yulian, 2004). This has been adopted by Armenia, Chile, and its elements are evident in Tanzania (Weidman & DePietro-Jurand, 2012). Particularly, the central government reduces the concentration of authority in the national capital, and in this sense it only shifts the authority for implementation of rules but not for making them (Rondinelli, Nelson & Cheema, 1984). The most important features in decentralization by deconcentration is the transfer of managerial accountability to the regional governments and ministry of education offices (Yulian, 2009), allocation and reallocation of budgets by the regional managers, and creation of regional bodies to advise regional managers (Weidman & DePietro-Jurand, 2012). The risky with regional bodies is becoming political in nature after acquiring penetrated political brains.

In another option, decentralization by devolution calls for permanent transfer of authority from the central government to more local units of government, for example provinces, municipalities or districts (Weidman & DePietro-Jurand, 2012). In this case, some of the funds might be provided by the central government to the local governments, funding education. In devolved powers, managers are appointed by elected officials at the local or regional levels; municipal or district educational officers who appoint school heads. Local governments are given power to allocate educational funds and determine the spending levels, while voters and financial sources question local officials for education service delivery (Weidman & DePietro-Jurand, 2012).

Moreover, decentralization by delegation or school autonomy is the administrative or legal transfer of responsibilities to elected or appointed school bodies such as school councils, school management committees, and school governing boards. This is common in El Salvador, where the communities manage schools, hire and fire teachers, maintain infrastructure and raise additional funds (Weidman & DePietro-Jurand, 2012). It is the form of school-based management in which school committees comprised of teachers, parents, and other community members monitor school’s performance, create endowments for the school, ensure that teacher’s salaries are paid regularly, develop and approve annual budgets and examine monthly financial statements (Barrera-Osario, Fasih, Patrinos. & Santibanez, 2009). Accordingly, communities have authority over curriculum development, procurement of school materials and monitor school infrastructure and teachers’ performance development. Sometimes, delegation may involve school principals or councils receiving powers to establish curriculum and decide on educational spending, receive and allocate funds, and school councils are appointed or elected with powers to name school principals (Paun & Hazell, 2009; McGinn & Welsh, 1999). Decentralization can also mean a transfer of authority to a private firm or individual, that is privatization as the case is in Mexico beginning 1979 (McGinn & Welsh, 1999).

3 Methodology

This study used a quantitative approach with questionnaire survey design. The questionnaires consisted of closed-ended questions for ratings and open-ended questions for explanations and opinions. The leading questions were: How is the autonomy of the heads of devolved secondary schools affected by the politicians and higher authorities’ professionals such as educational officers? How has the decentralization of the education system in 2009 changed secondary schools decision making? What are the secondary school heads’ suggestions for improving their autonomy in school decision making? A census sampling technique was done and involved 49 heads of public secondary schools in one district of developing country which practiced decentralization in 2009, who were found at their schools between April and May 2014. Data were recorded in questionnaires and were coded and analyzed using SPSS 16.0 software programme.

3.1 Validity and piloting for reliability testing

Data validity was ensured through proper questionnaire construction skills, appropriate content inclusion and set criteria for plausible items included and distractors discarded (Mugenda & Mugenda, 1999). Reliability was determined using Spearman’s rank correlation coefficients in the split half technique, by testing the consistence of the results at the significance level of 10% with exception of item 6 that was tested at the significance level of 2% (Kothari, 2004:381&303; Singh, 2006: 241). The filled in questionnaires (with 9 questions) were put into two groups by choosing the odd numbers on one hand forming Group A and putting the rest on the other hand forming Group B. The total responses per option in each group were computed and ranked. The differences between the ranks of the total responses (d i) from the two groups were determined.

The Spearman’s Rank correlation coefficient ‘r’ = 1─ (6∑ d i2 / [n (n2-1)])

Whereby; d is the difference between ranks for each pair of observation and n is the number of paired observations. The correlation between the two groups could be tested at the significance level of 10%, so the problem could be stated:

Null hypothesis: “There is no correlation between the responses made in Group A and responses made in Group B” i.e., μr = 0.

Alternative hypothesis: There is a correlation between the responses made in Group A and the responses made in Group B” i.e., μr > 0.

A significance level of 10% (or 0.10) means that there should be less than a 10% chance of making a mistake or being 90% certain that the responses were right. Starting with question number three (3), the table on the split-half process could be constructed:

Abbildung in dieser Leseprobe nicht enthalten

Source: Pilot Data (2014)

Abbildung in dieser Leseprobe nicht enthalten

Since n = 4, the distribution of responses per items is not normal and in this case the table showing the values for the Spearman’s Rank correlation could be used to accept or reject the guiding null hypothesis. Looking in the said table in row for n = 4 and column for the significance level of 0.10 the critical values for correlation are ± 0.8000, that is, the upper limit of the acceptance is +0.8000 and the lower limit of the acceptance is -0.8000. Since the calculated correlation coefficient r = 0.9750 is outside the limits of the acceptance range, then the null hypothesis could be rejected and the alternative hypothesis could be accepted, that there is correlation in the ranked data, that is, μr > 0. Therefore, the results in question three (3) of the questionnaire were consistent. The same procedure was adopted for testing the reliability in the next six (6) questions and the results were presented in summary below:

Summary of the reliability testing by Spearman’s Rank correlation in the split-half technique for data in questions 4 to 9

Abbildung in dieser Leseprobe nicht enthalten

Source: Pilot Data (2014)

The results summarized above show clearly that all of the questions in the questionnaire could produce consistent results since μr > 0 in all cases. The reliability of the results obtained by the research instrument administered to the respondents were confirmed.

4. Results and Discussion

4.1 How politicians and higher authorities’ professionals affect the autonomy of heads of devolved secondary schools in decision making

Firstly, the extent to which the politicians affect the autonomy of the heads of secondary schools was studied using a Likert scale of the semantic differential model explained by Nachmias and Nachmias (1996: 259) in which the respondents were given the options to respond to the conception of the measure on the rating scale defined by contrasting objectives at each end, that is, from ‘weak’ to ‘excellent’. The responses on the effect of the politicians on the autonomy of the heads of secondary schools were presented in Table 4:1. ‘Weak’ response could be explained as minimal influence or intervention while a ‘excellent’ response indicated that the intervention was maximum.

Table 4:1 Influences of politicians on the autonomy of the heads of secondary schools

Abbildung in dieser Leseprobe nicht enthalten

Source: Field data (2014)

From Table 4:1, the ‘very good’ response accumulated total frequency of one hundred and twelve (112) followed by the ‘excellent’ response with total frequency of ninety one (91), which in turn is followed by the ‘weak’ and ‘average’ responses with the total frequencies of forty four (44) and forty one (41) respectively, where summation of the response is done through all seven measured areas of secondary school decision making. Clearly, political influences seem to be ‘much, more and most’ happening in secondary schools decision making processes. It could be shown that political intervention existed especially in the areas of school enrolment capacity which accumulated a total frequency of twenty five (25) per single area of decision making, equivalent to fifty one percent (51%), and students’ discipline matters which accumulated a total frequency of twenty (20) per single area of decision making, equivalent to forty one percent (41%) of the total respondents’ population. The budget allocation, infrastructure improvement, development programs and personnel recruitment accumulated frequencies of nineteen (19) (39%), eighteen (18) (37%), fifteen (15) (21%) and fifteen (15) (21%) respectively, in the respective ratings of very good, very good, excellent and very good.

[...]