Power and Conflicts - A study of the Swedish school system at local level An outline for the doctorial thesis

Transcription

1 Power and Conflicts - A study of the Swedish school system at local level An outline for the doctorial thesis Work in progress Jon Nyhlén Ph. D Student Mid Sweden University Department of Social Sciences Östersund

2 Introduction... 3 The charter schools... 4 The Purpose of the study... 5 To study local power... 6 The three faces of power... 8 Governance... 9 Citizenship and participation Exit, voice and loyalty Method The selection of the cases The theory testing approach The theory developing approach Fuzzy-set Different types of fuzzy set Fuzzy sets and necessary conditions Fuzzy sets and sufficient conditions My study

3 This paper is an outline for my doctorial thesis. I am therefore very grateful for any comments. Introduction The Swedish schools system has since the mid 1970 s gone thru a lot of changes. The core of these changes has been a trend of decentralizations. Up until the middle of the 70 s the Swedish school system was very centralized. The starting point for these changes was the so called SIA-study (SIA-utredningen) which was initiated by the Swedish government. The conclusions from the study was that the Swedish schools had major internal problems mostly related to the work environment and to solve these problems the study proposed a decentralization of the school system. This meant that more power was moved from the national level to the local level. More specifically was that power over organizational matter of the schools was moved from the national level to the local level. This meant an empowerment of the local governments. These reforms were also made in order to increase the influence in the schools by the parents and the pupils. (Lidström & Hudson, 1995, p. 11) The reason why the Swedish school system was so centralized in the first place depends on the ambition to make education universal. In order to achieve this the school administration needed to be centralized. (Lidström & Hudson, 1995, p. 20) In a governmental bill from 1989 it was made possible for pupils to choose their own school. This freedom of choice should be satisfied as far as possible within practical and economical boundaries for the municipalities. In 1992 another governmental bill was presented, which proposed a major increase in the freedom of choice. In order to stimulate this freedom of choice-proposition the possibilities of starting a charter school was also increased, this by introducing a public financing system based on vouchers. The value of the voucher is decided in different ways. For the nine year compulsory school the value of the voucher is based upon the average cost for the pupils at the local governments own schools (Proposition 1991/92:95). For the comprehensive school the rules are a bit different. This because the pupils in the comprehensive schools often comes from surrounding municipalities. This is reflected in the voucher system. The value of the 3

4 voucher is based on the pupil s municipality of origin and their costs for the compulsory school. This is then related to a national price list which is based upon the national average cost for pupils in the Swedish compulsory schools. A result of these decentralization trends are that the Swedish school system has become divided. On the one hand we have the public schools and on the other hand the charter schools. What happens in some of the municipalities is that conflicts arise. The conflicts are often between the local government and the charter schools. These conflicts have very often its grounds in, for example, the access to public facilities such as swimming pools, football fields and so on. The charter schools in those municipalities states that they don t have the same access to these facilities as the public schools have. For the public schools, on the other hand, this with charter schools has led to a completely different situation. They now need to compete with other schools in order to attract pupils. My scientific problem is concerning some of the basic conceptions of political science. These concepts are power, influence, decentralization and organization. The problem originates in the changes of the Swedish administrative policy which in this case is the decentralized educational system. According to the Swedish political scientist, Stig Montin, the local political level is challenged by three perspectives, the first one is Communitarism, the other one is Functionalism and the third one is Governance. This leads to one of my theoretical starting points which are governance and how I m going to use this perspective is something that I will describe later on. (Montin, 2000, p. 6-8) The decentralizing trends during the last three decades has led to dual structures that competes with each other in power and influence over common resources and about the same pupils. In order to get access to these common resources different types of power resources is needed. The use of power theory is basic starting point for the theoretical approach of my thesis. The charter schools The first non-governmental school in Sweden was established This school was the St: Erik s Catholic School in Stockholm. Since then the number of schools has increased, especially since from the beginning of the 1990 s due to the change in legislation. In

5 about 0,9 per cent of all Swedish pupils were enrolled in independent Schools. In the number was 5,7 per cent ( ) In is 8 per cent of the pupils in the comprehensive school enrolled in a independent school and 13 per cent of the pupils in the upper secondary school.( ) It is in the urban areas of Sweden that has the highest concentration of independent schools. Charter schools can be either owned by cooperatives, trusts, by limited companies, voulentary organizations or by individuals. The figures of 2004/05 tells us something about the distribution. Corporate Company 52 % Voluntary organization. 15% Cooperative 14% Trust 14% Trading partnership 1% Private firm 3% Others 1% (Source: Fakta om friskolor, 2005) The charter schools are not allowed to charge fees for instruction so the schools are financed entirely through a voucher system, as I mentioned earlier. The independent schools are supervised by the National Agency for Education, just like the public schools. The independent schools are free to organise their own programs and schedules. The performance and quality of its education are however evaluated against a nationally provided syllabus. This evaluation is done by inspections and nationally provided tests. ( ) The Purpose of the study The purpose of this study is divided in two parts, the first purpose is; 5

6 The purpose of this study is with the starting point in some theoretical concepts study how we best can understand the divided school system on municipal level, which has developed due to the decentralization reforms and the conflicts created by this development. The second purpose is as follows; To study the consequences of these conflicts with the starting point in participation. The second purpose is dependent of the outcome of the first purpose. I will return later to the connection between these two purposes. To study local power Robert Dahl s famous study of New Haven differs in many aspects from the study that I m going to do but there are, however, some aspect worthy to consider. In his book, Who Governs? Dahl discusses in what way political decisions can benefit the citizens. Some benefits are devisable, which means that they could be traced to specific individuals. Such benefits are for example jobs and agreements. The second form of benefits are indivisible such as play grounds, schools, the military and so on. (Dahl, 1961, p. 52) In my case is it the indivisible benefits that are in focus. It is around these benefits that some of the conflict arises in some municipalities and the charter schools feels set aside by the local governments but in some cases they don t. The problem is that the charter schools feel that the public schools have exclusive benefits. These exclusive benefits have access to information and, as I mentioned earlier, access to playgrounds, sports arenas and so on. The result is an relationship were power is not equally distributed among the actors. Power can be seen as the amount of change an actor can introduce. This understanding rests on the notion of power as a will to do something despite resistance from other actors. This definition of power focuses on decision making processes and from that starting point Polsby describes three types of data that indicates the power among actors. These are; 1) Who participates in the decision making process? 2) Who wins and who loses on the actual outcome? 3) Who dominates the process? 6

7 By these indicators the researcher could decide who s got more power than others. (Polsby, 1980, p. 4) I mean that some of these indicators can be used in my case. Most relevant is number one and number three. But the focus is not decision processes but rather the form of organization. These indicators can be used as I compare the different municipalities. Is there difference in the type of organization that can explain different outcomes or is it that some actors are more dominant than others? For example in cases were the conflict levels are low the charter schools have better access to indivisible benefits? I think that Polsby s indicators could guide me in that matter. In his book, Community Power & Political Theory, Polsby states that many of the decisions concerning educational policy were made by public officials, more or less hierarchal. (Polsby, 1980, s. 79) This is also something that I need to take in to consideration. That is also why I use one non hierarchal theoretical approach but maybe I also need a theory that focuses on hierarchal structures of governing. The degree of hierarchy could be something that explains the differences in outcome, different degrees of conflict. The understanding in how differences in political influence comes up and survives could be reduced to two basic explanations. The most common understanding in political science is that these differences depend on one actor s lack of political resources. This one could be described as the liberal principal. The other one is the monopolistic explanation. In this perspective the exclusion from the political arena is a result of a conscious strategy from the political establishment. For me this means that I need to investigate the empirical matters concerning these two types of exclusion from influence. No matter what, the conclusion we can draw from these arguments is that resources is important in order to gain influence. The question will be if it is exclusion that cerates this lack of resources or if it is the other way round. But that is an empirical question. The resources in question are not only economic one it can also be personal contacts. 7

8 The three faces of power The concept of power rests on some common basic foundations. One of the basic foundations is the relationship between humans which could take the shape of either social structures or social actors. The other basic foundation is the notion of power as unequally distributed. This means that one actor A can affect actor B in a much greater extent than actor B can affect actor A. The third basic foundation is that a relationship is built upon mutual influence. If one actor has less power than other actor that will effect the relationship. (Johansson, 2002, p. 64) The three faces of power are in different ways related to these three basic foundations that I just mentioned. The first face of power is the pluralistic notion of power, represented by Dahl and Polsby among others. This understanding of power rests upon the notion that by studying the outcome of political questions where the there is different opinions you have captured the essence of power. The starting point for the pluralistic view of power is Robert Dahl s now famous definition: A has power over B to the extent that he can get B to do something I otherwise do. (Dahl, 1957, p.203). According to this definition B has to adapt to A s will. The driving force in this power relationship is therefore A. The second face of power concerns non-decisions or the power of absence. The driving theorists behind this was Peter Bachrach and Morton s. Baratz. This perspective means that it s not necessarily those who actually participate and has influence hat are the most relevant actors. In this perspective some actors has benefits that are exclusive just for them. Important are also dominating values, rituals and specific institutional procedures which can exclude actors from participating. (Bachrach & Baratz, 1962) The third face of power is about shaping peoples preferences so that consensus can be achieved. Steven Lukes is the driving force behind this theoretical development. Lukes means that power is when A exercises power over B if A gets B to do something against its will. A is also exercising power over B if A shapes, influence and decides B s will. This persective 8

9 seeks to bring different opinions together in areas ehre their interest should differ.(johansson, 2002, p ) Governance Governance can t be seen as a clear cut concept. On the one hand, governance could be seen as an analytic perspective and on the other hand as a new empirical phenomena. If we see governance as an analytic phenomenon it represents a shift in paradigm. The starting point is then that the social sciences have not been able to explain or describe empirical social phenomena in a satisfying way. In that light governance represents a new way in which we can describe how and why political processes organizes and governs. (Björk et al. 2003, p ) The ones that advocates governance as a new empirical phenomena sees governance as new forms of governing. The focus is then on the changed conditions for governing and on the changed processes for governing. In the light of these arguments governance could be seen as new methods for governing. Governance do not only focuses on the methods for governing. It could also be seen as a form of organizing public and private actors (Björk, et. Al., p ). What could be said is that governance, whether we focuses on methods or organization for governing, is that it represents a less hierarchal form of governing. Governance could also be seen as a descriptive and a normative concept. It s descriptive in the sense that it describes the reality as it is. The normative aspect means that it describes how thing should be. One of the problems with governance is that there is no clear line between the different forms of governance. What emphasises is that governance is a process where the public sphere meets the private and acts together and the resources is coordination towards a common direction. (Björk et. al, p. 128) Stig Montin (1993) has focused on the relationship between the private and the public sphere and developed this five criterions that illustrates the boundary between the two; 1. Initiative: The production of the service could be initiated by a private actor or by the public sphere. An example could be when the municipality decides to start a school or when a private actor, a private firm or a trust, wants to start a school. 9

10 2. Production: The service is produced by the municipality, the private sphere or in cooperation. 3. Financing: The service could be financed in different ways. One way is by government grants. Another way is by the private way with fees. A third way is a mix by the private and the public way. A fourth way that are becoming more common is services financed by sponsors. 4. Legal norms: There are legally mixed forms of private and public service providers which are constituted by the public companies. Companies that is fully owned by the municipality but operated as a private company. 5. Policy- steering: There are two different ways in which public norms and private norms could affect the service production. The service could be public in that sense that it is strongly governed by political or bureaucratic norms or it could be private in that sense that it is affected by market principles. These developments create conflicts between the new and the old norms, but it also creates new institutions. We could also say that the development has gone from a legal oriented public administration to a market oriented administration. The following list could be made, that illustrates the development, according to Montin, Old norms Legal security Equal treatment Equal standard Equality Solidarity Bureaucratic rules Citizens Oriented by needs New Norms Flexibility Local adaptation Decentralization New Leadership Competition Management by objectives User, Costumer Oriented by demand (Source: Montin, 1993, p. 47) This could be related to my research problem. Maybe the conflict between charter schools and the local government has it causes in the conflict between different norm-settings. Some 10

11 municipalities could have had problems adapting to this new order with decentralisation and competition. Citizenship and participation This part of the paper some theoretical aspects connected to my second purpose will be presented. I will start this chapter with an introduction of how the view upon the citizen has changed. In the end I will present some methods for acting and practice influence. The way we look on the citizen s role in local politics has changed over the years. In his article Medborgardeltagande, brukarinflytande och konsumentanpassing (2000) Torgny Klasson develops different models with the starting point in the changing role of the citizen. In the first model, the citizen-participation model, the citizen is seen as the central actor. This means that the model sees the citizenship as something universal, that the citizen has a set of rights, but it also constitutes some obligations for the citizen (Klasson, 2000, p. 40) The model also states that the citizen shall have an influence in political decisions both in the preparation of a political decision but also in the following phases, such as the decisionmaking process and the implementation of the decisions. This also leads to an education of the citizens in the democratic process and how it works. It also means that the politic becomes a common goal for the citizens (Klasson, 2000, p. 41) The ability to protest is by a voice function. This is a personal, open and direct way to show discontent. The citizen is also assumed to take the common interest into account when it acts (Klasson, 2000, p ) The second model presented by Klasson is the user-influence model. This models starts from the point of view that the users of public services has gone from a passive role to a active role. This means that the user of public services today could in a more effective way demand better services. This model just includes those who are users of public services. The result of this is that the model does not encourage the users to take the common interest into account when they try to influence politics. (Klasson, 2000, p. 44) 11

12 In this model the influence is only recognized for those who already have given rights, the right to health care for example. The citizen s involvement in politics is given legitimacy by their social rights rather then their political rights, which was the case in the previous model (Klasson, 2000, p. 45). A change in the law for the municipalities in 1988 gave the municipalities possibilities to create institutions where the users of the services could make their demands heard to politicians and professionals (Klasson, 2000, p. 46). The final model presented in Klasson s article is the consumer-adapting model. In this model the citizen is viewed upon as a consumer of welfare services. This means that the citizen participates and influences the politics by the freedom of choosing the alternative of welfare service. The model is also based on competition between the different service producers. This is because the more popular the service producer is, the more money it gets from the authorities (Klasson, 2000, p. 48) The direct form of citizen influence in this model is constituted by the fact that if the user of the service is not satisfied is the exit function used and the user chooses another service producer. (Klasson, 2000, p. 49) The three models could be summoned in this table: Actor Citizen Users Costumer Character of Interest Common interest Group interest Individual interest Ground of legitimacy Political rights Social rights Civil rights principal of organisation Territorial Sectors Market Type of influence Protest "Voice" Protest "Voice" Exit Decision principal Consesnus/Majoritan Consensus Costumer decision (Modified from Klasson, 2000, p. 50) As we can see, the view of the citizen has changed from a the citizen perspective were the citizen has a common interest and political rigthts to a user with a group interest with social 12

13 rights. The development ends in Klassons model with a actor that is seen as a consumer with an individual interest and civil rights. For me the most useful perspective would be the user perspective. This is useful for the second purpose. The following parts takes it s starting points in theories concerning aspects of participatory democracy. The empirical research concerning democratic participation draw a line between so called fact-knowledge and procedure-knowledge. Fact-knowledge is basic knowledge about the political system, number of seats in the Parliament for example. Procedure-knowledge refers to a form of knowledge that is not as concrete as the fact-knowledge. Procedureknowledge is knowledge about the democratic procedures and what to do in order to express opinions. (Jarl, 2004, p. 54) Why user participation? That depends on the fact that user participation is a form of local participation advocated by the participation theory. The theory stresses the importance of the possibility for the citizens to participate in areas outside the political sphere. The participation in the society should be spread to other spheres, like the social, cultural, and economic sphere. The participatory theorists also advocates that this participation should be realized at the local level. This because that issues that are dealt with on the local level presumes to be of great importance to the citizens. The third important reason to focus on user participation, according to the theory, is that the users presume to represent a collective of opinions.(jarl, 2004, p ) To summon this up one could argue that user participation is a good way to operationalize the participation theory. Exit, voice and loyalty As we seen in the previous parts, exit and voice are two of the key elements in citizens ability to become involved and practice influence. The Exit option is closely related to the economic way of thinking and acting. This means that when a member leaves the organization he is acting according to the exit option. If the member expresses his or her dissatisfaction to a subordinated or anyone in the management he or she is using the voice option. The voce option is categorized in relation to the political way of thinking. (Hirschman, 1970, p. 4, 15) 13

14 The exit function rests upon the believe that competition is the best way to create efficient organizations. If the members of an organisation are not satisfied, they will leave the organization for another organization. This also means that the members of an organization need to be alert. Exit demands action from the members. (Hirschman, 1970, p ) Voice, on the other hand, could be defined as an attempt to try to change from within rather than escape, which what is constituted by exit. Hirschman describes voice as the basic portion and function of the political system. (Hirschman, 1970, p. 30.) The starting point of voice is for example the decline in performance of an organisation or a firm. Like exit, voice demands active and alert members (Hirschman, 1970, p 31-32). A problem with the exit option is that once you have exited an organization you have lost the opportunity to use voice, but not vice versa. So in some cases exit can be seen as the final alternative when voice has failed (Hirschman, 1970, p. 37). Hirschman means that the existence of the exit option can reduce the use of voice. The loyalty option is the middle way between the exit and the voice option. (Hirschman, 1970, p ) The function of loyalty is to hold exit at bay and activate the voice option. Another important aspect of loyalty is that it limits the tendency of the most quality conscious members to be the first to exit. (Hirschman, 1970, p ) Method In the processes of selecting the best research approach it could be wise to consider some important aspects of my research. Am I as a researcher eager to understand the general of the phenomena or the specific? If we want to know more of the specific, the case study approach would be the best suitable otherwise, if we want to know more about the general the comparative approach would be the best suitable. I am more interested in understanding the general rather than the specific so for me the most suitable approach would then be the comparative approach. Another important aspect to consider is the role of the theory. Here we have two major roles we can give the theory in our research. The first one is a theory testing approach and the other one is a theory developing approach. I will be more detailed in my description of these two approaches in the next section of this paper. 14

15 The selection of the cases The theory testing approach I am going to use strategic selection of the cases and there are three major situations where a strategic selection of the cases is ideal. The first situation is where we only have a limited number of cases to study. The second situation is when the question we are interested in is of that nature that for example a statistical design would mean an enormous amount of resources to get the useful measurements of the phenomena we are interested in. The third situation is when we want to add to the analysis a survey of interesting cases (Essaiasson et al., p ). King, Keohane and Verba point out in their book Designing Social Inquiry the importance of accuracy in the selection of cases. The selection of the cases affects the outcome and the accountability and the strength in the results (King et al, p. 128). When it comes to the actual design of a theory testing approach there are two main strategies, those are Most similar systems design (MSSD) and Most different systems design (MDSD). MSSD is build around the belief that systems as similar as possible with respect to as many features as possible constitutes the optimal selection of cases in a comparative approach (Przeworski &Teune, p. 32). According to Przeworski and Teune this strategy reduces the number of experimental variables. The number of variables can still be big and unknown but MSSD as a strategy reduces them. The principal of MSSD is that the selection of the cases is made out of the independent variable by holding the other variables constant and establishing the value on the dependent variable (Esiasson et al., p. 112, Anckar, p. 4) This is not the only way to select the cases with MSSD, there is another possibility. If we only know the value on the dependent variable it is impossible to select the cases from the value on the independent variable. What is needed in this case is that we make sure that we have maximal variation on the dependent variable (King et al, p. 129, Esaiasson et al, p. 113). The other strategy with a theory testing approach is MDSD. With this strategy we select systems as different as possible. The result with this selection strategy is that we strengthens the belief to the investigation variable (Essaiasson et al., p. 114) 15

16 The theory developing approach When it comes to the selection of cases with a theory developing approach there are several different approaches. I will focus on two of them, Method of Agreement and Indirect Method of Difference. This is two methods developed by John Stuart Mill in the book A system of logic. The principal of selecting the cases with the method of agreement is that if you want to know why wars occur you need to study wars. The analytic task is then to find as many common factors as possible in all of the investigated cases. (Esaiasson et al. p. 126). The problem with this method, as Mills also was aware of, is the lack of variation. This did that he argued for the second approach, Indirect Method of Difference. If we want to know why wars occur it is not enough to just study wars, we need also to study cases where wars haven t occurred. The problem with this approach is to find these negative cases where wars never broke out (Esaiasson et al., p ) Fuzzy-set For analyzing the data I am going to use a fuzzy set approach. Since I am in the beginning of my research project, I have not yet been able to operationalize the theory, so the following part of this paper will just be a description of the fuzzy set approach. The fuzzy set approach is especially useful when we as researchers are dealing with categories that are imprecise. (Ragin, 2000, p. 3) Fuzzy set as a technique offers an interpretive algebra, a language that can be seen as half-verbal-conceptual and halfmathematical-analytical. The fuzzy set approach has a great value for social scientist in it s potential for intensifying the dialog between ideas and evidence in social research. This dialog is the intensive interplay between theory and data analysis. This means that there exists an intensive back-and-forth movement between the use of data to improve theory and the use of theory to guide the researcher in the exploration of data. (Ragin, 2000, p 4) A conventional set is dichotomous which means that an object is either in or out of the set. A fuzzy set on the other hand permits a membership in the interval between 0 and 1 which retain the states of full membership and full non-membership. It is, however, up to the researcher to 16

17 specify the scores for the fuzzy membership. (Ragin, 2000, p 6) Ragin describes the back-andforth movement between theory and data as a tailoring of fuzzy sets to theoretical concepts. (Ragin, 2000, p 7) Ragin also describes fuzzy set as a technique that combines both quantitative and qualitative assessments in a single instrument and that fuzzy set distinguish between objects that are more in versus those that are more out. This by using the crossover point 0,5. (Ragin, 2000, p. 8) Fuzzy set also makes it possible to operationalize multiple interpretations of a concept. For example, in political science it is often agued that democratic countries do not make war with other democratic countries. The problem with this is that countries vary in degree of democracy. The problem is therefore to know if a statement applies to all countries with at least some presence of democracy only to full-fledged democracies. By combining qualitative and quantitative assessments in one single instrument, fuzzy set makes it possible to address varied interpretations of social scientific concepts in an explicit manner (Ragin, 2000, p. 9) The above section was some general aspects of the use of fuzzy set and how it can be motivated. The following section will concern matters more specific about fuzzy set. As we have seen, by using fuzzy set instead of a dichotomous set it is possible to address two types of diversity, both the varying degree which cases can belong to sets and the idea of difference in kind (Ragin, 2000, p. 149) This means that diversity do not only exist in the different configurations of set membership that social phenomena exhibit but also in the degree to which they belong to such sets and configurations. For example, two countries can be described as both advanced industrial, capitalist democracies, as instances of sets, and the can also differ in the degree to which they belong to this intersection. The first aspect of differences in kind is captured by the notion of difference in kind and the many configurations of membership that arise from multiple distinctions. The second aspect of diversity is captured by the notion of degree of membership and is based on the idea that virtually all categorical distinctions in the social sciences also involve variation by degree. Difference in kind can be referred to as qualitative distinction and differences in degree of membership as quantitative distinctions. (Ragin, 2000, p. 149) 17

18 Different types of fuzzy set The three-value fuzzy set is the simplest form of this instrument. The third value added is the 0,5, which indicates objects that are neither fully in nor fully out of the set. The five value fuzzy set is still simple but, in Ragins terms, a more elegant form. This form of fuzzy set uses the numerical values 0, 0.25, 0.5, 0.75, and 1. Essential for the five-value fuzzy set, according to Ragin is, that it is separates between cases that are more in from those that are more out and it distinguishes between cases that are mostly in versus fully in and between cases that are mostly out versus fully out. The five-value fuzzy set is useful when the researcher have a substantial amount of information about the cases, but the evidences is not strictly comparable between the cases.(ragin, 2000, p ) The seven-value fuzzy set adds two more stages to the set. To visualise this I have used the table below. This table shows both the differences between crisp set and fuzzy set but also the differences between different types of fuzzy set. Crisp set Three value fuzzy set Five value fuzzy set Seven value Fuzzy set "Continuous" Fuzzy set 1=fully in 1=fully in 1=fully in 1=fully in 1=fully in 0,83= mostly but not fully in Numerical scores 0,75=more in than out indicating that degree of 0,67= more or less in membership is more "in" than "out" (0,5<x < 1) 0,5=neither in nor out 0,5=neither in nor out 0,5= neither in nor out 0,5= neither in nor out 0,33= more or less out 0.25=more out than in 0.17=mostly but not fully out Numerical scores indicating that degree of membership is more "out" than "in" (0 <x<0,5) 0=fully out 0=fully out 0=fully out 0=fully out 0=fully out (Sources: Ragin, 2000, p. 156) 18

19 This section will just deal with the basics of fuzzy set and necessary conditions. I don t have enough knowledge about the research object to discuss the technical specifics of the approach at this time. Fuzzy sets and necessary conditions Ragin describes focus on finding necessary condition as very important in social theory. This is because necessary conditions can bring clarity to large bodies of social science thinking. A example that is mentioned is Skocpol s 1979 study States and Social Revolutions. Skocpol s study gave at hand that state breakdown is a necessary condition for social revolution. This is important in two aspects, both for the development for theories of revolutions and for the conceptualization of social revolutions and the descriptions of types of revolution.( Ragin, 2000, p. 203) It is also important to focus on necessary conditions for another reason. It could give our politicians knowledge about, for instance under what conditions ethnic conflicts develop, so that they can be avoided in the future. Finding necessary conditions can, therefore, have a great practical impact on policy. (Ragin, 2000, p. 203) How can necessary conditions be expressed explicit? A factor that is necessary is a condition that needs for a specific occurrence to develop. The equation D=TG expresses two factors necessary. If stable democracies (D) shall develop is trust (T) between the citizens and economic growth (G) necessary conditions. It is the combination of the two that makes the condition. (Denk, 2002, p. 71) Fuzzy sets and sufficient conditions To study sufficiency is central when we are dealing with studies that have a high degree of casual complexity. When the causation is complex no single condition may be either necessary or sufficient. This is because causes are sufficient only in combination. (Ragin, 2000, p. 230) This is shown by the section below. 19

20 A sufficient factor is one that develops under the condition that the same situation develops during other conditions. The equation D=T+G contains two sufficient factors. Stable democracies ( D) develops in societies where the level of trust (T) is high. At the same time, stable democracies also develops in societies with high economic growth (G). (Denk, 2002, p. 70) When we talk about sufficient conditions it is necessary to talk about subsets. This means that the study of causal sufficiency addresses whether a cause or a combination of causes is one of many ways to reach an outcome. To illustrate the subset principle Ragin uses the following example: Consider two fuzzy sets relevant to students in a course; those who studied hard and those who performed well on the exam. To study hard is one of several ways to do well on the exam. Likely there will be individuals who perform well on the exam without studying hard. If we assume that all those that study hard perform well on the exam. It follows that membership in the fuzzy set studied hard will be a subset of membership in the fuzzy set performed well on the exam. The outcome will likely be that the relationship will be that those who studying hard also do well on the exam, but we will also have some student who will have a good result without studying hard. It would then be reasonable to argue that studying hard is sufficient but not necessary for performing well on the exam. My study In this part of the paper I will be more explicit in what I want to do in my doctorial thesis. The purpose with this study is do study how we should understand the educational system on local level with the starting point in the aroused conflicts and if these conflicts has any influence over user-participation. The relationship between my variables in part one of the study, with the starting point of my declaration of purpose will the relationship between my variables be the following: Dependent variable: levels of conflict Independent variable: organisational structure of the local school-system 20

21 In this model I present how I see the relationship between the theories in this paper: Power theory Governance Network Hierachy The basic theoretic foundation is power theory. This because the basic question in this research is power over common resources. If we see the relationship in the light of the first face of power one could say that we have a dominating actor, the municipality, that is the driving force. Its actions will affect actor B, the charter schools. The municipality acts in a way that will affect the charter schools. That could be sharing information for example. If we then see this problem in the light of the second face of power we could argue that specific institutional arrangements could be what cerates these conflicts. What I mean then is that the school system is not adapted to the new situation so the local government has trouble in handling this new situation. I also see that the power theory could be a theoretical link between the first purpose, were much of the focus is on structure, to the second purpose where more of the focus is on the actors in terms of users. The starting point in understanding the organisation of the school-system is taken in the theoretical perspective of governance. Governance focuses much on cooperation between the actors. But in some municipalities this doesn t seem to work. And why is that? To investigate this I have developed a model that describes how this might be studied. This model below is developed from the previous model but here the governance perspective is developed. Model over the types of governance A cooperative relationship A non-cooperative relationship Network (network theory, pluralism) Hierarchy (corporatism) Spreading of power Concentration of power Openness Closeness 21

22 In this model I try to do a theoretical distinction between municipalities where the cooperation between the local government and the charter schools works well and in those it don t. The model should be seen as the analytical tool for analysing the empirical problem expressed in the first purpose. The question that could be asked is; Are the municipalities that are characterized by conflict more hierarchal organized and has a concentration of power to a few actors or is it the other way around? I argue that relationship that works well could be understood as non hieratical and open systems with a distribution of power between the actors. The municipalities that are characterized by conflicts are understood in this model as municipalities that have a very hierarchal structure for governing s and a high amount of concentration of power to the local government. My question is if we can se governance in this way or if I should complement the governance perspective with a theory that focuses on hierarchy. As shown in the model above, one alternative that I have been thinking at is to add network theory and pluralism to the success cases and corporatism to the part that describes a hierarchical structure in the non success cases. But this is something that I need to work more with. For this investigation I am going to use a comparative approach and a theory testing one. Both the case study and the comparative approach are useful for me, but because I am more interested in understanding the general rather than the specific of the phenomena. As the strategy for selecting the cases I m going to use MSSD. This because I am interested in differences on the dependent variable. Another reason is that we only have a limited number of cases to study. As method for collecting the data I am mostly going to use interviews and studies of official documents. The technique for analysing the data will be Fuzzy set. This because fuzzy set is especially useful when we as a researcher are dealing with categories that are imprecise which I am doing. Fuzzy set also makes it possible to operationalize multiple interpretations of a concept. For me that is the varying degree of cooperation between actors and the levels of conflict. In the second part of the investigation I will develop the dependent variable and focus on the possibilities for the citizens to become involved and their ability to participate, both at the municipality level and on the school level. This is one part that I need to develop more and I am eager to get comments on how that could be done. 22

23 In the second part of the investigation I m changing the perceptive from the organisational matters to the user perspective. It is in this part that the theories of influence and Hirschman s exit and voice will be useful. What I want to investigate in this part is if the possibility for the users of charter schools is dependent of the level of conflict. This is also something I need to develop more, and I would be grateful for any comments. To connect the first purpose with the second the following table could illustrate that connection Level of Conflict High Low High Yes No Participation Low No Yes The table states that municipalities with high levels of conflict also show high levels of participation. The thought behind this is that conflict triggers the will to participate. Here I need a theoretical connection to the model that confirms this way of thinking. In part two I am, for the most part, going to use interviews to collect the data. I am also considering using a questionnaire but at this stage of the research project it is too early to say. 23

24 References Anckar, Carsten, (2004). Some reflections on the applicability of most similar systems design and the most different systems design in comparative research, Åbo, Åbo Akademi Bachrach, Peter & Morton S Baratz, (1962) Two faces of power American Political Science Review, V. 57: Björk, Peder, Göran Bostedt och Hans Johansson, (2003), Governance, Lund: Studentlitteratur Dahl, Robert A. (1957), The concept of Power, Behavioural Science, V. 2, Dahl, Robert A. (1961), Who Governs? Democracy and Power in an American City, New Haven: Yale University Press Denk, Thomas, (2002), Komparativ metod- Förståelse genom jämförelse, Lund: Studentlitteratur Esaiasson, Peter, Mikael Gilljam, Henrik Oscarsson, Lena Wängnerud, (2004). Metodpraktikan, Stockholm, Norstedts Juridik Friskolornas riksförbund, (2005), Fakta om friskolor, Mälartryckeriet, Stockholm Hirschman. Albert O. (1970), Exit, Voice and loyalty, Cambridge: Harvard University Press Johansson, Marcus, (2002) Exkludering av invandrare I stadspolitiken - Makt och maktlöshet I Örebro , Örebro: Örebro Universitet King, Gary, Robert O. Keohane, Sidney Verba. (1994), Designing Social Inquiry- Scientific Inference in Qualitative Research, New Jersey, Princeton University Press. Klasson, Torgny, ( 2000) Medborgardeltagande, brukarinflytande och konsumentanpassning, in Egevall, Lena & Torgny Klasson (red.) Demokrati i praktiken. Lund: Studentlitteratur Lidström, Anders & Christine Hudson, (1995) Skola i förändring -decentralisering och lokal variation, Stockholm: Nerienius & Santérus Förlag Jarl, Maria, (2004) En skola i demokrati- Föräldrarna, kommunen och dialogen, Göteborg: Göteborgs universitet Montin, Stig, (1993) Svenska kommuner i omvandling, Örebro: Novemus Montin, Stig, (2000),Between fragmentation and Co-ordination- The changing role of local government in Sweden, Public Management v. 2/ Polsby,Nelson W. (1963,1980) Community Power and Political Theory, second edition, Yale University Press: New Haven 24

25 Proposition 1991/92:95, Om valfrihet och fristående skolor Przeworski, Adam & Henry Teune, The Logic of Comparative Social Inquiry, Malabar, Robert E. Krieger Publishing company Ragin, Charles C., (2000), Fuzzy Set Social Science, The Univeristy of Chicago Press, Chicago. Internet