xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam a22 u 4500
controlfield tag 008 c20019999azu 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E11-00211
Educational policy analysis archives.
n Vol. 9, no. 14 (April 24, 2001).
Tempe, Ariz. :
b Arizona State University ;
Tampa, Fla. :
University of South Florida.
c April 24, 2001
Conceptualizing the process of education reform from an international perspective / Benjamin Levin.
Arizona State University.
University of South Florida.
t Education Policy Analysis Archives (EPAA)
xml version 1.0 encoding UTF-8 standalone no
mods:mods xmlns:mods http:www.loc.govmodsv3 xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govmodsv3mods-3-1.xsd
mods:relatedItem type host
mods:identifier issn 1068-2341mods:part
mods:detail volume mods:number 9issue 14series Year mods:caption 20012001Month April4Day 2424mods:originInfo mods:dateIssued iso8601 2001-04-24
1 of 17 Education Policy Analysis Archives Volume 9 Number 14April 24, 2001ISSN 1068-2341 A peer-reviewed scholarly journal Editor: Gene V Glass, College of Education Arizona State University Copyright 2001, the EDUCATION POLICY ANALYSIS ARCHIVES. Permission is hereby granted to copy any article if EPAA is credited and copies are not sold. Articles appearing in EPAA are abstracted in the Current Index to Journals in Education by the ERIC Clearinghouse on Assessment and Evaluation and are permanently archived in Resources in Education .Conceptualizing the Process of Education Reform From An International Perspective Benjamin Levin Manitoba Education and Training Winnipeg, Manitoba, Canada Abstract A great deal of comparative work on education refor m is now being done, but this work often lacks a clearly articulat ed conceptual frame. This paper, based on a study of change in five juri sdictions in four countries, develops a model of reform based on four interactive elements origins, adoption, implementation, and outcomes. Within each of these elements, questions and concepts from the relevant literature are developed with the intent of building a more compre hensive approach to the analysis of reform from political, organization al and educational perspectives. The past twenty years has seen an enormous amount of change in education policy in the industrialized English-speaking countries. I n the United States reform has been a constant since at least the publication of A Nation At Risk in 1983, though as Mazzoni
2 of 17(1993) points out, there was also a great deal of s tate level reform before 1983. Other countries, too, have made significant changes in ma ny aspects of education policy. England and New Zealand had particularly dramatic p eriods of change during the 1980s and early 1990s. Australian states and Canadian pro vinces, like U. S. states, have seen significant policy shifts, though, again like the U S., it is the rare reform that is actually taken up universally. All of this reform has been the subject of an enormous amount of discussion and analysis (Fowler, 1995). Some of the most controver sial reforms, such as increased assessment, decentralized governance and parental c hoice of school have generated vast amounts of literature. Many case studies of particu lar reforms have been done, whether at the local, state or national level. Scholarly work comparing education reform a cross countries has also developed rapidly in the last few years (e.g. Beare & Boyd, 1 993; Carter & O'Neil, 1995; Glatter, Woods & Bagley, 1997; Whitty, 1997; Whitty, Power & Halpin, 1998). It seems clear that such work is likely to grow as states or count ries are increasingly interested in policy directions elsewhere (Levin, 1998) and as sc holars consider the implications of internationalization. This comparative work is interesting and important, but much of it lacks a clearly articulated conceptual frame and is therefore relat ively a-theoretical. Studies typically do not explain why they have chosen to look at particu lar features of change, or how these features might be related to each other or to large r frames of reference drawn from the relevant literature. Yet comparative work is partic ularly challenging conceptually (e.g. Kohn, 1987; Archer, 1989) because of the myriad var iables and perspectives that could be used to explore the issues. This paper, which is part of a larger comparative project, develops a conceptual framework for studying educat ion reform across political jurisdictions with the intent of being useful to sc holars working in these areas. The most complete available effort to devel op an overall frame of analysis of education reform is, in my view, the work of Stephe n Ball (1990) on reform in England. Ball introduces a variety of frames in which to thi nk about education reform. He points out the important differences in thinking about ref orm from economic, political or ideological perspectives, and describes the differe nt emphases resulting from forms of analysis that focus more or less on structural, int eractional or discursive elements. Ball also recognizes the dangers that analysis can lead to an excessive determinism. Many accounts, he suggests fail to capture the messy realities of influence, p ressure, dogma, expediency, conflict, compromise, intransigence, resistance, er ror, opposition and pragmatism in the policy process. It is easy to be simple, neat and superficial and to gloss over these awkward realiti es. It is difficult to retain messiness and complexity and still be penetrating. (p. 9) In another work (Bowe, Ball & Gold, 1992), the authors develop a tripartite frame for thinking about education reform. They describe education policy as having three phases that are at least partly autonomous from eac h other Â– influence, text production and practice. Influence is the process of bringing policy into being, having to do with who shapes the nature of policy. Text production de als with the creation of policy as a product. Bowe, Ball and Gold point out that what ge ts produced as policy often deviates from at least some of the intentions of its promote rs. Third, they note that education practice, while it is influenced by policy texts, a lso has a degree of autonomy from them. They characterize these three elements as forming a "trajectory of policy."
3 of 17 Ball's work is important and has rightly be en frequently cited (though a large body of work on education reform continues to appear wit hout being linked to any explicit framework). However since it appeared there have be en relatively few efforts to extend or deepen the framework. In this paper I attempt to develop an approach to thinking about and studying education reform that builds on but also (I hope) provides a fuller basis for both analytical and descriptive work on e ducation policy.The Research Base The analysis in this article grows out of a research project that looks at education reform in five different settingsÂ—the Canadian prov inces of Alberta and Manitoba, the state of Minnesota, and the countries of England an d New Zealand. (To be precise, England is neither a country nor a province or stat e, but a part of Britain. However, because education in England and Wales differs from that in Scotland and Northern Ireland in important respects, my work has been lim ited to the former.) In each setting I looked at a major reform process from its inception to outcomes several years later. The study has used a variety of data source s, including official documents (position papers, records of legislative debates, newspaper a rticles), secondary analysis of data from other studies, and interviews with key partici pants in several of the settings. The study will be fully described in a forthcoming book (reference deleted for author anonymity). Stage Theory as a Basic Approach Every study of policy uses some version of a stage theory. There are many of these (a good overview can be found in Howlett and Ramesh 1995), all of which involve some series of stages moving from the identificatio n of a problem through the identification or adoption of particular strategies to issues of implementation and impact. The tripartite analysis of Bowe, Ball and Gold is o ne example. To an extent, the choice of stages is a matter of personal preference. It is vital to remember that although the division into component parts is useful for analyti cal purposes, in reality reform is not neatly divided in this way, nor can any set of head ings adequately represent the complexities of a reform process. In political anal ysis, discrete categories and periods are devices of the analyst, not the experience of t hose directly involved. The four components I use here are: Origins. Where did particular reform proposals come from? How did they become part of the government agenda, when so many ideas d o not? What role did various actors and interests play in the development of ref orm programs? 1. Adoption. How do policies as finally adopted or mad e into law differ from those originally proposed? What factors lead to changes b etween proposals and approval? Who supported and proposed various polici es, and to what effect? 2. Implementation. A considerable body of research, in education and other policy fields, lays out clearly the difficulties of moving from policy to practice. What model of implementation, if any, did governments us e to move their reforms into practice. What "policy levers" were used to support reforms? How did schools and school systems respond to reforms? 3. Outcomes. Interest here is on the available evidenc e as to the effects of reforms. Any political action may have a number of results, some of which were intended by policy-makers and others which were not. Because the reforms under study are 4.
4 of 17about education, the study gives particular attenti on to what may be known about how the reforms have affected student outcomes and learning processes in schools. This delineation draws more attention to th e political process of reform, since it highlights the degree to which policy ideas are sub ject to modification not only after they are promulgated, but at every step of the process. The framework also gives explicit attention to the process of implementation, which s ome other analyses tend to underplay.Developing the Framework Creating four major stages or elements is o nly the first step towards an adequate conceptual framework for a comparative policy study It is also necessary to provide much greater specification of the main consideratio ns within each of the four stages, so that general terms such as "adoption" or "implement ation" are fully developed conceptually. Each of the elements must also be gro unded in the relevant literatures. However before taking up the four main themes in th is way, attention has to be given to two general themes: the overarching importance of c ontext in comparative policy analysis, and the balance between linearity and con tingency in thinking about political processes.A first consideration: The importance of a historic al and cultural perspective Education reform is political work, and pol itical work can only be understood appropriately in a historical and cultural perspect ive. Reforms necessarily arise in particular social, economic, political and institut ional contexts. The way any reform program is conceptualized, developed, defended (and attacked), and implemented will owe a great deal to previous events and practices i n a given jurisdiction. A few examples can illustrate. The legacy o f social class distinctions in Britain, including a long history of elitism in education, c onditions the way all new reform proposals are seen (Whitty & Edwards, 1998). The de bate over such proposals as opting out (grant-maintained schools) or parental choice o ccurs in a system that for most of its history has provided educational opportunity based largely on class background. In the United States, on the other hand, concerns about ra ce relations have been at least as powerful as class issues in shaping debates about t he same kinds of issues. In Canada, with a long history of separate but co-existing ins titutions based on language and religion, concerns about choice are much more muted Â“Separate but equalÂ” means something quite different in Canada than it has in the U. S. The politics of language, religion and ethnicity can have powerful effects on education and are strongly conditioned by unique historical factors in each se tting. Institutions of government and political cu ltures are also powerful influences on education reform with quite variable effects. Feder al states will approach reform differently than unitary states. American political institutions, with their constitutional checks and balances, provide very different politic al dynamics than do parliamentary majority governments. Jurisdictions, whether countr ies, states, or provinces, can have very different political cultures and practices dep ending on whether they have a history of strong or weak executives, a single house or two a two-party system or a history of multiple parties and coalitions. For example, a two party state with a long history of polarized politics will have quite different politi cal processes than one with multiple parties and a history of seeking central ground.
5 of 17 Even those developments that are seen as ha ving international scope need to be viewed in a historical perspective. The current zea l for reform in education, for example, follows a long period of expansion of educational s ystems accompanied by a high level of confidence and substantial resourcing. Economic cycles have a strong impact on people's thinking about education; when economic ti mes are good, there tends to be more optimism about education (Krahn, 1996). Increa sed attention to consultation in policy-making (whether real or symbolic) and to the role of parents is very likely connected to an increasingly educated populace and growing skepticism in many quarters about the degree of discretion that should be afforded to professionals. One does not have to be a Hegelian to see the influence of dialectic in these developments. All of this means that the analyst should, in studying reform, pay careful attention to the way that new developments emerge from past e vents and practices. This is a particular problem in comparative work, in that the more jurisdictions one considers, the harder it is to develop a reasonable degree of fami liarity with the requisite background. The intended and the contingent: Conceptualising re form as a whole. The common view of reform tends to assume t hat political or ideological analysis leads to a reform program that in turn leads to cha nges in practice leading to particular outcomes. Politics is treated as largely an intelle ctual process embodying a relatively straightforward means-ends rationality. One body of work embodying these assumptions operates at a high level of abstraction, concerned with such matters as the changing role of the state and the impact of globalisation as det ermining forces in political events (e.g. Ball, 1998; Carter & O'Neil, 1995; Taylor et al., 1 997). Education reform in these treatments is often discussed as the implementation of a set of well-defined political views having to do with beliefs in the reduced role of the state or the primacy of markets over public provision. The same line of thinking ca n be found in many of the analyses and critiques of such recent policies as school cho ice (e.g. Lauder et al.; 1999). Reform is certainly driven in some important ways b y a linear, means-ends rationality. If there were no sense that an action would produce pa rticular consequences there would be no reason to undertake the action. In politics, careful calculation of consequences is of absolutely central importance, although the cons equences that are of interest include personal, partisan and symbolic outcomes as well as substantive policy consequences (Edelman, 1988). At the same time, it is important not to ov erstate the degree to which reform is driven by a straightforward rationality. Usually on e finds a high level of ambiguity and contingency in every aspect of the political proces s. At every step, multiple and conflicting influences come to bear, purposes chang e or are worn down by existing structures and processes, and circumstances change in ways that require modification of plans and actions. As Ball puts it: National policy making is inevitably a process of b ricolage: a matter of borrowing and copying bits and pieces of ideas from elsewhere, drawing upon and amending locally tried and tested approach es, cannibalising theories, research, trends and fashions and not inf requently flailing around for anything at all that looks at though it might w ork. Most policies are ramshackle, compromise, hit and miss affairs, that are reworked, tinkered with, nuanced and inflected through complex process of influence, text production, dissemination and, ultimately, re-creat ion in contexts of practice. (1998, p.126)
6 of 17 The literatures on organization change and on government policymaking support a model that goes beyond a simple means-ends rational ity to include attention to the inevitable ambiguity and contingency of the politic al process. Work in political science (such as Edelman, 1988, and Stone, 1988) remind us that that politics is shaped by many considerations other than policy, including the req uirements of staying in office and the vicissitudes of the moment. Symbolic and emotional issues are often at the forefront of political work. Work in politics also emphasizes th e degree to which the entire process of policy development and implementation takes plac e in a short-term context that is constantly changing, multifaceted, and very diffi cult to read but is also affected by long-term trends in government and society. Dror (1 986) describes this environment as Â“fuzzy gamblingÂ”, in which the rules change while t he game is played and Â“surprise dominatesÂ”. Dror is among writers (see also Lindblo m, 1990) who also point out that even in the best of circumstances human abilities t o understand problems and generate appropriate solutions are limited and often inadequ ate. Most of the time governments are operating under circumstances that are far from opt imal in this regard due to pressures of time, lack of information, and multiple competing i ssues. Work under the heading of neo-institutional ism (March & Olsen, 1989; Wilson, 1989) shows the powerful effect that institutional structures have on the political process, constraining the available political choic es and also shaping the way in which political decisions are put into effect, not necess arily consistent with the intentions of their originators. Institutions possess considerabl e ability to resist changes or to reduce their impact significantly. Research on policy effe cts (e.g. Elmore, 1995) also suggests that in many cases strategies for reform may focus on elements that cannot produce the kinds of changes that are really wanted or, to put it another way, that reforms focus on what can be done instead of on what might really ma ke a difference. The task of the analyst, then, is to consid er the ways in which policies are driven by a particular logic or ideology, but also the ways i n which they are shaped by other factorsÂ—historical, cultural, institutional, and po liticalÂ—that are far less predictable. At the macro level, long-term changes in societies and the role of the state are important. At the micro level, chance, in the form of individual personalities or unexpected events, is also an important consideration in understanding re form. Neither the importance of means-ends rationality nor the underlying contingen cy of life can be ignoredÂ—both must be accommodated in an adequate theoretical account. One way of thinking about this balance is t o think of the logic of policyÂ—actions leading to particular consequencesÂ—as the numerator in an equation. However the power of this numerator is affected by a denominato r that contains all the contingent elements. Sometimes the numerator is larger and pol icy is driven by careful strategy. Other times the denominator is larger and policy is primarily the result of unforeseeable elements.Considering the Four Stages With those points in mind, we turn to a fu ller discussion of the four elements of the proposed framework: origins, adoption, implementati on and outcomes. The origins of reform. The work of Kingdon (1994) provides a valua ble approach to looking at the origins of policy. Kingdon believes that political decision s emerge from the interaction of three streams: political events, problem recognition, and policy proposals. The balance of
7 of 17importance between political events, problem recogn ition and policy proposals, and the way the three streams interact will vary from setti ng to setting. Within each of these streams, additional wo rk has been done in this study to define key elements. Kingdon's framework also needs to be adjusted to take into account the differences between parliamentary systems and the U S governance model of separation of powers. Political events are often under-emphasize d in analyses of education policy because they are so unpredictable. The outcome of p olitical processes, however, can be affected strongly by such factors as stages in the political cycle, the internal dynamics of the governing party, the personalities of important actors, the nature of the relationships among key players, and unanticipated events or cris es. Problem recognition involves a variety of i nfluences bearing on political decision makers and on each other. One important set of infl uences is found within government itself, including both political and bureaucratic e lements. Within-government political sources of problem recognition can include individu als in key roles (such as ministers, governors, or members of a legislature), political parties and their associated bodies, legislatures, and central agencies of government (s uch as finance units or central policy units). Sometimes policy issues emerge not from the particular policy area but as a result of broader government agendas, such as a focus on r educing expenditure. Problem recognition also emerges from the apparatus of gove rnment as various agencies try to convince politicians of the importance of particula r issues or problems, or expand their own influence and budget, or manage external pressu res. External influences on the definition of pr oblems are wide ranging. They can arise through various consultative processes as well as f ormal and informal lobbying efforts by many different interests. An area like education which involves a great deal of money and affects a large number of people, is the subject of a great deal of lobbying from many different directions. As Lindblom pointed out many years ago (1980), because of its economic importance business plays a particularly significant role in shaping problem agendas. Research, polling and medi a reports are also a potential source of problem definition, though typically medi ated through some kind of lobbying process. Policy proposals are often, but not always, connected to problem definition, so many of the same sources are important in developin g or promoting policy proposals. Policy proposals have eventually to get the approva l or support of political leaders. However the actual ideas can come from many differe nt places. In fact, Kingdon suggests that policy proposals have so many differe nt versions and sources that the search for their origins is a vain task (1994, p. 7 1). Wherever they originate, policy proposals m ay be promoted by individual politicians or civil servants, political parties, g overnment agencies, lobby groups, think tanks, policy entrepreneurs, or from research of va rious kinds. The process of policy borrowingÂ—of taking i deas from one jurisdiction and applying them to anotherÂ—is also relevant to the di scussion of both problem definition and policy proposals. Much of the comparative work cited earlier argues that current education reform proposals have moved from one coun try to another. However careful examination is needed to determine whether referral s to other country's policies are anything more than rhetorical. Similar policy label sÂ—such as "choice"Â—may hide very great differences in policy content. This is so acr oss states in the U. S. (see Mintrom & Vergari, 1997) and provinces in Canada (Levin, 1998 ), and even more so across national boundaries (Halpin & Troyna, 1995). The entire process of shaping policy also o ccurs within a more general context of
8 of 17ideas and policy preferencesÂ—what Schon (1971) call ed "ideas in good currency." These ideas provide a taken-for-granted backdrop to polic y discussions, helping to determine the range of ideas that will even be considered, le t alone adopted. Changes in dominant ideas about the role of the state, for example, hav e been important in recent years in many English-speaking countries (Manzer, 1994). Ide as are themselves affected by and affect changing economic and social conditions. The re is a constant interaction between changes in material conditions, the way that people name or explain these, and the proposals that may be advanced to address perceived problems. These more fundamental ideas about policy a re often labelled as "ideological." Some argue that all policy is ideological by defini tion, but if so than the idea loses its analytic value. Others use the term "ideological" a s rather meaning the absence of common sense, and apply it to those policies with w hich they do not agree (e.g. Lawton, 1994). Again, it is not clear how such terminology is very helpful in any analytic sense. Manzer (1994) distinguishes between ideolo gy as a public justification or rationale for reform, and ideology as an actual constituting element of reform, whether so stated publicly or not. Any combination of these possibili ties could exist that is, reform programs that are justified and constituted ideolog ically, programs that are justified ideologically but are in practice more pragmatic, p rograms that are justified in pragmatic terms but are actually ideologically constituted, a nd programs that are neither justified nor constituted ideologically. It could also be arg ued that a policy could have effects that are ideological regardless of either justification or its constitution. There is a danger here of circular argument, in which every policy can be found to be ideological on at least some criterion (for a fuller discussion see Levin, in press). All of this suggests that discussion of the role of ideology in education reform needs to be well grounded empirically, and that ana lysts need to be clear about what they mean when they make claims about ideology and polic y. Adoption For purposes of this analysis, adoption is the process of moving from a policy proposal to an approved piece of legislation, regul ation, or policy. The literature in this area in regard to education reform is sparse even t hough policies often change in important ways from inception to final adoption. In the adoption process several elements collide. W hat began as a slogan or a conceptÂ—school choice, local management, open enrol ment, provincial testing, charter schoolsÂ—must be turned into a detailed scheme in th e form of legislation, regulations, or policy guidelines so that it can actually be put in to place in a large and complex system. Both administrative and political issues can result Many important policy initiatives begin as ideas that are not fully developed, so tur ning them into something workable may involve quite a bit of debate as to what the in tentions originally were and how they can best be realised. The debates can be political, in that opponents of a particular reform in and out of government may revisit their c oncerns as the details are worked out. At other times the issues will be administrative as the system tries to work out detailed procedures for managing large-scale changes. Althou gh all of these processes may be intertwined and often occur simultaneously, they ca n usefully be considered under the headings of internal political debate, bureaucratic accommodati on, and public political debate Internal political debate refers to discus sion within a government, among contending political factions. These debates can be motivated by substantive disagreements about policies, by arguments over the politics of action (such as whether
9 of 17the timing is right for a particular idea), or just as easily by rivalries and animosities between individuals or organizational units. The de velopment of detailed plans or the requirement to approve budgets or legislative propo sals can reopen internal disagreements about how strongly or how far a polic y should be pursued, or which part of the bureaucracy should manage it. These debates may continue long after a policy is officially announced, and even after its implementa tion. It is not always easy to know which people are going to be key to a policy's fate. Heads of government are vital, but they are not nec essarily the only essential actors. Ministers of finance and their deputies can be espe cially important if a new policy has financial implications Â– as they almost always do. Heads of other units may be important if they see a policy as affecting their own program s or plans or if their co-operation is needed to move a proposal forward. In the U. S. sys tem, individual legislators of any party can play a vital role in the adoption process regardless of the position of the executive. The political process of adoption often le ads to policy proposals that are vague or even contradictory. Politicians are often amateurs in the substantive policy field who may not understand the complexities of existing org anization, the impact of other competing agendas, or the difficulties that inevita bly arise in the attempt to move from a general idea to a specific set of procedures. Civil servants, on the other hand, as Wilson (1989) points out, are concerned to make the system work as smoothly as possible. They may have no personal commitment to a government's p urposes, but they do have to think about the procedures in detail, what could go wrong who will administer or manage the policy, how exceptions will be handled, what timeli nes are possible, what resources will be neededÂ—all things that may not be part of an att ractive political vision. As a policy proposal moves towards implementationÂ—or, as Fitz a nd Halpin (1991) describe it in their study of Grant Maintained School policy in En gland, Â“from a sketchy policy to a workable schemeÂ”Â—there will inevitably be a process of limiting and narrowing, of trying to rub off the sharp corners of policy that will create the most difficulty, of trying to make new policies at least partly consistent wit h existing practices. Much of this discussion takes place in a p ublic political arena in which conflicts over both intent and implementation are debated. Th e latter may include the "official" debate in parliament or a legislature as well as th e debate that goes on in public, through various consultation processes, the media, and with various interest groups. Through the entire process, all sorts of proponents and opponen ts of reform are trying to advance their position and counter opposing arguments, so r eforms are frequently accompanied by intense political disagreement. In the oppositio nal world of politics, even groups that rather like an idea may take a critical stance in p ublic. Public political debate involves an effort to "frame" thinking (Davies, 1999), or to shape discourses (Ball, 1990) about educational iss ues. The parties to the debate have various devices available to them in their efforts. Governments can use such means as press releases, white papers, speaking tours, legis lative committees and hearings, advisory groups and public consultation processes i n efforts to mould public opinion. Opposition groups have their own set of devices, su ch as public rallies and the media's interest in conflict and controversy, to get their views across. Public distrust of government is itself a weapon that can be used by o pponents. In Canada and the United States opponents have also sometimes used the court s to attack elements of various reform programs. A number of other devices can be u sed by any of the parties, such as issuing official statements and press releases, adv ertising, polling, research, and the citing of authorities. All sides may appeal to the views of supposedly neutral or objective third parties; research is often used in this way i n political debate.
10 of 17 The process of adoption is also influenced by the degree of commitment to a given program. Some reforms may be deeply important to go vernments, in which case a high degree of conflict may be tolerated before there is any willingness to make changes. Sometimes a government actively seeks conflict on a n issue as a way of convincing its supporters of its commitment. In other cases a spon sor may only be willing to expend a small amount of political capital before retreating from a position. These decisions are themselves influenced by the kinds of political eve nts discussed earlier, such as the timing in relation to an election, leadership rival ries, or other competing issues. The study of adoption is important because it shows the ways in which intentions are modified by political and administrative consid erations. Since reforms often end up looking rather different than was originally intend ed, an analysis of the sources of these changes should be an important part of an overall a nalysis of reform. Implementation In contrast to adoption, the literature on the problems of implementation is very large and quite well developed (e.g. Fullan, 1991; McLaughlin, 1987). A whole series of obstaclesÂ—some of them generic to policy implementa tion and others particular to schoolsÂ—stand in the way of policies being put in p lace as intended. Although the problems of implementation are well known, governme nts have tended to give relatively short shrift to these issues in the policy process. Perspectives on implementation have become more complex over time. While the greatest amount of attention has been given to ways of making implementation more effective, another body of work (e.g. Hargreaves, 1 994) takes the view that disputes over change are best thought of as political struggles, and that resistance may be well-justified. More recently there has been increa sing interest in ideas about organizational learning, and the need to see change as a process of testing and refining ideas on the basis of evidence about their impact a nd value. Factors affecting implementation can be tho ught of (Fullan, 1991) as pertaining to the change itself, to the setting where implementat ion is to occur, and to the wider context. The first heading would include the clarit y of the change and the degree of difficulty involved in implementation. Political di rection for change is often either vague or contradictory because of the need to reconcile d ivergent interest. Further complexity and lack of clarity occur because education does no t have a generally accepted core technology, or way of doing things, but depends gre atly on the values and approaches of individual teachers or administrators. Thus almost all policy is subject to extensive interpretation. The second category includes the degree of understanding of the proposed change, the level of commitment to it by relevant actors, a nd the various resources allocated to support change. Commitment to change is shaped not only by the skills of those involved, but also by their attitude to a given ref orm. Attitudes in turn are shaped by educators' views of the practicality of a proposal, and also by its fit with the existing culture of schools. These two elements also reinfor ce each other, such that culture shapes attitudes to practicality and practicalities of teaching also shape school culture. The third category includes the other press ures either supporting or inhibiting implementation, such as competing demands and commu nity support or opposition. Not only are education reforms themselves often multi-f aceted and sometimes inconsistent, but they take place in a context which is itself ch anging. Too many reforms happening too quickly may lead to increasing cynicism and res istance in schools. The availability or lack of resources also affects the willingness of s chools to adopt particular reforms.
11 of 17 The concept of "policy levers" or "policy i nstruments" (Howlett, 1991; McDonnell & Elmore, 1987) is an important part of this analys is. Governments have a number of means they might use to promote implementation. The most important of these levers are mandates (such as legislation or regulation), i nducements (money, recognition), capacity building (training, professional developme nt, research), system changing (reorganization, governance changes), and opinion m obilization (exhortation, public pressure). However in few cases do governments appe ar to develop comprehensive strategies to support implementation of their polic ies. Implementation cannot be taken for granted Many reforms end up leaving few lasting marks on the system they were designed to c hange. Consideration of the ways in which implementation is structured and supported is thus an important part of the overall analysis of reform.Outcomes Reform programs are always justified on th e basis of the outcomes they will yield. Most reforms, however, are justified on the basis o f a number of outcomes, and these are sometimes quite different from one another. Vigorou s debate already exists around the outcomes of such changes as more student assessment or increased parental choice of schools. As is always the case in education, the di scussion is made more difficult because the purposes of schools are multiple, somet imes mutually contradictory, and often very difficult to assess. Moreover, reforms m ay yield outcomes quite different from or in addition to those intended by advocates or feared by opponents. A framework for considering outcomes needs to include impacts on students, impacts on schools, and impacts on the broader soci ety. The most frequently cited reasons for education reform have to do with impact on students, with the most common outcome measure being some form of assessment of st udents' skill or knowledge in the various curriculum areas. However a variety of outc ome measures beyond academic achievement have also been used to assess the impac t of education policies. These include graduation rates, attendance rates, numbers of disciplinary problems, or rates of referral to special education. Students' assessments of the quality and va lue of their schools experience are an important, if seldom evaluated outcome indicator, i f only because they say something about motivation, which is absolutely critical to a ll other outcomes. Life-chance indicators are also very important, since the most important purposes of schools often have to do with what happens to students after they leave the institution. These could include such outcomes as postsecondary education rates, employment outcomes, interest in lifelong learning, income, and citizens hip indicators (such as propensity to volunteer, voting behavior, or criminality). A surprisingly large proportion of the rese arch on reform focuses on the impacts on schools rather than students. One of the most frequ ently assessed aspects of reform is its impact on teachers' work and their attitudes toward s their work. Outcomes related to work might include hours, time in and out of the cl assroom, attention to individual students, professional development activities, skil l levels or teaching practices. Indicators in regard to attitude include teachers' sense of effort, satisfaction and stress, among others. Reform is also held to have had diffe rent effects on administrators than on teachers, partly because governance changes have altered the work of administrators in important ways. However the relevant measures fo r administrators are generally similar to those used for teachers. Greater involvement of parents has been a g oal of most reform programs. Parents'
12 of 17active role in school governance and in their child ren's education as well as their sense of satisfaction with the school and their part in i t have been measured. Some reforms are intended to affect school programs. Curriculum changes or graduation requirements are obvious examples. Schoo l choice is also often argued on the basis that it will lead schools to diversify and im prove programs. Many reform programs have not given very much attention directly to teac hing and learning practices per se, with the possible exception of efforts to extend th e use of educational technologies. However improved teaching and learning practices ar e clearly central to the achievement of all school outcomes and so should be a key part of assessing any reform. Although changes in school organization such as devolution o f authority, are usually argued as means to achieve other more important ends, they co uld also be considered as outcomes in themselves. The larger social impact of reform has also been an important subject of debate. Economic, equity and social cohesion outcomes are a ll potentially important. Insofar as reform has frequently been justified on economic gr oundsÂ—that is, on the contribution of schooling to national economic successÂ—societal economic outcomes would be important indicators of the success of reforms. Suc h outcomes could include labor force participation rates, employment rates, earnings, an d productivity growth, not only for students but more generally. Many critics and some proponents of reform have been concerned about the potential of reform to increase inequity in society An important outcome measure is thus the extent to which reforms act either to redu ce or to increase the gaps in outcomes in society that are due to socioeconomic status, et hnicity, gender, or other demographic factors. A related issue is the degree to which ref orms serve to build or reduce an overall sense of community among people. Efforts to assess social cohesion have included measures of ethnic segregation, citizen participati on, or of attitudes such as tolerance. Many of these outcomes are clearly very dif ficult to assess. The assessment is made more difficult because policies are not the only fa ctors that produce outcomes Â– in fact, the evidence in education is that the most importan t shapers of many outcomes lie outside the school system. Reforms themselves may h ave a variety of unintended consequences, and because many reform programs are multifaceted, interactive effects are common such that developments in one area have strong effects on other elements. All of this suggests that analysts need to pay careful attention to their choice of outcomes and need to justify these in relation to t he particular reforms under study.Conclusion The elements of reform sketched here are, a s mentioned earlier, highly interactive rather than discrete. Considerations of implementat ion and outcomes may shape original intentions, just as perceptions of outcomes may mod ify policies and their implementation. Each of the four elements is import ant to understanding reform, but complex in its own right. Doing such analysis on a comparative basis is even more difficult. The study of education reform is complicated even in a single se tting, so trying to make comparisons across settings is fraught with additional difficul ties. One inevitably risks drawing comparisons without full knowledge of local circums tances, and seeing as similar what would, with closer analysis, look quite different. In these conditions researchers need to be especially careful to be clear about their presu ppositions and the frameworks within which they are assembling data and deriving finding s. My hope is that this paper will contribute to that process.
13 of 17AcknowledgmentsThis article is drawn from a larger study of intern ational education reform that has just been published as a bookÂ— Reforming Education From Origins to Outcomes (London/New York: Routledge-Falmer, 2001). The rese arch was supported financially by the Social Sciences and Humanities Research Coun cil of Canada. Particular thanks are due to Jonathan Young. Many other colleagues ar ound the world also contributed to the study. An earlier version of this work was pres ented at the American Educational Research Association, New Orleans, April 2000. All views and opinions are solely those of the author.ReferencesArcher, M. (1989). Cross-national research and the analysis of educational systems. In M. Kohn (Ed.), Cross-national research in sociology (pp. 242-262). Newbury Park: SAGE. Ball, S. (1990). Politics and policy making in education London: Routledge Ball, S. (1998) Big policies/small world: an introd uction to international perspectives in education policy. Comparative Education 34(2), 119-129 Beare, H. & Boyd, W.L. (eds.). (1993). Restructuring Schools London: Falmer Press. Bowe, R., Ball, S. & Gold, A. (1992). Reforming education and changing schools London: Routledge.Carter, D. & O'Neill, M. (eds.) (1995 ). International perspectives on educational reform and policy implementation. London: Falmer. Davies, S. (1999). From moral duty to cultural righ ts: A case study of political framing in education. Sociology of Education 72(1), 1-21. Dror, Y. (1986). Policymaking under adversity New York: Transaction. Edelman, M. (1988). Constructing the political spectacle Chicago: University of Chicago Press. Elmore, R. (1995). Structural reform in educational practice. Educational Researcher 24(9), 23-26.Fitz, J. & Halpin, D. (1991). From a "sketchy polic y" to a "workable scheme": the DES and grant-maintained schools. International Studies in Sociology of Education 1(2), 129-151.Fowler, F. (1995). The international arena: The glo bal village. In J. Scribner and D. Layton (eds.), The study of educational politics (pp. 89-102). Washington: Falmer Press.
14 of 17Fullan, M. (1991). The new meaning of educational change New York: Teachers College Press/ OISE Press.Glatter, R., Woods, P. & Bagley, C. (eds.) (1997). Choice and diversity in schooling London: Routledge.Halpin, D. & Troyna, B. (1995) The politics of educ ation policy borrowing. Comparative Education 31(3), 303-310. Hargreaves, A. (1994). Changing teachers, changing times New York: Teachers College Press.Howlett, M. (1991). Policy instruments, policy styl es, and policy implementation: National approaches to theories of instrument choic e. Policy Studies Journal 19(2), 1-21. Howlett M. & Ramesh,. M. (1995). Studying Public Policy: Policy Cycles and Policy Subsystems Toronto: Oxford University Press. Kingdon, J. (1994). Agendas, alternatives and public policies. 2nd edn New York: HarperCollins.Kohn, M. (1987). Cross-national research as an anal ytic strategy. American Sociological Review 52(4), 713-731. Krahn, H. (1996). School-work transitions: Changing patterns and research needs. Consultation paper for the Applied Research Branch of Human Resources Development Canada. Lawton, D. (1994). The Tory mind on education London: Falmer. Lauder, H., et al. (1999). Trading in futures: Why markets in education don't work Buckingham: Open University Press.Levin, B. (1998). An epidemic of education policy: (what) can we learn from each other? Comparative Education 34(2), 131-141. Lindblom, C. (1980). The policy-making process .2nd ed. Englewood Cliffs, NJ: Prentice-Hall.Lindblom, C. (1990). Inquiry and change New Haven: Yale University Press. Manzer, R. (1994). Public schools and political ideas Toronto: University of Toronto Press.March, J. & Olsen, J. (1989). Rediscovering Institutions New York: The Free Press. Mazzoni, T. (1993). The changing politics of state education policy making: A 20-year Minnesota perspective. Educational Evaluation and Policy Analysis 15(4), 357-379
15 of 17 McDonnell, L., & Elmore, R. (1987). Getting the job done: Alternative policy instruments. Educational Evaluation and Policy Analysis 9 (2), 133-152. McLaughlin, M. (1987). Learning from experience: Le ssons from policy implementation. Educational Evaluation and Policy Analysis 9(2), 171-178. Mintrom, M. & Vergari, S. (1997) Political factors shaping charter school laws. Paper presented to the American Educational Research Asso ciation, Chicago, April. Schon, D. (1971). Beyond the stable state New York: Norton. Stone. D. (1988). Policy paradox and political reason New York: Harper-Collins. Taylor, S., Rizvi, F., Lingard, R. & Henry, M. (199 7). Educational policy and the politics of change London: Falmer. Whitty, G., Power, S. & Halpin, D. (1998). Devolution and choice in education: The school, the state and the market. Buckingham: Open University Press. Whitty, G. (1997). Creating quasi-markets in educat ion: A review of recent research on parental choice and school autonomy in three countr ies. In M. Apple, (Ed), Review of Research in Education 22 (pp 3-48). Washington: AERA. Whitty, G. & Edwards, T. (1998). School choice poli cies in England and the United States: an exploration of their origins and signifi cance. Comparative Education 34(2), 211-227Wilson, J. (1989). Bureaucracy New York: Basic.About the AuthorBenjamin Levin, Ph.D.Deputy MinisterManitoba Education and Training162 Legislative BuildingWinnipeg, Manitoba, R3C 0V8E-mail: email@example.comBenjamin Levin is Deputy Minister of Advanced Educa tion and Deputy Minister of Education, Training and Youth for the Province of M anitoba, in which capacity he is the chief civil servant responsible for these areas. He is filling this position on secondment from The University of Manitoba where he is profess or of educational administration with particular interests in education policy, poli tics and economics. This is his third article for EPAA .Copyright 2001 by the Education Policy Analysis ArchivesThe World Wide Web address for the Education Policy Analysis Archives is epaa.asu.edu
16 of 17General questions about appropriateness of topics o r particular articles may be addressed to the Editor, Gene V Glass, firstname.lastname@example.org or reach him at College of Education, Arizona State University, Tempe, AZ 8 5287-0211. (602-965-9644). The Commentary Editor is Casey D. C obb: email@example.com .EPAA Editorial Board Michael W. Apple University of Wisconsin Greg Camilli Rutgers University John Covaleskie Northern Michigan University Alan Davis University of Colorado, Denver Sherman Dorn University of South Florida Mark E. Fetler California Commission on Teacher Credentialing Richard Garlikov firstname.lastname@example.org Thomas F. Green Syracuse University Alison I. Griffith York University Arlen Gullickson Western Michigan University Ernest R. House University of Colorado Aimee Howley Ohio University Craig B. Howley Appalachia Educational Laboratory William Hunter University of Calgary Daniel Kalls Ume University Benjamin Levin University of Manitoba Thomas Mauhs-Pugh Green Mountain College Dewayne Matthews Western Interstate Commission for HigherEducation William McInerney Purdue University Mary McKeown-Moak MGT of America (Austin, TX) Les McLean University of Toronto Susan Bobbitt Nolen University of Washington Anne L. Pemberton email@example.com Hugh G. Petrie SUNY Buffalo Richard C. Richardson New York University Anthony G. Rud Jr. Purdue University Dennis Sayers Ann Leavenworth Centerfor Accelerated Learning Jay D. Scribner University of Texas at Austin Michael Scriven firstname.lastname@example.org Robert E. Stake University of IllinoisÂ—UC Robert Stonehill U.S. Department of Education David D. Williams Brigham Young UniversityEPAA Spanish Language Editorial Board
17 of 17 Associate Editor for Spanish Language Roberto Rodrguez Gmez Universidad Nacional Autnoma de Mxico email@example.com Adrin Acosta (Mxico) Universidad de Guadalajaraadrianacosta@compuserve.com J. Flix Angulo Rasco (Spain) Universidad de Cdizfelix.firstname.lastname@example.org Teresa Bracho (Mxico) Centro de Investigacin y DocenciaEconmica-CIDEbracho dis1.cide.mx Alejandro Canales (Mxico) Universidad Nacional Autnoma deMxicocanalesa@servidor.unam.mx Ursula Casanova (U.S.A.) Arizona State Universitycasanova@asu.edu Jos Contreras Domingo Universitat de Barcelona Jose.Contreras@doe.d5.ub.es Erwin Epstein (U.S.A.) Loyola University of ChicagoEepstein@luc.edu Josu Gonzlez (U.S.A.) Arizona State Universityjosue@asu.edu Rollin Kent (Mxico)Departamento de InvestigacinEducativa-DIE/CINVESTAVrkent@gemtel.com.mx email@example.com Mara Beatriz Luce (Brazil)Universidad Federal de Rio Grande do Sul-UFRGSlucemb@orion.ufrgs.brJavier Mendoza Rojas (Mxico)Universidad Nacional Autnoma deMxicojaviermr@servidor.unam.mxMarcela Mollis (Argentina)Universidad de Buenos Airesmmollis@filo.uba.ar Humberto Muoz Garca (Mxico) Universidad Nacional Autnoma deMxicohumberto@servidor.unam.mxAngel Ignacio Prez Gmez (Spain)Universidad de Mlagaaiperez@uma.es Daniel Schugurensky (Argentina-Canad)OISE/UT, Canadadschugurensky@oise.utoronto.ca Simon Schwartzman (Brazil)Fundao Instituto Brasileiro e Geografiae Estatstica firstname.lastname@example.org Jurjo Torres Santom (Spain)Universidad de A Coruajurjo@udc.es Carlos Alberto Torres (U.S.A.)University of California, Los Angelestorres@gseisucla.edu