Building a Policy-Oriented Research Partnership for Knowledge Mobilization and Knowledge Transfer : The Case of the Canadian Metropolis Project

The aim of this paper is to examine government–university–community partnerships for knowledge mobilization (KM) and knowledge transfer (KT) in the area of immigration and settlement research using the illustrative case of the Canadian Metropolis Project. The Metropolis Project in Canada began in 1995 with the goal of enhancing policy-oriented research capacity for immigration and settlement and developing ways to better use this research in government decision-making. Core funding for this partnership was provided jointly by Citizenship Immigration Canada (CIC), a department of the Government of Canada and the primary social science granting agency, the Social Science and Humanities Research Council (SSHRC). As of 2012, and subsequent to three successful funding phases, the decision was made to end government and SSHRC core funding for this initiative, however, other non-governmental funding avenues are being explored. The longevity of this partnership and the conclusion of this specific initiative present an opportunity to reflect critically on the nature of such partnerships. This paper is an attempt to identify some of the key themes, issues and challenges related to research partnerships, KM and KT. Also, with the aid of an illustrative case, it aims to specify some of the possibilities and limitations of this kind of policy relevant knowledge mobilization. Special consideration will be placed on the context in which the demand for knowledge mobilization and knowledge transfer has emerged. This examination has considerable international relevance as the Canadian Metropolis Project offers the leading example of a research partnership in the field of immigration and settlement. OPEN ACCESS Adm. Sci. 2012, 2 251


Introduction
Building partnerships between government and nongovernmental actors has become increasingly important to modern governance.While the role of community-based organizations and the private sector in partnering with government is well examined, research partnerships between the state and nongovernmental actors have been rarer and generally neglected in terms of study.This is important for academic consideration because it allows us to better understand the broader range of partnership relationships the state has been engaged in.It may also signal government's innovative attempts to strengthen areas of weakness in the state, in particular with respect to enhancing policy capacity.
This paper offers an examination of one initiative, the Canadian Metropolis Project, to construct a research partnership between the Canadian federal state, academics and community organizations for knowledge mobilization and knowledge transfer for the purpose of enhancing policy knowledge capacity in the area of immigration, diversity and settlement.It is beyond the scope of this paper to provide a detailed history and analysis of the Canadian Metropolis Project itself.Rather our case will be used more selectively to highlight and illustrate various observations regarding the nature, value and challenges introduced by these partnerships and their impact on KT/KM and contribution to evidence-based policy capacity issues.
Two main arguments guide this paper.First, using the illustrative example (as differentiated from a traditional case study) of Canadian Metropolis Project, we argue that academic-government research partnerships are fostered by governments with an interest in bringing a stronger evidence-based approach to their policymaking decisions in areas of strategic importance with difficult policy problems.Such partnerships are viewed as particularly useful in periods when the government's own internal policy capacity has been weakened by such factors as expenditure restraint.However, the stability and longevity of such partnerships remain insecure due to the shifting importance of political factors in sustaining state commitments.Governments which are guided by conviction-based policy agendas have far less interest in supporting such partnerships and may even see them as threats to their own policy agenda.
Second, research partnerships contain power imbalances between actors as the various partners bring different expectations, values, cultures, capacities and needs to the relationship.For research partnerships to succeed, we contend, these differences must be acknowledged and accommodation for these differences made.Trust and understanding between the parties can only be build upon such foundations which is necessary for knowledge sharing critical to effective knowledge translation and knowledge mobilization.

The Research Approach
This work is informed by a broad political economy framework.Political economy concerns "the study of the social relations, particularly the power relations, that mutually constitute the production, distribution, and consumption of resources" [1].Consequently, political economy draws our attention to the structural location of the various actors and takes seriously the power relations embedded within the research partnership.Moreover, the political economy framework assists us in situating the partnership within the larger political and socio-economic context in which the Canadian Metropolis project existed, allowing for a deeper understanding of its ever evolving relationship.
This work also makes use of a participant-observer and practitioner-oriented perspectives.Co-author John Shields was a Director with CERIS-The Ontario Metropolis Centre (2002)(2003)(2004)(2005)(2006)(2007)(2008), and has been centrally involved with the Metropolis Project since its origins, at the regional Centre level.Co-author Bryan Evans brings the perspective of a policy practitioner having spent some sixteen years as a policy analyst and policy manager in the Ontario Government before moving to an academic appointment in the field of public administration and public policy.Consequently, the following analysis is based in part on a participant-observer approach and is informed by a policy practitioner perspective. 1 These different vantage points have naturally influenced our understanding of knowledge mobilization partnerships and knowledge transfer for policy effect.In particular, however, much of the knowledge and analysis of the Canadian Metropolis Project that follows is based on some sixteen years of participating in and observing this initiative from inside and close up and, hence, this informal knowledge informs this paper.
The Canadian Metropolis Project is used as an illustrative case in this study rather than as a more in depth case study approach.Consequently, the illustrative example of the Canadian Metropolis Project is integrated throughout the various sections of the paper.

The Canadian Metropolis Project: The Illustrative Case
Over the past decade and a half the Metropolis initiative has been the prime mechanism to achieve this goal in informing Canadian immigration and settlement policy.In 1995, led by the department of Citizenship and Immigration Canada (CIC), with support from other federal government departments, and in collaboration with the Social Sciences and Humanities Research Council (SSHRC), a project was launched with the objective to support a number of research Centres of Excellence across the country focused on policy relevant research in the area of immigration and settlement and its impact on major metropolitan centres in Canada.The stated goal of the Canadian Metropolis Project, in its initial phases, was: "to improve policies for managing migration and cultural diversity in major cities by: o Enhancing academic research capacity o Focusing academic research on critical policy questions, options and delivery mechanisms o Developing effective ways to use research in decision making" [3].
Over time this mandate was modified to move beyond immigration's impact only on large cities to include smaller centres and rural areas as well as a pan-Canadian comparative perspective.Four research Centres of Excellence based in Vancouver, the Prairie Provinces, Toronto and Montreal were 1 The authors are also engaged in ongoing related research examining the policy capacity of Canadian governments (see for example: Baskoy, Evans and Shields, 2011) [2].
established in 1996 (a fifth centre was added in 2004 in the Atlantic Provinces) for this purpose.In addition, CIC set up a National Metropolis Secretariat situated in Ottawa and staffed by career public servants to act as a liaison with the research centres and to help facilitate and nurture this governmentacademic-community research partnership.The funding of the academic centres of excellence was provided by SSHRC and the federal government, each sharing roughly half of the research budgets, but with funding dollars flowing through SSHRC as an institutional research grant to host universities in an effort to maintain a clear division of funding 'independence' from the federal policymakers themselves.The basic annual research grant provided to each regional centre was a modest $350,000 (with some yearly variation) with additional support leveraged from each of the participating universities.Metropolis Canada is also connected to a larger International Metropolis Project 2 which is a network of researchers, policymakers, and civil society organizations from some 22 countries concerned with sharing research, policy and practice on migration (see: http://international. metropolis.net/generalinfo/index_e.html).In fact the Canadian Metropolis Project stood as the largest and deepest national Metropolis participant.This paper restricts itself to the Canadian initiative.
The structure of the Canadian Metropolis Project was designed to help direct research toward policy relevant immigration, diversity and settlement issues while preserving academic autonomy in defining the research questions and academic freedom for scholars to draw their own policy centred research conclusions.In this sense Metropolis was an attempt to balance research excellence with relevance, a key goal of KT initiatives [5].
Government funding for this partnership has been terminated [6], as of April 2012, although efforts to find alternative sources of funding to keep the larger research network functioning has been ongoing.The longevity of this partnership and its recent termination presents an opportunity to reflect critically on the nature of such partnerships.This paper is an attempt to identify some of the key themes, issues and challenges related to KT/KM research partnerships.Special consideration will be placed on the context in which the demand for knowledge mobilization and knowledge transfer has emerged.This examination has considerable international relevance as the Canadian Metropolis Project offers the leading example of a research partnership in the field of immigration and settlement.

Knowledge Mobilization and Knowledge Transfer
An important starting point for our analysis is in clarifying the meaning of the core concepts: knowledge mobilization (KM) and knowledge transfer (KT).These concepts are relatively recent formulations of a long standing challenge respecting how to gather together, in user ready form, and then to effectively and efficiently transfer information/knowledge about a topic or area of interest to end users who may be at some geographical or organizational distance from the knowledge source.The literature does not offer a consensus on the distinction between KM and KT, with the concepts often used interchangeably.The 2005 Knowledge Mobilization Symposium held in Banff, Alberta observed that: "Knowledge mobilization involves making knowledge readily accessible-and thereby useful to any number of individuals and groups in society-by developing ways in which groups can work together collaboratively to produce and share knowledge" [7].We conceive KM as the stage in the process in which relevant knowledge is gathered together (mobilized) prior to formal transfer.
Dickinson maintains that: "Knowledge Transfer consists of efforts to provide decision-makers with the best available research findings to use in making policy and providing services.Its goal is to improve the quality of policy and practice outcomes" [8,9].Kramer speaks to the interactive nature of KT describing it as: "The process by which a body of research knowledge is presented in multiple formats to practitioners and decision-makers.The… parties are engaged in a sustained, intensive, interactive process that results in a transformation of the knowledge to the purposes of the organization" [10].For Zarinpoush and Gotib: "Knowledge transfer refers to the process by which knowledge is transferred to people and organizations that can benefit from it.[KT]… is about reducing the gap between what is known and what is used" [11].This part of the KT process is about knowledge exchange between academic and nonacademic stakeholders [12].
For its part, KT is not just a simple communications issue that can be accomplished by sending information in the form of a report or e-mail to a user since the knowledge that is to be mobilized is often complex, technical and discipline specific and hence not easily understood by the array of potential end users.Additionally, the knowledge to be accessed is not always in the form of readily available written documents but is often held more informally in the knowledge networks of community and government practitioners and academics.Making the knowledge policy relevant generally requires an ongoing dialogue with users to help transform and shape information for this purpose.
The issue of KM and KT has become even more challenging in the current period given the complexity of contemporary societies.For example, organizational structures like state bureaucracies, educational institutions, and nongovernmental organizations have become very large.We also live in an information rich age where the problem is often too much information to manage rather than too little.Information overload is a common complaint and a serious challenge for information users.Attempts to assess and condense vast amounts of what is often discipline specific research, what has come to be called "data smog" [13], is particularly daunting for the non-researcher who must attempt to filter this mass of information for use.It is within this general context that KM and KT have become increasingly significant.
In brief, KM and KT are initiatives designed to build ongoing and systematic exchange of social science knowledge between academic and non-academic stakeholders establishing networks, partnerships and infrastructure for knowledge creation, mobilization and exchange.There has been increased interest within policy communities in obtaining evidence and applying this to better inform policy and practice.In part, this is seen in movements toward the use of 'best practices' and 'evidencebased decision/policy-making'. Knowledge making, if it is to be more relevant and applied, is viewed as best done in partnerships-especially the joining of academics and practitioners linked to policy shops in government.According to Lomas: The clearest message from evaluation of successful research utilization is that early and ongoing involvement of relevant decision makers in the conceptualization and conduct of a study is the best predictor of its utilization.Similarly, research centres with ongoing linkages to and an accepted role in a specific jurisdiction's or organization's decision making, have greater influence than those without such links.Apparently, familiarity breeds pertinence not contempt [14].
Policy relevant research questions that can better direct research toward evidence-informed decision-making, are more likely to occur in this context.Moreover, the transfer of knowledge is also seen as more effectively accomplished through knowledge networks, in part because such spaces open up the opportunity to share information, research outcomes, and to air disagreements, between a broader set of constituencies of interests thus bringing new insights to knowledge formation.It allows for a more engaged and dialectal formation of research questions and understanding of research results for immigration and settlement public policy, public administration and service delivery.

The Increased Importance of KM and KT Today
Geoff Mulgan, writing in 2005 after 7 years with Tony Blair's Cabinet Office, concluded that "government must draw on independent knowledge" [15] as a means of creating a more strategic approach to policy problems.Quite simply there is now more evidence respecting what will work.This knowledge is not found primarily in government but is widely distributed in sites like universities, think tanks, civil society organizations, and international organizations.Herein we encounter a problem in that traditional sources of academic dissemination for research findings have not been effective at reaching audiences beyond the scholarly community.This represents a KT 'gap' where potentially applied knowledge fails to be connected to the policy process and the end-users working in that domain.The standard academic written dissemination outlets have been peer reviewed academic journals and books which have long publishing timelines and which are written primarily for very narrow audiences or peers almost always in highly discipline specific language.If societal impacts are forthcoming from such exercises they take a considerable time to filter through the policy advisory system.Certainly in the short to medium term academic information only becomes influential when, if you will, it gets a social life [16].When we recognize that the typical academic journal article has a very limited social life-a reading audience one could count on their fingers-the use and impact value of getting research to bigger and broader audiences become obvious.
There has, consequently, been a call by government and civil society for the academy to become more relevant-value for public investment in academic scholarship should include, according to this logic, a greater contribution of the application of the knowledge created to addressing society's problems and progress.The Canadian Metropolis Project expressed the point in the following manner: "Over the past two decades we have seen greater expectations on the part of publicly-funded universities throughout the world to have their research put to use by the societies… that support them.… One frequent call to the social sciences is that their research has a clear application in policy development" [17].Within the academy itself there has been a search for greater academic relevance to better addressing societal needs.The disconnect between much of the academic research and its timely application to the real world has been labelled the "relevance gap" [13].One way of framing the problem is the importance of striking, on the one hand, a balance between curiosity-driven research excellence and relevance, on the other [5].
It is also important to note in this regard that a transformation has taken place within SSHRC.SSHRC, as the primary granting council for social sciences and humanities research, has modified its mandate from an overriding focus on curiosity driven research for traditional academic journals to more strategic research financing of projects that address more socially relevant themes.As SSHRC describes it they have moved from a research granting council "preoccupied mainly with research production and training" to a knowledge council "also focused on systematically moving knowledge into active service for the broadest common good" [18]. 3 shift to a knowledge council orientation was viewed as an appropriate repositioning for SSHRC to address the realities of an "information economy".For SSHRC this was also a tactically important move in "an age of permanent fiscal crisis" [19] to help protect and hopefully expand its funding base from the federal government.SSHRC is placed in a better position to argue before government for enhanced resources when it is able to demonstrate its explicit societal relevance.In this respect the social sciences are disadvantaged compared to health and science and engineering funding councils and their more visibly direct spinoff benefits to human well-being and national productivity.In fact, SSHRC saw its funding allocation from the federal budget increase rather significantly during the formative years of Metropolis.Between the mid-1990s the SSHRC budget grew from around $100 to $332.3 million in 2011-2012, although since 2008 it has been under fiscal restraint [20].The movement by SSHRC to more grants awarded to policy-based work provided a powerful push/incentive factor in moving academic social science research to areas of specified policy concerns.This is an important mechanism for facilitating strategic mobilization of research knowledge toward specific fields of interest.Tellingly, by fiscal year 2011-2012 some $31.3 million or 9% of the entire SSHRC budget was dedicated to "connection and knowledge mobilization" [20].
Finally, many governments have become more willing to invest in knowledge mobilization initiatives in an effort to increase a diminished policy capacity within the state through the use of new 'partnerships'.Such initiatives have been viewed as cost effective in an era of government restraint with high value added potential.A community partner in the Canadian Metropolis network expressed this point well: "Metropolis was conceived as an academic project, a way of harvesting academic talent of universities and apply their work to government policy concerns in a time of government downsizing" [21].
It is important to note that in the case of the creation of the Canadian Metropolis Project, 1995-1996, came at a time of deep cuts by the Federal Liberal Government.Federal Program Review in 1994-1995 saw a 19% cut in state expenditures.The cuts were so significant that Finance Minister "Paul Martin spoke with pride of the fact that relative to the size of the economy, program spending [would] be lower in 1996-1997 than at any time since 1951" [22].Moreover, knowledge sectors in public service took particularly hard hits [23].Yet one of the very few programs to receive new funding at this time was the Canadian Metropolis Project.The likely reasoning is that this was seen as a cost-effective investment in knowledge mobilization useful to a government interested in evidence-based policymaking in a time of declining policy capacity.Out of a politics of government restructuring and fiscal crisis opportunity opened for the creation of a research partnership brings into focus the importance of timing and context for creating space for such initiatives.Significantly, the terminating of Metropolis funding also came during a period of austerity but from a government with a very different orientation to knowledge informed decision-making.

Prioritizing Immigration and Settlement Research
Immigration and settlement research has become a focus of KM and KT because of the importance that immigration has in public policy in Canada and elsewhere.However, it is significant to recognize that immigration and settlement is by no means the only area of policy interest to have been so privileged in Canada.Health and community care, social exclusion/inclusion, child poverty and homelessness, indigenous government, social capital and civil society, and the environment and global warming are among other areas that have also been prioritized by granting councils and government.The following factors are in particular what made immigration and settlement a salient one for policymakers:  Globalized economies have given greater emphasis to large labour force movement between national political boundaries. Increased competitiveness means that nations are engaged in a global search to attract high human capital assets. Demographic change (ageing national populations) has made immigrants important for population and labour force stability/growth. Immigration and national security have become increasingly closely tied. Canada is a country which is nearly alone in having explicitly placed immigration at the centre of its economic growth strategy. New immigrants to Canada have seen their economic performance decline in recent years causing concern about the potential emergence of a racialized underclass and raising questions more generally regarding the positive or negative economic and social benefit that newcomers bring.And;  Canada as a settler society has always been a country of high immigration and, consequently, immigration policy has been a long standing central issue of policy concern in the country.This is in part reflected in the support for multiculturalism equity/diversity initiatives in Canadian public policy [24,25].
All these factors, from a policy vantage point, contribute to the prominence of immigration and settlement issues.
It must be noted that the movement toward KT/KM was first advanced in a substantive manner in the area of health in the 1990s.There are numerous KT initiatives in the health field and a growing body of literature analyzing this phenomenon.Moreover, a culture has emerged within the health disciplines that incorporate KT into its research practices [8][9][10].More generally for the social sciences, however, KT/KM is more recent and has not become normal practice within its research domains.The Metropolis initiative was in fact the most substantive institutionally-based experiment in KM/KT within the social sciences in Canada.The institutional linkage is important because while ad hoc approaches to academic-government research relationships can be beneficial, in the end they are inherently "unreliable and unstable" [8].

Changing Roles for the State in the Policy Process
In the 1980s and 1990s the dominant paradigm that guided government shifted in the West from one founded on Keynesian understandings about the role of the state in society to a neoliberal one.In a break from the past, Keynesianism adopted as a principle of the state's role in society the "imperative to act" [26] for addressing economic and social problems.This approach embraced a belief in the capacity of the state to understand complex problems and devise activist policy remedies.Consequently there was a heavy investment in state capacity building, including substantive investment in internal policy capacity.
By contrast the neoliberal state shed its faith in the overarching capacity of the state as a hands-on policy engineer.In many respects the state (and certainly the activist state) came to be identified as one of the 'problems' and not the 'solution' to society's problems.There was under this paradigm an emphasis on the creative force of the market and to some degree of independent civil society.This set the stage for a significant disinvestment in the state and its administrative structure, including its policy infrastructure [27].
An aspect of this change, following Osborne and Gaebler's (1992) call to 're-invent government', was a switch in the focus of the state from rowing (service delivery) to steering (management and policy) [28].But even with a steering mandate the neoliberal state disinvested in in-house policy shops especially with respect to efforts directed around policy research [2].The neoliberal idea was that the state should adopt a more laissez-faire approach and concentrate on government deregulation.In this regard Painter and Pierre observe: An overarching objective in many countries is to 'roll back the state' and allow other actors to play a greater role.Given the preference for a minimal role by the state, policy capacity is not a top priority, perhaps not even an issue worthy of inquiry, because it is typical of a state model of times past [29].
Part of the response to these developments was an attempt to draw upon policy relevant research from outside the state-research which in many respects was viewed as less self-interested than that emerging from the state bureaucracy itself.This position is in part a refection of the influence of public choice theory in public administration and public policy.Public choice views society as composed of self-interested maximizing individuals [30].This logic was extended to cover public servants who were also seen as fundamentally not guided by an 'ethos of public service' but rather by 'narrow self-interest' including the desire to expand their influence and control in public policy realm.Hence, there was an increased use and dependence on the work of think tanks, research institutes and foundations, academia, consultants and special political advisors as a way to counter the influence of an 'entrenched policy bureaucracy' and overcoming the so-called 'Yes Minister' syndrome [31].
As the neoliberal paradigm has waned thinking about the state and policy capacity has continued to evolve.One development has been the emergence of the notion of 'governance' over 'government'.Unlike government, governance extends beyond the simple workings of the machinery of the state.What government once did alone is now seen as being performed by a wide range of public, private, non-profit, national and/or international bodies [32].Some have even suggested that we are witnessing a change in the state's role from that of policy researcher to that of policy manager.This also suggests that policy in the current period is being done much less from closed vertical policy silos and more from an open horizontal policy process.
Such ideas have been very much picked up in the New Public Governance (NPG) literature.Just as NPM was an evolution of traditional public administration, new public governance represents an evolution of NPM and governance.Building on Kickert (1993) and Rhodes' (1997) definition of governance as the machinery of self-organizing inter-organizational networks, Osborne (2006) explains that NPG is "predicated upon the existence of a plural state and a pluralist state and it seeks to understand the development and implementation of public policy in this context [33][34][35].At its core, NPG posits "both a plural state, where multiple inter-dependent actors contribute to the delivery of public services and a pluralist state, where multiple processes inform the policy making system" [35].Bovaird (2006) and Teicher et al. (2006) explain that NPG "lays emphasis on the design and evaluation of enduring inter-organizational relationships, where trust, relational capital and relational contracts act as the core governance mechanisms" [35][36][37].
While we must take with some caution the argument that we have moved from a model of 'governing from the centre' (Savoie, 1999) to a notion of 'shared governance' (Pierre, 2000) as promoted by NPG, it does to some extent reflect reality [38,39].However, this is not a one way movement as the emergence of the Conservative Party to power in Ottawa has witnessed a movement away from pluralism in goverance and toward the practice of more ideologically-oriented politics.
The movement toward NPG in Government has been especially promoted by Third Way advocates and political leaders-a trend that has been more advanced in Europe than in North America."Shared governance entails collaboration among a wide range of actors from the public, private and voluntary sectors, and a transformation of the state's role from one of exercising direct control and operating through hierarchies to one of working through networks" [40]-which embraces a more inclusive and collaborative model.This also suggests greater involvement of parties beyond the state in the policy setting and policymaking process.The role of research partnerships with KM and KT for policy affect would, under this model, play a more central place for government.The return, however, to severe public sector austerity with the wake of the 2008 financial crisis calls into question more cooperative forms of governance.More value charged neoliberal models of government may hold advantaged positions in such an environment.

The Policy Process and Rational Decision-Making
At its most basic level public policy is "a course of action or inaction chosen by public authorities to address a given problem or interrelated set of problems.… The general character of public policy… is that it is a guide to action, a plan, a framework, a course of action or inaction designed to deal with problems" [41].Many models have been developed in an attempt to understand public policy and the public policy process [42,43].The definition of public policy and many of the approaches to its study suggest that policymaking is a thoroughly rational, logical and linear science.Moreover, there is also often the assumption that policymaking is a "neutral endeavour" where evidence/research is gathered, assessed and applied to solve problems in a rather technocratic fashion [44].Hence, a general understanding has been that if only policymakers had the right or better information they would make optimal policy decisions.
There is no denying that there is an underlying rationality and logic behind how policy decisions come to be made and that academic study can help make sense out of this process.It is not, however, our intent to explore the policy process in detail here.The point to be made is simply that how policy comes to be made in the more popular understanding is not so straightforward and immediately rational, especially to those who stand somewhat outside the process itself.
Policymaking is as much an art as it is a science.Notwithstanding all the talk "about 'evidence-based research' and 'academic knowledge transfer', the reality is that policy is not solicited or implemented within a rationalist framework that these kinds of concepts tend to imply" [45].By its nature the process is highly political which alone adds a considerable degree of unpredictability and volatility to the enterprise defying the pure rationalist models of policymaking [46].The involvement of politicians, bureaucrats, lobbyists and interest groups introduces a significant human element to this enterprise.From an NGO perspective Richmond's observations in this regard are cogent: "it is… useful to think of the initial process of soliciting policy input and in the later stages of setting and implementing policy recommendations as sharing the characteristics of being highly political, volatile, conjunctural, and 'irrational' in the traditional academic sense of scientific inquiry" [45].The policy process, especially at the policy decision-making stage, can also be very secretive and thus more difficult to determine what factors (or research) informed key decisions.Moreover, in order for significant changes in policy direction to occur generally "policy windows" first need to open up-these do not occur very frequently-and in the absence of a policy window research rarely has much influence on policy decision-making.As Bunker reminds us: "Political and ideological commitments which bind top level political actors together are not hospitable to independent explorations of fundamentally new appreciations of the experienced world" [48].It is only when that world view comes into crisis that questioning becomes more acceptable and research which points in alternative directions can become influential.Researchers interested in informing policy need to be aware of "the distinction between rational and sensible decisions".Those who "fail to acknowledge the influence of these political and institutional factors" [14] are bound to find the world of policymaking frustrating.
The complexity of the policymaking process has been expressed as one where there is "need to understand that there are many sorts of evidence, that sensible decisions may not reflect scientific rationality, and that context is all important, particularly with policies related to services and governance" [49].Aside from academic evidence policy makers are informed by such sources as experience, anecdote, public opinion [13,50,51], and information that is provided in consultations with relevant professional organizations and stakeholder groups.
The idea and practice of rational policymaking received a boost with the emergence of evidence-based policymaking.This approach came to the fore in 1997 with the election of New Labour in Great Britain which promoted the policymaking philosophy of "what matters is what works" over explicitly politically driven policy orientations [52].In the 1980s and 90s at the high point of conviction politics research in many areas of policy was very often ignored or dismissed [53].
Evidence-based approaches have arguably brought more rationality into the policy decision-making process and with it a renewed interest in research.However, by its nature the policy process remains a highly political one with all the other influences which inform policymaking as noted above in play.Consequently, even in an environment more encouraging of the use of research, the concept of 'evidence-based' policymaking is misleading.It would be more accurate to speak in terms of 'evidence-informed', 'evidence-influenced' or 'evidence-aware' policymaking [52].Those engaged in policy relevant research dissemination must be aware of the limits this kind of process imposes for the use of their work.It remains the case that policymaking is in many respects like making sausages-it is messy and it is not easy to tell (partly because of secrecy) what ingredients went into the making.This makes it especially difficult to determine the actual impact that research/evidence plays in policy making.

The Three Policy Communities
With respect to research partnerships and KM/KT it is important to recognize the structure of the three basic communities involved in the policy process.The First Community consists of policy decision-makers (politicians and senior civil servants that actually formally make policy decisions).The Second Community are academics and others (think tanks, policy institutes, NGOs, etc.) who are engaged in the creation of knowledge and information.The distance between these first two communities is rather wide.However, there is a Third Community that provides a bridge between the first two communities, consisting of the knowledge brokers.The knowledge brokers are "those who work in government and whose work is intended primarily to support the efforts of decision-makers" [54].
Knowledge brokers are mid-level public service policy analysts and advisors inhabiting the various policy units located within each ministry.Typically it is at this level that the most direct engagement takes place with knowledge creators.In turn, it is this section of policy professionals who are directly engaged with developing policy background papers, briefing notes, house notes, and slide presentations for use by senior management.In other words, they are the first line of knowledge users and translators.Consequently, it is to this group that KT products should be directed and where ongoing and intensive researcher links to government need to be most concentrated.
The Canadian Metropolis Project greatly strengthened the connections between the Second and Third Communities allowing for a better flow of distilled knowledge creation to decision-makers.Because of the way that knowledge is absorbed within government (which can be haphazardly and inconsistently) it is important, for maximum effect, that the linkages between the 2 nd and 3 rd communities be regular, strong and preferably institutionalized.This greatly opens up the possibilities that such policy research will actually make a difference.This is particularly important because: Knowledge utilization depends on disorderly interactions between researchers and users, rather than linear sequences beginning with the needs of researchers or the needs of users… The more sustained and intense the interaction between researchers and users, the more likely utilization will occur [15,54].
This gives emphases to the importance of researcher engagement with the third community to maximize policy relevant knowledge transmission.KM/KT, as the foregoing discussion of 'players' implies, is above all a social process.The purpose is to assist in the practical matter of improving policy formulation so that the best and widest scope of research on a given issue can inform the policy development.And this objective is not an end-point deliverable, but rather a process of iterative engagement between the different actors.The Metropolis Project was a working example of such a research partnership.

The Role of Community Based Organizations in KM/KT
Building 'partnerships' with community-based organizations (CBOs) has come to be seen as increasingly important by universities and governments.At a basic level the value of CBOs to knowledge mobilization and knowledge transfer is their organic and grounded connection to the communities that are directly affected by government policies and programs.In the area of immigration, for example, CBOs are charged with the delivery of a large proportion of the settlement services offered to newcomers, most often funded by government contracts [47] and renewed austerity has exacerbated the situation.CBOs are also engage in various types of advocacy and educational activities for immigrant communities.CBOs are, consequently, on the frontline of where policy and programming meet affected communities and as such CBOs are in a unique position to understand and translate the impacts of policy and programming on, for example, immigrant populations from a community perspective.As well, the CBO role as service deliverer means that many of these organizations are in a strategic location to assess the effectiveness of policy and programming from a frontline deliverer/community practitioner vantage point.
CBOs are involved to various degrees in research in the field.Through community reports and newsletters, action research projects, advocacy activities, submissions to commissions and government bodies, sponsored community-based research initiatives, service contract proposals and evaluation reports, and the like, considerable grounded information/research and analysis is created by CBOs.This so-called 'grey literature' produced by the CBO community has grown incredibly rapidly over the last number of years.The growth of this literature is especially remarkable given the context of government downloading of service delivery and the magnification of under addressed social problems that has increased the work load of the community sector to near crisis proportions [55].This grey literature is now commonly accepted as making a valuable contribution to our understanding of policy relevant issues and questions by the broader policy community.In addition to a more formally written set of documents, in the form of grey literature, vast amounts of analytically useful information are held more tacitly in the 'person knowledge banks' of CBO employees and volunteers.There is also vast amounts of largely untapped administrative data that nonprofit service deliverers have that could also prove to be a valuable information source.Moreover, community-based researchers are increasingly part of research teams together with academics engaged in more traditional, although still policy and practice relevant, scholarly activity.Much of Metropolis-based research and publishing is reflective of this collaborative model of study.The involvement of CBOs in research networks adds substantively to knowledge mobilization and transfer.
CBOs are both physically, socially and emotionally close to the communities they service and there are generally high levels of trust between them.CBOs have a level of involvement with these populations which does not exist elsewhere.One of the strategic advantages of academic and government 'partnering' with CBOs is the unprecedented access to grassroots populations for in-depth study that this opens up.The 'lived experience' of these populations is able to be accessed in a manner that was rarely possible before.New and innovative knowledge generation relevant to policy concerns has been fostered because of these 'partnerships'.Much of this work, although certainly not all, is qualitative in orientation.Moreover, because of the closeness of CBOs to their communities, partnerships with CBOs also provide important new communication pathways in terms of knowledge dissemination to these grass-roots communities.

The Meaning of Partnership
The idea of partnership as a central organizing principle of government emerged as part of New Public Management reform of public administration in the late 1980s and 1990s [27,56].The general idea is that the state should not act alone but in 'partnership' with civil society actors in the process of governance.The practice of partnering has been especially popular between government and nonprofit agencies for the delivery of publicly supported services.Provision of settlement services for newcomers in Canada is a prime example of this practice [47].
Much of the partnering arrangements to date have been involved with service delivery and other formal relationships involving programs and projects.The formal agreements and partnerships for service delivery almost always involve a financial relationship between government and third parties.In fact service contracts have come to define these 'partnerships' as contractually driven, with government funders narrowly determining the terms and conditions of the contracts.In this sense these relationships are not equal partnerships between parties of similar bargaining power but a business relationship.In this business deal there is a buyer who issues a contract for services and sellers of services who competitively bid on contracts (generally nonprofit agencies).Consequently, these relationships are by definition one-way contractual ones with government in control.
The knowledge mobilization and knowledge transfer 'research partnership' also has an important contractual/financial element to it but these relationships tend to be far less asymmetrical in terms of power and control by government.The case of the Canadian Metropolis Project illustrates how these kinds of partnerships often unfold.
In the case of the Metropolis Initiative the initial call in 1995 for proposals to establish research Centres of Excellence in Canada focused on policy-oriented immigration and settlement study issued by SSHRC (an arm's length government research funding agency for the social sciences and humanities) in collaboration with CIC and other federal government departments.The direct state bureaucratic interest was mediated by SSHRC which had an established set of research expectations and operating procedures which were modified to fit this new 'partnership' initiative.
Unlike other directly contracted research projects involving academics that the state sometimes engages with and where the dimensions of the research undertaken is tightly proscribed-the Metropolis initiative was more open ended with far greater scope for the researchers engaged in the initiative to shape the nature and scope of the research undertaken as long as it is directed toward the broad mandate and Memorandums of Understanding of the initiative.In short, research autonomy is enshrined in this model, although overall research outcomes were regularly evaluated by both SSHRC and CIC, ensuring 'accountability'.Moreover, subsequent funding renewals were subject to renegotiation among the key players with resulting modification of the research mandate to bring it 'better in tune' with the various interests of the partners and in particular the funders.
The 1995 call for proposals went out to universities across Canada, but, implicitly at least, targeted those institutions of higher learning found in the largest immigrant-receiving cities in the country.These universities were compelled to form their own cross-university alliances with a clustering of key immigrant researchers and younger faculty who were positioned to move into the area of immigration research.A core grouping of researchers was mobilized that would be capable of rapidly expanding the base of immigration and settlement scholarship in Canada.These groupings of cross university immigration researchers, with the explicit support of their universities, would submit proposals to SSHRC in a bid to host a Centre.The universities involved in the bids were required to commit significant resources of their own to make their bids competitive.At the same time some of the major community organizations involved with immigrant communities in their metropolitan areas and regions were also drawn into respective university alliances and asked to be contributing partner institutions.
Clearly, however, while CIC and SSHRC valued the added bonus of community 'partners', they viewed the main partnership as being with the academic communities involved with the Centres.It was the academics who, after all, were seen as providing the recognized research credibility for the project.Hence, in this sense, the allied community organizations became second tier partners for the funders.This understanding of the situation was also the perception of many of the community-based participants in Metropolis as well and it is reflected in the following sentiment drawn from this group in their evaluation of the Metropolis Project.For example, one community member expressed the widely held CBO view that: "The purpose of NGO involvement is to help academics do better research; the NGO involvement is not a goal in itself, but a means to an end, a subordinate goal.This is felt" [21].Significantly many of the academics see the community organizations, if not in completely equal research partnership positions, as very important and hence the community's role in research is acknowledged and structurally recognized in significant ways in the institutions governing the Centres and their research.Nonetheless, the community sector, while believing the Metropolis Initiative to be valuable to be a participant in, do not by-and-large see it as an equal partnership [21].
Overall, it is fair to state that partnership in the Canadian Metropolis Project is far deeper and more meaningful than that found in the service contract relationships that the state often holds with third parties.This partnership involves real give and take among the parties.Nonetheless, the strategic and structural positioning of the three main partners (funders, academics and community) has resulted in an uneven and tiered set of partnering relationships between the parties.

The Canadian Metropolis Project: An Experiment in Research Collaboration
As noted, the Canadian Metropolis Project is a forum for undertaking KM/KT respecting population migration, diversity, and immigrant integration.The stated goal was to directly link research in these areas as a means to improve public policy for managing migration and diversity by: (1) enhancing academic research capacity; (2) focusing academic research on critical policy questions, options and delivery mechanisms; and (3) developing effective ways to use research in decision-making [3].
Overall, the Metropolis research partnership has been much deeper and more meaningful than that found in the more typical service contract relationships that the state often holds with third parties.This partnership involves real give and take among the parties and thus constitutes a true partnering relationship.Nonetheless, the strategic and structural positioning of the three main partners (funders, academics and community) has resulted in an uneven and tiered set of partnering relationships between the parties with academics being more senior partners compared to CBOs, for example.
In concrete terms, the KM/KT process embodied in the work of Metropolis entails a number of avenues through which research networking and transfer occurs.The Centres of Excellence were the engines driving the process, as it is within these structures that the KM/KT axiom-the "more sustained and intense the interaction between researchers and users, the more likely utilization will occur" [54,57]-is actualized.Here a multi-layered range of networking and dissemination activities are carried out.
Networking opportunities are also presented through the organization of national, international and graduate student conferences, focused forums held in Ottawa and the regions, and through less ambitious day long research retreats and forums.Dissemination of research is carried forward through the publication of Working Papers, magazines-such as Our Diverse Cities and Canadian Diversity, seminars, bulletins, and publication of a Metropolis academic periodical, the Journal of International Migration and Integration.In addition, more traditional academic modes of dissemination such as publication in scholarly books and journals and presentations at academic conferences that occurred outside of the Metropolis framework.Moreover, each regionally-based research centre organized its KM/KT activities around specified research domains.National priority leaders for six domains identified by federal funders were charged with facilitating a pan-Canadian orientation to the centres' work and with promoting this work with federal partners.
The construction of such an elaborate structure by Metropolis made it "something of an experimental site for how research gleaned at the academic and community level could be more effectively transmitted into the tightly knit realm of actual policy-making" [58].It must be noted that the movement toward KM/KT was first advanced in health research in the 1990s.Consequently there are numerous KT initiatives in health related research endeavours and a growing body of literature analyzing this phenomenon.Moreover, a culture has emerged within the health disciplines which incorporate KT into their research practices [7,10].More generally for the social sciences, however, KM/KT is more recent and has yet to become normal practice within its broad research domains.The Canadian Metropolis Project was in fact the most substantive institutionally-based experiment in KM/KT within the social sciences in Canada.The institutional linkage is important because it builds institutional coherence and stability into the academic-community-government research relationships [8].

Conclusions: Challenges and Opportunities
KM/KT, and the working example Metropolis presents, are important to the process of building toward a more strategic and evidence-based model of policy development.Political considerations are necessarily central to the policy process [59] but the promise of KM/KT is to ensure that the best information and knowledge available on a given issue of policy relevance can be presented and evaluated within the political context.
Aside from political criteria, however, other factors can impede academic-practitioner partnerships.Smith identified five such obstructions: Time frame: the lag between problem identification and completion of quality research on the problem; Resources: the lack of sufficient resources to effectively research the problem area; Accessibility: research may focus on more obscure areas of study.Also research may be inconclusive and/or researchers may disagree on evidence and conclusions; Jargon: academic language and writing styles that are inaccessible to non-academics; and, Resentment: practitioners may believe their knowledge and contribution are not sufficiently valued by academics thus hindering the development of trust in the relationship between academic and government officials [13].
In the final analysis, however, politics does reign dominant.For a government that does not have an interest in 'evidence-informed/based' policymaking such research partnerships is one that will be of little value or even see as a threat.The Canadian Metropolis project to the current federal Conservative Government is at best a frill. 4 In addition to these challenges, the building of research partnerships composed of diverse communities of researchers and government also confronts other issues.Shifting political priorities as a result of unforeseen issues, changes in government, or even a rotation of senior ministers, can result in 'hot' issues suddenly either being added to or falling off the agenda.Moreover, governments operate within a culture of secrecy with respect to policy development and decision-making.The result is that there may be less than full inclusion of non-governmental researchers.And, there are rather different intellectual cultures separating the spheres of academic researchers and decision-makers.As Smith again observes, "(s)cholars and practitioners do 'think differently' because of the requirements of their environments and purposes.Their incentive structures and values are structured in a manner that encourages such differences" [13].These differences are structural in nature and must be recognized, valued and meaningfully incorporated into the operating mechanisms of the research partnership.
With respect to the question of measuring the KM/KT impact of Metropolis, challenges are raised.In terms of KM there has clearly been a large increase in the volume of studies on immigration, diversity and settlement in Canada since the start of Metropolis in 1996.Much of this work has been conducted by Metropolis-affiliated researchers and available as Working Papers and other research reports on Metropolis Centre websites.However, the effectiveness of KT on policy decision-makers is far more difficult to determine.
Given the closed nature of public policymaking and the various channels through which academic knowledge is transferred and filtered through to decision makers, a process that generally obscures the 4 The Conservative Government in Ottawa's movement away from evidence informed policymaking was vividly illustrated by its controversial decision to end the mandatory long-form census collection in Canada a major source of information for public decision-making.As Roger Gibbins observed of this move: "I would argue that in a knowledge-based economy, paying more for poorer data makes no sense.We need more rather than less evidence-based policy design, and the movement away from the current long-form census is movement towards policy impoverishment.It is also a signal to the world that we don't take ourselves seriously, that we're content to rely on hearsay and guesswork" [60].Similarly commenting on the cancelation of the Metropolis Project one of the community partners to the Toronto Metropolis centre, John Campey, contended that the Federal Government by doing this was attempting "to undermine the ability of the government and community to argue the case based on facts, which is truly frightening."And he continued: "If you want to make policies based on opinions instead of what the facts are, you get rid of the facts" [6].
origins of policy informing knowledge, it is difficult to identify with any precision the impact particular research has on policy outcomes.This point is confirmed by CIC Metropolis Secretariat insider John Biles: Knowledge uptake within a policy environment is almost impossible to measure.In particular, the challenges include the extraordinary range of sources that are used in the development of policy, the numbers of iterations of documents written by committees, and the absence of references or bibliographies in most of that work [61].
Moreover, it is rare that any one research report makes a difference but rather it is the ongoing contacts and dialogue around policy issues and research evidence that has a penetrating impact on policy thinking.Studies indicate that linking researchers to users, as in the case of Metropolis, enhances government uptake [57].It is evident from many informal conversations with immigration policy brokers in Canadian government that Metropolis research has meaningfully influenced policy at various levels of the state.An internal federal government survey of relevant policy staff, in fact, revealed that 79% have in fact made use of Metropolis research in their work [61].The general assessment of Metropolis Project insiders is that Metropolis research has made a difference but the extent of the difference remains an open question.However, one hard statistic that is revealing is that the network of websites constituting the Canadian Metropolis Project received just fewer than 15 million visits between April 2007 and March 2008 with most visitors accessing Working Papers and other immigration research products [62].This is an impressive KT achievement.
Examples of areas where Metropolis research impact have clearly been felt are: (1) Research evidence that foreign skilled professionals who have had access to bridge training and mentoring programs in their areas of expertise have significantly better labour market outcomes in achieving better paid skills commensurate employment [44]; and (2) Studies also identified the integration advantages that foreign born students in Canadian universities have over other high human capital immigrants [63].
In each of these cases Metropolis-based research evidence helped to inform government policymakers resulting in new programming and policies promoting bridge training and mentoring [64] and changes to immigration rules which targeted foreign students in Canadian universities with preferential tracks to immigration [65].
Research evidence sometimes, however, can have unintended policy impact independent of what academic researchers themselves may have intended.For example, the Federal Government has long been focused on what the data was showing regarding the economic integration of newcomers and this was targeted as a priority area for Metropolis research.Study after study over the life of the Metropolis project documents the greater difficulties that newer waves of immigrants from all immigration categories were having in securing employment, and in particular in finding skills commensurate employment.Many researchers concluded that this demonstrated the need for enhanced settlement services supports for newcomers to assist in the difficult integration process [66].But the national Conservative Government guided by a neoliberal policy perspective used this research evidence to justify a significant restructuring of immigration policy in Canada.It brought in an extensive 'guest worker' program with no pathway to citizenship to address low skill employment gaps, and the government weakened non-economic class immigration recruitment in favour of immigrant applicants considered to be 'labour market ready' and not in need of much settlement support.It is a model that de-emphasizes family and humanitarian class immigration [67].
There are real opportunities in research partnerships that can be exploited to the benefit of all participants, and most importantly society as a whole.The Metropolis case provides something of a template as to how these partnerships can be constructed and managed.The achievements and limitations of this working experiment in research partnering is worthy of study.It offers valuable lessons for forging even deeper and more meaningful institutionalized KM/KT relationship which may be replicated within Canada but also internationally.
As noted above, KM/KT is a process, not an end-state.The challenges identified are all amenable to resolution as long as all the parties are aware of such challenges and are prepared to work through them.Building trust between the various participants is key, and trust can only occur where there is opportunity to work and learn collaboratively.Through the process of engagement, the research partners learn about their different styles, cultures and personalities, for that matter.As with anything human, it is far from perfect but as Metropolis amply demonstrates, it certainly can be successful.Above all, sustaining a durable and productive research partnership requires that all parties are dealt with equitably and with respect.Identifying and building upon the areas in which each partner's mutual self-interest intersects is especially fruitful for constructing creative and enduring partnerships.It is also true that partnerships have a life span, ending because of changed circumstances, especially in the realm of government where vagaries of politics often intervenes.In the case of the Canadian Metropolis Project this included the defeat of a Liberal government in favour of a Conservative one that had less interest in evidence-based policymaking and where the politics of the new austerity made expenditures on such initiatives much less attractive than the past.