Early childhood education and care reform in Canadian provinces: understanding the role of experts and evidence in policy change.

White, Linda A. ; Prentice, Susan

Numerous advanced industrialized countries have recently introduced or expanded policies to support the employment of parents of young children. The biggest policy investors have traditionally been the Scandinavian welfare states and France; but all EU countries are challenged to meet the EU's Barcelona targets to increase maternal employment and to provide child care for at least 33 per cent of children under age three and 90 per cent of children age three to age of mandatory schooling (RAND Europe 2014:1). The range of investments and policy instruments varies significantly: Germany and most Nordic countries guarantee a legal right to child care for each young child (European Commission 2014: 11); whereas the UK and New Zealand fund a certain number of hours of "free ECE" or "free ECEC" for each child (Government of the UK 2015; Ministry of Education New Zealand 2015). The "liberal" welfare states of the UK, Australia, and New Zealand have expanded ECEC services, usually delivered by private market actors (Penn 2014; White 2012; see also Esping-Andersen 1990).

Canada is a laggard in public funding and delivery of ECEC services, even when compared to other liberal welfare states (OECD 2015), but several provincial governments are actively considering or adopting full-day kindergarten (FDK) programs. Such enthusiastic cross-Canada embrace of one specific policy instrument is counter-intuitive since FDK is more expensive to deliver than tax breaks or subsidizing parents' child care expenses. FDK requires capital commitments as well as significant salary costs for teachers and/or trained early childhood educators. There is little scientific agreement that FDK (as opposed to half-day programs, for example) is the best means to achieve early childhood developmental outcomes (White, Prentice and Perlman 2015). Some Canadian initiatives propose a broader range of policy instruments, raising the question of why provincial governments are latching on to this single policy prescription.

Choosing to invest in FDK comes as Canadian ECEC investment remains weak. Provincial and territorial governments continue to underfinance and underprovide child care services compared to most OECD countries (OECD 2015). Outside Quebec, the bulk of ECEC funding comes as federal tax breaks for working parents, block funding transfers to provinces, child care fee subsidies for low-income families, and some operational and capital funds (Ferns and Friendly 2014). The number of regulated child care spaces in centres and family child care homes has scarcely grown over the past decade, except in Quebec (CRRU 2013: 5).

The wave of kindergarten adopters, in contrast, is unmistakable, although each jurisdiction has a slightly different approach. Until recently, only three provinces had provided FDK: New Brunswick developed a voluntary public-school based program for 5-year-olds in 1992, making it mandatory in 1998; and Quebec and Nova Scotia initiated their voluntary full-day programs in 1997, with Nova Scotia making it mandatory in 1998 (CNLEEC 2010). Governments in British Columbia, Ontario, and Prince Edward Island have introduced FDK, and it is on the agenda in Alberta and Manitoba. Ontario has implemented full-day junior kindergarten; Quebec is in the midst of expanding junior kindergarten to a greater number of disadvantaged regions; and the Newfoundland and Labrador government recently announced it intends to implement FDK by 2016 (MECYFS 2014).

A second feature of this wave of FDK adoption is how provinces are proceeding. Each adopting government has appointed an expert advisor, task force, or commission to review policies and make specific policy recommendations, allowing us a key source of insight into the policy-making process. The documents generated from these processes allow us to track the evolution in policy thinking from governments first announcing an initiative and/ or appointing a commissioner or expert, to issuing a formal government response to those expert reports. This approach allows us to explore the relationship between expertise and the experts themselves, the evidence upon which they draw, and any policy shifts resulting from their analysis.

The comparative public policy literature hypothesizes three policymaking models: rational learning, policy emulation, and policy legitimation. Using NVivo to assist us in coding government documents and expert reports, we examine the FDK process in BC, Alberta, Ontario, Quebec, and PEI. These cases represent the universe of Canadian early learning initiatives actively on the policy agenda since 2000. We tracked the language used to justify program investment and analyzed the frames used, which allowed us to assess rationales and changes in government thinking, and may reveal more than would post-hoc interviews, in which policy makers might rationalize policy decisions taken.

Our analysis uncovers that none of the standard decision making models fully accounts for the adoption of FDK across Canada. Instead, we argue that path dependent policy making largely informs policy uptake and contours how learning occurred. Although some of the expert advisors, most notably Charles Pascal in his report to the Ontario government recommended more comprehensive reforms within a broader set of family policies, other provincial governments simply took up FDK. There are few signs that provincial elected officials actively debated or contested the principles undergirding the FDK proposals. Given that provinces have a longstanding historic responsibility for education, a well-understood and highly institutionalized sector, political remedies appeared restricted to possibilities easily incorporated into provincial education mandates, and paid for by own-source revenues rather than federal transfers.

Evidence and policy decision making

Public administration theory suggests three clusters of rationales which might explain the uptake of FDK in Canada. First, decision makers can approach a policy problem in a "puzzling" fashion (Heclo 1974). Haas (2004: 576) argues that "useable knowledge" can emerge when decision makers recognize "the limits of their abilities to master new issues and the need to defer or delegate to authoritative actors with a reputation for expertise." Under conditions of genuine uncertainty and where decision makers are endowed with epistemic (expert) authority, actors can approach decision making with the goal of learning. This model presumes that experts and their ideas matter, and that policy uptake may be affected by the quality of the expertise, who the experts are, and their sources of expertise. We characterize this model as "power listening to truth": it predicts a fairly close relationship between what experts recommend and what policymakers implement. Here we would expect to see reliance on scientific studies and evidence, using language such as "the science tells us." This form of policy learning involves persuasion, which Gormley (2011: 979) defines as "conversion to a new opinion." We label this approach rational learning. Rational learning is generally considered to be the archetypical approach to evidence-based decision making.

Recent scholarship has provided more nuanced arguments regarding expert influence on policy outcomes. Lindvall (2009: 708), for example, argues that experts rarely influence the goals policy makers pursue, but instead can influence the selection of instruments. Goal setting can be best understood as a political process while experts play a larger role in helping political decision makers overcome "specific intellectual problems" (Lindvall 2009: 708). Boswell (2009: 166) argues that knowledge is more "likely to be used in areas characterised by scientific uncertainty and the potential for risk construction, rather than [in] more traditional conflicts over distribution or values." In these latter situations, the public could react negatively or even question experts' policy advice. Endowing political actors with "epistemic authority" is important since "The perception that an individual, party or government possesses reliable, relevant and detailed knowledge creates confidence that their decisions will be well founded" (Boswell 2009: 167). Haas (2004: 576) also notes that a degree of autonomy from politics is key to truth speaking to power: he argues that "The more autonomous and independent science is from policy the greater its potential influence." That includes how experts are selected (for example, by merit), whether their work is subject to peer-review, and to what extent they are independent from sponsoring agencies.

The second approach draws on policy diffusion and emulation models to predict a "bandwagon effect," where so much support for a policy builds that governments have an incentive to adopt it, either for competitive reasons or because they come to believe such policies are important part of what it means to be "modern" (Simmons, Dobbin and Garrett 2008). We label this approach "policy emulation." Experts and policy makers may conduct a jurisdictional scan of practices in other provinces in order to inform their own policy decision making. Of course, emulation can also reflect true policy puzzling and learning; indeed, the literature on policy diffusion acknowledges that policy decisions are often made in an interdependent fashion and a decision taken by one government can condition how other governments make their decisions (Simmons, Dobbin and Garrett 2008; Weyland 2006). What makes this process more "imitative," however, is if policy makers or experts justify their actions based on an "everybody's doing it" rationale, rather than by an appeal to science or evidence.

Boushey's (2010) research reveals that policy diffusion can have an almost viral quality. Jurisdictions may be more susceptible to the diffusion of innovation, either because of the presence of what Boushey labels "transfer agents" or carriers of innovation; particularly receptive host conditions; or because of the virulence of the ideas themselves (Boushey 2010: 10). Emulation predicts an S-shaped pattern of policy diffusion, beginning with a few early adopters, then rapid diffusion as other governments jump on the policy bandwagon; geographic clustering; and similarity within diversity (Weyland 2006: 18-19). The embrace of FDK across many Canadian provinces exhibit these imitative characteristics of adoption: some provincial governments were early adopters; then others seem to jump on the policy band wagon. Observing patterns, however, does not confirm why governments are adopting these similar policies.

Third, apparent reliance on expert advice may simply be a post-hoc justification of decisions already taken for partisan or competitive reasons (Boswell 2009; Weiss 1977). We label this model policy legitimation (Hall 1993). Here experts play little role beyond reinforcing conclusions reached in the political sphere. Enthusiasm for popular policy ideas and programs could be understood as an instrumental rational vote-getting strategy by governments (CP 2009). In such cases, "politicians make less effort to understand and critically reflect over the information provided to them ... and are less inclined to change their opinions even if good arguments are presented to them" (Lundin and Oberg 2014: 25). When policy makers do draw on scientific studies, often it may be to justify a predetermined policy preference rather than arising from genuine belief in the validity of the evidence (Weiss 1977).

Case selection and research methodology

To uncover the decision processes in FDK-adopting jurisdictions, we examined publicly available government documents that announced a government initiative and/or the appointment of a commissioner, other expert, or task force; the release of the expert report; and the ensuing government response. In each case, we first examined the background and expertise of the expert(s). We then analyzed the documents the experts and their task forces or commissions produced to determine how they are justified. We paid close attention to the language used in the reports and to the government responses to see whether rational learning, policy emulation or policy legitimation rationales predominated. Through analytic coding, we identified key phrases that correspond to various rationales behind policy interventions as well as key differences in the use of evidence and attention paid to evidence.

We discerned rational learning at work when experts/scientists made

explicit reference to scientific data to justify a policy intervention. Experts and policy makers in these cases we coded as learning if they used language such as "the research shows" or "studies show" and cited scientific studies. We also expected to see government officials citing the credentials of the commissioners in their responses in an appeal to scientific authority under a rational learning model.

Evidence of emulation was found in language based on a rationale that a jurisdiction should follow the example of other jurisdictions or in the use of language such as "we will be left behind" or "we want to be a leader" or "we don't want to be a laggard."

Evidence of policy legitimation/justification of a prior policy opinion was signalled either by language such as "we are doing this given our political goal of" or "we are doing this because the public likes/wants it." We would see few references to scientific arguments in particular and instead language such as "we have no choice but to ..." We also did not expect to see many references to scientific studies if the evidence from those studies contradicted government opinion. Oppositional discourse would likely be framed in evidentiary terms (for example, "the science does not justify an intervention or experts advise against it"); or in instrumentally rational terms (for example, "this expenditure is not rationally justified because the costs outweigh the benefits or it is electorally unsalable"); or framed in terms of principled beliefs ("this goes against" kinds of value articulations).

Analysis of provincial initiatives

PEI: learning from a weak evidence base; some emulation

Pre-2000, PEI was the only jurisdiction where kindergarten was offered by private child care centres and community-based programs, instead of school boards. In 2000, the province implemented a publicly funded half-day kindergarten delivered through for-profit and not-for-profit child care centres with a curriculum designed by the Department of Education and Early Childhood Development (Mella 2009: 5). Six years later, the Premier's Task Force on Student Achievement (2006) called for more rigorous requirements in the kindergarten program in terms of curriculum, hours in the school day, a longer school year, as well as the integration of francophone kindergartens into the public school system (Mella 2009: 5; Simpson 2010: 108). The province responded by implementing a more comprehensive, integrated kindergarten curriculum.

In April 2008, the then-newly-elected Liberal government announced early childhood would move from the Department of Social Services and Seniors to the Department of Education (renamed the Department of Education and Early Childhood Development). It committed to bringing the half-day kindergarten system into the public school system. The Premier appointed former provincial Conservative party leader Patricia Mella as the Public Kindergarten Commissioner (Premier's Office of PEI 2008).

In announcing Mella's appointment, the PEI government stated the decision to bring kindergarten into the public school system was "guided by the wealth of recent research knowledge about the early childhood years" and that the goal was to "identify the model that provides an optimal combination of learning outcomes, access, and family strengthening" (Premier's Office of PEI 2008). The Commissioner's mandate included: whether kindergarten should be half-day or full-day; mandatory or voluntary; as well as issues of staffing and other considerations (Mella 2009: 1). Mella was asked to explore evidence and engage in public consultations "so as to ensure a sound base of evidence and public understanding and consensus for the plan" (Mella 2009: 1).

Commissioner Mella released her report in July 2009. It proposed the delivery of FDK in schools, justifying it as "strongly supported" in the research, as well as being a trend both nationally and internationally (2009: 8). In 2010, PEI launched policy changes in line with the Commissioner's principal recommendations to provide mandatory FDK in public schools, a play-based curriculum, and a system led by teachers with a B Ed (Mella 2009; Premier's Office of PEI 2009). Celebrating the first year of FDK in public schools, Education Minister Doug Currie repeated the program's rationale: "Our goal is to have a system that will help prepare Island children for a lifetime of learning.... Education is a priority for our government and we know that early learning is pivotal to a child having success in life" (Premier's Office of PEI 2011).

Mella further recommended that PEI conduct a comprehensive review of child care. In response, the province asked Kathleen Flanagan (a former provincial Director of the Children's Secretariat) to lead the review. Flanagan's report, sub-titled Early Learning in PEI: An Investment in the Island's Future, was released in April 2010 (Flanagan 2010). It provided detailed recommendations as to how to support the early learning sector in the province after the creation of kindergartens in public schools. Again, the PEI government followed the consultant's recommendations, announcing a Preschool Excellence Initiative in May 2010, 1 month after the release of the Flanagan report.

Public Kindergarten Commissioner Mella's covering letter to the Premier stressed that the "report is based on research evidence." In accepting her report, Premier Ghiz declared that it was "based on the most current research" and extensive consultations (Premier's Office of PEI 2009). Yet a close read of the report shows a lack of extensive citation of primary research. The report's appendix provides a selected bibliography, but the vast majority of material cited was a distillation of primary research and policy documents such as the McCain-Mustard (1999) report, a policy document prepared for the Ontario government, and "grey literature" such as working papers and public presentations. Just a few published studies were cited. Nevertheless, according to the Commissioner (Mella 2009: 17):

The weight of evidence of better outcomes for children is very strong. All Island children will benefit from a well-designed, high quality, developmentally appropriate full-day kindergarten program, and those who will benefit most are the children facing disadvantages and challenges. Island children who have special needs, who are socio-economically disadvantaged, who do not speak either official language, who have previously attended an early learning program--these children will gain the greatest benefit from a full-day kindergarten program. As those children live in all parts of Prince Edward Island and come from all walks of life, a universal approach is essential to ensure that all children have an equal opportunity to benefit from kindergarten.

The lack of cited evidence informing these conclusions is striking.

Mella was not a typical "expert," but rather a long-time politician, although she had a teaching background and was a strong supporter of kindergarten as Leader of the Opposition. Flanagan, in contrast, was a former provincial bureaucrat and early years consultant. The government's embrace of the Mella report's recommendations suggest the report was used to legitimize a policy decision, rather than to inform policy and was done for competitiveness reasons. In a 2013 interview with the Globe and Mail, Premier Robert Ghiz stated, "If we want to compete with the Chinas and the Indias in the world, it needs to start with the education system" (Anderssen 2013).

Kathleen Flanagan's (2010) subsequent report on child care was much more grounded in evidence: it included an extensive research bibliography with recommendations grounded in a comparative survey of best practices in other provinces. The government subsequently announced an evaluation as part of changes to the child care sector, including creating Early Years Centres, temporary wage subsidies for child care workers, and new act and regulations (PEI Department of Education and Early Childhood Development 2013).

Ontario and BC: rational learning with selective uptake

The 2003 election of Liberal government under Premier Dalton McGuinty launched a number of reforms to education and children's policy. The new Premier committed to improving relations with teachers as well as improving provincial educational standards such as literacy and numeracy rates and high school graduation rates. The government also committed to reducing classroom sizes in the primary grades and to improving children's early years experiences (Glaze and Campbell 2007). McGuinty, it was quipped, was the "Education Premier."

In late November 2007, the government announced the appointment of Dr. Charles Pascal as the Ontario Special Advisor on Early Learning to "recommend the best way to implement full-day learning for 4- and 5-year-olds" (Ontario Office of the Premier 2007). Pascal's report, issued in June 2009, recommended full-day early learning for 4- and 5-year-olds as well as integrated services to support all young children (Pascal 2009b). In addition to his report and recommendations, Pascal released a compendium of research that informed his recommendations (Pascal 2009a). The report cites a number of pilot projects and experiments in creating a "seamless day" in programming for children.

In response, the government announced that it would phase in FDK for all 4- and 5-year-olds across the province, beginning in September 2010 (Ontario Ministry of Education 2009). The government stated it would staff classrooms with a teacher and an early childhood educator. It further announced that parents would have the option to enrol their child in an extended day program before and after regular school hours, in schools with sufficient expressed demand (Ontario Office of the Premier 2009). But the Province did not implement the core idea of seamless day programming, with schools as hubs for early years programs.

In announcing Pascal's appointment and the FDK initiative, the Ontario government made extensive references to research evidence. Pascal, a professor of psychology at OISE (Ontario Institute for Studies in Education at the University of Toronto) and a former Deputy Minister in the Ontario government in Education and Community and Social Services, garnered much respect. A companion compendium of research evidence was released in conjunction with his report that provided a list of research informing the study. Pascal had engaged in numerous consultations with early years experts. Yet, the provincial government only chose to pay attention to the FDK part of his recommendations, demonstrating selective uptake of the report and its evidence.

The Liberal government continues to root its support of FDK on grounds of evidence. The Ministry of Education website encourages parents to enrol their children in the voluntary program because "research shows that children who participate in full-day kindergarten get a solid foundation for future learning" (Ontario Ministry of Education 2014b). The Ministry also cites recent evaluations of the roll-out of FDK - despite the mixed results of those evaluations (Ontario Ministry of Education 2014a; Rushowy 2013).

BC had embraced earlier federal funding initiatives. These included the Early Childhood Development Agreement signed in 2000, providing a federal transfer of $2.2 billion over 5 years to programs deemed part of a "child development" agenda: healthy pregnancy, birth and infancy; parenting and family supports; early childhood development, learning and care; and community supports (Friendly and White 2007); and the 2003 Multilateral Framework Agreement on Early Learning and Child Care with the federal government providing $900 million over 5 years to support early learning and child care initiatives (Friendly and White 2007). Relying on 2004 federal Liberal commitments (cancelled in 2006 by the federal Conservatives), BC launched a province-wide consultation process to implement: "life-long literacy, early learning, libraries and health promoting schools, as well as community development and access ..." (BC MCFD/Ministry of Education 2006). The government also began to express concerns about school readiness. In 2005 the government expanded the Ministry of Education's mandate to include early learning. The 2005/2006 annual report contained a joint preface of the Minster of Children and Family Development and the Minister of Education, which lamented that "Research shows that nearly one out of four BC kindergarten students is not developmentally ready to start school ..." (BC MCFD 2006: 3).

In the 2008 Speech from the Throne, Premier Gordon Campbell's Liberal government announced a new Early Childhood Learning Agency, composed of people from the Ministries of Education and Children and Family Development and led by Education. The Agency was to conduct a feasibility and cost study on implementing FDK for children aged five and extending kindergarten to children ages three and four. BC's process did not involve an external commission, but was led internally in a commission-like model. In 2008, the Ministries of Health and Children and Family Development issued a broad early learning curriculum framework produced by a group of researchers and child development advisors (BC Ministry of Health/MCFD and Early Learning Advisory Group 2008).

The Agency issued its report in 2009 (Early Childhood Learning Agency 2009). The government first put implementation on hold, citing cost concerns (Steffenhagen 2009). But in the August 2009 Speech from the Throne, it announced phasing in FDK for 5-year-olds. Given cost concerns, though, it would delay implementation for children younger than age five until FDK for 5-year-olds was completely rolled-out (Hyslop 2012). That implementation was completed in September 2011 after 2 years, but the government has not yet announced a timeline for implementing FDK for younger children.

The BC report and subsequent documents make constant references to the evidence base for early years interventions. The Early Childhood Learning Agency's (2009: 2) report declares "A growing body of research shows that quality early learning programs can have a range of benefits for all children" and cites several studies. It notes that "Many jurisdictions in Canada and around the world either have, or are moving to provide, early learning programs for 3 and 4-year-olds and many offer full day kindergarten for 5-year-olds" (Early Childhood Learning Agency 2009.). Yet BC has only selectively implemented FDK for 5-year-olds, despite its own statements backing FDK for younger ages.

Quebec: mixed results

Quebec family policy is distinct (Le Bourdais 1994). Well-placed observers point to the important role played by social movement actors in the province, and because it is highly attuned to European social policy developments (Barrere-Maurisson and Tremblay 2010; Tremblay 2012; Jenson 2001, 2009). Over the 1980s and 1990s, concerns about declining fertility, poverty, the need to increase women's labour market participation, and strengthen the social economy (and reduce rates of social assistance) led to dramatic changes to Quebec's family policy (Jenson 1998, 2009; Tremblay 2009). These changes culminated in major reforms outlined by Premier Lucien Bouchard and Minister of Education Pauline Marois in the 1997 White Paper on family policy Nouvelles dispositions de la politique familiale: Les enfants au coeur de nos choix (Gouvernement du Quebec 1997).

Quebec's 1997 White Paper outlined a series of family policy initiatives, including a new child benefit program for low-income families, funding for new early childhood education and child care services, and a new parental leave plan (Gouvernement du Quebec 1997). The government's first step was to expand from part day to FDK, making all 5-year-olds eligible (as well as a small portion of 4-year-olds from disadvantaged backgrounds, mostly in the city of Montreal). That same year, the government began to directly fund child care programs for children aged 0-4 years, as well as to school-age programs for 5-12 year olds (Tougas 2002). This was coupled with the gradual roll-out of across-the-board parent fees of $5/day for children in child care centers and regulated family child care (later increased to $7/day) as well as capital funding to encourage expansion of non-profit Centres de la petite enfance--small local networks of non-profit centre-based and family day care (Tougas 2002: 4-5). In March 2013, in a second round of reform, the Parti Quebecois government decided to expand kindergarten programs to all disadvantaged 4-year-olds across the province (Dougherty 2013). Those programs are now only available for a small portion of the population of disadvantaged children.

While the 1997 reforms in Quebec may have been politically driven and electorally popular, they also were grounded in evidence-based decision making. The 1997 reforms stemmed from an expert group that evaluated family policy from an employment and human capital development, as well as a family policy perspective. Two years before, the government had appointed a group of experts with a broad mandate to recommend redesign of the province's social assistance programs and labour market policies (Jenson 2009). In 1996 the Premier brought together other actors including employers, labour union representatives, and other nongovernmental groups for a Summit on the Economy and Employment. The group of experts did not come to a consensus on all issues but agreed on the importance of new services for children (Jenson 2009: 53).

Soon after, the Premier announced the new family policy. In his 1997 speech, Bouchard expressed concern about high rates of grade repetition and school failure. There was, he said, a "direct relation between the length of time in an ecole maternelle and a reduction in the rate of having to repeat a primary grade" (Miville-Deschenes 1997). The Premier linked the new family policy to the goal of increasing educational outcomes in the school system, as did the Minister of Education Pauline Marois, quoted in the 1997 press release citing "American researchers" who found 6 to 1 returns on investment of early years programs (Miville-Deschenes 1997). But while the Quebec government explicitly labeled its programs "educational child care" (Ministere de la Famille et de l'Enfance, Quebec 1997; Ministere de la Famille et des Aines, Quebec 2007) and educational goals underpinned its ECEC policy (Ministere de la Famille et de l'Enfance 1999), subsequent governments expanded family child care and the range of for-profit providers with researchers uncovering poor quality child care overall (Baker et al. 2008; Japel 2008; Japel, Tremblay and Cote 2005). Research also uncovered that wealthier families, rather than the poorest families, are the biggest users of regulated child care.

The Parti Quebecois decision to expand junior kindergarten in disadvantaged regions of the province seems rooted in rational learning and evidence-based policy making. In 2009, the Ministry of Education identified a 13-point plan to curb school dropout rates, including programs to improve school readiness of very young children in disadvantaged regions. The report endorsed increasing the number of child care spaces and providing funding for child care services for families receiving income assistance (Ministere de L'Education, du Loisir, et du Sport 2009). A 2012 report, though, raised concerns that many vulnerable children were not in regulated child care centres but were, rather, over-represented in poor-quality child care (Capuano, Bigras and Japel 2014; Conseil superieur de l'education 2012). Child care subsidies would not reach those children and would not expose them to high quality preschool experiences; kindergarten, in contrast, would be free for them and their families. The Quebec government thus agreed to expand 4-year-olds' access in March 2013 (Dougherty 2013). The government seemed to acknowledge that quality "educational child care" may need to be provided in school settings.

Alberta (and Manitoba): opposition leads to policy failure

The Alberta government was historically unfavourable toward ECEC programs. In the mid-1990s, it cut $30 million in public funding for its voluntary kindergarten program (Government of Alberta Education Ministry 1993). Municipal school boards made up the difference in several ways, in some cases charging parents for the service (Mitchell 1994: D3). By 1996, the Progressive Conservative government under Ralph Klein reinstated full funding for kindergarten, in recognition of the public backlash against the cuts (Laghi 1996: A1).

In 2002, the Ministry of Learning appointed the Alberta Commission on Learning to conduct a comprehensive review of the province's education system from kindergarten to Grade 12. Chaired by Patricia Mackenzie, a former teacher, Edmonton city councillor, and business person, the nine-member commission had a broad mandate to review classroom conditions, hours of instruction, and related issues, as well as the impact of globalization, technology, and demography on education (Alberta Education 2014; N.A. 2002). In June 2003, the Commission issued a final report with 95 recommendations, including funding for FDK for 5-year-olds and a phase-in of junior kindergarten for 4-year-olds (Alberta's Commission on Learning 2003).

The government did not initially reject the Commission's recommendations (Government of Alberta 2003). However, after the report's release the government encountered widespread public resistance. The oppositional sentiment is aptly captured by a statement from the Alberta Federation of Women United for Families: "At 3 years old, you need care from your parent. You don't need peers or teachers or whatever ..." (Mahoney 2004: A2). Alberta's Learning Minister, Lyle Oberg stated "I was assuming that this was going to be very easy, but it's turned out to not be really easy" (Mahoney 2003: p. A12). In the end, the government entirely disregarded the Commission's kindergarten recommendations.

The FDK idea did not gone away, however. In 2009, a joint report by the Alberta education ministry and the Learning Commission chair, Patricia Mackenzie, restated a number of recommendations from the 2003 report, including a renewed call for junior and senior kindergarten (Alberta Education and Mackenzie 2009). In 2012, then-Premier Alison Redford announced her government would begin a new consultation on the Alberta Education Act, including a review of "the learning benefits of full-day kindergarten" and the operational issues surrounding possible implementation (Government of Alberta 2012). In January 2013, the Alberta government announced a delay in implementing provincially funded all-day kindergarten (Cuthbertson 2013) and Premier Redford has since resigned.

Public opposition to FDK in Alberta scuppered provincial funding, leaving responsibility for funding of FDK initiatives to local school boards (Hammer 2011). In Manitoba, in contrast, opposition to FDK comes from the New Democratic Party (NDP) government itself despite, as Puxley (2014) reports, the government having amassed a 15-year file on FDK pointing to the benefits of the program. Internal polling by the Education Ministry in September 2013 found 61 percent of Manitobans surveyed want FDK to be a government priority. In Manitoba, as in Alberta, some school districts have been lobbying for funding so as not to have to pay the full cost of the programs. Nevertheless, Education Minister James Allum has declared that "I think it's fair to say that the jury is still out on the entire value of full-day kindergarten, either from an academic stance, an emotional stance or a social stance" (Puxley 2014).

Discussion and conclusion

This review of five provincial initiatives and a sixth shadow case finds striking similarities. In Quebec, PEI, Ontario, Alberta and BC, provincial governments initiated a commission or commission-like process. Despite varying mandates, each produced reports with recommendations proposing to modernize kindergarten in light of new demographic realities, new scientific findings, and international best practices. And in all cases, provincial governments embraced the idea of FDK, although ultimately the Alberta government backed away due to public opposition, and Quebec's initiative included low cost child care as well. Given that our sample included jurisdictions led by conservative, liberal, and social-democratic parties, this apparent consensus on FDK as a policy option is striking.

The commissions in Ontario and Quebec were led by high-profile academics. In contrast, PEI's two commissions were headed respectively by a provincial politician and a former provincial bureaucrat. The Chair of Alberta's process had a teaching background and was a former municipal politician and businessperson. BC's initiative was directed by internal bureaucrats. In all cases, however, Commissioners or committee chairs were regularly referred to as "experts"--a sign that expertise on early childhood care and education is not seen as restricted to those with direct professional qualifications.

Each provincial reform explicitly used the language of evidence in their policy proposals. Expert evidence about the benefits of early learning and care was requested when commissions or their equivalents were established, and became a central plank--if only marginally in some cases of the content of reports and recommendations, and was later appealed to by governments which received the reports and agreed to act on them. Many of the reports used language such as "the science tells us" or "research shows" to discuss the cognitive and social benefits of early years education and broader socioeconomic benefits of such interventions.

Close examination reveals, though, that what passes muster as "evidence" is broadly construed. Many commission reports drew only on already synthesized documents, such as the McCain-Mustard (1999) report. In all cases, with perhaps the exception of Quebec, the true act of "puzzling" through contemporary needs for early learning and care was highly circumscribed. A broad consideration, one that holistically and comprehensively addressed early childhood care and education, seemed beyond the reach of most commission actors.

Even when commissioners pointed to broader policies, such as the seamless day as the Pascal report did in Ontario, governments did not always implement programs recommended by their own experts. In most provinces there was selective uptake of the evidence while in news releases and premiers' statements, program investments are justified in terms of "creating a stronger province" or economy.

We found next to no evidence of provincial emulation of Quebec, with its broad population-based family policy. Instead, commission reports and governments mainly relied on evidence from small-scale longitudinal studies from the USA such as the Perry Preschool project or the US Department of Health and Human Services' Head Start Impact Study to inform their analysis, which are not easily scaled to mass delivery.

Finally, the provincial cases reveal an additional explanation for the widespread embrace of FDK; it appears that path-dependent policy making largely shapes how learning occurred and evidence was used. FDK seems to the "hammer" for all of the "nails": child development, socio-economic disadvantage, parental labor market participation, economic competitiveness, and so on. Ironically, only in Quebec, with its more recent evidence-grounded rationale for expanding junior kindergarten to children in disadvantaged regions, do we see evidence of true "learning," in the sense of updating based on expert evaluation.

References

Alberta's Commission on Learning. 2003. Every Child Learns, Every Child Succeeds: Report and Recommendations. Edmonton: Alberta Learning.

Alberta Education. 2014. Commission membership. Available at http://education.alberta.ca/ department/ipr/archive/commission/about/membership.aspx.

Alberta Education and P. Mackenzie. 2009. Every Child Learns, Every Child Succeeds: Fifth Anniversary Retrospective on Alberta's Commission on Learning Report. Edmonton: Alberta Learning.

Anderssen, E. 2013. "What PEI and Quebec can teach the rest of Canada about improving child care." The Globe and Mail (24 October). Available at http://www.theglobeandmail. com/life/parenting/pei-and-Quebec-offer-lessons-on-improving-child-care-across-canada/ article15071996/?page=all.

Baker, M., J. Gruber, and K. Milligan. 2008. "Universal Childcare, Maternal Labor Supply, and Family Well-being." Journal of Political Economy 116: 709-745.

Barrere-Maurisson, M.-A. and D.-G. Tremblay, 2010. Conciliation Travail-Famille: Le Role des Acteurs. Lecons d'une Comparison Entre la France et le Quebec. Documents de travail du Centre d'Economie de la Sorbonne. Paris: Maison des Sciences Economiques.

BC MCFD (Ministry of Children and Family Development). 2006. British Columbia's Annual Report 2003/2004: Activities and Expenditures on Early Childhood Development and Early Learning and Child Care. Victoria: The Ministry.

BC MCFD/Ministry of Education. 2006. "Summary of community consultations." Available at http://www.mcf.gov.bc.ca/childcare/pdfs/consultation_summary.pdf.

BC Ministry of Health/MCFD and Early Learning Advisory Group. 2008. British Columbia Early Learning Framework. Available at http://www2.gov.bc.ca/gov/DownloadAsset? assetId=245C9B82FFF94171BB61818A53F0674A.

Boswell, C. 2009. "Knowledge, legitimation and the politics of risk: The functions of research in public debates on migration." Political Studies 57 (1): 165-186.

Boushey, G. 2010. Policy Diffusion Dynamics in America. New York: Cambridge University Press.

Canadian Network for Leadership in Education and Early Learning and Care (CNLEEC). 2010. Early Learning/Kindergarten Programs and Ministry/Department Integration of Education and Child Care in Canada. Vancouver: CECD.

Capuano, F., M. Bigras and C. Japel. 2014. "Kindergarten for four-year-olds: A measure to promote school and social success in children from disadvantaged backgrounds." In Encyclopedia on Early Childhood Development, edited by Richard E. Tremblay, Michel Boivin, Ray DeV. Peters. Montreal: Centre of Excellence for Early Childhood Development and Strategic Knowledge Cluster on Early Child Development.

CP (Canadian Press). 2009. "Majority of Canadians Would Like to See Full-Day Kindergarten: Poll" (25 June). Available at http://www.childcarecanada.org/documents/child-carenews/09/07/ majority-canadians-would-see-full-day-kindergarten-poll-ca.

Childcare Resource and Research Unit (CRRU). 2013. The State of Early Childhood Education and Care in Canada 2010: Trends and Analysis. Toronto: CRRU.

Conseil superieur de l'education. 2012. Mieux Accueillir et Eduquer les Enfants d'age Prescolaire, une Triple Question D'acces, de Qualite et de Continuite des Services. Available at http:// www.cse.gouv.qc.ca/fichiers/documents/publications/Avis/50-0477.pdf.

Cuthbertson, R. 2013. "Promise of all-day kindergarten pushed back: Timeline for 50 new schools also delayed by 'bleak' fiscal outlook." Calgary Herald (24 January). Available at http://www.calgaryherald.com/news/PromiseH-kindergarten+pushed+back/7863279/ story.html.

Dougherty, K. 2013. "Quebec to introduce kindergarten for disadvantaged 4-year-olds." The Gazette (15 March).

Early Childhood Learning Agency, B.C. 2009. Expanding Early Learning in British Columbia for Children Age Three to Five. Victoria: Ministry of Education.

Esping-Andersen, G. 1990. The Three Worlds of Welfare Capitalism. Princeton: Princeton University Press.

European Commission. 2014. Key Data on Early Childhood Education and Care in Europe, 2014 edition. Luxembourg: Publications Office of the European Union.

Ferns, C. and M. Friendly. 2014. The State of Early Childhood Education and Care in Canada 2012. Toronto: CRRU.

Flanagan, K. 2010. Early Learning in PEI: An Investment in the Island's Future. The early years report prepared for the Department of Education and Early Childhood Development; Department of Community Services, Seniors, and Labour; the Department of Health and Wellness; and the Executive Council Office.

Friendly, M. and L.A. White. 2007. "From multilateralism to bilateralism to unilateralism in three short years: Child care in Canadian federalism 2003-2006." In Canadian Federalism: Performance, Effectiveness and Legitimacy, 2nd edition, edited by G. Skogstad and H. Bakvis. Toronto: Oxford University Press, pp. 182-204.

Glaze, A. and C. Campbell. 2007. Putting Literacy and Numeracy First: Using Research and Evidence to Support Improved Student Achievement. Paper presented at the American Educational Research Association annual meeting (11 April). Available at http://www.edu.gov. on.ca/eng/research/li tNumfirst.pdf.

Gormley, W.T. 2011. "From science to policy in early childhood education." Science 333 (6045): 978-981.

Gouvernement du Quebec. 1997. Nouvelles Dispositions de la Politique Familiale. Les Enfants au Coeur de nos Choix. Quebec: Secretariat du Comite des priorites du ministere du Conseil executive.

Government of Alberta. 2003. "Government supports 84 commission recommendations." News Release (4 December). Available at http://education.alberta.ca/department/newsroom/news/archive/2003/december/20031204.aspx.

--. 2012. "10-point plan for education: Backgrounder." News Release (10 January). Available at http://alberta.ca/acn/201201/31784C87E013F-07B2-09FF-E745E76FlB9DFClF.html.

Government of Alberta Education Ministry. 1993. Meeting the Challenge: An Education Roundtable Workbook. Edmonton: The Ministry.

Government of the UK. 2015. Free Early Education and Childcare. Available at https://www. gov.uk/free-early-education.

Hall, P. 1993. "Policy paradigms, social learning and the state: The case of economic policymaking in Britain." Comparative Politics 25 (3): 175-196.

Haas, P. 2004. "When does power listen to truth? A constructivist approach to the policy process." Journal of European Public Policy 11 (4): 569-592.

Hammer, K. 2011. "Alberta struggles with all-day kindergarten." Globe and Mail (14 June). Available at http://www.theglobeandmail.com/news/toronto/kindergarten/alberta struggles-with-all-day-kindergarten/article583266/.

Heclo, H. 1974. Modern Social Politics in Britain and Sweden: From Relief to Income Maintenance. New Haven: Yale University Press.

Hyslop, K. 2012. "BC child care advocates in Geneva to complain to UN." The Tyee (6 February). Available at http://thetyee.ca/News/2012/02/06/Child-Care-Advocates-in-Geneva/.

Japel, C. 2008. "Factors of risk, vulnerability, and school readiness among preschoolers: Evidence from Quebec." IRPP Choices 14 (16): 1-42

Japel, C., R.E. Tremblay and S. Cote. 2005. "Quality counts! Assessing the quality of daycare services based on the Quebec longitudinal study of child development." IRPP Choices 11 (5): 142.

Jenson, J. 1998. "Les reformes des services de garde pour jeunes enfants en France et au Quebec: Une analyse historico-institutionnaliste." Politique et Societes 17 (1-2): 183-216.

--. 2001. Family policy, child care and social solidarity: The case of Quebec. In Changing

Child Care: Five Decades of Child Care Policy and Advocacy in Canada, edited by Susan Prentice. Halifax: Fernwood Publishing.

--. 2009. "Rolling out or backtracking on Quebec's childcare system? Ideology matters."

In Public Policy for Women, edited by M. Griffin Cohen and J. Pulkingham. Toronto: University of Toronto Press, pp. 49-70.

Laghi, B. 1996. "Alberta to start spending surplus: Klein to reverse kindergarten cut." The Globe and Mail (29 January): p. A1.

Le Bourdais, C. 1994. "Quebec's pro-active approach to family policy: 'Thinking and acting family'." In Canada's Changing Families: Challenges to Public Policy, edited by Maureen Baker. Ottawa: Vanier Institute of the Family, pp. 103-125.

Lindvall, J. 2009. "The real but limited influence of expert ideas." World Politics 61 (4): 703-730.

Lundin, M. and P. Oberg. 2014. "Expert knowledge use and deliberation in local policy making." Policy Sciences 47 (1): 25-49.

Mahoney, J. 2003. "Junior-kindergarten proposal sparks division." The Globe and Mail (5 December): A12.

--. 2004. "Debate rages on early education." The Globe and Mail (3 January): A1-A2.

McCain, M. and F. Mustard. 1999. Reversing the Real Brain Drain. Early years study final report. Toronto: Children's Secretariat of Ontario.

Mella, P.J. 2009. Every Child a Better Future: Report of the Public Kindergarten Commissioner. Charlottetown: Government of PEI.

Ministere de L'Education, du Loisir, et du Sport. 2009. I Care About School. Quebec: The Ministry.

Ministere de la Famille et de l'Enfance. 1997. Educational Programs for Child Care Centres. Quebec: The Ministry.

--. 1999. Family Policy in Quebec: Another Step Towards Developing the Full Potential of Families and Their Children. Quebec: The Ministry.

Ministere de la Famille et des Aines, Quebec. 2007. Meeting Early Childhood Needs: Quebec's Educational Program for Child Care Services Update. Quebec: The Ministry.

Ministry of Education New Zealand. 2015. 20 hours ECE. Available at http://parents. education.govt.nz/early-learning/early-childhood-education/20-hours-ece-2/.

Mitchell, A. 1994. "Alberta's 'missing children'." The Globe and Mail (12 November): D3.

Miville-Deschenes, C. 1997. "La politique familiale: Les enfants au coeur des choixs du gouvernement." News release, Cabinet de la ministre de l'Education du Quebec (23 January).

N.A. 2002. "Newly appointed commission reviews Alberta's education system." Airdrie Echo (17 July). Available at http://www.airdrieecho.eom/2002/07/17/ newly-appointed-commission-reviews-albertas-education-system.

Newfoundland and Labrador Ministry of Education, Child, Youth and Family Services (MECYFS). 2014. Full-Day Kindergarten to Be Implemented Province-Wide in 2016. News Release. Available at http://www.releases.gov.nl.ca/releases/2014/edu/0327n12.htm.

OECD (Organisation for Economic Cooperation and Development). 2015. OECD Family Database. Available at http://www.oecd.org/els/family/database.htm.

Ontario Ministry of Education. 2009. "Phasing in full-day learning for four- and five-year-olds." Backgrounder. Available at http://www.edu.gov.on.ca/eng/document/nr/09.10/ bg1027.html.

--. 2014a. "Full-Day Kindergarten Study Evaluation." Available at http://www.edu. gov.on.ca/kindergarten/theresear chisin.html.

--. 2014b. "Why Should I Enrol My Child?" Available at http://www.edu.gov.on.ca/ kindergarten/whyshouldienrolmychild.html.

Ontario Office of the Premier. 2007. "McGuinty government moves forward on full-day learning for four- and five-year-olds." News Release (27 November). Toronto: Office of the Premier.

--. 2009. "Ontario moves forward with full-day learning: McGuinty government putting kids and parents first." News Release (27 October). Toronto: Office of the Premier.

Pascal, C. 2009a. An Updated and Annotated Summary of Evidence: A Compendium to: with our Best Future in Mind: Implementing Early Learning in Ontario. Toronto: Queen's Printer for Ontario.

--. 2009b. With our Best Future in Mind: Implementing Early Learning in Ontario. Report to the Premier by the Special Advisor on Early Learning. Toronto: Queen's Printer for Ontario.

PEI Department of Education and Early Childhood Development. 2013. "Evaluation provides recommendations for early learning and child care system." News Release (4 September). Available at http://mwmccain.ca/_media/cms_page_media/2014/10/3/ pressrelease-evaluation-provides-recommendations-sept413.pdf.

Penn, Helen. 2014. "The business of childcare in Europe." European Early Childhood Education Research Journal 22 (4): 432-456.

Premier's Office of PEI. 2008. "Pat Mella appointed public kindergarten commissioner." News Release (6 May). Available at http://www.gov.pe.ca/newsroom/index. php?number=news&dept=&newsnumber=5703.

--. 2009. "Government commits to full-day school-based kindergarten in 2010." News Release (20 July). Available at http://www.gov.pe.ca/newsroom/index.php? number=news&dept=&newsnumber=6428.

--. 2011. "1,400 students complete first year of full-day public kindergarten." News Release (24 June). Available at http://www.gov.pe.ca/newsroom/index.php? number=news&dept=&newsnumber=7855.

Puxley, C. 2014. "Manitoba won't budge on all-day kindergarten despite internal push." The Globe and Mail (2 February). Available at http://www.theglobeandmail.com/news/ national/education/manitoba-wont-budge-on-all-day-kindergarten-despite-internal-push/ article16653635/.

RAND Europe. 2014. Use of Childcare in the EU Member States and the Progress Towards the Barcelona Targets. Short Statistical Report no. 1 prepared for the European Commission. Santa Monica: RAND.

Rushowy, K. 2013. "Academic impact of full-day kindergarten fades with time, provincial report finds." The Toronto Star (18 October). Available at http://www.thestar.com/ yourtoronto/education/2013/10/18/academic_impact_of_fullday_kindergarten_fades_with_time_provincial_report_finds.html.

Simmons, B.A., F. Dobbin, and G. Garrett. 2008. The Global Diffusion of Markets and Democracy. New York: Cambridge University Press.

Simpson, C. 2010. "Doing it right! The changing landscape of kindergarten in PEI." In Research in Early Child Development in Prince Edward Island: A Research Monograph, edited by R. Doiron and M. Gabriel. Charlottetown, PEI: Centre for Education Research, University of Prince Edward Island, pp. 101-126.

Steffenhagen, J. 2009. "Explanded kindergarten delayed." The Vancouver Sun (4 February). Available at http://blogs.vancouversun.com/2009/02/04/expanded-kindergarten-delayed/.

Tougas, J. 2002. Reforming Quebec's Early Childhood Care and Education: The First Five Years. Toronto: Childcare Resource and Research Unit.

Tremblay, D.-G. 2009. "Quebec's policies for work-family balance: A model for Canada?" In Public Policy for Women, edited by M. Griffin Cohen and J. Pulkingham. Toronto: University of Toronto Press, pp. 271-290.

--. 2012. Conciliation Emploi-Famille et Temps Sociaux, 3rd edition. Quebec: Presses de l'Universite du Quebec.

Weiss, C.H. 1977. "Research for policy's sake: The enlightenment function of social research." Policy Analysis 3 (4)): 531-545.

Weyland, K. 2006. Bounded Rationality and Policy Diffusion: Social Sector Reform in Latin America. Princeton: Princeton University Press.

White, L.A. 2012. "Must we all be paradigmatic? Social investment policies and liberal welfare states." Canadian Journal of Political Science 45 (3): 657-683.

White, L.A., S. Prentice, and M. Perlman. 2015. "The evidence base for early childhood education and care program investment: What we know, what we don't know." Evidence and Policy 11 (4): 529-546.

Linda White is Interim Director, School of Public Policy and Governance, and Associate Professor, SPPG and Department of Political Science, University of Toronto. Susan Prentice is Professor, Department of Sociology, University of Manitoba. Financial support for this research was provided by the Social Sciences and Humanities Research Council of Canada grants # 410-2008-0630 and #410-2010-1906. The authors also gratefully acknowledge the research assistance of Katherine Jin.