Policy analytical capacity and "Eastern" styles of policy analysis: evidence from West Java Province, Indonesia

An emerging body of scholarship suggests that "Western" notions of policy analysis may not be relevant in "Eastern" jurisdictions, and that non-Western countries, particularly in Asia, may have their own local policy analytical style or tradition. However, in many Asian countries, little is known about the work that public sector policy analysts do. Using data from a survey and focus groups, this article investigates policy analysis and analytical capacity in the provincial government of West Java, Indonesia. We find ample signs of policy analytical activity as it would be understood by Western scholars, with little evidence of any specific Asian style.


Introduction
Overwhelmingly, scholarship on policy analysis focuses on Western countries, especially the United States, countries in Europe, and to a lesser extent, Canada, Australia, and New Zealand. However, attempts are increasingly being made to shine light on other traditions of public policy and administration. While the notion of non-Western traditions of public administration is not new (see for instance, Weber, 1994Weber, [1918, pp.156-159), more interest has developed in recent years (e.g. Drechsler, 2013;Peters, 2021).
In conjunction with the literature on non-Western traditions of public administration, a parallel body of work has suggested there might be a more general spatial dimension to public policy, with various countries or regions operating within distinguishable policy "styles" (Bayerlein et al., 2021;Howlett, 1991;Zahariadis et al., 2021). In line with analyses of Asian traditions in public administration (e.g. Haque, 2019), Mukherjee and Howlett (2016) have proposed an "Asian" policy style-one that features a high level of 1 3 corporatism, openness to devolving responsibility to subordinate levels of government, and especially, low levels of policy analytical capacity.
Questions therefore arise as to whether there is a particular Asian style of public policy making or public administration, what that style might look like, and to which countries or contexts it might extend. While some authors have suggested various characteristics that might be common in Asian public administration contexts (e.g. Berman, 2010), a general consensus has not yet emerged. One important, and yet, unanswered concern relates to the function of policy analysis in Asian contexts: are there features of policy analysis in Asian bureaucracies that make this activity fundamentally different from the way it is practiced in Western countries? Many observers (e.g. Colebatch, 2004Colebatch, , 2006Drechsler, 2018;Haque, 1996) have indeed argued that this is the case-although empirical studies in this vein are sparse.
In this study, we use a survey and focus group data to investigate policy analysis activity and analytical capacity in the government of West Java Province, Indonesia. Indonesia is the fourth most populous country in the world and has over four million public servants serving approximately 270 million people (Statistics Indonesia, 2021). Indonesian law stipulates the role of public servants as planners, supervisors, and implementers of general government tasks and national development through the delivery of policies and professional public services, free from political intervention, corruption, collusion, and nepotism. To carry out these main tasks, public servants must have adequate qualifications, competencies, and capacities to engage in public policy processes. Many of these qualifications, and the activities for which they are intended, are in theory similar to those pursued in Western bureaucracies. As will be elaborated below, our survey and focus group data suggest that in our case study of West Java, government policy workers interpret their roles, competencies, and activities in a way that aligns closely with Western conceptions of policy analysis, with little indication of any particular "Eastern" or "Asian" style.
A multitude of studies have looked at policy analytical capacity in Western developed democracies (e.g. Elgin & Weible, 2013;Newman et al., 2017;Page & Jenkins, 2005), but very little work has been done to examine the policy analytical activity of government organizations in the developing world. This is a notable knowledge gap, especially as developing democracies are increasingly being tasked by the international community with responsibility for managing complex policy issues with international ramifications, like global migration, climate change, control of infectious disease, and poverty.

Policy analysis and policy analytical capacity
The notion that policy should be informed by the best available evidence has become increasingly popular among many observers since at least the early 2000s. According to this point of view, public policy decisions that are based on "intuition, ideology, or conventional wisdom", rather than research evidence, can "go seriously astray" (Banks, 2009, p. 4). A wide variety of actors, including academics (Cartwright & Hardie, 2012), practitioners (Chalmers, 2003), politicians (O'Malley, 2014), administrative organizations (European Commission, 2010) and international organizations (OECD, 2013) have all issued calls for policy decisions to be better informed by rigorous evidence. Supporters of evidence-based policy often argue that policies that are not informed by available lessons on inputs, outcomes, and stakeholder values are vulnerable to policy failure (e.g. Dunlop, 2017).
Evidence-based policy has, however, also been the subject of considerable debate. A critical perspective, one that questions the entire evidence-based policy paradigm, has emerged in parallel to the more instrumentalist school of thought outlined above. Some critics, who have called into question the fundamental ability of the public sector to base policy decisions on evidence, argue that proponents of evidence-based policy simply do not understand the complex political nature of the policy process (Cairney, 2016, p. 4). Others argue further that evidence-based policy would be socially undesirable, as it would undermine democracy and deliberation (Holmes et al., 2006), or could be used by powerful elites to suppress particular viewpoints (Packwood, 2002), suppress particular research methodologies (Neylan, 2008), or oppress minorities (Maddison, 2012). Many authors note the basic contestability of evidence, with some extreme viewpoints denying that objective evidence can ever be obtained for subjects that are relevant to policy making (Luton, 2007).
Although debates in this area have become somewhat ill-tempered at times (see Newman, 2017 for an overview), there is a more moderate perspective that has also made a contribution to the discourse. In this middle-ground approach, policy decisions are understood to be informed by evidence but are also influenced by partisan considerations, values, human judgement, emotions, political strategy, and a host of other factors that are all apparent and legitimate in the context of a developed democracy. Authors who adopt this point of view argue that there are multiple sources of information that must be processed by decision makers, and the focus for scholars and practitioners should be to support decision makers to make the most purposeful and intelligent use of information possible (Parkhurst, 2016). The emphasis of this perspective is therefore on improving the use of information for policy decision making, not on eliminating political or other influences from the policy process.
Our approach here is informed by this middle-ground perspective on evidence-based policy, or as some have called it, evidence-informed policy (Nutley & Webb, 2000, p. 20). Public policy decisions are the result of a complex and dynamic process that incorporates numerous voices, conflicting and competing interests, power imbalances, game-playing, and risk-taking, in addition to strategic decision making and planning. While the debates surrounding evidence-based policy are certainly relevant, it is still possible to accept as fact that there is a part of the democratic policy making process that consumes information and transforms it into useful advice for decision makers. Following others (e.g. Adachi, 2017), we refer to this link in the policy decision making chain as policy analysis. We also assume from the outset that at least some policy analysis takes place in dedicated administrative units within public sector organizations (Weimer & Vining, 2017, p. 30).
Accordingly, Howlett and his co-authors (e.g. Oliphant and Howlett, 2010;Wellstead et al., 2011) have developed the concept of policy analytical capacity to represent the ability of public servants to collect and process information for the purpose of informing policy decision making. Policy capacity is itself a confusing and amorphous term that has been used to refer to a variety of concepts, from the quality of policy development (Anderson, 1996) straight through to the level of control the state has in governing society (Painter & Pierre, 2005). Policy analytical capacity, on the other hand, is specifically meant to capture the capacity of public sector employees to turn information into policy advice. Howlett (2015, p.174) defines policy analytical capacity as the "ability of individuals in a policy-relevant organization to produce valuable policy-relevant research and analysis on topics asked of them or of their own choosing" and identifies numerous factors that contribute to policy analytical capacity, which can be condensed into four categories: 1. individual policy workers' knowledge of the policy process and the needs of policy decision makers; 2. individual policy workers' skills, including research skills and statistical analysis skills; 3. an organization's support for advice creation, which can include effective leadership, professional rewards for quality policy analysis, team building and information 1 3 sharing, and strategic long-term planning; and 4. physical resources, such as information technology or library subscriptions. The implied argument of much of this literature is that greater policy analytical capacity results in better quality policy advice, which in turn can produce more strategic, more effective, and more sustainable policy outcomes (Mendez & Dussauge-Laguna, 2017).
But is policy analysis performed the same way in different national contexts? Starting in the 1980s and early 1990s, scholarship on "policy styles" developed around the notion that bureaucracies in different national jurisdictions might operate under varying sets of customs and traditions, which are linked to the unique political histories and policy legacies within those countries (Howlett, 1991;Richardson, 1982). The focus of much of this research is on the selection of policy instruments, especially the mix of public sector intervention with private markets and other network actors (Mukherjee & Howlett, 2016). However, more recently, some of the academic debate in this area has shifted to national styles in administration and policy analysis (Bayerlein et al., 2021;Berman, 2010;Tao, 2018). As in other areas of public policy and administration scholarship, studies on national styles have tended to focus on Western countries, particularly the United States, the United Kingdom, and Canada (e.g. Cairney, 2018;Halligan, 2021;Howlett & Lindquist, 2004).
Again, more recently, prominent voices have been calling for scholars to pay more attention to other jurisdictions, especially in Asia (Bice et al., 2018), and some authors already have done so (e.g. Berman, 2010Berman, , 2011. However, while some have pointed to "non-Western" (Bice et al., 2018), "Asian" (Mukherjee & Howlett, 2016), and "Eastern" (Ugyel, 2016) styles, definitions are still somewhat ambiguous, and agreement on what core elements might be characteristic of a non-Western paradigm of public administration has not yet been reached.
Conversely, the most frequently discussed aspect of "Western" public administration is Max Weber's (2009aWeber's ( [1921) depiction of the administration of government as a neutral, professional machine, mechanically and proficiently implementing policy decisions that were made elsewhere in a separate political branch of government. According to Weber, "the genuine official…will not engage in politics. Rather, he should engage in impartial 'administration. ' " (2009b [1919], p.95). Drechsler (2018, p.22) notes that this Weberian conception of an ideal public service is one of the core paradigms of Western public administration. By contrast, Berman (2010, p.2) argues that in Asia, "Different conceptions exist about the 'political neutrality' of civil servants. Countries do not prioritize or define 'democracy' in exactly the same way" as they do in Western countries (Berman's comparator is the United States). Tao (2018), in writing more specifically about countries with a Confucian tradition, argues that while Weber idealized professionalism, some Asian countries emphasize morality instead, and might bend Weber's rules on political neutrality to suit. Weber himself noted that in his conception of public administration, the bureaucrat's personal morals were of no consequence, asserting instead that "The honour of the civil servant is vested in his ability to execute conscientiously the order of the superior authorities, exactly as if the order agreed with his own conviction. This holds even if the order appears wrong to him and if, despite the civil servant's remonstrances, the authority insists on the order" (2009b [1919], p.95).
Others have argued that policy analysis in non-Western countries might just not be as well resourced as it is in the West. Several observers have described weaknesses in policy analytical capacity in Asian jurisdictions, for instance in the Philippines (Saguin et al., 2018), Hong Kong (Cheung, 2007), India (Bali & Ramesh, 2021), and Vietnam (Phuong et al., 2018). Colebatch (2004) has gone as far as to argue that in Southeast Asia, policyrelevant employees in public sector administrative organizations do not conduct policy analysis at all. Elsewhere, Colebatch (2006) has argued that the concept of policy analysis itself is a mainly American construct and may not be applicable to jurisdictions outside the United States.
Taken together, the existing scholarship on this subject suggests that policy analysis and public administration activities as they are understood in the West, in which professional and politically neutral public servants collect and process information, craft evidencebased advice for political decision makers, and faithfully implement policy decisions for the delivery of public sector interventions, may not be an accurate depiction of what goes on in non-Western bureaucracies, especially in Asia. However, as mentioned earlier, there is little empirical data on how well this depiction of Asian public administration corresponds with reality, and these claims need to be investigated further.

Policy analytical capacity in Indonesia
Indonesia is a prime candidate for investigating policy analysis in the developing world. It is a regional powerhouse in Southeast Asia, with a gross domestic product roughly equivalent to 35% of the entire ASEAN bloc (ASEAN, 2020). Major domestic policy issues, including addressing religious extremism, managing responses to natural disasters like earthquakes and volcanoes, conserving the natural environment, poverty reduction, and dealing with urbanization, are areas where Indonesia has the potential to set positive examples for the rest of the developing world. Moreover, Indonesia has a strong administrative culture and an extensive administrative machine inherited from its authoritarian past. An active and official policy of decentralization since 2002 has effectively dispersed this administrative culture across the country and through virtually every policy area (Tjiptoherijanto, 2007, p. 35). In short, Indonesia is large, it has an expanding sphere of influence, it has a pervasive bureaucracy and a strong bureaucratic tradition, and it is a major player in a variety of international policy areas. A better understanding of policy analysis in Indonesia is essential, because Indonesia is responsible for many policy areas of regional and global significance, and because the public sector's capacity for policy analysis will greatly influence and contribute to how well they can approach these tasks.
Our data come from West Java, Indonesia's most populous province, with a population of about 50 million inhabitants (about 18% of the Indonesian population) throughout 18 regencies and 9 cities (BPS-Statistics Indonesia, 2018). Annual population growth rate, average household size, employment rate, average net income, happiness index, poverty rate in urban areas, human development index, and a variety of other indicators for West Java are all about the same as the national means, so West Java likely represents something of an Indonesian average.
As in other developing countries, academic scholarship on public administration in Indonesia is scant, and very little is known about policy analysis in the Indonesian public sector. A large proportion of international scholarly studies on public administration in Indonesia focus on administrative reform, and in particular, the elimination of corruption (e.g. Blunt et al., 2012;van Eeden Jones & Lasthuizen, 2018). Another tranche of studies deals with the outcomes of specific policy decisions or decisions in particular policy areas, such as security (Fealy & White, 2016) or economic development (Kuswanto et al., 2017). A smaller number of studies investigate citizen satisfaction with public policies or social services (e.g. Benny & Abdullah, 2011).
There are some more targeted studies on public administration in Indonesia, and many of these report deficiencies in the ability of the Indonesian public sector to conduct 1 3 policy analysis. Tjiptoherijanto (2007), for instance, presents some details on the technical resources of the Indonesian public sector as part of his investigation into possible avenues for public administration reform. Studies by Hillman (2013) and Tumanut (2016) look at more specific questions of structural problems within the Indonesian bureaucracy. A report by Zhang (2015), conducted on behalf of the Australian Department of Foreign Affairs and Trade, examines evidence-based policy in Indonesia, and assumes from the outset is that analytical capacity is low. A more comprehensive assessment of policy analytical capacity in Indonesia, undertaken by the Knowledge Sector Initiative (also sponsored by the Australian Department of Foreign Affairs and Trade), found that capabilities of individual policy workers in the Indonesian national government was high but that the level of resources available to them was low, resulting in major deficiencies in policy analytical capacity (Pellini et al., 2018). Datta et al., (2016, p.16), also from the Knowledge Sector Initiative, found that Indonesia's lower levels of policy analytical capacity are also evident at the subnational and local levels. Apart from describing weaknesses in policy analytical capacity, the factor that these studies all have in common is that they take a very Western approach to public administration in Indonesia, including language and notions of evidence-based policy, policy analysis, and policy analytical capacity that are usually discussed in Western public administration contexts.

Methodology
Bureaucracies are by nature closely guarded institutions, whose inner workings are rarely made public. Therefore, the best way to address questions about policy analysis in the Indonesian public sector would be to ask public servants directly about their perceptions of policy work. Accordingly, we conducted a survey of policy-relevant public servants across the government of the Indonesian Province of West Java, and convened focus groups of policy analysts and managers from the province's central strategic administrative organization, Badan Koordinasi Pemerintahan Pembangunan (or BKPP, the Agency of Coordination, Government-Administration, and Development) and from the office of the Governor. Our survey asked a variety of questions about respondents' work experience, background and training, self-assessment of professional skills and abilities, and attitudes towards the practical use of research in the creation of advice used for policy making. Focus groups explored similar questions, but in greater depth. The survey materials and focus groups were all conducted in Bahasa Indonesia, and all translations reported here are our own translations.
It should be noted that our results do not represent any kind of objective assessment of policy analytical capacity in the West Java government (nor were they intended to do so). On the contrary, these results are subjective by design. They are the perceptions and personal perspectives of the respondents on their own function and capabilities as public sector policy workers. Importantly, despite the open nature of some of the survey questions, there was a high degree of internal consistency in responses, and survey responses largely agreed with focus group discussions, indicating that the research questions likely were interpreted in broadly the same way by many respondents.
The survey was modelled after a study that used a similar methodology to learn about policy capacity in Australia . The Australian study was, in turn, based on a series of studies from Canada (e.g. Amara et al., 2004;Howlett & Newman, 2010). Although this methodology now appears to have a respectable pedigree, we acknowledge that it is by no 1 3 means a perfect instrument for obtaining the information we are looking for. As in previous studies, we did not attempt to capture a probability sample of respondents, which would have been impossible for two reasons: first, we could not determine the population of possible participants with any degree of certainty, because there is no way of ascertaining the complete set of all West Java public servants who do policy analysis as part of their work responsibilities. Some public servants are dedicated policy analysts, but others are not and yet still devote some of their time to analysis activities. Job titles can be confusing and uninformative. Instead, we relied on a provincial agency (the West Java Board of Research and Development) to select potential candidates for us, based on internal human resource information. At the time the survey was conducted, the West Java Provincial Government had around 12,000 public servants. The West Java Board of Research and Development sent out 1200 paper surveys on our behalf in late 2017, of which 391 were returned, for a response rate of about 33%. Secondly, for obvious reasons, we could not compel people to respond to the survey, so the set of responses will inevitably suffer from self-selection bias. However, as was done in previous studies, we have tried to mitigate these limitations as much as possible by writing survey questions that would only be relevant to people who conduct policy analysis as part of their job roles (and hence only these people would be able to complete the survey). Furthermore, as alluded to above, our survey responses were triangulated with our focus group discussions to ensure internal validity of response data. For a more detailed discussion of the limitations of this particular methodology, see Vesely (2013).
Focus groups were conducted on five occasions. Each session lasted on average about two hours and included on average 12 participants, for a total of 60 individual participants over the 5 sessions.
It should also be noted that this is intended to be an exploratory study. As has been already discussed, the information previously available on this topic is very sparse. In addition, the size of the Indonesian public sector, even if we only consider the national government, is immense. It would be extremely challenging for a first study in this area to attempt to address policy analysis across the entire Indonesian government. Moreover, Indonesia is still in the process of transitioning to a fully developed democracy, with over thirty years of authoritarian rule having ended only as recently as 1998. A more comprehensive analysis would include the dimension of time, by also taking into consideration the historical development of policy analytical capacity over the last two decades.
Instead, we chose a more modest pathway. We aimed to begin the research agenda by examining policy analysis at the provincial level, at one point in time (roughly twenty years since the end of authoritarian rule). Indonesia has three official levels of government: the national (or "central") government; 34 provinces (including several autonomous regions like Aceh and Yogyakarta); and about 500 regencies and cities. Provinces in Indonesia are responsible for significant areas of public policy, including education, health care, social welfare, public works and spatial planning, and housing and residential areas, and are therefore in need of professional policy analysis to inform decision making in these important policy areas.

Results
Respondents were asked to provide information about their personal and professional backgrounds (see Table 1 for a summary). 59% of respondents identified as male, and 39% as female. This is roughly consistent with the gender balance in the general West Java public service, which at the time of the survey was about 56% male and 44% female. The median age was 46. This is an older group, especially compared with national averages. For instance, across Indonesia, about half the population is under 30. Of course, an older group also suggests a more experienced group. In fact, two-thirds of respondents reported having more than 10 years of experience working for the government, and 38% reported having more than 20 years of experience.
Education levels were high. 84% of respondents reported having completed at least one university degree. A large number of respondents-30%-had a master's degree as well, although only 1% of respondents reported having completed a PhD.
Do West Javan policy workers conduct policy analysis? At first glance, the answer would seem to be yes. Forty-five per cent of survey respondents claimed to spend at least a quarter of their work time on "data analysis and management". When asked, on a 5-point Likert scale ranging from never (0) to every day (5), how often they "gather information and data related to policies", the mean of responses was 3.28, which falls somewhere between "every month" and "every week". When asked more directly, on the same 5-point Likert scale, how often they "analyse data related to policy", the mean of responses was 2.23, which falls between "several times a year" and "every month". A summary of these responses is given in Table 2.
Furthermore, survey responses are distributed among levels of education in a way that further supports the notion of analytical capacity. For instance, for the question asking, "How often are you involved in analysing data related to policy?" the mean for respondents with a PhD was 3.75 (nearly "every week"), whereas the mean for respondents with no higher education was 1.5 (between "every year" and "several times a year"). These results were found to be statistically significant at the 95% confidence level using the nonparametric Kruskal-Wallis test for ordinal data. The implication here is that these responses are not random; respondents with higher levels of education see themselves in roles that involve greater application of analysis. A breakdown of relevant survey questions according to level of education is shown in Table 3. For all of these questions, higher levels of education are associated with increased policy analysis activity.
These results from the survey, suggesting that policy analysis is a core activity for many West Javan public servants, are echoed in the focus group responses. Focus group respondents noted the importance of primary data collection, as well as data analysis for strategic planning, and especially the benefit of a good relationship with BPS (Badan Pusat Statistik, or Statistics Indonesia, the national statistics agency) for acquiring data useful for the creation of policy advice. One participant reported that their unit collaborated frequently with the University of Indonesia, to organize seminars on economic policy. According to another participant, "Our job is to carry out coordination, facilitation, monitoring, and evaluation" (Participant 4, Garut Office). With respect to the use of research for informing policy advice, another participant commented that "in practice, in the process of policy formulation, we involve many parties, one of them are universities, especially for things that are very strategic" (Participant 3, Governor's Office).
Focus group respondents also saw themselves largely as implementers of decisions made elsewhere, in a traditional (Western) Weberian sense. According to one participant, "We… don't formulate any policies. We act as executors only… The policy has been formulated and given" (Participant 4, Bogor Office). According to another, "we provide the recommendations, but it is entirely up to the Secretary or the bureau to follow up our recommendations" (Participant 2, Garut Office). A third participant put it more bluntly: "From an institutional point of view, [our] role is strategic. However, the policy maker…is the Governor" (Participant 5, Purwakarta Office).
In the focus group sessions, discussions of a division of labour between policy decision making and implementation often led straight into expressions of dissatisfaction with the level of authority given to policy analysts and their managers. Numerous focus group participants said they felt their work unit's authority regarding policy was unclear, or that the authority they actually wielded was far less than what they felt they ought to have. Comments of this nature were abundant, with participants noting frequently that their policy recommendations were routinely ignored by political decision makers. Some participants were quite animated on this subject, such as one who declared, "Praise to God, since I arrived here in 2016, there were 17 recommendations that I conveyed to the Governor, none of which was followed up" (Participant 2, Garut Office).
Survey respondents were, on average, fairly optimistic about the ability of research evidence to inform policy decisions. When asked if they agree with the statement, "Policy decisions are based on research data and supporting evidence", 91% of respondents selected "agree" or "strongly agree". However, the concerns noted by focus group participants were also apparent in the survey responses. 57% of respondents agreed or strongly agreed with the statement, "Policy decisions are based on what is politically acceptable", 88% agreed or strongly agreed that "Policymaking is influenced by budget considerations", and 85% agreed or strongly agreed that "Policymaking is influenced by crises that occur in society". Only about half (53% per cent) agreed or strongly agreed with the statement, "I have the power to ensure that policies are based on research". Moreover, survey respondents with ostensibly more authority (as indicated by their job role) did not report stronger agreement to this sentiment, with about 57% of assistant managers, managers, and directors all responding agree or strongly agree (see Table 4). This suggests that these attitudes are not a product of a respondent's position of authority, as the trend in responses does not change across varying degrees of management-level responsibility. Accordingly, a Kruskal-Wallis test on this set of data did not yield a statistically significant result, indicating no statistical difference between these groups-which further demonstrates a divide between administrative policy analytical activity and political decision making, as described in traditionally Western accounts of the policy process.
Regarding barriers and enablers of evidence-based policy making, focus group participants most frequently pointed to human resource deficiencies, particularly adequate staff numbers and appropriate skills, education, and training. When asked what was needed to support their agency's role in policy making, one participant answered, "What is [our agency]'s expectation? It's a simple thing, such as qualified human resources and suitable training" (Participant 9, Bogor Office). Another participant commented that although "technical training is important", it was rare for their agency's staff to receive adequate training (Participant 8, Bogor Office). According to a third participant, adequate and appropriate education is crucial, especially at the leadership level (Participant 1, Purwakarta Office).
Survey respondents were asked to assess their own policy analysis skills and training. 62% of respondents indicated that they had not undertaken research training, and 66% indicated that they had not undertaken training related to policy analysis. Accordingly, respondents' self-confidence in their own professional policy analysis skills was not particularly high. For instance, while 57% of survey respondents reported confidence in their ability to "interpret the results of statistical analysis", only 36% agreed or strongly agreed that they were "an expert in applying the results of research".
Again, these results are not random. Respondents who reported having undertaken skills training were more likely to indicate confidence in their professional policy analysis skills. The majority (52%) of respondents who said they had not received research training also expressed low confidence in their ability to interpret statistical data, whereas 72% of those who had undertaken research training did report confidence in their ability to interpret statistical data. Likewise, 76% of respondents who said they had not undertaken policy analysis training also reported low confidence in their ability to apply the results of research, whereas a majority (57%) of respondents who had undertaken policy analysis training indicated a high level of confidence in their ability to apply the results of research. Figure 1 shows these results in side-by-side comparison. Discussion and conclusion Contra Colebatch (2004), the results from our survey and focus groups challenge the notion that policy analysis is dramatically different in a developing Asian country than it is in a Western democracy or that it does not take place at all in this context. Colebatch's (2004: 182) assessment that policy analysis as it is discussed in a Western bureaucratic context "bears little resemblance to the governmental process as it is experienced" in the developing world, particularly in Southeast Asia, appears to be something of an overstatement. Many of our survey respondents indicated that they devote significant time to policy analytical activities, including research, collecting and analysing data, and assessing policy options. Focus group respondents provided rich discussions of their work conducting research and analysing data, connecting with academics and other external research sources, creating advice for decision makers, and general policy analytical activities such as monitoring and evaluation of existing policies and programs. Both focus groups and survey responses noted the importance of human resource capacity for improved policy analysis, including adequate numbers of staff and appropriate education, skills, and training. These factors are commonly cited in the scholarly literature on policy analysis, and correspond to Howlett's (2015) features of policy analytical capacity (knowledge of the policy process and the needs of policy decision makers; individual research and analysis skills; organizational support; and physical resources). Moreover, our focus group data suggest that at least some West Java public servants have good networking capacity with academics, especially for the formulation of very strategic policies. This finding corroborates previous research on the impact of policy networks on policy capacity. For example, also using data from West Java, Hudalah et al. (2010) demonstrated that policy networks with academics can mobilize knowledge between policy actors and create sustainable program implementations. In the Indonesian government system, there is a long-standing tradition of hiring academics with certain skills as policy Fig. 1 Self-assessment of skills, training, and confidence in professional abilities: Side-by-side comparison entrepreneurs, or what is known as "Academic Administrative Entrepreneurs" (Wicaksono, 2020). Their expertise supports and influences the process of bureaucratic reform, although their role is limited to changes that have low political risks. This indirectly serves as a catalyst in increasing the policy capacity of public servants (Fatonie, 2020).
Most importantly, we did not find evidence of any particular style or tradition of policy analysis in our data. Our focus group respondents largely described their activities as conforming to a traditionally Western Weberian paradigm of public administration: they see a stark division between administrative formulation and implementation activities, in which administrative employees act as advice creators and loyal implementers only, while tough policy decisions are made by decision makers with political mandates. Further to this, focus group participants expressed substantial dissatisfaction with the level of authority for decision making they were given in their roles as policy analysts. Survey responses also indicate a perceived limit to an individual administrator's ability to influence policy decisions, with no apparent practical difference among respondents at different levels of management authority. This is not to say that an "Eastern", Asian, or non-Western policy style does not exist-it may, and Mukherjee and Howlett's (2016) hypothesis of a "convergence towards a common Asian style" of public policy should be pursued through further research. While we did not observe characteristics of policy analytical activity that would suggest a particular style in West Java, it is possible that other areas of policy decision making (for example, instrument selection or institutional design), or other geographical or political contexts might reveal these differences more clearly. In particular, future comparative research might reveal differences or similarities in policy analytical activity between countries within a regional bloc that we could not have uncovered in this study.