(PDF) Supporting 'Doing Development Differently' Effectively: Analysing Attributes, Networks and Impacts of Donor-Supported Paci

Asia & the Pacific Policy Studies - ORIGINAL ARTICLE OPEN ACCESS Supporting ‘Doing Development Differently’ Effectively: Analysing Attributes, Networks and Impacts of Donor‐ Supported Pacific Development Coalitions Aidan Craney1 | Dan Chamberlain2 | Chris Roche3 1 Department of Social Inquiry and Centre for Human Security and Social Change, La Trobe University, Bundoora, Australia | 2Australian Research Centre in Sex, Health and Society, La Trobe University, Bundoora, Australia | 3Centre for Human Security and Social Change, La Trobe University, Bundoora, Australia Correspondence: Aidan Craney (

) Received: 22 December 2023 | Revised: 27 October 2024 | Accepted: 9 January 2025 Keywords: doing development differently | donors | locally led development | pacific development | thinking and working politically ABSTRACT Despite growing appreciation for the complexity of development and the need for adaptability, there remains a limited evidence base of how, where and why adaptive approaches to development improve development outcomes. This paper examines the organisational and political attributes of support provided to locally led development initiatives in the Pacific region supported by an Australian government development programme, as well as the extent to which these attributes were present in initiatives deemed by programme staff to be more or less successful. Findings include that positive outcomes are more likely to emerge from partners led by women working in politically stable environments. Rather than endorsing donors to support only such projects, we share a combination of findings to offer insights into how the impacts of development initiatives working in politically tricky settings may be understood holistically and with nuance and flexibility, leading to better project design and evaluation. 1 | Introduction Adaptive approaches to development have gained traction in recent years along with the assertion of the importance of cultural and political contexts in determining developmental outcomes. Thinking and Working Politically (TWP) demands adaptability in programming focused on sustainable change rather than discrete project outputs (TWP CoP 2015). Advocates of Problem‐Driven Iterative Adaptation (PDIA) promote local leadership in design and facilitation of development interventions, with a commitment to ongoing learning and experimentation (Andrews, Pritchett, and Woolcock 2012). Similarly, the Doing Development Differently (DDD) community acknowledges that ‘Donors donʼt actually “do” development’, advocating strengthened capacity of local governments, organisations and networks to achieve developmental reform (Wild et al. 2016). Beyond this, these approaches all identify that creating the conditions for positive social change is complex. Progress is not linear, processes must be adaptable and there exists no singular ‘correct’ model. Andrews, Pritchett, and Woolcock (2012) have been particularly vocal about the need for development interventions to avoid engaging in practices that take a copy and paste approach to similar challenges being faced in very different settings. They argue that ‘successful’ development interventions require an understanding of localised political economy and that cultural, economic and political contexts prohibit the possibility for any one‐size‐fits‐all model. Projectised models of development—emphasising the value of This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. © 2025 The Author(s). Asia & the Pacific Policy Studies published by John Wiley & Sons Australia, Ltd. Asia & the Pacific Policy Studies, 2025; 12:e70013 https://doi.org/10.1002/app5.70013 1 of 12 There is a small but important catalogue of literature that comments on aid and development effectiveness in the Pacific that intersects with conversations about how development initiatives could be improved through greater recognition of ‘small p’ politics. Although it has been suggested that poverty rates would be higher in the Pacific (Feeny and McGillivray 2008) without the high aid levels that make the region the highest receiver of development assistance in the world (Dornan and Pryke 2017), development interventions in the Pacific are noted as being ‘less effective on average than projects in the rest of the developing world’ (Wood, Otor, and Dornan 2022, 4). Studies that have focused on development effectiveness in the Pacific have noted foundational constraints, such as countries having small populations that are geographically isolated (Dornan and Pryke 2017; Wood, Otor, and Dornan 2022), as well as structural challenges related to development economics following an orthodoxy that does not adhere to sustainable environmental nor cultural development for Pacific peoples (see Meki and Tarai 2023; Ravuvu and Thornton 2016). Drawing on PDIA principles, Wood, Otor, and Dornan. (2022) argue that improved development effectiveness in the Pacific is contingent on donors adapting their practices to better suit the particular contexts in which Pacific development occurs, as well as fostering adaptation within interventions. This paper takes an ‘inductive theory‐building approach’ (Dasandi, Marquette, and Robinson 2016, 6) in an attempt to highlight how an understanding of social, economic and other contexts influence the likelihood of success of development initiatives in the Pacific. It does so by examining the impacts of an Australian Aid programme—the Pacific Leadership Programme (PLP)—that was designed to support developmental leadership and locally led development interventions in the Pacific Islands region. It did this by providing funding and/or technical assistance to a range of projects led by local communities, civil society organisations, peak body groups and Pacific governments on discrete reform measures. We examine interventions, described here in generic terms as ‘coalitions’, because of the focus on collective action to identify the factors that contributed to greater and lesser success of several development interventions assisted by PLP, as well as the impact that PLPʼs support had on achieving greater or lesser success of these 2 of 12 coalitions. From 2008–2017, PLP worked with 77 different organisations based in Samoa, Solomon Islands, Tonga, Vanuatu and region wide. These included the private sector, churches, womenʼs organisations, youth groups, traditional and local level leadership, civil society organisations, national and regional leadership forums and organisations, regional organisations and programs, and others. The research this paper describes emerged from discussions between the authors and PLP staff in late 2016 and early 2017, who were aware that PLP was undergoing a formal ‘parachute’ evaluation that was unlikely to capture some of the complex, context‐rich impacts of the programmeʼs work. Further, we feared that the lessons would be lost if PLPʼs funding was not renewed.1 Seeing benefits for its internal learning and for its partners, PLP agreed to the proposed study. The aim of the study and this paper is to offer insights into how the impacts of development initiatives working in complex and politically tricky settings may be understood holistically and with nuance and flexibility, rather than being evaluated against indicators identified at project design stage which may or may not be appropriate. Further, we aim to stimulate thinking about how engaging in partnerships and working in complex settings can be better designed in ways that both promote greater likelihood of initiative success and build in flexibility for nuance in understanding the impacts of partnerships and their initiatives throughout their life cycle. In doing so, we contribute to the literature on TWP and the adaptive school of development by documenting how innovative and open approaches to collecting and analysing information about development initiatives can provide deeper understanding of their impacts (see Aston et al. 2022; Eyben et al. 2015; McCulloch and Piron 2019; OʼKeefe et al. 2014). We in particular explore how both contextual and intervention factors combine in determining the relative success of the initiatives we explored. This paper is organised as follows. The first section describes the three data collection approaches that contributed to our methodology, with an acknowledgement of the challenges of definitively capturing the influence donors and even implementing partners can have in shaping social, economic and political change. The following section discusses results from our research regarding which attributes appear to have contributed to greater or lesser success of a coalition, with a particular focus on sub‐regional geography, the gender composition of coalitions and the extent to which coalitions expanded their networks following support from PLP. The third section discusses the results in light of contemporary debates of adaptive development. The paper concludes with a summary of policy and programming implications of the results, with the key messages highlighted for practice and academic audiences. 2 | Methodology The research employed three discrete methods to attempt to understand the impact that PLPʼs partnerships had on coalitions it supported. The first method was a survey of key partners to assess the social, economic and governance impacts that partners self‐identified their coalitions contributing to in their areas Asia & the Pacific Policy Studies, 2025 20502680, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app5.70013 by Aidan Craney - La Trobe University Library , Wiley Online Library on [30/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License interventions that are predictable, have easily measurable outcomes and are timebound—run counter to recognition that effective development is typically complex, non‐linear and occurring over uncertain time frames (Scott, 2023). This throws up a number of challenges for the design, implementation, and monitoring and evaluation of contextually specific developmental initiatives aimed at achieving positive change in complex and politically tricky (Aston et al. 2022). These challenges include the difficulties of fixing specific objectives and timelines at the outset of programs; defining in detail the processes by which changing objectives might be achieved; measuring changes in relationships within initiatives as they evolve, and the impact of these; and summarising and communicating what has been achieved and how—and, from this, distilling lessons learnt which might guide future action. These challenges contribute to the lack of a strong evidence base supporting how and why thinking and working politically matters (McCulloch and Piron 2019). 2.1 | Survey of Pacific Leadership Programme Partners A survey instrument was used to gather insight into the value PLPʼs key partners placed on the partnership between their coalition and PLP and to what extent outcomes would have been achieved in the absence of such support. The survey was distributed to the 28 coalitions that staff identified as key partnerships throughout PLPʼs operations. 28 responses were received, representing 24 of the key partners.2 The survey sought to gather information from each of these key partners in three broad subject areas—the context relating to their partnering with PLP, the value‐adding impacts that their work has had on their communities that they ascribed at least somewhat to the support of PLP, and the characteristics of the key partnersʼ organisational structure, development foci and the political capital of their members. Key partners were not asked to self‐assess the success or otherwise of their coalition. Respondents were asked to provide long‐form information about the context of the formation of their coalition, what prompted their partnership with PLP and what factors are required for a positive relationship between donor and recipient. These responses were thematically coded and compared. The survey instrument requested information of each respondent about the positive and negative social, economic and governance impacts of their work and the extent to which they have been reliant on PLPʼs support. This specifically included asking about positive and negative intended and unintended consequences of partnering with PLP, information that we identified would be unlikely to be captured anywhere without else. Where possible, respondents were asked to provide examples. These responses were thematically coded and compared. Further questions were asked about whether key partners were ‘Entirely’, ‘Mostly’, ‘Partially’ or ‘Not at all’ dependent on PLP support for their daily operations. The characteristics of the key partners were captured with the intention of comparing these with findings generated through the qualitative comparative analysis. 2.2 | Qualitative Comparative Analysis The presence of specified attributes in more and less successful partnerships between PLP and significant project partners was examined using qualitative comparative analysis, which facilitates investigation within single datasets and across multiple complementary datasets (Aston et al. 2022). Two datasets were available for analysis. One was the aforementioned survey responses, completed by representatives of 24 of PLPʼs 28 key partners in March/April 2017. The second dataset also looked at these 28 coalitions and was collated from two sources: (a) PLP project documentation, such as project progress reports, and (b) a card sorting exercise carried out with PLP staff in early 2017. Current and former PLP staff engaged in a guided card sorting exercise, where they grouped each of the 28 key partners into categories or ‘more successful’ and ‘less successful’. The distinction between more and less successful was determined by the PLP staff involved, following the open card sorting principle of avoiding pre‐determination (Paul 2008, 8), and in recognition that it is almost impossible to judge social change initiatives as comprehensively successful or unsuccessful. Staff identified 17 key partners as more successful and 11 as less successful and then discussed attributes common to two or more of the key partners in either category, drawing on project documentation for triangulation. Combining these attributes with other information on the partners that had been previously gathered by PLP, such as thematic focus areas, a total of 46 attributes were registered (see Appendix A). These responses were analysed using EvalC3ʼs predictive modelling capacities.3 Analysing this data comparatively allowed us not only to see the presence or absence of these attributes for each key partner but to compare the combination of presence and absence of attributes across coalitions against indicators of greater and lesser success. 2.3 | Social Network Analysis A social network analysis (SNA) (Bodin, Sandstrom, and Crona 2017) was undertaken of 21 of the 28 significant project partners4 PLP worked with. SNA was used to produce descriptive analyses of the connections between actors involved in the coalitions across different regions, and the characteristics of the network structures. Network structures were created using datasets provided by PLP which indicated the network reach of the 21 examined coalitions upon the beginning of their partnership with PLP and as at time of data collection or how far it reached upon termination of partnership, if applicable. These data were analysed with a comparison of the two time events to determine the difference of the network structure since partnering with PLP commenced. The data were then analysed for ties per actor and coalition heterogeneity. Ties per actor were taken as a proxy measure for improved ability to engage in collective action (see Bodin, Sandstrom, and Crona 2017). Coalition heterogeneity was also measured to gauge the variety of skills, experiences and linkages that appeared necessary for a coalition to achieve their objectives. This measure looks at the 3 of 12 20502680, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app5.70013 by Aidan Craney - La Trobe University Library , Wiley Online Library on [30/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License of interest, as well as the extent to which the coalitions were dependent on PLPʼs support. The second method was a qualitative comparative analysis—drawing on project documents and a card sorting exercise that included identifying coalitions as ‘more’ or ‘less’ successful—to explore the combinations of contextual and intervention factors which might explain success or lack of it. The third method was a social network analysis, exploring the degree to which the changing shape of the networks and alliances contributed to success or otherwise. The methods were conducted independently of one another, with findings shared in the Results section. These findings were then cross‐analysed for broader insights into how developmental change can be supported in politically tricky contexts in the Pacific, reflected upon in the Discussion and Conclusion. This ‘bricolage’ approach has been identified as particularly useful in researching complex development interventions (Aston and Apgar 2022). 2.4 Limitations | Within the survey, limitations existed regarding the diversity and potentially the motivation of respondents. Survey respondents were all, to the best of our knowledge, individuals representing their coalition. The results therefore are liable to be indicative of individual rather than organisation‐wide perspectives. As in any survey the responses provided by individuals may represent amongst other things personal biases or agendas, gaps in knowledge or misunderstandings as to the purpose of the questions being asked. Further, individuals responding may not have been members of their coalitions for the whole time period being explored. For these reasons there were multiple respondents for some key partners and some conflicting responses. A potential limitation of the card sorting dataset is that it has attempted to quantitatively analyse data drawn from a qualitative method. Card‐sorting techniques can capture experiential knowledge by asking participants to identify and articulate attributes that data share (Coxon 1999). Some identified attributes may reflect individual biases of participants. This limitation was minimised by the involvement of multiple people in the gathering of the information by consensus techniques, triangulation with other data and comparison of the data collected through EvalC3. All three datasets were limited by the relatively small number of key partners that were available to be examined. Identifying trends in a small pool of actors is possible, but minor variations within any select few actors can accentuate differences in results. These issues are compounded further when the data is disaggregated by geographic focus. Cross‐referencing the findings across three distinct datasets mitigates some of the potential bias presented, however variations in results and their interpretation must be acknowledged. 3 3.1 approaches to development. Three examples offer an insight into these achievements. One countryʼs Chamber of Commerce and Industry5 reported being able to drive agendas for greater inclusion of women and youth in decision‐making processes and their needs being impacted by social and economic policy. The Women in Shared Decision Making (WISDM) coalition in Vanuatu drove legislative change for womenʼs quotas on municipal councils that resulted in both men and women voting for female candidates (see Rousseau and Kenneth 2017). And the Green Growth Leadersʼ Coalition has opened discussions at high levels about what development should look like in the Pacific region, with these discussions being reflected in the Vanuatu 2030 Peopleʼs Plan (see Craney and Hudson 2020). Positive economic consequences of PLPʼs support that were mentioned included challenging gender norms and greater recognition of youth needs and capabilities. For example, a private sector peak body successfully lobbied for the Women in Business Council to be included in the Fiji Commerce and Employers Council and Women and Youth Entrepreneurs Council. Further, the national youth council of one Pacific country that received support from PLP was invited to provide input into national budgetary measures impacting youth, while another was provided extra funding to expand their operations. Positive governance consequences of PLPʼs support mentioned included increased access and exposure to decision‐making processes and extending measures of public accountability and anti‐corruption. These were reflected in multiple examples. WISDM, the national youth councils and the Chamber of Commerce and Industry mentioned above were all recognised as peak bodies and invited into national decision‐making structures. One countryʼs civil society peak body saw increased accountability from elected representatives regarding parliamentariansʼ discretionary funding due to collective civil society action in villages it has worked with. The private sector peak body previously mentioned partnered with the United Nations Development Programme Pacific Office to co‐draft an anti‐corruption code for the private sector. Negative consequences were also raised but were far less prevalent. They related to risks inherent in challenging the status quo, changing mindsets so that people can embrace the unfamiliar, and expectations that civil society organisations have the capacity and/or legitimacy to represent and respond to the needs of their constituents. Such difficulties have been documented elsewhere, including in Ward and vanHeerde‐Hudson 2016 account of attempts to ratify the Convention on the Elimination of All Forms of Discrimination Against Women in Tonga. Results | | Self‐Reported Discrete Successes The partner survey provided evidence of numerous examples of positive change reported by the coalitions since engaging in partnership with PLP. These include matters related to changing policy and practice, challenging social norms, increased influence and exposure of coalitions and being intimately engaged in new 4 of 12 There is a strong case that PLPʼs role was a major contributor to the successes reported. 23 of the 24 respondents to the survey noted that they were mostly or entirely reliant on donor funding. This is not to suggest that partnership with PLP was the only necessary form of support for these organisations or their successes. As is discussed below, a number of extrinsic and intrinsic attributes for each key partner were associated with more and less success. Asia & the Pacific Policy Studies, 2025 20502680, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app5.70013 by Aidan Craney - La Trobe University Library , Wiley Online Library on [30/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License diversity of coalition membership. To achieve this, all network actors were coded into classifications of domestic governance body, multilateral/regional agency, international donor government, international non‐government organisation (NGO), locally‐based NGO/civil society organisation, faith‐based organisation, and private sector‐focused. Examining the networks by these categories allowed for insight into the diversity of their networks, providing insights into where network scope and strength may be associated with greater success of a coalition. The following sub‐sections investigate each of the datasets in more detail. These begin by looking at the preponderance of greater or lesser success with respect to geography and gender of leadership. This is followed by a discussion of the attributes that are commonly present or absent in more and less successful partnerships as identified through the use of EvalC3 prediction modelling. Finally, the shape of greater and lesser successful partners is discussed as analysed through the social network analysis undertaken. 3.3 | Success by Geography The success of partners in Polynesian countries compared with Melanesian and regional partners is notable (see Table 1). In Tonga, all supported partnerships were deemed by PLP staff to have been more successful. The levels of success drop slightly to Samoa and further to Solomon Islands, though more than half of the cases from these countries were still deemed more successful. Less successful were partnerships based in Vanuatu or with a regional focus. Numerous possible reasons exist for the fairly stark divergence in success levels. Difficulties in achieving successful developmental outcomes may be a result of political, social and/or environmental factors as well as the operations of the specific coalitions or PLP. Also, Samoa and Tonga are much more culturally homogenous than Vanuatu or Solomon Islands. Further, the political systems of the Polynesian partner countries are more stable than their Melanesian counterparts. Positions of leadership are held by a relatively exclusive elite resulting in leaders being more intimately connected to and knowledgeable of other leaders across various categories (Barbara and Haley 2015, 51). Geography and population size offer other reasons for difficulties in achieving success in Vanuatu and Solomon Islands, which have greater populations and are spread over a greater number of islands than Samoa and Tonga. Though this appears on face value to add a layer of difficulty to promoting change through coalitions in Melanesia, Barbara and Haley argue that this may in fact provide opportunities. As the political landscape is less stable, more entry points for agile networks to exploit may open up as the political landscape shifts (2015, 45–47). Another possible explanation as to why Tonga and Samoa achieved higher levels of reported success than other countries TABLE 1 | Success by geography. Coalitions may be that the projects PLP partnered in these countries were quite similar. For example, four key partners have counterpart organisations in both countries. This may reflect the attributes of the individual organisations or simply be indicative of the greater levels and intensity of networking between Tonga and Samoa, compared to Vanuatu and Solomon Islands. Interestingly, the one Samoan key partner without a Tongan counterpart actually had one in Vanuatu—both of which were deemed less successful. Intriguingly, regional initiatives were twice as likely to be judged less successful than more successful. Exploration of attribute sets across these coalitions showed the least consistency for association of greater and lesser success. As such, the reasons for inconsistencies across these key partners is unclear and requires further research. The lesson from studying success by geography is not that donor money is best spent on development initiatives based on Polynesia. Indeed, the lower rates of success in Melanesia may speak to a greater need for development assistance in those countries. However, we do suggest that expectations need to be modified depending on country and issue contexts. As a caveat to these data, it should be noted that while the sample sizes of partners disaggregated by geography are similar they remain small. For key partners based in Samoa and Solomon Islands the 20% difference in more successful partnerships, for example, consists of a singular case variance. 3.4 | Success by Gender Strikingly, coalitions led by women were strongly associated with greater success. In 13 of the 16 cases where a coalition was led by a woman it was deemed a greater success. In comparison, only 4 of the 12 coalitions headed by men was deemed a greater success (see Table 2). McLeod (2015) has already demonstrated that womenʼs leadership in civil society spaces in the Pacific is more widespread than in other spheres, yet little is known about the impact of this leadership. Spark and Corbett (2018) argue that this is at least in part a result of the unwelcoming environment for women leaders in formal political spaces, which means women choose to lead in less formal realms. Fletcher, Brimacombe, and Roche (2016) find that women civil society leaders in the Pacific utilise a combination of skills to promote inclusivity and challenge established norms in manners that can successfully promote social change. The reasons for the greater association of women leaders with success may be related to how women promote the expertise of those around them, generate support beyond their direct environment, or some combination of such and other factors. With reference to the PLP‐supported WISDM coalition, Rousseau and Kenneth noted the ability of women leaders in the Pacific to # Deemed more successful % Regional 2 out of 6 33% Samoa 4 out of 5 80% Tonga 6 out of 6 100% Coalitions TABLE 2 | Success by gender. # Deemed more successful % Vanuatu 2 out of 6 33% Led by women 13 out of 16 81% Solomon Islands 3 out of 5 60% Led by men 4 out of 12 33% 5 of 12 20502680, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app5.70013 by Aidan Craney - La Trobe University Library , Wiley Online Library on [30/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 3.2 | Evaluating Greater and Lesser Coalition Success Of additional interest is the finding that beyond the sole attribute of womenʼs leadership in association with more success is that this is particularly true in coalitions whose staff/members have cross‐sectoral knowledge (evident in 11 of the 13 more successful coalitions led by women) and were prepared to expend political capital (evident in 12 of the 13 more successful coalitions led by women). This finding suggests projects are more likely to achieve success if the staff or membership base demonstrate both commitment to promoting developmental change as well as an understanding of how social change happens. This finding supports TWP, DDD and PDIA philosophies (Andrews, Pritchett, and Woolcock 2012; Honig 2018; McCulloch and Piron 2019) regarding the need to tailor interventions in ways that are congruent with the political economic context as opposed to following standardised formulae. 3.5 | Factors Present in More Successful Cases From the card‐sorting exercise with PLP staff, 46 attributes were described as being present in two or more of the partners. Attributes fell broadly into three groups: process, which included the focus of the coalition and its operations; influence, which included the skills, knowledge and reach of key members/staff; and outcomes, which included evaluations of the coalitionʼs impact over the course of the partnership. Each partner was then assessed against each attribute for its presence or absence. This was then cross‐referenced against determinations of greater and lesser success for each partner (see Appendix A). The attributes were further analysed to determine their prevalence by presence or absence by geographic breakdown (see Table 3). Responses from the partnership survey regarding the factors that contributed to a positive working relationship appear to reinforce PLPʼs adaptive approach to supporting developmental change. Themes of trust, flexibility and understanding the context of the development environment in the Pacific were emphasised as attributes supporting the notion that person‐to‐ person connections are a crucial factor in positive relationships of this nature. Further, practical support measures related to technical assistance, training, mentoring and networking were also highlighted. These responses are supported by the association of more success with attributes such as shared values and equal commitment between PLP and the key partners. Further results of note include the association of greater success with positive attributes related to organisational capacity, leadership and influence in Solomon Islands. These findings suggest that increased exposure to networks and willingness to expend political capital with them were positively linked to achieving success. At an organisational level, the association of success with good governance and the ability to increase staff/membership size suggests internal processes were important to 6 of 12 achieving positive outcomes. It should also be noted that though staff/member size was positively associated with greater success in all cases, this did not extend to growth in financial turnover or donors. 3.6 | Factors Absent in More Successful Cases The absence of attributes in more successful cases provides further insight into the dynamics of key partners. Though there are fewer highlighted attributes to examine (see Table 4), the results are still valuable. Unsurprisingly, an absence of fraud and/or mismanagement was positively associated with all more successful key partners. Interestingly, all more successful key partners also reported that their partnership was not contingent upon a key PLP staff member. The PLP staff identified this as an attribute which may enhance the partnershipʼs strength but also risked destabilising it if the individual staff member were to leave PLP or be unavailable for any reason. That the absence of this attribute is associated with more successful partnerships suggests that these key partners were not heavily reliant on regular support from PLP to achieve their organisational objectives, that they had created solid relationships with multiple PLP staff members or a combination of the two. Less apparent, but of note, are the findings that Vanuatu‐based and regional‐focused key partners were more likely to be associated with success when their donor networks and financial turnover had not increased by very much. This seems to support Barbara and Haleyʼs (2015) suggestion that strategically targeted coalitions may be more likely to be successful in Melanesian countries because the regularly shifting political landscape provides more opportunities for this sort of reform initiative than in more stable political situations. Given this is not a finding replicated throughout all sites it should not be taken as a suggestion that these attributes are unimportant to all development projects. These findings do suggest, however, that increasing network size is not a necessity for the success of all development initiatives. 3.7 | Shape of More and Less Successful Networks To understand the network influences which were associated with greater and lesser success in partnerships, the significant connections of the key partners were identified by PLP staff. Country programme managers utilised project documentation and their knowledge of the functioning of 20 of the 28 key partners—14 of which were deemed more successful and 6 which were less successful7—to note the significant connections each key partner had at the beginning of their partnership with PLP and when the partnership ceased. These connections were coded by organisational background, as outlined in the earlier discussion of coalition heterogeneity. Comparing the aggregated profiles of the number and form of connections of more and less successful partnerships displays a clear pattern of greater local support for more successful partnerships (see Table 5). More successful key partners had an average of almost four times as many connections with domestic Asia & the Pacific Policy Studies, 2025 20502680, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app5.70013 by Aidan Craney - La Trobe University Library , Wiley Online Library on [30/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License wield both ‘hard’ and ‘soft’ influence (2017). Of course, this is not to suggest that the reasons for greater success are uniform across the cohort of women leaders and coalitions, as each leader, coalition and context is different. Attributes Regional Samoa Tonga Vanuatu Solomon Islands Grew in influence X X X X X Networking influence increased X X X X X Networks increased in size X X X X X Shared PLP‐Partner values X X X X X Focused on developmental reform X X X X Organisation had a clear purpose X X X X Had a broad stakeholder base X X X X Equal commitment between PLP and partner X X X X Supported by persons of influence (outside org) X X X X Responded to a specific need X X X X X Strategic direction (positive) X X X Politically savvy X X X Members/staff were persons of influence X X X Strong, well‐connected individual leader X X X Governance (positive) X X X Members/staff were prepared to expend political capital X X X X X X X X X Members/staff were recognised leaders in their field/s X Cross‐sectoral knowledge Strong government support Has broadened its objectives Partnership was PLP time intensive X X X Purpose was aligned with DFAT objectives X Partnership recommended by DFAT/AusAID X Strong external support (civil society) X Led by a woman X Good reporting X Grew in membership/staff size X Financial turnover increased X Number of donors increased X Present in all more successful and absent in all less successful coalitions X Present in all more successful and some less successful coalitions X Present in all more successful and all less successful coalitions X governance bodies upon partnership completion as less successful key partners and roughly two‐thirds more average connections with domestic NGOs/civil society organisations (CSOs). Conversely, international support was not strongly associated with more success in PLPʼs partnerships. Average connections with international donor governments was relatively similar at beginning and end of partnerships. Interestingly, average connections with international NGOs was much higher at both beginning and end of PLP partnership with less successful key partners than it was with more successful key partners. These results suggest that domestic support is more likely to be associated with success in achieving developmental change in the Pacific region. This supports iterative approaches to development, such as TWP and PDIA, which promote the need to understand political and cultural contexts when engaging in development interventions, as these are more likely to be innate in local actors. Alternatively, it may reflect a willingness from some international NGOs to support organisations and coalitions with less technical capacity, influence and likelihood of achieving success in the short‐term, precisely because they are working in challenging political environments and require greater assistance. The one datapoint that may challenge the association of local support and success is the higher average connection of more successful key partners with regional and multilateral bodies 7 of 12 20502680, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app5.70013 by Aidan Craney - La Trobe University Library , Wiley Online Library on [30/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License TABLE 3 | Attributes present in 100% of successful cases per location.6 Attributes Regional Samoa Tonga Vanuatu Solomon Islands Fraud/mismanagement present X X X X X Partnership contingent on key PLP staff X X X X X Financial turnover increased X Number of donors increased X Coalition broadened its objectives X X X X Divergence of interests X Grew in membership/staff size X X Partnership recommended by DFAT/AusAID X Absent in all more successful and present in all less successful coalitions X Absent in all more successful and some less successful coalitions X Absent in all more successful and all less successful coalitions X TABLE 5 | Total number of connections of more successful key partners at beginning and end of PLP partnership. Connections present at beginning Connections present at end Average at beginning Average at end Domestic governance body 30 62 2.14 4.43 Regional and multilateral 12 27 0.86 1.93 International Donor Govt 9 24 0.64 1.71 International NGO 1 8 0.07 0.57 Domestic NGO/CSO 7 31 0.50 2.21 Faith‐based organisation 1 3 0.07 0.21 Private sector 12 14 0.86 1.00 Total connections 72 169 5.14 12.06 than less successful key partners have (see Table 6). Whilst the regional agencies replicate local knowledge, it should be noted that they make up significantly less of the noted connections than multilateral bodies. Explaining the association of these agencies with more success would require further within‐case analysis, as the links with more success may be explained by a number of potential variables. This may include the large number of local staff within these offices, more adaptive approaches to developmental change existing in multilateral bodies than from international donor governments or issues‐ focused international NGOs, or that multilateral bodies themselves have highly interconnected networks which may expose partners to individuals and coalitions with diverse and useful skillsets. 4 | Discussion The results of the multi‐sited, mixed method analysis of the attributes of more and less successful project partnerships which PLP engaged provide an indication for the potential of this form of approach for examining and investing in coalition‐ 8 of 12 focused development reforms in the Pacific region in the future. The combination of attributes present or absent in key partners, the reach and form of their networks, and the examples of social, economic and governance change that they have been able to achieve with PLPʼs support provide some insight into how developmental change takes place in the Pacific region. The findings presented range across the spectrum from the obvious to the more surprizing. Identifying that a lack of fraud and mismanagement is associated with greater likelihood of success confirms what would be intuitive for most people. The association of womenʼs leadership with success in itself is not surprizing, but the extent to which it was so much more closely associated with greater success than partners led by men provides pause to consider what this may say about leadership styles and the role of women in promoting positive social change, outside of formal political spaces. Further, when viewing these findings in light of the assertions by McLeod (2015) and Spark and Corbett (2018) that women in the Pacific are more likely to exercise leadership in the civil society sector, further possible implications arise. If womenʼs leadership leads to positive outcomes in the only sector where it is prominent, then perhaps Asia & the Pacific Policy Studies, 2025 20502680, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app5.70013 by Aidan Craney - La Trobe University Library , Wiley Online Library on [30/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License TABLE 4 | Attributes absent in 100% of successful cases per location. Domestic governance body Connections present at beginning Connections present at end Average at beginning Average at end 3 7 0.50 1.17 Regional and multilateral 2 3 0.33 0.50 International Donor Govt 5 10 0.83 1.67 International NGO 4 7 0.67 1.17 Domestic NGO/CSO 4 8 0.67 1.33 Faith‐based organisation 5 5 0.83 0.83 Private sector 1 2 0.17 0.33 Total connections 24 42 4.00 7.00 greater space should be open for such leadership in other spheres. It would appear that the public and private sector would benefit from investing time and resources into understanding how best to incorporate and promote these skills for the benefits of each sector, as well as providing pathways for womenʼs leadership. Disaggregating coalitions geographically, the greater level of cultural homogeneity in Polynesian partner countries provides a soft basis to expect higher levels of success with key partners in these countries, though the extent to which this was evident over Melanesian and regional partners was notable. Less expected was the finding that support from international NGOs would be more strongly associated with less successful partners. As previously mentioned, this should not be taken as evidence of poor partnership choices by international NGOs. It may well be that such organisations are supporting initiatives seeking to influence change over a longer time span or in more politically difficult areas. Rather than act as a justification to avoid partnerships less inclined to short‐term success, these findings should prompt reflection from donor and partner agencies about what forms of change they hope to induce and what success looks like for them. Achieving consistent ‘success’ by supporting initiatives with small‐target interventions that do not drive positive change is anathema to supporting developmental reform measures. The strong association of local support with greater levels of success is important, though. This finding mirrors the growing consensus within the emerging literature on driving social change which promotes the value of local context and knowledge (see Booth and Unsworth 2014; Roche et al. 2020; Tawake et al. 2021; Wild et al. 2016). This is not to suggest that international partnerships are not beneficial—indeed, Spark, Cox, and Corbett (2019) find that such relationships are a strategic attribute of women leaders in the Pacific—but to highlight the importance of local capacity and relationship building. The lesson here for international agencies may be that they should look at how they can be in supportive networks as donors and providers of technical assistance, rather than seeking to work in one‐on‐one relationships with local development actors. Further, the correlation between local support and greater success adds weight to the importance of the survey data provided by the key partners on the attributes of PLP they felt enhanced its value as a donor and partner. The key partners highlighted characteristics of trust, flexibility, understanding context and person‐to‐person connections. This is to complement professional support, with the key partners also noting the importance of assistance by means of mentoring, training and technical assistance. The repeated responses referencing interpersonal skills imply that technical skills are not sufficient to providing appropriate levels of support and that, indeed, soft skills and trusted relationships are equally valuable, in line with adaptive approaches to development (TWP CoP 2015). A successful relationship between a central actor and supported partners seems to require: a) an alignment of values between that actor and the coalition (100% of more successful coalitions, 55% of less successful ones); b) an equal commitment from both parties (88% of more successful coalitions, 18% of less successful ones); and c) coalitionsʼ functioning not being contingent on PLP staff. Achieving this combination requires time and commitment from both parties, typically beyond the typical donor programme timeframes of three‐to‐five years. Beyond this, the results also indicate that increased network reach is associated with more successful outcomes. More so, exposure to networks across multiple and various categories across public, private and non‐profit sectors was associated with greater success, suggesting an increase in network exposure may result in increased influence. Further, being linked to others and their networks can allow access to a broader set of resources, links and influence (see Scott & Thomas, 2017). This implication is further supported by the finding that staff/ members with cross‐sectoral knowledge who were willing to expend political capital are associated with greater success, particularly when their coalition or organisation was led by a woman. These findings are not intended to be considered as definitive. The sample size of PLPʼs key partners is small—smaller still when disaggregated by geographic focus—and though input was sought from both PLP staff and representatives of the key partners discussed, this study would have benefited from access to project documentation and a baseline hypothesis to test against. This is where we believe the greatest utility in these findings lays. Instead of being viewed as stand‐alone conclusions to add to an evidence base of how change happens and the 9 of 12 20502680, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app5.70013 by Aidan Craney - La Trobe University Library , Wiley Online Library on [30/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License TABLE 6 | Total number of connections of less successful key partners at beginning and end of PLP partnership. agency and the coalitions themselves and seek to capture both the intended and unintended developmental consequences of the intervention. Combining a context centred focus with the ability to adapt to changing environments and emerging information may prove an important shift in leading to more effective and sustainable development in the Pacific region. 5 Conflicts of Interest | Conclusion Despite all caveats, the findings from this study offer insights for practitioners to consider when working on development initiatives that are reform‐based and/or politically tricky— particularly in the Pacific. Further, as has already been mentioned, the purpose of this exercise is not to attempt to generate definitive findings of how coalitions must be supported to make change happen in the Pacific region. Rather, the findings help to indicate what combinations of contextual and organisational factors may contribute to more and less successful partnerships between donor agencies and locally led and owned networks, coalitions and organisations, and to assist practitioners in thinking about how to design, support and evaluate complex and challenging development initiatives. It is important to highlight that there is no one size fits all answer. There are different ways to achieve the same thing. Support needs to be tailored to specific coalitions dependent on their location, the issues they are exploring, the networks they are already part of and the skills they have at their disposal. And while there do seem to be a number of organisational attributes which, although they do not ensure success, may indicate whether a collation with these attributes might be a ‘better bet’ than those without them, our analysis also indicates that predicting failure can be as difficult as predicting success—that is, there is no one explanation for failure either. Our findings have important implications for designing, implementing and monitoring and evaluating programs that seek to support reform coalitions in the Pacific and beyond. This includes designing programs in ways which are ‘context centred’ as much as ‘intervention‐centred’ (Cartwright 2017)—that is, they recognise that the particular political economy of the context and issue will play an important role in determining success. Further, initiatives should be implemented in ways that not only tailor to context and the issue, but which build from existing relationships and build coalitions of people with complementary attributes appropriate to achieving an initiativeʼs aim. Finally, monitoring and evaluation should assist in the understanding of network evolution and shifts in coalition attributes, track the ongoing relationship between the supporting 10 of 12 Acknowledgements The authors would like to thank the staff of the Pacific Leadership Program and members of their key partners who generously contributed to the data collection of this paper. We also wish to thank Rick Davies for his kind and willing assistance in utilising EvalC3 for the data analysis. Open access publishing facilitated by La Trobe University, as part of the Wiley ‐ La Trobe University agreement via the Council of Australian University Librarians. At time of data collection, Aidan Craney was the research coordinator for the Pacific Leadership Programme. This position ceased with the closure of the programme at the end of 2017. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. Endnotes 1 PLPʼs funding was discontinued at the end of 2017. 2 One key partner had three survey respondents, one had two respondents and all others had a single respondent. Four key partners did not provide a response. 3 EvalC3 is a web‐based tool ‘for developing, exploring and evaluating predictive models of expected outcomes’ (Davies). The tool utilises binary data to generate evaluable prediction models based on the configurations of attributes of cases in a dataset. The performance of these prediction models is assessed using concepts from Qualitative Comparative Analysis—that is, necessity and sufficiency relationships (Rihoux and Ragin 2009)—and from supervised machine learning— that is, the Confusion Matrix and Classification Accuracy (Kotu and Deshpande 2015). For our study, we used EvalC3 to identify and attempt to understand why certain sets of attributes (see Appendix A) of the partnerships and their context may be more or less likely to be associated with a greater or lesser success of the coalitionʼs developmental influence (noting patterns of necessity and sufficiency, not associating relationships of correlation). We did this by assigning binary codes across each of our attributes (assigning 1 for presence, 2 for absence), which EvalC3 then analysed, identifying patterns in relationships across attributes that related to greater and lesser success. 4 Seven of these were excluded due to lack of data on the size of beginning and final networks. 5 Partner organisations are anonymised in this paper except in circumstances where the support of PLP and outcomes relevant to this paper have already been published. 6 Data is presented disaggregated by location and only highlighting attributes 100% associated with success as this provides a better snapshot of attributes across locations than can be seen in a list of the presence and absence percentages across all locations without geographic disaggregation. Given the small sample sizes in each location (i.e., 5 or 6 coalitions per location), this disaggregation also maintains fidelity to Asia & the Pacific Policy Studies, 2025 20502680, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app5.70013 by Aidan Craney - La Trobe University Library , Wiley Online Library on [30/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License role of coalitions in association with this, the findings presented offer the beginnings of a middle range theory (Merton 1968; see also Dasandi, Marquette, and Robinson 2016) that potentially allows for the construction of a grounded hypothesis of how developmental change might or can happen in the Pacific region, which could be further developed and tested. As such, the results discussed offer some building blocks for a framework for future coalition‐based projects to be evaluated against. These findings provide insights useful for practitioners and academics seeking deeper understanding of why Pacific development initiatives have lower success rates than in other parts of the world (see Wood, Otor, and Dornan 2022), and particularly with relation to the lack of examples regarding the success or otherwise of politically informed development programs from the Pacific region (Denney and McLaren 2016, 1). 7 Eight key partners did not have their connections documented due to lack of paperwork, particularly when partnership had ceased multiple years ago, or because the key partners were part of multi‐level coalitions where connections and membership could not be cleanly articulated. Of these eight partners, three had initially been adjudged more successful and five as less successful. References Andrews, M., L. Pritchett, and M. Woolcock. 2012. Escaping Capability Traps Through Problem‐Driven Iterative Adaptation. Washington, DC: Center for Global Development. CGD Working Paper 299. Aston, T., and M. Apgar. 2022. The art and Craft of Bricolage in Evaluation, Practice Paper. Brighton: Institute of Development Studies. Aston, T., C. Roche, M. Schaaf, and S. Cant. 2022. “Monitoring and Evaluation for Thinking and Working Politically.” Evaluation 28, no. 1: 36–57. https://doi.org/10.1177/13563890211053028. Barbara, J., and N. Haley. 2015. “Analytical Framework to Engage With Developmental Leadership in the Pacific.” Pacific Leadership Program. Bodin, A., A. Sandstrom, and B. Crona. 2017. “Collaborative Networks for Effective Ecosystem‐Based Management: A Set of Working Hypotheses.” Policy Studies Journal 45, no. 2: 289–314. https://doi.org/10.1111/psj. 12146. Booth, D., and S. Unsworth. 2014. Politically Smart, Locally Led Development, Discussion Paper. London: Overseas Development Institute. Cartwright, N. 2017. “Evidence for Policy Prediction: Intervention‐ Centred or Context‐Centred?” LIDC Development Debate, London School of Hygiene and Tropical Medicine, March 29th: 2017. https:// www.lshtm.ac.uk/newsevents/events/evidence‐policy‐prediction‐ intervention‐centred‐or‐context‐centred. Coxon, A. P. M. 1999. Sorting Data: Collection and Analysis. Thousand Oaks: SAGE. Craney, A., and D. Hudson. 2020. “Navigating the Dilemmas of Politically‐Smart, Locally‐Led Development: The Pacific‐based Green Growth Leaders’ Coalition.” Third World Quarterly 41, no. 10: 1653– 1699. https://doi.org/10.1080/01436597.2020.1773256. Dasandi, N., H. Marquette, and M. Robinson. 2016. Thinking and Working Politically: From Theory Building to Building an Evidence Base, DLP Research Paper 37. Birmingham: Developmental Leadership Program. Davies, R. n.d. EvalC3. www.evalc3.net. Denney, L., and R. McLaren. 2016. Thinking and Working Politically to Support Developmental Leadership and Coalitions: The Pacific Leadership Program, DLP Research Paper 41. Birmingham: Developmental Leadership Program. Dornan, M., and J. Pryke. 2017. “Foreign Aid to the Pacific: Trends and Developments in the Twenty‐First Century.” Asia & the Pacific Policy Studies 4, no. 3: 386–404. https://doi.org/10.1002/app5.185. Eyben, R., I. Guijit, C. Roche, and C. Shutt. 2015. “The Politics of Evidence and Results in International Development: Playing the Game to Change the Rules?” Practical Action Publishing. Kotu, V., and B. Deshpande. 2015. Predictive Analytics and Data Mining: Concepts and Practice With RapidMiner. Waltham: Elsevier. McCulloch, N., and L. H. Piron. 2019. “Thinking and Working Politically: Learning From Practice – Overview to Special Issue.” Development Policy Review 37, no. S1: O1–O15. https://doi.org/10.1111/dpr.12439. McLeod, A. 2015. Women’s Leadership in the Pacific, Developmental Leadership Program State of the Art Paper, 4. Birmingham: Developmental Leadership Program. Meki, T., and J. Tarai. 2023. “How can Aid be Decolonized and Localized in the Pacific? Yielding and Wielding Power.” Development Policy Review 41, no. 2: e12732. https://doi.org/10.1111/dpr.12732. Merton, R. K. 1968. Social Theory and Social Structure. New York: Free Press. O’Keefe, M., J. T. Sidel, H. Marquette, C. Roche, D. Hudson, and N. Dasandi. 2014. Using Action Research and Learning for Politically Informed Programming, Research Paper 29. Birmingham: Developmental Leadership Program. Paul, C. L. 2008. “A Modified Delfi Approach to a New Card Sorting Methodology.” Journal of Usability Studies 4, no. 1: 7–30. Ravuvu, A., and A. Thornton. 2016. “Beyond Aid Distribution: Aid Effectiveness, Neoliberal and Neostructural Reforms in Pacific Island Countries.” In Assessing the Impact of Foreign Aid, edited by V. Jakupec and M. Kelley. London: Elsevier. Rihoux, B., and C. C. Ragin. 2009. Configurational Comparative Methods: Qualitative Comparative Analysis (QCA) and Related Techniques. Thousand Oaks: SAGE Publications. Roche, C., J. Cox, M. Rokotuibau, P. Tawake, and Y. Smith. 2020. “The Characteristics of Locally Led Development in the Pacific.” Politics and Governance 8, no. 4: 136–146. https://doi.org/10.17645/pag.v8i4.3551. Rousseau, B., and D. Kenneth. 2017. Supporting Coalition‐Based Reform in Vanuatu. Birmingham: Developmental Leadership Program. Scott, C. 2023. The Project in International Development: Theory and Practice. Abingdon: Routledge. Scott, T. A., and C. W. Thomas. 2017. “Winners and Losers in the Ecology of Games: Network Position, Connectivity, and the Benefits of Collaborative Governance Regimes.” Journal of Public Administration Research and Theory 27, no. 4: 647–660. https://doi.org/10.1093/jopart/mux009. Spark, C., and J. Corbett. 2018. “Emerging Women Leadersʼ Views on Political Participation in Melanesia.” International Feminist Journal of Politics 20, no. 2: 221–235. https://doi.org/10.1080/14616742.2016.1189680. Spark, C., J. Cox, and J. Corbett. 2019. “Gender, Political Representation and Symbolic Capital: How Some Women Politicians Succeed.” Third World Quarterly 40, no. 7: 1227–1245. https://doi.org/10.1080/01436597. 2019.1604132. Tawake, P., M. Rokotuibau, J. Kalpokas‐Doan, A. Mua Illingworth, A. Gibert, and Y. Smith. 2021. Decolonisation and Locally Led Development. Canberra: Australian Council for International Development. TWP CoP. 2015. “The Case for Thinking and Working Politically: The Implications of ‘Doing Development Differently.” https://twpcommunity.org/ wp‐content/uploads/2018/02/the‐case‐for‐thinking‐and‐working‐ politically.pdf. Feeny, S., and M. McGillivray. 2008. “Do Pacific Countries Receive Too Much Foreign Aid?” Pacific Economic Bulletin 23, no. 2: 166–178. Ward, O., and J. vanHeerde‐Hudson. 2016. “Lessons Learned on Thinking and Working Politically for the Ratification of CEDAW in Tonga.” Pacific Leadership Program. Fletcher, G., T. Brimacombe, and C. Roche. 2016. Power, Politics and Coalitions in the Pacific: Lessons From Collective Action on Gender and Power, Research Paper 42. Birmingham: Developmental Leadership Program. Wild, L., M. Andrews, J. Pett, and H. Dempster. 2016. Doing Development Differently: Who we are, What we’re Doing and What We’re Learning. London: Overseas Development Institute. Honig, D. 2018. Navigation by Judgment: Why and When Top‐Down Management of Foreign Aid Doesnʼt Work. Oxford: Oxford University Press. Wood, T., S. Otor, and M. Dornan. 2022. “Why are Aid Projects Less Effective in the Pacific?” Development Policy Review 40, no. 3: e12573. https://doi.org/10.1111/dpr.12573. 11 of 12 20502680, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app5.70013 by Aidan Craney - La Trobe University Library , Wiley Online Library on [30/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License the prevalence of attributes in relation to greater success. A full attributes list with percentages of prevalence can be found in Appendix A Table A1. TABLE A1 | Correlation of attributes and greater or lesser success. Present in more successful coalitions Present in less successful coalitions Focused on local issues 100.0% 72.7% Grew in influence 100.0% 36.4% Shared PLP‐Partner values 100.0% 54.5% Network size increased 100.0% 54.5% Networking influence increased 100.0% 27.3% Clear purpose 94.1% 72.7% Broad stakeholder base 94.1% 45.5% Respond to specific need 94.1% 81.8% Passionate and committed 94.1% 63.6% Governance (positive) 88.2% 54.5% Focused on reform 88.2% 45.5% Politically savvy 88.2% 45.5% Members recognised as leaders in field/s 88.2% 63.6% Members prepared to expend political capital 88.2% 54.5% Partner at time of data collection 88.2% 27.3% Equal commitment between PLP and partner 88.2% 18.2% Supported by persons of influence (outside org) 88.2% 54.5% Members are persons of influence 88.2% 45.5% Strong, well‐connected individual leader 88.2% 45.5% Strong government support 82.4% 27.3% Cross‐sectoral knowledge 82.4% 81.8% Positive strategic direction 76.5% 36.4% Led by a woman 76.5% 27.3% Processes led to outcomes 76.5% 27.3% Strong external support (civil society) 76.5% 63.6% Broadened its objectives during operation 70.6% 36.4% Purpose aligned with DFAT objectives 64.7% 36.4% Good reporting 64.7% 9.1% Partnership was PLP time intensive 58.8% 45.5% Grew in membership/staff size 52.9% 18.2% Strong organisational capacity 47.1% 18.2% Financial turnover increased 41.2% 18.2% Number of donors increased 41.2% 27.3% Focused on womenʼs leadership 41.2% 9.1% Focused on community leadership 35.3% 63.6% Focused on political‐bureaucratic leadership 35.3% 0.0% Staff wages were secure 35.3% 36.4% Focused on Melanesian leadership 29.4% 36.4% Focused on future developmental leadership 29.4% 45.5% Partnership recommended by DFAT/AusAID 29.4% 45.5% Focused on private sector leadership 23.5% 9.1% Interests diverged between PLP and partner 17.6% 63.6% Focused on local governance support 17.6% 9.1% Focused on regional issues 11.8% 18.2% Fraud/ mismanagement present 0.0% 36.4% Partnership contingent on key PLP staff 0.0% 18.2% 12 of 12 Asia & the Pacific Policy Studies, 2025 20502680, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/app5.70013 by Aidan Craney - La Trobe University Library , Wiley Online Library on [30/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Appendix A