Strengthening Institutional Isomorphism in Development NGOs? Program Mechanisms in an Organizational Intervention

Nongovernmental organizations (NGOs) in international development struggle between being actors in the mainstream or representatives of alternatives to it. However, many NGOs all over the world align with the mainstream and are increasingly similar to each other. This homogenization results from institutional isomorphism, which is affected by their aspirations to be legitimate vis-á-vis the international field. Consultancies are among the main practices to promote normative isomorphism, but little is known about their micro-level dynamics. Drawing on the notion of program mechanisms in realistic evaluation, we scrutinize how external facilitators in organizational development processes enable normative isomorphism. As a result of analysis of interventions in three Finnish development NGOs, we identify program mechanisms of convincing, embedding, and consolidating. Our findings show how organizational development activities contribute to the direction of change toward normative isomorphism and argue that a detailed analysis of intervention mechanisms would be useful for self-reflection in any field of activity.


Introduction
One of the main questions concerning the roles of nongovernmental organizations (NGOs) in international development is whether they are agents for innovative, critical transformation or actors that are easily co-opted by mainstream development trends (Banks, Hulme, & Edwards, 2015;Bebbington, Hickey, & Mitlin, 2008;Mitlin, Hickey, & Bebbington, 2007). Instead of being unique and searching for local alternatives, development NGOs all over the world tend to resemble each other, to apply uniform "buzzwords" such as empowerment and participation (Cornwall, 2007), to have similar organizational forms (Tvedt, 1998), and to use shared standardized approaches in their project management (Kerr, 2008). On one hand, this similarity has been attributed to power as hegemony of managerialism (Girei, 2016) or considered to be a result of technologies of governmentality (Mueller-Hirth, 2012) within the donor-recipient relationship pointing out managerial practices that straightjacket organizations in developing countries and reproduce the asymmetric North-South power relations (Wallace, Bornstein, & Chapman, 2006). On the other hand, the apparent similarity is seen as a result of homogenizing tendencies stemming from the international organizational field that affect all development NGOs, notwithstanding their position in the aid architecture (Claeyé & Jackson, 2012;Tvedt, 1998Tvedt, , 2002Tvedt, , 2006. Although the North-South power dimension has been extensively researched (Ebrahim, 2003;Girei, 2016;Groves & Hinton, 2004;Lister, 2000), and the effects of global managerialism and mainstream project management tools as technologies of governmentality have been well-documented (Girei, 2016;Kerr, 2008) in the field of development studies, the relationship between the international field and NGOs in donor countries has received less attention.
To address this gap, we draw from the sociological studies of development organizations (Fejerskov, 2016;Watkins, Swidler, & Hannan, 2012) and organizational institutionalism (Greenwood et al., 2008;Scott, 2008), and address the dilemma between innovativeness and co-option with the concept of institutional isomorphism (DiMaggio, 1991; 688725S GOXXX10.1177/2158244016688725SAGE OpenKontinen and Onali DiMaggio & Powell, 1983;Scott, 1995). Following the central argument in organizational institutionalism (DiMaggio & Powell 1983), we contend that organizations tend to change to adopt common directions to gain legitimacy in a particular organizational field. The organizational legitimacy, in this instance, refers to organizations' alignment to cultural norms (Scott, 1995), and organizational assumptions of desirable practices within a socially constructed system (Suchman, 1995) rather than to existing laws and regulations (Edelman, 1992;Edelman & Suchman, 1997). Furthermore, we see organizational legitimacy as a consequence of interaction between macro-level institutional structures and micro-level actors (Nicholls, 2010;Powell & Colyvas, 2008).
Consequently, we understand the process of isomorphism not as determined by the field but as reflexive interaction between development NGOs and the international field (Claeyé & Jackson, 2012;Nicholls, 2010;Ramanath, 2009). We focus our examination of isomorphism and legitimacy on normative isomorphism related to professionalization of the field (DiMaggio & Powell, 1983) and so-called "professional legitimacy" constructed vis-á-vis the shared views and practices of professions (Deephouse & Suchman, 2008, p. 52). In parallel to the processes of professionalization in nonprofits in general, the legitimate expertise of development NGOs is often perceived as a technical matter, associated with mastery of common management practices and the expertise therein rather than expertise in substantial issues (Hwang & Powell, 2009;Lister, 2003;Thrandardottir, 2015). Moreover, NGOs working in a variety of countries tend to focus on top-down legitimacy for a global system rather than bottom-up legitimacy for their beneficiaries (Walton, Davies, Thrandardottir, & Keating, 2016).
The international system provides NGOs with normative ideas of appropriate goals and required project management approaches. Workshops, training, and organization development processes are the main means of spreading these ideas (Tvedt, 1998). Accordingly, trainers and external consultants have been identified as mediators between the institutional pressures and organizations (Hwang & Powell, 2009). However, how mediation takes place in actual interactions is less analyzed. Therefore, there is a need to identify ways in which institutional isomorphism is strengthened in the contexts of intentional and episodic organizational change facilitated by external change agents such as consultants (Weick & Quinn, 1999). In this article, we contribute to understanding of these micro-level processes through a self-reflection and an analysis of how our own conduct as external facilitators in an organizational intervention in support of monitoring and evaluation has contributed to isomorphic tendencies in Finnish development NGOs.
To explore the dynamics within our organizational intervention, we turned to the scholarship of realistic evaluation (Pawson, 2011;Pawson & Manzano-Santaella, 2012;Pawson & Tilley, 1997) which focuses on the context-mechanism-outcome configurations and social change mechanisms triggered in interventions (Pawson & Tilley, 1997). Realistic evaluation research can focus on organizational actors, intervention design, context, process, or outcomes (Nielsen & Abildgaard, 2013;Soininen, 2013). In this article, we scrutinize the intervention design and its implementation. To analyze the connection between intervention and isomorphism, we utilize especially the methodological notions of program mechanisms that are explanations of how intervention activities produce changes (Pawson & Manzano-Santaella, 2012).
Against this background, the main questions we address in this article are the following:

Research Question 1:
Through what kinds of program mechanisms does organizational intervention trigger change in the direction of normative isomorphism? Research Question 2: What kinds of changes within organizations have consolidated normative isomorphism and have thus contributed to the professional legitimacy vis-á-vis the organizational field?
To examine these questions, we selected three professional NGOs among the 10 Finnish development organizations that participated in an intervention called "Evaluation for Everyday Use" over a time span of 1½ years in 2008-2010. On the basis of qualitative research material including taperecorded organizational workshops and the documents produced at different stages of the project, we analyze the dynamics within these interventions processes.
In what follows, we first revisit the concept of normative isomorphism and especially its implications for development NGOs. Then, we describe our methodology, case organizations, and research material used. After that, we present our findings in regard to the analysis of program mechanisms and normative isomorphism. In conclusion, we discuss the implications of our results for understanding the mediating role of external facilitators and consultants in the interaction between an international field and organizations and the organizational legitimacy in development NGOs in particular.

Institutional Isomorphism and Legitimacy in the Context of Development NGOs
In this section, we discuss the implications of normative isomorphism and legitimacy within the field of development NGOs. We contend that isomorphic organizational change in NGOs results from a search for organizational legitimacy (Beckert, 2010;DiMaggio & Powell, 1983;Greenwood et al., 2008;Meyer & Rowan, 1977;Powell & DiMaggio, 1991;Scott, 2008) vis-á-vis the international organizational field of development NGOs (Claeyé & Jackson, 2012;Tvedt, 1998). The concept or organizational field, in general, refers to a "community of organizations that partake of a common meaning system" (Scott, 1995), and further, to the "organizations that constitute a recognized area of institutional life" (DiMaggio & Powell, 1983, p. 148). Following Tvedt (1998) and Fejerskov (2016), we contend that international development can be conceptualized as a recognized field in which organizations typically define their objective as "development of others" (Olivier de Sardan, 2005), and create common meanings across global divides through using shared "buzzwords" (Cornwall, 2007), using standardized project management practices (Kerr, 2008), and entertaining a specific resource base allocated to development aid (Tvedt, 1998). The global objectives such as sustainable development goals agreed at the United Nations (UN), the shared measures of effectiveness as defined by the Organisation for Economic Co-Operation and Development's (OECD) Development Assistance Committee (DAC), sets of joint management tools such as the logical framework approach (LFA), and overall frameworks such as the human rights based approach (HRBA) all contribute to what are appropriate goals and practices within the field.
Institutional isomorphism in its coercive, mimetic, and normative forms characterizes any field (DiMaggio & Powell 1983). Consequently, institutional isomorphism affects the direction of organizational change in individual organizations in such a way that organizations within a field tend to be homogenized (DiMaggio, 1991;DiMaggio & Powell, 1983;Scott, 1995). However, recent literature has argued that the homogenizing effects of the field do not determine organizational change, but leave room for organizational agency, translation, and negotiation (Beckert, 2010;Claeyé & Jackson, 2012) that take place in concrete micro-level social situations (Powell & Colyvas, 2008). For example, NGOs can respond to institutional pressures differently as a result of path-dependency in regard to their history and to their prevalent organizational values (Ramanath, 2009). The space is however limited, as development NGOs often depend on funding received from public development aid budgets. In this situation, the coercive pressures are often significant. For example, for the Finnish development NGOs, those pressures include Finnish laws in regard to associations and foundations, and when it comes to state aid funding allocated to NGOs through the Ministry for Foreign Affairs, they have to follow certain accountability procedures, in adherence to the State Audit Department and its strict guidelines on financial monitoring and the legal requirements posed by the Act on Discretionary Government Transfers.
In addition to reacting to coercive pressures and gaining legitimacy in legal terms within any national context, development NGOs are to a great extent subject to tendencies of normative isomorphism deriving from the international field. Consequently, their legitimacy is defined within a field that includes organizational landscapes both in developing and donor countries (Tvedt, 2002). Although the early accounts of organizational legitimacy perceived it as a survival strategy (Meyer & Rowan, 1977), the later contributions have emphasized legitimacy as alignment with the practices, perceptions, and actions assumed desirable and proper vis-ávis any socially constructed system with particular norms and values (Suchman, 1995), and results from the interaction between macrostructures and micro-level organizational actors (Nicholls, 2010). Therefore, the view of normative isomorphism and professional legitimacy (Deephouse & Suchman, 2008) as being aligned to the worldview and "professional ideals" (Hwang & Powell, 2009) is central in organizational change.
However, the definition of appropriate, professional practices within international development is challenging. The field is thematically very divided, and "professional development NGOs" can engage with anything from building sanitation systems in remote villages to offering microcredit schemes to urban women, to high-level lobbying for progressive human rights legislation in national and international arenas. Therefore, like in nonprofits in general (Hwang & Powell, 2009), the professionalization in development NGOs typically refers to expertise in management, and the institutional pressures are often directed toward the shared management practices rather than any substantial expertise in different activities (Lister, 2003). Moreover, there seems to be a tendency within development NGOs to seek legitimacy merely upward vis-á-vis the global system, rather than downwards in regard to the perceptions and experiences of those who are supposed to benefit from NGOs' projects (Walton et al., 2016), and this tendency contributes to homogenization rather than variation.
In cases of more established professional fields, the standardized practices are usually acquired in a required formal education (DiMaggio & Powell, 1983). However, there is no formal education for international development. There are a wide variety of academic development studies programs, but in principle, a person can be employed in a development NGO with (or without) any kind of education. Therefore, the professional networks, guidelines, and workshops play a significant role in creating normative isomorphism (Beckert, 2010;DiMaggio & Powell, 1983). The use of "proper practices" such as the LFA and indicator-based monitoring (Wallace, Bornstein & Chapman, 2006, pp. 32-33) occurs in training, workshops, and capacity building everywhere in the world (Claeyé & Jackson, 2012, p. 614;Kühl, 2009). The trainers and consultants function as mediators between the field and organizations (Hwang & Powell, 2009). In our case, our employer at the time, a Finnish umbrella organization, Kepa, provided training and organizational development services to Finnish development NGOs, and thus, was an important source of normative isomorphism. This was exemplified by the organizational intervention we analyze in this article, offered to the selected member organizations for free to promote the quality of the work of the Finnish development NGOs.
Although the consultants and external facilitators are mediators between field and organization, they are, similar to academic researchers, often embedded in the same normative system and rhetorical agendas (Tvedt, 2002). The literature has pointed to the fact that when consultants and organizations circulate in the same meaning systems, the possibilities of organizational interventions to produce innovative and critical ideas are delimited (Fenwick, 2003). This observation supports the argument that workshops, organizational development, and capacity building enable and support normative isomorphism. However, less is known about the micro-level processes of how this happens, and what kinds of interactions take place between consultants and organizations. To address this need for knowledge, we focus on the ways an organizational intervention supported normative isomorphism in Finnish development NGOs.

Methodology and Research Material: Analysis of Program Mechanisms in an Organizational Intervention
In this section, we describe our methodological approach we use to analyze our intervention, the original design of the organizational intervention under scrutiny, and three selected case NGOs. As the need to analyze the intervention from the perspective of isomorphism arose long after the interventions had ended, we opted for a retrospective reflection and evaluations of the process. Methodologically, our research is situated in studies of evaluation of organizational interventions (Nielsen & Abildgaard, 2013). As the analysis of our own conduct is by no means unproblematic, we searched for a methodological approach that could provide necessary analytical distance, and be something more conceptual than simple reflection on the interventionists' role typical to action research (Stringer, 2014). To be able to critically analyze our own intervention, we decided to apply the methodological tools provided by realistic evaluation (Holma & Kontinen, 2011;Pawson & Tilley, 1997;Soininen, 2013). Such evaluation typically examines the context-mechanismoutcomes configurations, and is particularly interested in mechanisms triggered by interventions in a certain context generating outcomes, in an effort to find out what kinds of interventions work for whom in which context.
As we were not primarily interested in the outcomes, but rather in the process of intervention, we draw especially from the methodological notion of program mechanisms which refers to the different ways in which the intervention triggers changes in the context (Pawson, 2011;Pawson & Tilley, 1997;Soininen, 2013). The program mechanisms are often implicit, "processual explanations of how the activities work" in producing the change in the course of an intervention (Pawson & Manzano-Santaella, 2012, p. 187;Pawson & Tilley, 1997, pp. 74-75). In our case, we set out to analyze specifically the ways in which the intervention supported those outcomes that consolidated mainstream ideas rather than initiated new and alternative ones. The mechanisms are not equal to activities such as workshops. Instead, they refer to the processes within or triggered by the activities which are not directly observable but result from a systematic analysis of the material processes of intervention (ibid.).
According to realistic evaluation, the intervention always takes place in a certain context that refers to, for example, the institutional locations, geographical space, and socially constructed understandings of the environments (Dahler-Larsen, 2001;Kazi, 2003;Pawson & Tilley, 1997). We define the intervention context in two ways. First, the context refers to the three NGOs participating in the interventions, and second, to the institutional field of international development for which these NGOs wish to gain legitimacy. Moreover, the realistic evaluation contends than an intervention, through its program mechanisms, triggers context mechanisms, "choices and capacities" which lead to change in behavior (Pawson & Tilley, 1997, p. 66). They take place in contexts in which people engage in meaning creation, action, and decision making (Astbury & Leeuw, 2010), and thus, guide the analytical attention to the micro-level interaction and action.
When it comes to a practical method to analyze program mechanisms, the realistic evaluation has been quite pluralistic tradition (Kazi, 2003). In our research, we opted for a qualitative case study (Yin, 2013) where three NGOs represented cases from a similarly structured intervention. We first considered them to be cases of the same phenomenon (Patton, 1990), namely, organizational change in a course of an organizational intervention. Second, on the basis of our long professional experience, we also considered these three NGOs as typical cases (Patton, 1990) of Finnish middlesized development NGOs with established history in development cooperation, funding from the Ministry for Foreign Affairs, and professional staff. Third, the cases were also selected on the basis of the amount and accessibility of the data, as these NGOs had voluntarily applied to be participants in the organizational development and consented to data collection during intervention processes (Yin, 2013).
The case selection poses limitations to the research. First, the NGOs were self-selected as they applied to participate in this particular organizational development offered by the umbrella organization, and second, based on our selection criteria, they were quite similar NGOs and thus, did not provide much variation. However, we contend that the strategy of selecting similar cases does not hinder us from analyzing the program mechanism, which is the main focus of this particular article. The similarity supports the examination of the typical ways through which the isomorphism was encouraged. We will next describe the intervention design and the selected case organizations in detail.

The "Evaluation for Everyday Use" Intervention
The Evaluation for Everyday Use intervention (Kontinen, 2012) was an organizational development process in support of monitoring and evaluation, offered by KEPA, an umbrella organization for some 300 Finnish NGOs engaged in international development cooperation. The intervention was initiated in 2008 to address the demand from KEPA member organizations, in the context of increasing external pressure, to improve knowledge creation systems in development NGOs. For some 20 years, KEPA had provided in-house training for its member organizations. The Evaluation for Everyday Use intervention was a new concept where facilitators from KEPA went to individual organizations to guide change processes instead of inviting individuals to be trained. The intervention implemented in 10 NGOs in total included six workshops where particular needs of the organization were analyzed. Specific developmental tasks were defined and implemented over a 1½-year time span. The novel approach was also considered to allow learning, innovation, and criticism of the mainstream demand of international development through providing space for internal debate in the organizations.
The design and implementation of the intervention was a learning process for us. Rather than implementing any clearcut organizational development model, or action research methodology, the design of the intervention was an eclectic combination of different principles. The first principle was to contextualize and historicize monitoring and evaluation activities in each individual organization. This principle was based on the notion of a zone of proximal development (Engeström, 2001) in an organizational context. The principle implied that new practices in organizations are always embedded in the current ones, with a potential to take the next steps within the zone of more advanced practices. Accordingly, it makes no sense to introduce ready-made models or practices from other organizations. Instead, it is better to identify with the staff the potential next step to be taken. The second principle was to consider tensions and contradictions in current practice as sources of organizational learning and innovation (Engeström, 2001). The third principle included the need to identify clear developmental tasks, and their joint prioritization for concrete implementation was taken from the approach of the European Foundation for Quality Management (EFQM, 2013). This approach had successfully been used by second one of us in a job related to internal planning, monitoring, and evaluation systems. The fourth principle was to hold a quite open agenda of "walking together" with the organization and flexibly addressing their needs during the intervention process. This aligned with an organizational development approach of process consultation (Schein, 1999) widely utilized in organizational development. The fifth principle was inspired by the approach of appreciative inquiry (Cooperrider & Whitney, 2001) in that it emphasized positive visions rather than problems, challenges, and tensions. All in all, our organizational intervention was based on the principles of historicity, paying attention to tensions and contradictions, as well as strengths and positive visions, and working with the organization to identify the next step in its zone of proximal development. The overall design is presented in Table 1.
Thus, the basic logic behind the intervention was that with the external facilitation, NGOs would be able to better analyze their current monitoring and evaluation practices, and come up with innovative developmental tasks and new practices that would enable them to learn better about their practices in their program countries in Africa, Asia, and Latin America, and this would result in better development projects. The first three workshops were structured in a similar way in each organization, whereas the rest were tailored depending on the requirements for accomplishing the selected developmental task.

Research Material on Three Selected NGOs
The selected case organizations were among the most professional ones with some two to four decades of history in development cooperation. They had established organizational structures, practices, and partnerships with southern NGOs. All three NGOs employed between 10 and 15 personnel at their headquarters in Finland, and their annual budgets were in the area of three million euros. They were heavily dependent on the Finnish government's aid budget, with approximately 75% to 85% of their spending originating from there. They continuously implemented development cooperation projects in Africa, Asia, and Latin America. Despite their differences in geographical and thematic focuses, their development cooperation was organized as projects implemented with partner organizations in the south. Two of the organizations had coordinators and a small office in the partner countries.
The first, the International Solidarity Foundation, was founded in the 1970s during the third-world solidarity movement. It gradually developed from a movement into a professional development NGO that implements development projects related to sustainable livelihoods and gender equality in Uganda, Somaliland, and Nicaragua. The implementation of the programs is conducted by local partner organizations Identification of strengths and weaknesses in NGO monitoring and evaluation 3 Prioritization of needs and selection of the developmental task 4-10 Working with and accomplishing the selected developmental task step-by-step Source. Kontinen (2012).
supported by a resident Finnish coordinator. The second, Interpedia, was established in 1974 mainly as an organization to coordinate international adoptions. The organization gradually started to conduct small-scale development activities and implement child-sponsorship programs. It specializes in child rights, and supports development projects in several countries in Africa, Asia, and Latin America. It does not have field staff, except in Ethiopia, and the partner organizations are responsible for project implementation. The third organization, the Finnish Refugee Council, founded in 1965, has a long tradition of working with refugees in postconflict areas as well as in Finland. Its development activities revolve around adult literacy in countries such as Uganda, Sierra Leone, Liberia, and Thailand. The organization has a Finnish coordinator in the partner countries, but relies on partner organizations for actual implementation of the development projects.
During the interventions, we continuously collected material that, as per our agreement, with the NGOs, we would be allowed to use in academic publications after the interventions. The research material for this article consisted of six audio-recorded workshop discussions in each organization. The workshop data were supported by saved email communications; NGO documentation considering planning, monitoring, and evaluation of their activities; and our own reflections during the interventions. In the analysis, we used all the material in our effort to analyze the process. The methodological notions of outcomes and program mechanisms guided our reading of the research material. In our analysis, we first reconstructed the intervention process in each organization to be able to describe the main steps taken and outcomes achieved. Second, we scrutinized the interaction between us as facilitators and the NGO staff, especially during the workshops to analyze what kinds of program mechanisms were realized. In what follows, we will first briefly describe the intervention processes and then proceed to the results concerning intervention mechanisms.

Intervention Processes and Outcomes
The structure of the intervention and the content of the first three workshops were similar in each organization. The first workshop started by introducing the project, presenting the general international debate around evaluation in development NGOs, and drawing a historical timeline of the activities of the NGO. The second workshop focused on joint analysis of organizational strengths in monitoring and evaluation, and identified possible needs for further development.
In the third workshop, these needs were revisited and prioritized, and one of the tasks was selected to be addressed in this particular project. In the subsequent workshops, the selected tasks were carried out, and thus, were somewhat different in each organization. The number of participants in the regular workshops varied between four and 10. Typically, these included project coordinators, representatives from the financial management department, program directors, and at times, the executive director of the NGO. However, in each of the NGOs at some point, a larger 1-day workshop with participation of a wider range of staff from both Finland and partner countries was arranged to further the developmental task.
The International Solidarity Foundation has its roots in the European Solidarity movement of the 1960s and 1970s. The timeline drawn in the workshop clearly indicated how its activities started by organizing demonstrations and making small gestures of solidarity such as sending hoes to Zimbabwe. After the 1980s, it expanded significantly, and started a number of development cooperation projects in Uganda, Nicaragua, and later in Somaliland with funding from the Ministry for Foreign Affairs. At the time of our organizational intervention, the NGO had a well-established, LFA-based project planning, implementation, and monitoring manual in use. It also had a development cooperation program that was divided into gender equality and decent work. The strengths and weaknesses were analyzed in a large workshop where also the staff from the program countries participated. Among 20 identified strengths were the existing program and tools, as well as good partner relationships and resources allocated for learning and development. Twenty developmental tasks identified included need for more experience sharing and learning, learning and benchmarking from other organizations, monitoring tools and plans, establishing a monitoring system at the level of organization, and defining program indicators. The latter was selected as the task for the intervention. Two program coordinators went through all the indicators in ongoing projects, and sketched a draft for program indicators on the basis of this effort.
In Interpedia, the historical analysis revealed how it started in 1974 as a small organization taking care of international adoptions in Finland, but has gradually grown into a middlesized NGO with development cooperation and advocacy work in parallel with the adoption function. The development cooperation started with small-scale projects, such as importing some local handicrafts from developing countries to be sold in Finland. After receiving increasing development cooperation funds from the Ministry for Foreign Affairs, a quite rapid expansion of development activities took place, and the number of projects implemented with partner organizations in global south grew, and they were planned and implemented in a more professional way. In the early 2000s, Interpedia started to work on a development cooperation program to polish its organizational objectives and explicate its indicators. However, the program was not finished due to some intervening issues. In the analysis of developmental tasks, three out of the identified nine were related to the unfinished program. The strengths identified included things such as willingness to learn and a clear thematic focus. In the prioritization workshop, the finalizing program and its indicators were selected for further development. The organizational process continued by facilitators' feedback on the clarity of the draft program document, comparing the new project proposals from the field with the objectives of the program, and discussing the issue in a 1-day workshop with the staff. In the workshop, it was decided that the main objective for the program was the "balanced childhood for the children of Interpedia's cooperation countries," and the program aligned with the Declaration of the Rights of the Child . It was suggested that the subobjectives and indicators could also be developed on the basis of the international standards, and not invented from scratch. During the process of revising the program document, the partner organizations were consulted, and finally, an experienced volunteer from a Nepalese partner organization came to Finland to finalize the program document on the basis of the discussions.
The Finnish Refugee Council, with over 50 years of organizational history, works with refugee-related issues both in Finland and in developing countries. In its development cooperation projects, it has focused on supporting adult literacy and conflict prevention among refugees in different countries, for example, in Sierra Leone and Liberia, by implementing individual projects with partner organizations. In the analysis of the strengths, the staff identified altogether 12 strengths, including good motivation, good personnel, and existing tools and guidelines for monitoring and evaluation. The 12 developmental tasks included practicalities, an attitude of learning and not control in regard to monitoring, and the systematization of knowledge creation at the level of the entire organization. The last point included combining the work with refugees in Finland with the development projects in developing countries. There were already many tools, but they were scattered among different parts of the organization, partners, and individuals, and bringing these together into one shared manual was selected as the main developmental task. The partners and country representatives from Liberia and Sierra Leone were asked to present and share their monitoring tools, and the draft of the guidelines was discussed in a 1-day seminar with extensive participation from the staff and the partner organizations. The manual was finalized by the program staff located in the headquarters and distributed within the organization.
In retrospect, we can identify the main outcomes of the interventions in three selected organizations as follows: (a) defining program-level indicators based on LFA logic in International Solidarity Foundation, (b) producing a polished development cooperation program with initial ideas for indicators and guidelines in Interpedia, and (c) putting together organizational guidelines for monitoring in the Finnish Refugee Council. All the outcomes reflected responses to both managerial and practical needs in monitoring, and continued already ongoing processes in the NGOs. Moreover, each organization had a different stand toward the general trend in development NGOs to move from implementing individual projects to defining organizational strategies and programs. The International Solidarity Foundation had already implemented a program-based approach, but still lacked program-level indicators for following its work. Interpedia had initiated the drafting of a program, but the process had been interrupted due to lack of personnel resources. The Finnish Refugee Council did not have a program, and at this point, did not even consider drafting one to be relevant but started to create connections between its different sectors in the developed manual. One can confidently argue that such outcomes related to organizational management tools such as program documents and indicators exemplify the consolidation of the mainstream rather than challenging the isomorphic. The interventions principle of moving into the zone of proximal development was thus realized, but the direction seemed to be toward institutional isomorphism with the international field. In what follows, we will answer the question of how this happened.

Program Mechanisms That Enabled Institutional Isomorphism
In this section, we present the findings of our analysis of the program mechanisms. We asked through what kinds of program mechanisms the organizational intervention triggered change in the direction of normative isomorphism. As a result of our analysis, we identified three such main intervention mechanisms: convincing, embedding, and consolidating. In what follows, we first describe each mechanism at turn, and then reflect on the changes triggered by comparing the three case organizations.

Convincing: Evaluation Is Important!
We named the first program mechanism identified as convincing. By "convincing," we refer to elements of the intervention that resulted in NGO staff being convinced of the importance and relevance of the intervention. Thus, the program mechanism of convincing triggered acceptance, motivation, and engagement in the NGOs. Convincing took place partly before the intervention while recruiting applications to participate from the NGO, and partly at the beginning of each intervention when motivating individual members of the staff to take part in the workshop discussions. The mechanism of convincing was enacted by the umbrella organization that provided the service, the external consultants, and by a few-or one-enthusiastic member of staff in each organization. It addressed two main features in the organizational contexts that potentially hindered organizational change: (a) the potential skeptical attitudes toward the organization's intervention itself and (b) the critical stances toward the monitoring and evaluation activities in general.
In each organization, a few members of staff were highly motivated to participate in the process, whereas others hesitated. Monitoring and evaluation, especially in its development cooperation projects, was part of the organizations' normal work, but some staff members experienced monitoring and evaluation mainly as an extra burden caused by external demands from the donors. They were partly conceived as control mechanisms and unnecessary performance that took time from the real and meaningful work. The mechanism of convincing tamed such criticism and created willingness in the participants to see monitoring and evaluation as important and worth further development.
One example of the activity that realized the mechanism of convincing was the introductory lecture during the first workshop. In that lecture, we presented the current general debates on monitoring and evaluation in development NGOs. In the presentation, the accountability, developmental, and knowledge creation motivations behind evaluation were discussed (Chelimsky & Shadish, 1997), followed by a presentation of the idea of four generations of evaluation: measurement, description, judgment, and negotiation (Guba & Lincoln, 1989). The emergence of participatory modes of evaluation in development cooperation was emphasized (Chambers, 2008), and the particular challenges in NGOs, including the culture of activism, strong value base, and learning-by-doing rather than reflecting, were presented (Edwards, 1997). Furthermore, we introduced the methodological dilemmas of constructing causal relationships, struggling with attribution, ideas of contribution analysis (Mayne, 2001), as well as the emerging trend of program theory evaluation (Chen, 1990) in development cooperation.
Even if the presentation of the recent trends was presented in a "matter of fact" style, a further analysis showed how we had quite straightforwardly created a norm according to which the "up-to-date" NGOs at international scale perceive monitoring and evaluation learning rather than accountability, discuss attribution and program theories rather than struggle with the traditional project indicators, and appreciate continuous reflection within organizations. In the presentation, for example, we used humor that seemed to implicitly downplay the "old" and opposite perspective as old-fashioned or even naïve. In all the NGOs, at the end of the first workshop and after the presentation, the participants, including those who had been hesitant, according to their immediate feedback and reflections on the later phases of the process, were at least to some extent motivated and convinced to improve monitoring and evaluation for the sake of learning.

Embedding: Situating Intervention in the Current Debates and Practices
The second identified mechanism was named embedding. This category was inspired by the literature of organizational knowledge and implementation of new technologies and models in which embedding refers to the ways in which general knowledge is translated to fit with the existing organizational practices and structures (Sydow, Lindkvist, & DeFillippi, 2004). Although the mechanism of convincing strengthened the motivation and zeal of the participants, the mechanism of embedding triggered changes in organizations' own activities by placing the general terminologies and ideas used within the current situation in the organization. The mechanism of embedding operated partly during the first, second, and third workshops. It spoke to the potential challenge of considering the organizational interventions as an irrelevant "consultant package," decoupled from the everyday challenges of the organizations. The members of staff had extensive experience of participating in different workshops and training courses. Moreover, some of the organizations had also previously used consultants to introduce new systems, or approaches providing experience of interesting workshops with little content connected to the actual work of the organization. As one of the coordinators commented, when describing his previous experience in regard to strategy development workshops, Maybe, we have different perceptions of the strategy; you say it would be beneficial for us, but it seems that the only objective is to develop rules for us, so it would not benefit us, but would force us to do certain things. (August 27, 2009) The program mechanism of embedding addressed these doubts and bridged the existing practices in the respective organizations with the general challenges in the field. The mechanism of embedding was realized particularly through activities such as using tools to analyze the existing organizational practices and to frame an agenda for discussions. The main tools included PowerPoint slides introducing the current debates concerning evaluation in general, a timeline by which the history of the organization was investigated, and a cross-table of focuses and levels of monitoring and evaluation used in the analysis of current organizational practices. Joint construction of the timeline of the flipchart showed the ways in which the organization had changed during its years of existence. In all three organizations, the timeline revealed significant growth and professionalization of development cooperation during the previous decade, with future prospects for further expansion. This observation supported the need for the organizations to gain further legitimacy as proper professional development NGOs, distancing themselves more clearly from the voluntarism and the do-it-yourself follow-up practices used in the past.
In addition, the current assessment practices, being already quite professional, were analyzed in regard to their focuses on individuals, projects, programs, organizations, and networks, and the related levels of activities, outputs, outcomes, and impacts. These dimensions directly reflected the mainstream practices and vocabularies in development organizations. For example, the levels used correspond directly with the object hierarchy of LFA. Not surprisingly, most of the existing organizational practices concentrated on monitoring of activities at the project level, and there were few practices and tools for analyzing the impact at the level of programs, organizations, and networks. The design of the analysis tool (see Figure 1) where networks and impacts were presented in the top-right box indicated this being the "most advanced" slot toward which NGOs should proceed to be legitimate.
The PowerPoint slides presented in the first introductory workshop depicted general challenges in evaluation in any field as follows: (a) objectives and strategic level, (b) indicators and operational level, and (c) utilization and learning. According to our presentation, the general evaluation challenges include vague identification of objectives, and with this lack of objectives, it becomes impossible to design indicators to measure them. Challenges presented also included the observation that evaluation results are rarely used in redirecting organizational action, and thus are not useful for learning purposes. Clearly, the joint analysis of the strengths and developmental challenges in the particular organizations that followed the presentation was inspired by the very terminology presented in these PowerPoint slides. Discussion of the challenges in each organization, not surprisingly, revolved around the unclear organizational objectives, lack of well-defined indicators, and nonexistence of organizational routines for learning from evaluation.
All in all, the mechanism of embedding included practices in which the NGO staff analyzed problems in their daily work vis-á-vis the tools and vocabularies provided by the facilitators. In this process, the uncertainties, challenges, and difficulties, as well as good practices and successes experienced by the staff were translated into the general vocabulary used for monitoring and evaluation in the field of development NGOs. Notions such as "objectives," "indicators," and "learning" were applied to organizations' own situation. Moreover, a certain structure of object hierarchy in project work and monitoring inspired the analysis of ongoing activities. In this translation process, the intervention vocabularies gained concrete meaning in every particular organization, and triggered a perception of relevance. As shown earlier, such project management vocabulary was in use in the organizations already before the intervention, and thus, embedding consolidated its relevance to experienced challenges that were not all previously articulated through this particular vocabulary.

Consolidating: Advancing Within the Mainstream
The third mechanism identified was consolidating. This mechanism continued the processes triggered by the embedding.
Although embedding encouraged the identification of needs for improvement in accordance with the mainstream vocabulary and trends, the mechanism of consolidation, for its part, strengthened the aspiration to improve the existing systems and to complete the developmental tasks identified. Consolidation refers to the ways in which the intervention strengthened proceeding along the current mainstream trends when completing the selected tasks. The mechanism of consolidating operated in two ways. First, it consolidated the selected task as an everyday activity of the organizations, and second, it consolidated the organizational change to adhering to the trends in the organization's field.
In the first regard, consolidation triggered a mechanism in which the staff started to commit time and energy to the selected tasks as part of their job description. For example, the two program coordinators in the International Solidarity Foundation spend days reviewing the existing project indicators to come up with feasible program indicators, the staff in both headquarters and field offices in the Finnish Refugee Council make a remarkable effort to collect all the existing monitoring guidelines and practices, and Interpedia invited the staff for a 1-day seminar to improve the program document and its indicators, and assigned a person from their Nepalese partner organization to complete the document. Thus, over a time span of 1 year, the developmental task identified in the intervention became part and parcel of daily work, and ceased to be an "extra" task conducted in the contexts of a particular intervention.
The intervention activities that realized this program mechanism included pushing decisions on the task selection, and structuring the work in between the three last workshops in such a way that the selected task was actually conducted. The prioritization and selection of the developmental tasks was difficult, and in each NGO, there was a lively discussion about different potentialities. Often the debate lasted until the last minute of the time allocated to the workshop, and that is when we often pushed for, and consolidated a certain decision, saying for example, "So, would it be these indicators then?" At the same time, some innovative examples such as developing video reporting instead or written reports were often quickly by-passed, often with joint laughter accompanied with comments of the impossibility of the ministry accepting such an idea. Moreover, the accomplishment of the tasks was heavily structured and scheduled by the facilitators. We volunteered to review and comment on documents or indicator drafts, and searched for additional information.
At the end of each workshop, the next steps were agreed, and partly as a result of the previous mechanism of convincing and embedding, the staff was determined and motivated to proceed, and it was ensured that the tasks discussed in the intervention workshops did not remain decoupled from the daily work. In this aspect of consolidation, our role was important. As a director of one of the NGOs stated after the process, .
. . maybe, we should put a photo of the facilitators on our wall, as a deterrent, I wish we could be as effective in other processes as well, when you were not there, our pace decreased [laughter]. (November 14, 2011) Second, consolidation included ways in which the accomplishment of the selected tasks made use of the already existing models in the international field. In practice, consolidation with the field showed in the ways the NGOs opted to move from a project to program approach and perceived indicators as the foundational tool in monitoring and evaluation, worth of improving rather than abandoning. These organizational development cooperation programs exemplified the strengthening of strategic thinking in development NGOs, and required clear definitions of the organizational mission, vision, and goals. In Interpedia and the International Solidarity Foundation, the consolidated developmental tasks revolved around the development cooperation program. As indicated in one of the workshops in Interpedia, now, we have a large number of projects that are only loosely connected, we would like to see it the other way around. It is our goal to proceed towards program-based thinking. (March 3, 2010) Solidarity remained with its existing programs but revised and further developed the indicators at the program level. In contrast, the Finnish Refugee Council abandoned the program approach at the time of the intervention as explicated by the director: "We will continue on the basis of projects . . . I do not see any reason to put the activities into a program model" (May 6, 2009). This was connected with the observation that the strategic objectives of the organization should first be defined: "As, we are to develop this monitoring system, we should define our strategic objectives first, what we want to do, and then follow those" (May 6, 2009).
A second isomorphic trend supported by the consolidation was the emphasis put on indicators as the main means of measurement. The need for clarifying organizational objectives proceeded hand in hand with that of developing indicators. Each organization used indicators in individual projects. However, there was dissatisfaction with the quality of these indicators, and with the extent to which the organizations and partners were actually able to use them as tools for monitoring.
The discussion around indicators reflected the mainstream project management view of having a baseline study that sets indicators which can then be compared with measurements at the completion of the project. The development of programlevel indicators was prioritized in the first organization whereas in the two other organizations, the actual improvement of indicators was to follow the clarification of objectives. In these endeavors, organizations not only analyzed their already existing objectives and indicators but also sought models from the international field. In one of the workshops in Interpedia, the structure of their program's objectives and indicators changed to align with the international field following the suggestion given in a group work report in one workshop: We were thinking how most of our activities come from The Convention of Child Rights . . ., the convention is based on three Ps-Protection, Participation, and Provision-so maybe these things, three things could be the main things and Interpedia's activities would be under them, then you don't have to like renew the plan all the time. (August 23, 2010) All in all, the consolidating encouraged the organizations to join the mainstream trends in international development. Rather than being critical toward the mainstream and question the entire indicator-based approach to evaluation, the intervention supported sharpening the use of the indicators, and also promoted the idea of a program approach with welldefined organizational rather than project-bound objectives.

Program Mechanisms, Outcomes, and Normative Isomorphism
The program mechanisms identified concerned mostly the ways the workshop discussions and tools used guided the process in certain directions, and motivated and encouraged the members of staff to engage in joint development of monitoring and evaluation tools rather than criticizing or dismissing them in their daily work. In this section, we revisit our second research question, and consider what kinds of consequences there were for the organizational behavior. The overview of the outcomes and the ways that the process supported normative isomorphism are presented in Table 2.
We have shown how the program mechanisms of convincing, embedding, and consolidating all enabled organizational development in the direction of normative isomorphism. As all the NGOs were already aligned with the coercive donor demands, the new improvements were merely related to strengthening their perception of being a professional organization. The mechanism of convincing drew partly from the donor demands and guidelines in regard to monitoring and evaluation, but most importantly, triggered ideas according to which a "legitimate development NGO" is the one involved in improving their monitoring and evaluation practices in a spirit of continuous learning, in contrast to the activist culture of not-so-professional NGOs who focus on action without systematic reflection. The mechanism of embedding, for its part, encouraged to label and name the current organizational dilemmas and needs with the mainstream terminology. It increased the experienced relevance of monitoring and evaluation while staff's uncertainties in regard to knowledge creation were jointly defined as needs to develop clearer objectives and better indicators to be used in the contexts of the existing object hierarchy of LFA rather than enabling the identification and naming of the challenges with a completely different vocabulary, and presenting alternative visions of change. The mechanism of consolidation supported the actual development of new material documents, practices, and tools in the mainstream direction, for example, resulting in accomplishment of tasks that previously had been left unaccomplished because they were not prioritized over other tasks.
The program mechanisms led to different outcomes in each organization. In solidarity, the outcome was to define program-level indicators based on the LFA logic already used in separate projects; in the Finnish Refugee Council, the outcome was a new guidelines and collection of tools for monitoring within the entire organization; and in Interpedia, the outcome was a development cooperation program with indicators adapted from the international child rights. In solidarity's case, opting for the program indicators was aligned with the normative trend in the field where "professional" development NGOs are supposed to have a program approach rather than implement separate projects, and furthermore, they should be able to measure the progress of a program through measurable indicators. In contrast, the Finnish Refugee Council refused to opt for a program approach but rather emphasized the mainstreaming of monitoring activities and tools in all the projects both in Finland and in developing countries. In so doing, it conformed to normative pressures in creating an organizational understanding that in a "professional organization," all activities and projects are monitored in a systematic way by using explicit tools. Interpedia, for its part, returned to its unfinished development program, and struggled with its finalization until it had the idea to utilize the goals and indicators used internationally in organizations focused on child rights. Therefore, at the end, each organization took the next step in the processes that had already been started before the intervention. None of the organizations came up with suggestions for radical change in the direction of ongoing organizational change. Although a wide variety of potential ideas were articulated in the joint analysis of strengths and needs, the final choice was made in the direction of the mainstream. Through different kinds of background material and tools designed on the basis of mainstream conceptualizations offered by external facilitators, the results of the joint analysis resonated extremely well with the international field. Thus, despite its effort to be open and innovative, our intervention did not invite nor encourage alternative or critical ideas. Had the background discussions, PowerPoint presentations, or analysis tools been totally different, or had the interventionists forbidden any "conventional" answers in the discussions, the results of the intervention and the direction of organizational change might have been different. However, the intervention process itself initiated new kinds of practices, such as inviting the partners from the global south to participate in the design of the systems, and creating spaces for organizational discussions attended by a wider variety of staff. Previously, the development of monitoring systems had been mostly the task of the coordinators in headquarters, and different sections of organizations, such as development cooperation and financial management, had worked separately without systematic joint reflection.
In the end, the organizational intervention did not result in much change in the organizational behavior in terms of the monitoring and evaluation practices and vocabularies but rather reaffirmed the already existing trends of normative isomorphism. The intervention provided encouragement and practical support regarding how to proceed with the ideas for consolidating isomorphism, rather than facilitating identification of new or radical ideas. Our self-reflections revealed that, apparently, we were very hesitant to initiate criticism of the mainstream, or to go against the standards, which would potentially hamper the legitimacy of the case NGOs vis-á-vis the organizational field, and diminish their possibilities for further funding circulating in the international development.

Conclusion
We set out to analyze through what kinds of mechanisms an organizational intervention in development NGOs strengthened their alignment with the mainstream of the international field, and what kinds of changes within organizations consolidated normative isomorphism. Inspired by realistic evaluation (Nielsen & Abildgaard, 2013;Pawson & Tilley, 1997), we identified program mechanisms of convincing, embedding, and consolidating. Through these mechanisms that occurred in overlapping phases of the process, and in a quite similar way in each organization, the intervention triggered a staff commitment to developing their monitoring and evaluation, an acknowledgment of relevance of selected developments to their work, and a determination to accomplish the developmental tasks identified. Furthermore, the mechanisms triggered change in which the NGOs improved their work in alignment with the general trends of shifting from project to program-based work, of using measurable indicators in monitoring from project to program level, and of utilizing monitoring approaches common in the field. At first glance, our results might seem self-evident. However, the findings allow a number of theoretical, methodological, and practical conclusions to be drawn. First, the analysis of the three case organizations confirmed how processes of normative isomorphism shape organizations. The cases showed how quite different organizations, which originally had engaged with mobilizing demonstrations and rallies for international solidarity, with activities of refugee empowerment, or coordination of international adoption several decades previously, were now quite similar to each other in terms of how they were organized, how they received funding from the Ministry for Foreign Affairs, and especially how they managed their development cooperation activities. The intervention did not challenge the ongoing isomorphic tendencies, but rather accelerated the isomorphic processes through strengthening the portrayal of certain kinds of legitimate development NGO, and aligning with the normative isomorphism of the international field through professional use of standardized project management tools or internationally defined objectives. Through different activities, the intervention convinced the NGO staff to follow paths aligned with isomorphic trends, embedded these trends in the experienced challenges in the everyday work of the organizations, and consolidated tasks related to these trends to be part and parcel of staff duties. The intervention, in accordance with the account of professional legitimacy, focused on professional practices of management perceived as desirable rather than on the substance of development work (Deephouse & Suchman, 2008;Hwang & Powell, 2009).
Second, our findings point to the power ambiguities related both to the facilitators-organization relationship and the field of development NGOs in general. First, our role as external change agents turned out to be that of a mediator between the organization and the field, in which we facilitated the translation of general terminologies into organizational terms (Hwang & Powell, 2009). In that role, we exercised power through selecting the information to be delivered, the tools used, and support given to accomplishment of the selected tasks, thus through providing certain kinds of knowledge and influencing its use (Gaventa & Cornwall, 2008). However, as our initial aspiration was also to encourage criticism and innovation, we argue that the most relevant power relation was not that between us over organization, but the power of international field over both the facilitators and organizations. Both were subject to similar isomorphic pressures as members of the same international field (Fenwick, 2003), and thus rather reconstructed than challenged the criteria of legitimacy. The power effect of the field was also seen in the way we privileged managerial practices, knowledge, and skills. Furthermore, their relevance was merely discussed in regard to the donor demands, and the organizational learning, compared with the few reflections on the need to show accountability toward the beneficiaries, such as the women in literacy groups in refugee camps in Sierra Leone, the members of cooperatives in Nicaragua, or the civil society organizations in Nepal. The intervention, typical to the field, thus concentrated on topdown rather than bottom-up legitimacy (Walton et al., 2016).
Third, coming back to the general question concerning the role of NGOs as either strengthening or challenging the mainstream development (Banks et al., 2015;Mitlin et al., 2007), we showed how NGOs having backgrounds in rather radical engagement in solidarity, have over the years become professionalized and mainstreamed. Although this change has been largely due to demands from their donor, the analysis showed how improvement was not only due to the demands or regulations of the donor but also stemmed from the aspiration to be legitimate "professional development NGOs," and thus conform to the normative isomorphism in the field. Efforts for legitimacy were further enabled and strengthened by the organizational intervention.
Finally, we conclude that the combination of organizational sociology and realistic evaluation enabled us to scrutinize, in a novel way, the microprocesses of strengthening normative isomorphism in the process of organizational development, considered as a practice among the most typical ways of introducing field-level ideas to organizations. Although we explicitly chose an analytical focus different from concentrating on the hegemony of managerialism (Girei, 2016), or the monitoring and evaluation as technologies of governmentality (Mueller-Hirth, 2012;Kerr, 2008), we suggest that the analysis of organizational legitimacy in combination with the phenomenon of power should receive more attention in the future. As shown in our article, the power challenges in development NGOs are not only those between Northern and Southern partners but also those between the donor NGOs and the field. More research is needed in a variety of national and geographical contexts to understand how the international field creates normative isomorphism in development NGOs of different sizes and focuses.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: This work was supported by Kone Foundation (Grant 28-1336) and the Academy of Finland (Grant 260701).