Leadership and Behavior in Humanitarian and Development: Transnational Non-Governmental Organizations

Does leadership matter in the governance of civil society organizations? In particular, do the CEOs of humanitarian and development NGOs exhibit different leadership styles and perceive their work environments in different ways as the literature suggests. To explore this question we interviewed 96 CEOs—32 from humanitarian NGOs and 64 from development NGOs. In the process we found support for the descriptions of the leadership of these two types of organizations extant in the research and practitioner literatures. Those in charge of humanitarian NGOs were more likely to challenge the constraints in their environments, to be interested in influencing what was happening, to want to affect outcomes, and to be focused on addressing the needs of those in the communities facing the crisis, disaster, or emergency. They viewed themselves as having short time in which to respond and chose to communicate and act informally as well as to only collaborate with other organizations if pushed. Providing direct aid and service were high priorities as was advocacy to secure the funding necessary for completing their task. In contrast, CEOs leading development NGOs focused more on respecting and working within the constraints of their positions, being adaptable and flexible in working on having an impact—in effect, being interested in building coalitions and achieving consensus as well as indulging in compromise with the intent of solving the endemic problems that they were there to address. They had a longer time perspective than their humanitarian counterparts and were willing to work within fairly hierarchical structures as well as with a variety of types of collaborators to reach their goals.


Introduction
Does leadership matter in the governance of civil society organizations? For all the practitioner story telling and scholarly case studies that exist in the literature, little effort has gone into systematically exploring how leadership might matter in understanding the impact of non-governmental organizations (NGOs) on the challenges that face civil society. To begin to remedy this situation, the present study explores what leaders of humanitarian relief and international development NGOs are like and how they perceive their governance challenges. How do CEOs of these two types of NGOs view the world and their organizations' place in it? We chose these two types of NGOs because of the growing attempts in these two communities to "bridge the gap between emergency humanitarian aid and long-term development aid that is essential to help people survive disasters and get back on the path to self-reliance and dignity" (Gabaudan, 2012, p. 1).
In the past decade there has been increasing examination of the "numerous conceptual, architectural, and political divides that prevent effective linkages between humanitarian and development aid" (Bennett, 2015, p.1) 1 . As crises and disasters have become more complex and costly as well as persistent, emergency aid is needed to begin with but is soon followed by the cry for longer term solutions that often call into question the behavior of the humanitarian organizations already involved. Consider such disasters and crises as what happened at the Fukushima Daiichi nuclear power plant in Japan, as an outgrowth of the Syrian civil war, to the "boat people" in the Mediterranean, and as a result of the Haiti earthquake as well as the zika virus and ebola outbreaks. Is it feasible to "create a shared space where both humanitarian and development actors can co-exist and apply different approaches and tools to address the range of problems that protracted crises entail" (Bennett, 2015, p. 1)? To answer this question, we need to know more about the leadership of these two types of NGOs. Indeed, the question postulates that the leaders of these two types of organizations deal with problems using different approaches and tools. Supposedly humanitarian relief organizations have a short-term focus on saving lives and providing goods and services allowing those involved to deal with the immediate aftermath of a disaster or crisis. They are responding to the event and the people affected by the event. They are not responsible for dealing with the underlying causes of what is happening and they do not have to interact with government officials in the process, particularly if such government officials are part of the problem. In contrast, development organizations are focused on dealing with the underlying problems be they poverty alleviation, postconflict reconstruction, or institution building. And these organizations have to work with the local and national governments and citizens in accomplishing such goals.
The leadership literature would suggest that the leaders of humanitarian-focused NGOs would be interested in gaining as much control over the situations in which they find themselves as possible, given that they are risking their own security in the process (Bass & Bass, 2008;Elgie, 2015;Fiedler & Garcia, 1987;'t Hart, 2014). Moreover, we have learned in crisis situations there is contraction of authority to the top-to that leadership able to deal with the situation and meet the needs of the people and communities in the throes of the disaster (Boin, 't Hart, Stern, & Sundelius, 2005). There is need for a quick decision and flexibility in responding to whatever situations should arise. Indeed, there is not time to build consensus or do the work entailed in developing collaboration. As Knox Clarke (2013, pp. 18-19) has observed, with time of the es-sence "relatively autocratic decision making is most appropriate". Taking the initiative and being entrepreneurial are traits to be prized. As a head of a humanitarian organization is quoted as saying: "My greatest advantage was that I did not come from a bureaucracy" (Knox Clarke, 2013, p. 20)-such structures in a humanitarian setting are considered an anathema to initiative-taking and 'free thinking'.
Development-focused NGOs, in contrast, are built around compromise and consensus. At its most fundamental, development involves reducing material want and enhancing people's ability to live a life they consider good and to do so for the broadest range of a population. It is a long-term process and often includes community organizing, poverty reduction strategies, and new forms of microfinance (Lewis & Kanji, 2009). In essence, the ultimate goal of development-oriented NGOs is to put themselves out of business in a particular setting-to facilitate marginalized groups gaining self sufficiency. Building collaborations and empowering others becomes the 'name of the game'. Such efforts are time consuming and generally depend on the receptivity of others. Leadership is viewed as a partnership with stakeholders, as shared, allowing for more diversity of views and perspectives, interest in confronting disagreements, and the building of consensus. Leaders in such settings operate "primarily as facilitators", focusing on group process and insuring participation and accountability (Knox Clarke, 2013, p. 66).
These views of the leadership styles envisioned for those leading humanitarian NGOs and those in charge of development NGOs are in stark contrast with one another. Do leaders in these two sectors, in fact, exhibit such different orientations to their work? Do they perceive their work environments as being as different as the literature suggests? In what follows, we will explore this question based on interviews with 96 CEOs of transnational NGOs engaged in humanitarian relief and international development. We are interested in gaining their perspectives on their leadership styles and their definitions of the environments in which they work.

Method
This study attempts to move beyond more traditional research on transnational NGOs which has generally centered around a particular sector, organization, or issue campaign, often involving case studies or small focus groups composed of organizational leaders. The data here come from face-to-face, in-depth interviews with 96 CEOs of transnational NGOs working on either humanitarian relief or international development issues with operations in multiple countries. The interviews were conducted as part of a National Science Foundation study (Grant No. SES-0527679) that focused on understanding the governance and leadership of transnational NGOs. The interviews lasted on aver-age an hour and a half and focused on the challenges these leaders saw their organizations as facing in a globalizing world. All participants were guaranteed confidentiality and interviews were digitally recorded and transcribed after the fact. For a description of the study in more detail, see Transnational NGO Initiative (2010). The term "transnational" connotes that an organization has sustained relations with other societal actors that cross borders and boundaries.
It is important to note that the particular NGOs and CEOs in the study were selected from the Charity Navigator 2 database and from among organizations with activities in multiple countries. The particular sample that was selected was from the resulting set of NGOs and was chosen so as to be representative of these organizations with regard to sector, size, and fiscal health. The Charity Navigator database was used because it offers a structured comparison of transnationally engaged NGOs across a variety of sectors along with size, efficiency, and capacity indicators for each. This selection procedure resulted in 32 transnational NGOs doing humanitarian work and 64 engaged in development. All these transnational NGOs were registered in the US so as to gain Internal Revenue Service 501(c)3 tax exempt status and access to US government funding such as from USAID as well as charitable donations from US citizens and foundations.
The humanitarian organizations in this study focused on crises and disasters and immediate aid. Many had relief in their names. The development organizations focused on endemic problems like hunger, poverty, gender, education, environment, and health issues. The interview protocol asked questions about personal and organizational attributes; organizational goals, strategies, and activities; organizational effectiveness focused around particular incidents where the CEOs viewed their organization had been effective; communication concerns; networks and partnerships; and leadership and professional engagement.
The interview protocols were content analyzed using ATLAS.ti, a software system that facilitates finding common themes in such material (see ATLAS.ti, 2016). The interview protocols were also content analyzed using the Leadership Trait Analysis software located on the Profiler Plus platform (see Social Science Automation, 2016). Whereas ATLAS.ti assists researchers in finding common themes in the material, the Leadership Trait Analysis (LTA) assesses interview protocols for indications of seven traits frequently associated with leaders and leadership in the research literature. With ATLAS.ti, we were interested in how the CEOs described their organizations, the functions they viewed their organization as serving, their assessments of the organization's effectiveness and accountability as well as their discussions regarding collaboration and partnerships (Transnational NGO Initiative, 2010). In the current study, we were interested in four of the LTA traits: ability to control events, need for power, sensitivity to contextual information, and focus on solving problems versus insuring the inclusiveness of others (see Hermann, 2005Hermann, , 2009.
In both types of content analysis, an assumption is made that the more frequently leaders use certain words and phrases in their interview responses, the more salient such content is to them. Of interest is how much the CEOs focus on a particular idea and description with regard to ATLAS.ti. In LTA, we are interested in the percentage of time leaders could use certain words and phrases that they, indeed, do. The use of computer software reduced coder bias and insured consistency in the coding.

Leadership Style
Leadership style gives us hints regarding how leaders of transnational NGOs are likely to interact with stakeholders, donors, and those around them-how they are likely to structure the decision-making process, from whom they will seek advice, and the kinds of contexts they are likely to prefer (e.g., Kille, 2006;Kowert, 2002;Lecy, Mitchell, & Schmitz, 2012;Mitchell, 2005). In other words, leadership style suggests how important it is to leaders to exert control and influence over the environments in which they find themselves and the constraints that those environments pose as opposed to being responsive to the situation and working with the demands of stakeholders and donors. We focus here on four aspects of leadership style. Data for all four were available as a result of the LTA content analysis of our interviews with the CEOs.
The first aspect is often referred to as locus of control or belief in one's ability to control what happens. How much control do leaders perceive they have over the situations in which they find themselves; how likely is it that individuals and organizations can influence what happens? Based on the theoretical work of Rotter (1993), his locus of control personality inventory, and extended research using the inventory, the focus of the LTA content analysis is on verbs or action words. An assumption is made that when leaders take responsibility for planning or initiating an action, they believe that they have some control over what happens. The focus here is on actions proposed or taken by the CEO or his/her organization as discussed in the interview. The score on this trait is determined by calculating the percentage of times the verbs in an interview response indicated that the speaker or the organization he/she leads took responsibility for planning or initiating an action. The overall score is the average percentage across the interview-ee's answers to the questions raised in the interview.
The second aspect of leadership style that we examined is the need for power. As Winter (2005a) has observed, this is the desire to influence or have an impact on other persons or groups. Is the speaker attempting with the proposed action to establish, maintain, or restore his/her influence. As with the previous trait, coding for need for power and influence focuses on verbs. Is the speaker attempting with this proposed action to establish, maintain, or restore his or her power? Some of the conditions where need for power is scored are when the speaker (1) proposes or engages in a strong, forceful action such as a verbal threat, an accusation, or a reprimand; (2) gives advice or assistance when it is not solicited; (3) attempts to regulate the behavior of another person or group; (4) tries to persuade or argue with someone else so long as the concern is not to reach agreement or avoid disagreement; (5) endeavors to impress or gain fame with an action; and (6) is concerned with his or her reputation or position. Once again the focus is on verbs or actions proposed or taken by the leader or a group with whom he or she identifies. A score is determined by calculating the percentage of times the verbs in an interview response indicate that the speaker-or a group with whom the speaker identifies-has engaged in one of these behaviors. The overall score for any leader is the average percentage across the interviewee's answers to the questions in the interview.
The third aspect of leadership style studied here explores how sensitive to contextual information the leader is. Does the leader have a sense of what needs doing in the context or is he/she interested in understanding the nature of the situation before acting? This trait builds on the work of Suedfeld (see Suedfeld, Guttieri, & Tetlock, 2005) on integrative complexity. We are interested here in the leader's use of conditional words and phrases versus those that are more black and white in nature. Consider the difference between such words as 'approximately', 'for example', 'possibly', 'it depends' and words like 'absolutely', 'all', 'certainty', 'irreversible'. Dictionaries of such words have been developed based on thesauruses from around the world. Scores on this trait are the percentage of conditional words to the total number of conditional plus absolute words in a particular interview response. The overall score for any leader is the average percentage across the interviewee's answers to the questions in the interview.
The fourth aspect of leadership style focuses on how much a particular leader focuses on solving problems versus on developing collaborative relationships in a group or organizational setting. This distinction forms the basis for Fiedler's Least Preferred Coworker Scale and his contingency model of leadership which explores in what contexts leaders interested in solving problems are more effective and in which those interested in collaboration and inclusiveness are more effective (Fiedler & Garcia, 1987). Here again the focus in LTA is on words. Consider, for example, words such as 'accomplishment', 'plan', proposal', and 'recommendation' which are more focused on problem solving and accomplishing the task at hand. In contrast, words such as 'consensus building', 'negotiation', 'identity', 'concern', and 'colleagues' are more focused on building relationships and inclusiveness. The score for this aspect of style is determined by calculating the percentage of words focused on problem solving relative to the total number of problem-solving versus relationship-building words in a particular interview response. The overall score is the average percentage across all interview responses. Table 1 shows the results of the LTA analysis for the CEOs of humanitarian and development organizations regarding their perception of their ability to control what happens. Scores were divided at the median to determine what was low and what high. Those CEOs who had a score at the median were categorized so as to make the N in each category as close to 50% of the total as feasible. The data in Table 1 indicate that there is a significant relationship between sector and the CEOs perception regarding control. CEOs of humanitarian NGOs are more focused on controlling what happens while those in development NGOs are relatively low in this concern.

Ability to Control What Happens
The literature suggests that leaders who believe that they can influence what happens are generally more interested and active in the decision-making process (see, e.g., Hermann, 2014;Hermann & Gerard, 2009;Keller, 2005;Kille, 2006). They want to maintain control over both decision making and implementation to insure that things happen. In many instances, just like the people they are helping, their lives, too, are often on the line. Moreover, as in any crisis situation, they are accountable for what happens as the citizenry turns to them for help. There is not time to involve many others; the short decision time requires action. Taking the initiative and being entrepreneurial must be built into the situation if anything is going to happen.
In contrast, leaders who are low in belief that they can control what happens, like the development CEOs, tend to be more reactive to situations, more willing to empower others to participate. Indeed, they are comfortable working in situations where they can foster collaboration; they have no desire to be in control nor do they see the benefit in it. They are interested in partnerships where all are held accountable and rise or fall together. Such leaders want to participate and lead in contexts where they perceive there is at least a 50% chance of success. The data show a match to the differences in behavior that the literature suggests exist in these two NGO sectors. Table 2 shows the results of the LTA analysis for the CEOs of humanitarian and development organizations regarding their need or desire for power and influence. Scores on this trait were divided at the median to determine what was low and what high. Those CEOs who had a score that fell on the median were categorized so as to make the N in each category as close to 50% of the total as feasible. The data in Table 2 show that the CEOs of humanitarian and development NGOs are mirror images of one another when it comes to need for power. This relationship approaches significance. CEOs of humanitarian NGOs are roughly 60-40 more likely to have a high need for power while CEOs of development NGOs are roughly 60-40 likely to be low in need for power.

Need for Power
The literature suggests that leaders with a high need for power-like the majority of the humanitarian CEOs-work to facilitate having power and influence in their environments and to appear a winner (e.g., Winter, 2005aWinter, , 2005bWinter, , 2010. They are good at sizing up situations and sensing what tactics will work to achieve their goals. Indeed, they can be highly skillful in behind the scenes negotiations. Moreover, they are generally daring and charming-the charismatic leader. But such leaders are likely to set up rules to ensure conformity to their ideas-rules that can change abruptly if the leader's goals or interests change. Indeed, leaders high in need for power often test the limits before adhering to a course of action, bartering and bargaining up until the last moment in order to see what is possible and what the consequences will be of pushing further toward their goals. These leaders are more skillful in such negotiations when they can interact directly with those involved; without face-to-face interaction, such leaders can misjudge the assumptions the other party is making and how far they are willing to go.
When need for power is low, as it tends to be for the majority of the development CEOs studied here, leaders have less need to be in charge; they can be one among several who have influence. It is perfectly okay with them that others receive credit for what happens. Indeed, empowering others is important for such a leader. These leaders are willing to sacrifice their own interests for those of the group since in their view what is good for the group is, in truth, good for them. In effect, they become agents for the group, representing the group's needs and interests in policymaking. And, in turn, such leaders share responsibility and accountability with other members of the "team" for what happens. Table 3 shows the results of the LTA analysis for the CEOs of humanitarian and development organizations for sensitivity to contextual information. Scores on this trait were divided at the median to determine what was low and what high. And, as before, those CEOs who had a score that fell on the median were categorized so as to make the N in each category as close to 50% of the total as feasible. The data in Table 3 show that CEOs in the humanitarian sector are a little more likely to be sensitive to contextual information than those in the development sector but the relationship is not significant. This leadership trait does not differentiate between the two types of CEOs. Table 4 shows the results of the LTA analysis for the CEOs of humanitarian and development organizations regarding their orientation to their stakeholders. Are they  more focused on solving the problems facing such stakeholders or are they more interested in insuring that their stakeholders' concerns are taken into consideration in what happens? Scores on this trait were divided at the median to determine who showed evidence of a focus on problem solving and who on stakeholder concerns. Those CEOs who had a score that fell on the median were categorized so as to make the N in each category as close to 50% of the total as feasible.

Orientation to Stakeholders
The data in Table 4 approach significance and suggest that while the CEOs of humanitarian organizations are about equally split in their focus on stakeholder concerns and solving problems, the CEOs of development NGOs are more likely to be oriented toward solving the problems of their stakeholders. The literature on this aspect of leadership style suggests that for leaders who emphasize the problem, moving a set of stakeholders toward a goal is their principal purpose for being in the setting whereas for those who emphasize the concerns of stakeholders, establishing and maintaining relationships as well as keeping the loyalty and morale of those stakeholders high are the central functions of leadership (see, e.g., Bass & Bass, 2008;'t Hart, 2014;Hermann, 2014). It appears that the CEOs of humanitarian NGOs are almost as likely to be focused on the morale of those in their care as solving the problems that gave rise to the crisis or disaster in the first place. That is not the case for CEOs of development NGOs. They are almost two-thirds as likely to focus on problem solving as relationships with stakeholders. Given the types of problems development organizations tackle, such an approach may make sense. But to solve such problems, they are going to have to enlist the aid and trust of the stakeholders which does not seem their first concern. Perhaps without a plan of action, it is impossible to bring the stakeholders along.

CEO Background
What else can we learn about the leaders of humanitarian and development organizations from their responses regarding the challenges that they face? Are the differences described in the NGO literature regarding these two types of organizations evident in the leaders who run them? For this analysis we are going to examine the leadership style variable that differentiated them most clearly, that of belief in their ability to control what happens. Examining only those CEOs high in this belief for the humanitarian relief organizations (N = 23) and only those low in this belief for the development organizations (N = 42), it is possible to explore what is associated with each of these leadership styles-the differences in background among the leaders, how they perceive their organizations to operate, and the challenges they believe they face. We will start this analysis by examining their backgrounds, notably experience and education. Table 5 shows background factors for the CEOs low and high in belief that they can control what happens, who represent the leadership of development and humanitarian relief organizations respectively. Interestingly, there was no real difference between the CEOs of organizations from these two sectors with regard to experience. Roughly 50% were in their first ten years as leader of the organization. Knox Clarke (2013) has commented about the high burn out rate in those leading humanitarian organizations. And, yet, the data show that around 40% of the leaders in both sectors have been in their positions for more than a decade.
There is a significant relationship between level of education and belief in control. Although there is close to a 50-50 distribution of CEOs with a college degree versus having an additional professional degree among those leading humanitarian organizations, almost three-quarters of those leading development organizations had advanced degrees. Do the problems being dealt with by CEOs in the development sector require more education; do these organizations attract persons with more education; or do those involved in the development sector return for more education in order to try to better understand endemic problems? These are questions worthy of further study. We do know that the CEOs whose data are reflected in Table 5 from both types of NGOs came to their positions from other civil society organizations or from the public sector (around 70% for each group) and were brought into the organi-zation to lead it rather than rising to their positions from within (some 85% for each group). These data suggest that those selected as CEOs came with an already formed belief regarding how much control was necessary in the particular setting rather than gaining it after they arrived in their leadership position.

Perceptions Regarding Organizational Structure and Function
Although Table 6 indicates that development-focused and humanitarian-oriented NGOs are about equally likely to see providing direct aid and services as a major goal of their organizations, there do appear to be some significant differences in perceived function and structure between the CEOs of these two types of NGOs. Interestingly a little over one-half of the CEOs of both types of organizations do not view advocacy as a goal on which they are focused. For 17%, however, of the humanitarian CEOs interested in controlling what happens, it is a primary goal. Such behavior may be a need because both types of NGOs view securing funding as an obstacle to achieving their goals, but humanitarian organizations view finding funding as an even more severe obstacle than do those running development organizations. Almost all of the CEOs of humanitarian NGOs viewed securing funding as an obstacle whereas only two-thirds of those heading up development NGOs viewed it as such. Securing funding for emergency operations may be a cause for frustration but wanting to be able to control what happens may make problems surrounding finding funding even more frustrating, particularly since some funding is needed up front to get relief aid started.
Part of the frustration of the humanitarian CEOs may come from the short time that they perceive they have in which to engage in action. Over twothirds of the CEOs of humanitarian NGOs with their interest in controlling what happens perceived their time frame for action was short. Such was not the case for those in charge of development NGOs. Some 77% of them viewed the time frame as long-term or a combination of short and long-term depending on the nature of the particular situation. The latter group has less interest in controlling what happens and, perhaps, more opportunity to be flexible in the situation than those operating in a crisis, disaster, or emergency setting.
Not only did the CEOs of humanitarian NGOs believe themselves operating under a short-term time frame, they prefer to do so informally and to be part of organizations that are primarily non-hierarchical in nature. In their minds, it is easier to control and exert influence over what is happening if communication is kept informal and those necessarily part of decision making are easily interacted with so that problems can be met quickly and decisively in situations that are constantly changing. The CEOs leading development NGOs prefer more formal methods of communication as well as to operate in hierarchical organizations. In such organizations not believing that one has to control over what happens may pay off-it may facilitate participating on teams and building the kinds of collaborations often demanded in working in development-focused settings.
At issue here is whether CEOs with the particular predispositions identified here were hired because their leadership styles were a fit to the organizational culture or the CEOs sought out the organizations that matched their styles. In the political arena, researchers have found that leaders opt to run for positions that are compatible with their ways of exercising leadership and that they have more appeal to constituents with similar preferences for a particular leadership style (for a review, see Hermann, 2014).

Perceptions Regarding Collaboration
In the course of the interviews with the CEOs, they were asked to talk about collaboration and with whom they were likely to collaborate. Table 7 presents the results for the CEOs from humanitarian relief organizations who were interested in controlling what happens and those leading development organizations with a predisposition to work within the system as leaders, not having to be in control of what happened.
Interestingly, for all but other NGOs, 50% or more of the CEOs leading humanitarian NGOs did not view these other types of organizations as worthy collaborators-that is, a majority or more were not inclined to collaborate with international organizations (IOs), corporations, or governments. Indeed, they often talked about such collaborations as being forced upon them and as a constraint they tried hard to avoid rather than welcome. Such collaborations were viewed as hindering rather than helping them achieve their goals. This view was almost 100% for collaboration with IOs who bore the brunt of this concern for the humanitarian relief CEOs. They perceived that they were often forced to collaborate with IOs when such organizations were put in charge of coordinating international activities. In these leaders' minds, with such coordination, they had less control than usual over what was happening which affected their ability to do their tasks in a timely fashion. The exact opposite was the case for the CEOs of development NGOs who appear to have welcomed collaboration. Fifty percent or more viewed their organizations as being quite willing to collaborate with and across these various types of institutions and as benefiting from such collaborations. The smallest percentage-right at 50%-occurred for IOs. These CEOs wondered how an organization could call itself effective in the development arena without collaborating across the institutions also involved in development. To achieve one's goals meant engaging the resources of these other types of institutions and building partnerships with them.

Conclusions
This paper began by asking if leadership mattered in the governance of civil society organizations. In particular, we were interested in the leaders of two different types of organizations which the literature suggests exhibit different leadership styles and perceive their work environments in very different ways. Indeed, there is growing interest in "bridging the gap" between these two communities: those organizations engaged in humanitarian relief and those organizations tackling the endemic problems involved in international development. We were interested in studying if the leadership styles envisioned in the literature for those leading humanitarian NGOs and those in charge of development NGOs were as starkly different as the literature portrayed them. To explore this question we talked with 96 CEOs representing these organizations-32 leading humanitarian-oriented NGOs and 64 leaders of development-focused NGOs. We systematically interviewed these leaders to learn more about what they were like and their views of their organizations' challenges. In the process we promised the leaders anonymity so we cannot indicate the particular leaders interviewed nor the organizations they represented. Those in charge of humanitarian NGOs were more likely to challenge the constraints in their environments, in effect, to be interested in controlling what was happening, to want to influence the outcome, and to be focused on addressing the needs of those facing the crisis, disaster, or emergency. They viewed themselves as having a short time in which to respond and, thus, chose to communicate and act informally as well as to only collaborate with other organizations if pushed. Providing direct aid and service were high priorities as was advocacy to secure the funding necessary for completing their task. In contrast, CEOs leading development-oriented NGOs focused more on respecting and working within the constraints of their positions and the settings in which they found themselves-in effect, being interested in building coalitions and consensus as well as indulging in compromise with the intent of solving the endemic problems that they were there to address. They had a longer term time perspective than their humanitarian counterparts and were willing to work within fairly hierarchical structures as well as with a variety of types of collaborators to reach their goals.
Considering the contexts in which these two types of organization operate, this difference in style seems almost self-evident. In fact, in the interviews the CEOs of humanitarian NGOs talked about usually being the first on the ground in crisis situations, engaging quickly in organizing the setting in order to help those affected. They generally were looking for immediate impact and for asserting control over what is often a chaotic environment. The very term 'development', however, suggests being in for the long-term, as those leading development NGOs argued, and working within the environment on the ground to help those involved to both define what they want to see happen and to work toward such goals. When they find themselves in the same setting, the CEOs of these two types of organizations often see themselves as working at cross purposes, although-as we just learned-three-quarters of them view collaboration with other NGOs as relevant to achieving their missions There is an increasing cry from those monitoring the humanitarian and development sectors for leaders to bridge the gap between short-term emergency aid and long-term development aid to help "people survive disasters and get back on the path to self-reliance and dignity" (Gabaudan, 2012, p. 1; see also Bennett, 2015;Knox Clarke, 2013). We wondered if the two groups of leaders that we did not focus on in this study might be helpful in bridging this gap. The 9 (or 28%) of those leading humanitarian NGOs who were low in their interest in controlling what happens and the 22 (or 34%) of those heading development NGOs who were just the opposite-interested in controlling what happens. Each of these groups was deviant from the norm for the kind of NGO that they led. These are the leaders that we did not look at in examining background, perceptions of structure and function, or interest in collaboration.
An examination of their interviews and organizational missions suggests that these CEOs already are involved in playing a bridging role. The nine CEOs leading humanitarian NGOs with low interest in controlling what happens appear to be part of organizations that identify areas where development is needed through providing relief in crisis and disaster situations. In contrast, the 22 CEOs heading up development NGOs with high interest in controlling what happens assist in crises in areas in which they already are doing development work. As a result, the nine leading humanitarian organizations plan to spend time in the area once the immediate crisis is attended to and are interested in continued interaction and in building collaborationsin working with the development NGOs. And the 22 CEOs leading development organizations feel the need to get out in front in crisis situations to insure their ongoing efforts at development do not get compromised during the emergency. Both sets of leaders are interested in working with the leaders of the other type of organization and show evidence of the leadership style more prominent in that other kind of NGO. They have reason to act as facilitators between the two types of organizations as their leadership styles 'match' those of the CEOs in the opposite sector.
In essence, leadership does appear to matter in the governance of civil society organizations. At least CEOs of humanitarian and development NGOs perceive some differences in their work environments and show evidence of different leadership styles. But we have just started to systematically study such leaders. There is more to do including exploring if CEOs with deviant leadership styles, as we observed here, can facilitate bridging the gap between the two roles often present in crisis situations, that of dealing with the immediate situation and that of working on underlying problems.