Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=lpad20 Download by: [The UC San Diego Library] Date: 15 February 2017, At: 11:37 International Journal of Public Administration ISSN: 0190-0692 (Print) 1532-4265 (Online) Journal homepage: http://www.tandfonline.com/loi/lpad20 Collaborative Crisis Management and Leadership in the Public Sector Naim Kapucu & Yusuf Ustun To cite this article: Naim Kapucu & Yusuf Ustun (2017): Collaborative Crisis Management and Leadership in the Public Sector, International Journal of Public Administration, DOI: 10.1080/01900692.2017.1280819 To link to this article: http://dx.doi.org/10.1080/01900692.2017.1280819 Published online: 07 Feb 2017. Submit your article to this journal Article views: 24 View related articles View Crossmark data
Transcript
Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=lpad20
Download by: [The UC San Diego Library] Date: 15 February 2017, At: 11:37
Collaborative Crisis Management and Leadershipin the Public Sector
Naim Kapucu & Yusuf Ustun
To cite this article: Naim Kapucu & Yusuf Ustun (2017): Collaborative Crisis Managementand Leadership in the Public Sector, International Journal of Public Administration, DOI:10.1080/01900692.2017.1280819
To link to this article: http://dx.doi.org/10.1080/01900692.2017.1280819
Collaborative Crisis Management and Leadership in the Public SectorNaim Kapucua and Yusuf Ustunb
aSchool of Public Administration, University of Central Florida, Orlando, FL, USA; bMinistry of the Interior, Ankara, Turkey
ABSTRACTThe goal of this article is to discover how leadership competencies affect the perceived effective-ness of crisis management. The study, based on a self-reported survey of executive public leadersin Turkey, found that the core leadership competencies have a positive relationship with theeffectiveness of crisis management. Among task–oriented, people–oriented, and organization-oriented categories of leadership behaviors, task-oriented leadership behaviors were found withthe highest level of impact on the effectiveness of crisis management. The study demonstratedthe importance of the core leadership competencies in the effectiveness of crisis leadership. Thehypothesis testing with the covariance structure model supported the positive impact of the coreleadership competencies on the effectiveness of crisis management. This study contributes to theliterature on leadership during crisis situations, and also provides proposals for public managersand practitioners to increase their effectiveness in leading their organizations during crises.
Various types of crises impact the operations of organiza-tions, from small local nonprofit organizations to interna-tional agencies, and even governments. The numerouscrises varying in size, duration, and complexity haveincreased the importance of leadership in managing them.More comprehensive and professional preparation forlarge-scale crisis management is considered to be one ofthe primary objectives of public management in providingsecurity for its citizens (Farazmand, 2007; Heller, 2012;Kapucu & VanWart, 2006). The public now expects effec-tive public sector leadership in crises more than they did inthe past (Ink, 2006). The lack of leadership skillsmay lead toinadequate crisis management, whichmay cause loss of lifeand property (Murphy & Dunn, 2012). There are manyexamples of these all over the world such as well-knownHurricane Katrina in 2005 in the United States. Theresponse to Hurricane Katrina was an example of poorcrisis management caused by lack of the public leadership(Farazmand, 2007).
Managing crisis demands inter-organizational colla-boration and collaborative leadership skills. In Turkey,observed shortcomings of collaboration in response tocrises prompted a reorganization of the disaster manage-ment system for a centralized system (Hemansson, 2016). Aleader’s capacity is best tested in a crisis situation (Klann,2003). Crises creates sensitive environments in which
leaders may have to make sudden and effective decisionsusing limited information. In these kinds of environment,emotions and instincts may quite easily override intellectand logic. In order to reduce the impact of these challengingevents, every competent leader must take a number ofactions prior to, during, and after crises.
The study focuses on public administrators’ leadershiprole in crisis management. The literature, including theUnited Nations (UN) standards, recognizes natural andman-made crisis contexts. In this study, both man-madeand natural disasters are included. The study examines thefollowing research questions to better understand the roleof effective crisis leadership in dealing with crises. What isthe role of effective leadership in managing crises anddisasters? How do the public administrator’s leadershiptraits and skills impact the effectiveness of crisis leadership?How do the public administrator’s leadership behaviors(task, people, and organization-oriented behaviors) influ-ence the effectiveness of a crisis management?
The study builds on and contributes to earlier stu-dies on collaborative leadership in crisis managementin the public sector. Although earlier studies haveexamined collaboration in crisis management, theydid not pay much attention to collaborative leadershipat the time of crisis in the public sector. Therefore, thisresearch provides valuable knowledge for scholars andpractitioners in understanding the importance of theleadership competencies to accomplish effective crisis
CONTACT Naim Kapucu [email protected] School of Public Administration, University of Central Florida, HPA II Suite 238B, Orlando, FL 32816-1395,USAColor versions of one or more of the figures in the article can be found online at www.tandfonline.com/lpad.
INTERNATIONAL JOURNAL OF PUBLIC ADMINISTRATIONhttp://dx.doi.org/10.1080/01900692.2017.1280819
management in the public sector, especially during theresponse phase of a crisis. Before establishing the con-ceptual framework of this study, a brief overview of theTurkish administrative system is provided in the nextsection.
Context of the study
Turkey experiences different kinds of natural and man-made disasters frequently. Roughly 92% of the land ofTurkey is a potential earthquake area. Turkey has lostthousands of citizens in earthquakes and other naturaldisasters within the last two decades (Hemansson, 2016;Unlu, Kapucu, & Sahin, 2010). The legal framework isalso of vital importance so as to better understand theTurkey context.
Turkish constitution indicates that, based on the“devolution of wider powers” principle, the centraladministrative structure of Turkey is divided into pro-vinces and other lower administrative levels in terms ofthe geographical location and economic conditions, andaccording to the requirements of public service. Turkeyhas 81 provinces, and 919 districts under these provinces(Kapucu & Palabiyik, 2008). The “devolution of widerpower” principle gives some of the powers of the centralgovernment to provincial organizations; therefore, itbecomes possible to carry out the central government’sauthority in the provinces by means of the governors whoare the representatives of the central governments. Thegovernors (vali), the highest public officials in a province,canmake decisions on some issues determined by law andexecute those decisions by their own authority. The pro-vinces are subdivided into districts, headed by a districtgovernor (kaymakam). In the Turkish administrative andcrisis management system, the province, and district gov-ernors are the principal responsible public administratorsto manage any type of crisis that occurs in their provinceor district jurisdiction (Kapucu, 2010).
A leadership competency framework: leadershipaction cycle model
Collaborative leadership focuses on the administrators’behaviors which facilitate a productive interaction andmobilize network participants to find effective solutionsfor problems (McGuire & Silvia, 2009). It is expected fromleaders that they openly express their ideas, inspire peopleto mobilize, and concentrate on problems and results.However, collaborative leadership requires a differentstyle of leadership. In this approach, the leader mustguarantee and guard the progression of collaboration,ease of interaction, and struggle patiently with frustra-tions that may arise during the functioning of
collaboration. Collaborative leaders guide rather thancontrol and concentrate on motivating rather than direct-ing network participants (Carter, 2006).
The term “competence” is generally understood asincluding the education, knowledge, abilities, and experi-ence of individuals who use themwhile performing a task.When assessing competencies for a position, there shouldbe a connection between the competencies and the suc-cessful way a specific mission should be performed(Moore & Rudd, 2004). Competencies are developedbased on a leadership experience, and literature are usedincreasingly and commonly in the public and privatesectors. For instance, the results of research conductedby Bolden, Gosling, Marturano, and Dennison (2003)indicate that 29 different competency frameworks havebeen used by companies such as Lufthansa and Shell fromthe private sector and organizations such as the SeniorCivil Service from the public sector in Turkey.
The leadership action cycle model, developed by VanWart (2004), is a leadership competency framework thatconcentrates on public sector leadership and can be usedfor all levels of government. As a multidimensional lea-dership model, it integrates many leadership researchtrends developed by scholars in public administration.The model was established based on three types of leader-ship competencies: traits, skills, and behaviors. In themodel, a leader is supposed to possess inborn traits andlearned skills. According to the model, leadership beha-vior can be used as a foundation in assessing leadershipeffectiveness. VanWart (2004) identifies 37 competencieslinked to administrative or managerial leadership.
Kapucu and Van Wart (2008) indicate that these 37generic competencies may change and adapt dependingupon the mission of the organization, the position of theleader, and environmental requirements such as the crisisitself. Even though there are significant similarities in thewider view of leadership, the requirements and core com-petencies needed to achieve desired results differ underspecific circumstances. Using the same theoretical frame-work, Kapucu and Van Wart (2008) implemented a qua-litative investigation among senior emergency managersin the public sector and pointed out 12 competencies,from a field of 37, as core competencies for leadershipeffectiveness during the response phase of a crisis. Thesecore competencies are decisiveness, flexibility, communi-cation (informing), problem-solving, managing innova-tion and creativity, personnel planning, motivating,building andmanaging teams, decision-making, network-ing and partnering, scanning the environment, and stra-tegic planning (Kapucu & Van Wart, 2008).
The impact of good leadership on diminishing thecatastrophic effects of big disasters/crises was theorizedby Kapucu and Van Wart (2006) in another study on the
2 N. KAPUCU AND Y. USTUN
subject of catastrophic hurricanes in the 2004 during theAtlantic hurricane season in Florida. According to theresults of their study, leaders have a significant effect interms of minimizing the harmful consequences of cala-mitous events by using their leadership competencies. Onthe contrary, leaders may worsen the results of a crisis ifthey either do not have or do not use adequate leadershipcompetencies (Boin, T’ Hart, Stern, & Sundelius, 2005).
Van Wart and Kapucu (2011) claim that “crisis man-agement does not necessarily require all the same compe-tencies of charismatic or transformational leadership asthey are articulated in the literature” (p. 495). While someaspects of transformational leadership, such as inspira-tional motivation and idealized influence, overlap withcrisis management, other aspects, such as intellectualstimulation, might be not applicable for crisis manage-ment. A crisis does not provide enough time to crisismanagers to be able to make long-term changes. Twelveleadership competencies that are identified as effective incollaborative crisis management represent the leadershipcompetencies explained by transformational and colla-borative leadership theories.
Leadership traits, skills, and behaviors
The traits approach aims to determine leaders’ personalfeatures based on three primary categories: physical,social, and individual characteristics (Lussier, 2002).According to the traits approach, a leader must possessvarious personal features, which are relatively innate orlong-term dispositions, and different from other groupmembers (Drummond, 2000).While traits are consideredas inborn characteristics of a leader, skills are accepted asindividual attributes that can be developed by learning(Boin et al., 2005). Skills are mostly practical and gainedattributes. Education, experience, and training influencethe improvement of leadership skills and competencies.Although there are certain inborn leadership skills, suchas verbal communication skills, some famous leaders did
not have those skills inherently, but developed them byeducation (Van Wart, 2011).
The behavioral approach is based on followers’ percep-tion, according to which leaders can be defined by theirbehavior rather than the characteristics of leaders.Therefore, leadership is not a component specific to theindividual, but rather it is considered a behavioral stylethat stems from a leader’s relations with followers (Yukl,2002). In the literature regarding behavioral approachsome scholars define behavioral structure and activitiesof leaders, while others examine the differences betweenbehavior structures of effective and ineffective leaders.
In the leadership action cycle model, Van Wart (2004,2011) categorizes leadership behaviors as task-oriented,people-oriented, and organization-oriented. Task-oriented behavior gives weight to careful supervision offollowers in order to achieve appropriate working meth-ods and success. The main focus of this method is toestablish a well-defined model of the organization, com-munication processes, and transaction methods betweenleader and followers (Bloisi, Cook, & Hunsaker, 2003).People-oriented leadership behaviors include activitiessuch as mutual trust in interpersonal relationships, com-munication, and respect for the opinions of and caringabout the emotions of subordinates. Leaders with thesebehaviors show more interest in the needs and desires offollowers and act in this direction (Yukl, 2002). Finally,organization-oriented behaviors mostly focus on outsideperspective, system approach, organizational culture, andorganizational change. Organization-oriented behaviorsinvolve “scanning the environment, strategic planning,articulating the mission and vision of the organization,networking and partnering, performing general manage-ment functions such as human resources and budgeting,decision making, and management of organizationalchange” (Van Wart, 2011, p. 234).
The conceptual model of crisis leadership was devel-oped based on the literature (Figure 1). The frameworkincorporates 12 key leadership competencies as inde-pendent variables and some extraneous individual
Figure 1. Leadership action cycle model.
INTERNATIONAL JOURNAL OF PUBLIC ADMINISTRATION 3
differences among leaders as control variables, whichaffect the perceived effectiveness of collaborative lea-dership in crisis management during crisis situations.In this study, the concept of perceived effectiveness ofcollaborative crisis leadership refers to the collaborativeleadership effectiveness of district and deputy gover-nors of Turkey during crisis situations. The modelassumes that acquisition of these competencies by aleader positively influences the perceived effectivenessof collaborative crisis leadership.
Core leadership competencies in managingcrisis
This study tests five hypotheses to determine the role ofleadership traits and skills and three types of leadershipbehaviors on the effectiveness of crisis leadership. Inother words, this study aims to analyze the relation-ships between core leadership competencies (leadershiptraits, skills, and behaviors) and the effectiveness ofcrisis leadership. Therefore, it is hypothesized in thestudy that:
H1: There is a positive relationship between coreleadership competencies and the perceivedeffectiveness of collaborative crisis leadership.
Traits and skills
Two leadership traits, “decisiveness and flexibility,” andone leadership skill, “communication (informing),” willbe examined in this group of competencies.
DecisivenessLeader decisiveness indicates the degree to which aleader desires to make decisions and to act decisively.Decisiveness has been determined to be an importanttrait that a leader should have and it is theoreticallypaired with assertiveness. While a decisive leader’sbehavior is expressed with clarity and precision indecision-making, indecisive behaviors are seen as areason for organizational failure (Williams, Pillai,Lowe, Jung, & Herst, 2009). Powerful and decisiveleadership is becoming a particularly important requi-site when crisis situations occur (Yukl, 2002).
FlexibilityThis trait can be understood as the ability and readinessto react in considerably different ways based on differ-ent situational necessities. A leader should have a wide-ranging response collection and ability to apply the
correct response that is suitable for situational require-ments (Zaccaro, Gilbert, Thor, & Mumford, 1992).
CommunicationCommunication can be defined in general as transmis-sion of information, ideas, and emotions from one toothers (Barrett, 2006). In the end, all managementactivity is based on the functioning of a communicationprocess because any kind of managerial decisions, theirresults, and the vision of an organization’s future can berealized and be meaningful if they are transferred toemployees and other stakeholders (Barrett, 2006).Based on the prior studies, it is hypothesized in thestudy that:
H2: There is a positive relationship between leader-ship traits and skills and the perceived effective-ness of collaborative crisis leadership.
H2a: There is a positive relationship between leader-ship traits and skills and the perceived effective-ness of collaborative crisis leadership throughtask-oriented leadership behaviors.
H2b: There is a positive relationship between leader-ship traits and skills and the perceived effective-ness of collaborative crisis leadership throughpeople-oriented leadership behaviors.
H2c. There is a positive relationship between leader-ship traits and skills and the perceived effective-ness of collaborative crisis leadership throughorganization-oriented leadership behaviors.
Task-oriented behaviors
Two task-oriented behaviors will be examined throughthe use of problem-solving and managing innovationand creativity.
Problem solvingThe literature provides diverse definitions about theconcepts of problem and problem-solving competency.One definition of problem is that it is a perceiveddifficulty or obstacle, a gap between a result and adesired situation, or perhaps an undesirable situationthat is able to be resolved with little difficulty (Evans,1991). According to Yukl (2002), problem-solvingdescribes work-related problems, analyzes problems ina timely and logical manner, recognizes reasons forproblems and finds permanent and coherent solutions,
4 N. KAPUCU AND Y. USTUN
performs decisively to accomplish solutions, and solvessignificant problems or crisis.
Managing innovation and creativityAccording to Amabile and Amabile (1983), creativity is tobe able to develop new and useful ideas, while innovationis the successful realization of creative ideas in an organi-zation. After all, creativity occurs at the individual level,while innovation occurs at the organizational level. Theinputs of organizational innovation are comprised ofindividual characteristics of the persons who created theorganization as well as features of teams and the organi-zation. To transform these inputs to innovative behaviorsand innovative products, it requires a culture and envir-onment that supports innovation (Woodman, Sawyer, &Griffin, 1993). Based on these explanations, it is hypothe-sized in the study that:
H3: There is a positive relationship between task-oriented leadership behaviors and the perceivedeffectiveness of collaborative crisis leadership.
People-oriented behaviors
The second group of behavioral competency consists ofpeople-oriented behaviors, which are team building,planning and organizing personnel, and motivatingfollowers in crisis leadership.
Team buildingA team can be defined as a distinguished group thatconsists of two or more people with a specific role ortask, acting in a compatible manner for common andvalued objectives, goals, and missions (Salas, Dickinson,Converse, & Tannenbaum, 1999). A team is formed bya small number of individuals with complementaryskills. These people act in accordance to commonobjectives and performance goals and they have mutualresponsibilities within this context (Katzenbach &Smith, 1993). In other words, a team is a collection ofpeople coming together for a specific purpose. Team-building behaviors of a leader provide better relation-ships and communication among followers, and ulti-mately increase the effectiveness of the organization.
Planning and organizing personnelThe leader’s competency in planning and organizingpersonnel refers to his/her ability to identify long-termgoals and policies to utilize personnel and personnelroles. The main concern in planning and organizingstaff is to find the best way to distribute and regulatework in order to encourage staff. This behavior involves
employing staff and continually improving their effec-tiveness by using “training, development activities, per-formance appraisal (when it is robust), social eventsrelated to work, team building, and recognition andrewards activities” (Van Wart, 2004, p. 197). Humanresources planning and organization is a dynamic pro-cess, which provides well-trained and satisfied employ-ees in the short-term and stability and continuity inorganizational culture in the long-term.
MotivatingMotivation is generally an internal attitude that evokescertain behaviors (Spector, 2000). The concept of moti-vation contains some factors that mobilize humanbehaviors, and determines the direction and period ofthose behaviors. These factors can be internal andexternal motivating factors. There is intrinsic motiva-tion for employees when they perform a specific jobwhich is engaging and exciting for them. On the otherhand, if they perform the same job for rewards such asmaking money, or getting a promotion or fame, themotivation is extrinsic. Sometimes, personal satisfac-tion means a lot more than money, but monetaryawards are also important (Dessler, 1997). Eitherintrinsic or extrinsic motivation can inspire people’ssense of enthusiasm and persistence (Daft, 2000).Thus, it is hypothesized in the study that:
H4: There is a positive relationship between people-oriented leadership behaviors and the perceivedeffectiveness of collaborative crisis leadership.
Organization-oriented behaviors
The last group of leadership behaviors is organization-oriented behaviors. These behavioral leadership competen-cies include networking and partnering, decision-making,scanning the environment, and strategic planning.
Networking and partneringNetworks are considered as a different managementstructure alongside markets and hierarchies (Rhodes,1997). Markets are multicenter, atomized, and chaoticstructures. States or companies are hierarchical and cen-tralized bodies. However, networks have more pluralisticgovernance forms. Networks, as an alternative form ofsocial and political organization approaches, challengetraditional forms of social organization and managementunderstanding. A leader’s task in a network is to providecollaboration among the people and organizations.Therefore, the network will be constructed, and the flow
INTERNATIONAL JOURNAL OF PUBLIC ADMINISTRATION 5
of resources, such as information, expertise, and technol-ogy among interdependent organizations, can be ensured.
Decision-makingThere is no doubt that every crisis is unique and mayrequire different types of effort and intervention(Kapucu & Garayev, 2011). Therefore, in crisis situa-tions there is always some level of uncertainty whichcauses fear and stress within the community. Accordingto Kapucu and Van Wart (2006), “catastrophic disastersare characterized by unexpected or unusual size, dis-ruptions to the communication and decision-makingcapabilities of the emergency response system itself,and an initial breakdown in coordination and commu-nication” (p. 280). Especially in the response phase,public managers must perform many different taskssuch as search and rescue operations, coordinationbetween rescue teams and other stakeholders, providefood and shelter, and most importantly, they mustmake rapid and accurate decisions with limited infor-mation in a time-sensitive environment.
Scanning the environmentEnvironmental scanning refers to searching for oppor-tunities and risks that may come from outside of aparticular entity or organization. Government organi-zations mostly gather information from other govern-mental entities, private sector organizations, customers,and legislative and supervisory agencies that have influ-ence on the organization. Effective leaders can assesswhether the information is vital and pertinent, andassure that this information is obtained from varioussources rather than a single source. This competency iscrucial for a leader especially in dynamic environments.If external factors overwhelm a leader, the whole orga-nization may be in danger (Van Wart, 2004).
Strategic planningThere are multiple definitions of the concept of strategicplanning that covers common features of planning andstrategy. Bryson (2011) describes strategic planning as amethodical endeavor to yield fundamental decisions andactions which shape and lead the organization’s objec-tives, actions, and working methods. Based on theseexplanations, it is hypothesized in the study that:
H5: There is a positive relationship between organi-zation-oriented leadership behaviors and theperceived effectiveness of collaborative crisisleadership.
Method
To examine how leadership competencies influence theperceived effectiveness of crisis management in thepublic sector, this research utilized structural equationmodeling (SEM) in analyzing the data. An online ques-tionnaire was developed through Qualtrics SurveySoftware and was used to collect data. The surveyincluded a total of 70 questions, including demographicquestions and open-ended questions.
This research analyzed the issue at the individuallevel. The unit of analysis of this research was territorialstate representatives such as the district and provincegovernors, deputy province governors, administrativeinspectors, and high- and middle-level bureaucrats ofTurkish Ministry of Interior who were the study popu-lation of this research. Province deputy governors,administrative inspectors, and Ministry of Interiorhigh- and middle-level bureaucrats are appointed bythe central government from the pool of district gov-ernors who have completed a specific time span inpublic service. Therefore, they are considered to haveadequate knowledge and experience in crisis leadership.
An e-mail including a link to the survey was sent toall of 2,095 territorial state representatives. Since thesurvey was sent to the entire population, any specialsampling method was not used. Three hundred andthirty people responded to the survey. However, 29responses were excluded from the data for furtheranalysis because they did not complete in more than50% of the survey questions. The final dataset of thestudy was comprised of 301 responses.
The SEM was used in this research. SEM is definedas “a very general statistical modeling technique widelyused in the behavioral sciences” (Hox & Bechger, 1998,p. 1). SEM is one of the common ways to test hypoth-eses on relationships among observable and non-obser-vable (latent) variables (Hoyle, 1995). There are severalviews regarding the necessary sample size for SEManalysis in the related literature. This study followedthe recommendation of Boomsma and Hoogland(2001) that 200 cases are an appropriate sample sizefor a proper SEM analysis. With its sample size of 301,this study had an adequate sample size for the analysis.
SPSS Statistic, version 17, and AMOS software wereutilized to run the analyses. In the first stage, descrip-tive statistics were performed by using SPSS to under-stand general characteristics of the data, and assess thegeneral picture. And then, in addition to descriptivestatistics in the form of frequency distribution, correla-tion analysis was performed to identify relations amongstudy variables and the possible multicollinearity
6 N. KAPUCU AND Y. USTUN
problems between indicators of each latent construct.As a common issue, multicollinearity occurs when twoor more variables are highly correlated.
The SEM is conducted to validate the theoreticallydriven model (Wan, 2002). The SEM mainly consists ofthe measurement model and the structural model(Byrne, 2010). Confirmatory factor analysis (CFA) wasutilized for each latent construct in order to examinewhether the hypothesized measurement models fit thedata. When the model did not fit, the model wasrevised until reaching a fit model. The covariance struc-ture model (CSM) was developed after validating themeasurement models of the latent variables. The CSMconsists of exogenous, mediating, endogenous latentvariables, and control variables. A CSM was utilizedto examine the structural relationships between coreleadership competencies (leadership traits and skillsand leadership behaviors) and the effective crisis lea-dership of Turkish province and district governors.
Results and discussion
In terms of the study population’s professional posi-tions, the frequency and percentage distributions ofthe target population were as follows: 171 provincegovernors constitute 8.2% of all population, 498 dep-uty province governors represent 23.7% of total, 866district governors represent 41.3% of target popula-tion, 191 administrative senior inspector constitute9.2% of total, 134 Ministry of Interior high- or mid-dle-level bureaucrats represent 6.4% of all population,and 236 candidate district governors constitute 11.2%of total target population of survey. In terms ofrespondents’ professional positions, district governorsconstitute the largest respondent group with 135 validresponses (45.9%). Deputy province governors consti-tute the second largest group (56; 19.0%) of therespondents. Fifty-two respondents listed their profes-sional position as an administrative senior inspector,which is 17.7% of all respondents. Thirty-three inter-ior ministry high- or middle-level bureaucrats validlyresponded to the survey questions, which represents10.9% of all responses. With 16 valid responses, can-didate district governors constitute 5.4% of respon-dents. Due to the low proportion of provincegovernors relative to other positions, it is not surpris-ing that only three province governors completed thesurvey. Another reason could be the province gover-nors’ heavy work-related responsibilities. This groupaccounts for 1.0% of the study participants.
In this study, .75 was used as the multicollinearitythreshold. In total, four multicollinearity problems were
detected. Therefore, four indicators were excluded fromthe models in order to solve multicollinearity problem.All of the control variables were not significant atp < .05, thus the control variables were also removedfrom the CSM.
Reliability tests the extent to which a scale is able tomeasure the intended issue. For reliability, a test mustbe repeatable in and transferable to similar research(Trochim, 2016). In order to confirm the reliability ofthis study’s scales, Cronbach’s alpha coefficient analyseswere performed. Table 1 indicates the Cronbach’s alphavalues, calculated by SPSS.
The table shows that all constructs achieved excellentmeasurement reliability with the lowest value of .932for task-oriented behaviors, and the highest value of.945 for leadership traits and skills.
Leadership traits and skills
This study hypothesized a positive association betweenleadership traits and skills, and the perceived effective-ness of crisis leadership. The only exogenous latentvariable of the study, leadership traits and skills, wasmeasured by 14 items, which reflect different attributesof leadership traits and skills.
Survey participants were asked to indicate to whatextent they agreed that the leadership traits and skills ofthe district or province governor whom they had givenan opportunity to examine, investigate, or observe clo-sely in a crisis situation impacted the effectiveness ofcrisis management. First five indicators of this variableevaluated decisiveness trait of the leaders. In general,more than half of the respondents supported that thedistrict or province governors had clarity and precisionin their decisions during the crisis (58.8%), had a highlevel of self-confidence when making a decision (60.1),did not lose his/her self-control under stress (63.3%),made decisions independently, when appropriate, byconsidering himself/herself the primary decision-maker (52.8), and they did not hesitate to use initiative,if necessary, by taking into account possiblerisks (60.7%).
Table 1. Cronbach’s alpha scores of measurement models.Number of
Effectiveness of crisis leadership 11 11 .936 .936
INTERNATIONAL JOURNAL OF PUBLIC ADMINISTRATION 7
Regarding flexibility, a majority of the respondentsagreed or strongly agreed with the statements that thedistrict or province governors had a capacity to reactwith distinctive methods to different situational neces-sities (64.4%), adapted to different needs (such asadapting to an extremely stressful working environ-ment) when needed (68.8%), and they can diagnosethe situation quickly and determine the proper formof behavior that will achieve a positive result (61.0%).
In the model, the only skill of the leaders, which isespecially important in crisis leadership, was commu-nication. The respondents supported the statements ofall seven indicators. According to the respondents, thedistrict or province governors communicated with sta-keholders regularly, as needed (62.8%), developed andexecuted external and internal communication withstakeholders (victims, organizations, and the media)in the crisis (49.1%), utilized information and commu-nication technologies (ICT) in order to maintain aprecise and constant flow of information (48.5%),chose appropriate communication channels and meth-ods (Internet, TV, radio, and such) in the crisis(44.6%), identified barriers for listening to the staffand other stakeholders in the crisis (50.4%), reducedbarriers for listening to the staff and other stakeholders(52.9%), and they involved all stakeholders in crisiscommunication plans (50.9%). Overall, responses tothe indicators of leadership traits and skills accumulatewithin strongly agree and agree responses. The dis-agreed and strongly disagreed responses are low, andgenerally constitute around 20% of total respondents.
The factor loading values of all the indicators weregreater than .40 (from .51 to .88); hence, none of themwere excluded from the traits and skills measurementmodel. Among the 14 indicators, the indicator “diag-nosing the situation quickly” has the strongest impacton the latent variable of leadership traits and skills, witha regression coefficient of .88, followed by the indicator“using distinctive methods” with a regression coefficientof .80. The other indicators had moderate regressionweights except the indicator “utilized ICT,” which hada noticeably lower regression coefficient value (.51)than other indicators.
Task-oriented leadership behaviors
The task-oriented leadership behaviors variable is thefirst of three mediating variables of this study. Thislatent variable was designed to measure the perceptionsof Turkish province and district governors as to whatextent task-oriented leadership behaviors play a med-iating role between leadership traits and skills and theperceived effectiveness of crisis leadership. The research
hypothesis was supported with a standard regressioncoefficient of positive .33, which means there is a sta-tistically significant association between these two con-structs as specified in the hypothesis testing.
A measurement model was created by utilizingeight indicators to analyze if the hypothesized modelfits the data. Two task-oriented leadership behaviors,problem-solving and managing innovation and crea-tivity, were evaluated by the respondents. The state-ments regarding problem-solving were supported byalmost half of the respondents. Participants wererequested to indicate to what extent they agreed thatthe governors showed adequate problem-solving beha-viors during a crisis. Respondents supported that thedistrict or province governors defined the problem andformulated responses in a crisis situation (48.8%),developed a systematic approach in analyzing pro-blems (36.8%), generated alternatives by creating alist of options to solve problems and chose the bestoption (48.1%), and they promoted collaborative pro-blem-solving (58.5%). The respondents reportedweaker support to the statements of managing innova-tion and creativity compared with their previousanswers. Only 38.5% of all respondents approved thatthe district or province governors created an organiza-tional culture of innovation and creativity by encoura-ging and rewarding followers who intend to makechange and achieve successful results. They benefitedfrom the creative and innovative ability of the staff andpartner institutions (49.3%), had willingness to takerisks and to consider new and untested approaches attimes of crisis (44.9%), provided a welcoming atmo-sphere in which followers did not feel any pressure(38.3%), and they provided the tools and opportunitiesfor learning and innovation (42.3%).
The results of CFA implied that all indicators in therevised measurement model have significant factorloadings at p < .05, and the factor loadings of indicatorsranged between .67 and .87. While the indicator “ben-efiting from the creative and innovative ability of thestaff and partner institutions” had the highest factorloading with the value of .87, the indicator “generatingalternatives” had the lowest factor loading with thevalue of .67.
People-oriented leadership behaviors
The people-oriented leadership behaviors variable is thesecond latent mediating endogenous construct in thisstudy. The measurement model of people-oriented lea-dership behaviors is specified with 11 indicators. The11 items reflect different attributes of people-orientedbehaviors. Participants were asked to indicate their
8 N. KAPUCU AND Y. USTUN
evaluation about three core people-oriented leadershipbehaviors, team building, planning and organizing per-sonnel, and motivating, which are crucial for effectivecrisis leadership.
Some statements of team-building behaviors, com-pared with other questions, received the lowest agreedresponses. The respondents think that the district orprovince governors enhanced group identity by creat-ing a group mission, vision, common interests, andshared values among participating organizations in acrisis (60.6%), encouraged the staff to work as a team(42%), selected the proper number of people with well-balanced capabilities for the best group structure(40.6%), built teams with special training, skills, andcompetencies in the crisis (31.6%).
The majority of the respondents approved the state-ments about leadership behaviors of the district orprovince governors in planning and organizing person-nel. The respondents think that the district or provincegovernors arranged the division of labor according tothe duties and responsibilities of the staff (58.4%),scheduled personnel by using negotiation and percep-tions of fairness (56.7%), matched staff preferences andcompetencies to the work as much as possible (53.9%),and they evaluated and supported the staff’s perfor-mance and helped them perform better (59.2%).
More than half of the respondents supported thatthe district or province governor established a positiverelationship with the staff by making them feel thattheir contribution was important (55.6%), appreciatedthe staff’s efforts in a timely and appropriate manner(48%), explained how rewards and significant commen-dations are distributed and used them to motivatefollowers (64.6%). Almost half of the respondentsthink that the district or province governors evaluatedfairly the staff’s contribution to the crisis response team(46%), and just 35.3% of them think that the district orprovince governors explained rules and procedures toensure that subordinates had understood the conse-quences of deviations, and executed punishment whendeviations occurred.
Organization-oriented leadership behaviors
The latent construct of organization-oriented leader-ship behaviors is the last mediating variable of thestudy and is measured using 15 items. This variableconsists of networking and partnering, decision-mak-ing, scanning the environment, and strategic planning.Almost three out of four respondents (74.4%) agreed orstrongly agreed that the district or province governorwhom the respondents had given an opportunity toexamine, investigate, or observe closely in a crisis
situation periodically contacted external stakeholders,politicians, and other strategic allies. The respondentsobserved that the district or province governors devel-oped long-term relationships with stakeholders(48.1%), constantly exchanged information with otherorganizations in the network during a crisis (67.5%),and they were open to partnerships during crisis inter-vention and answered to collaboration needs of othersat the maximum level (58%).
A vast majority of respondents agreed with thestatements, which evaluated decision-making beha-viors of the leaders. According to the respondents,the district or province governors made decisionswith limited information under time pressure inresponse to crises (64.8%), made quick decisions dur-ing a crisis compared with routine management(76.5%), sought counsel from others in analyzing thesituation (61.9%), reacted differently during the crisis(although nervous, became more focused and solutionoriented) (67%), and they detected problems correctlywithout losing sight of the complete picture and madecorrect decisions by considering the possible conse-quences (54.6%).
The third important organization-oriented leader-ship behavior for crisis leadership is scanning the envir-onment. The respondents evaluated this leadershipbehavior with three questions. More than half of allrespondents (55%) agreed or strongly agreed that thedistrict or province governor identified and used multi-ple relevant sources of external information, and fol-lowed up on significant external trends, such as newdevelopments in technology (51%). Almost half of therespondents also agreed that the district or provincegovernor reflected on the significance of external trendsfor the organization (45.8%).
The agreement rates of the respondents with thestatements about strategic planning behaviors of theleaders were not as high as previous behaviors. Thetotal percentage of respondents who agreed or stronglyagreed that the district or province governor collectedsystematic and comprehensive data from the staff andstakeholders was 42 of all respondents. Just 38.4% oftotal respondents think that district or province gover-nors regularly reviewed the mission and capabilities ofthe organization for strategic planning. According to34.3% of total respondents, the district or provincegovernor developed a step-by-step comprehensive stra-tegic plan for crisis management. Overall, except for thelast four indicators, agreement responses exceed dis-agreement responses for each indicator of thisconstruct.
The critical ratios of parameter estimates for therevised model show that all regression coefficients were
INTERNATIONAL JOURNAL OF PUBLIC ADMINISTRATION 9
significant at p < .05 (CR > 1.96). All indicators hadsubstantially high factor loading values, from .51 to .85.
Perceived effectiveness of collaborative leadershipin managing crises
The survey used 11 items to measure the perceivedeffectiveness level of crisis leadership. These items indi-cate different attributes of leadership effectiveness incrises. The majority of survey respondents agreed orstrongly agreed that the district or province governorswhom they had an opportunity to examine, investigate,or observe closely in the crisis situation showed effec-tive collaborative leadership in the crisis. Regardingfacilitating crisis management functions, the cumulativepercentage of survey participants who either agreed orstrongly agreed is 50.5%. Half of the respondents (50%)either agreed or strongly agreed that the district orprovince governor successfully implemented crisismanagement plans in mobilizing his/her own personneland resources, while 28.4% disagree or strongly dis-agree. Nearly half of the respondents (48.8%) acceptedthat the district or province governor successfullyincluded emerging resources in the implementation ofcrisis management plans. Overall, agreement responsesexceeded disagreement responses for each of the 11indicators. The results indicate that almost half of the
respondents showed their agreement for each item. Inother words, the district or province governors demon-strated effective collaborative leadership in a crisiswhich was examined, investigated, or observed closelyin the crisis situation by the respondents.
This study tested five hypotheses and three sub-hypothesis to determine the role of leadership traitsand skills and three types of leadership behaviors onthe perceived effectiveness of crisis leadership. Whetheror to what extent these hypotheses are supported can beexamined based on the results of SEM analysis. Thecovariance structural model, which is revised accordingto the results of SEM, is presented in Figure 2. Therelationship of the effectiveness of crisis leadership witheach of the control variables was insignificant. Based onthe theoretical framework and literature review, thefollowing hypothesizes were tested in this study:
H1: There is a relationship between core leadershipcompetencies and the perceived effectiveness ofcollaborative crisis leadership.
The outcomes of SEM analysis supported the firsthypothesis of this study. The relationships among corecompetencies and the effectiveness of crisis leadershipwere all positive and statistically significant at the .05level. Consequently, it is safe to claim that core
Figure 2. Covariance structural model.
10 N. KAPUCU AND Y. USTUN
leadership competencies positively influence perceivedeffectiveness of collaborative crisis leadership.
H2: There is a positive relationship between leader-ship traits and skills and the perceived effective-ness of collaborative crisis leadership.
To assess whether leadership behaviors mediate therelationships between leadership traits and skills andthe effectiveness of collaborative crisis leadership, thesignificance of the structural paths from leadershiptraits and skills to each leadership behaviors waschecked. As the sub-hypotheses H2a, H2b, and H2cindicated, the results of revised SEM demonstrate thatleadership traits and skills have significant and positiverelationships with task-oriented, people-oriented, andorganization-oriented leadership behaviors.
H2a: There is a positive relationship between leader-ship traits and skills and the perceived effective-ness of collaborative crisis leadership throughtask-oriented leadership behaviors.
The results of revised SEM show that leadershiptraits and skills has a significant and positive relation-ship with task-oriented leadership behaviors, with aregression coefficient value of .932; and that a positivecorrelation exists between task-oriented leadershipbehaviors and the effectiveness of crisis leadership,with a correlation coefficient of .331 at p < .05. Thisimplies that task-oriented leadership behaviors partiallymediate the relationship between leadership traits andskills and effectiveness of collaborative crisis leadership.
H2b: There is a positive relationship between leader-ship traits and skills and the perceived effective-ness of collaborative crisis leadership throughpeople-oriented leadership behaviors.
The results of revised SEM show that leadership traitsand skills have a significant and positive relationship withpeople-oriented leadership behaviors, with a regressioncoefficient value of .937; and that a positive correlationexists between people-oriented leadership behaviors andthe effectiveness of crisis leadership, with a correlationcoefficient of .262 at p < .05. This result indicates thatpeople-oriented leadership behaviors partially mediatethe relationship between leadership traits and skills andeffectiveness of collaborative crisis leadership.
H2c. There is a positive relationship between leader-ship traits and skills and the perceived
effectiveness of collaborative crisis leadershipthrough organization-oriented leadershipbehaviors.
The results of revised SEM show that leadershiptraits and skills have a significant and positive relation-ship with organization-oriented leadership behaviors,with a regression coefficient value of .954. In addition,a positive correlation exists between organization-oriented leadership behaviors and the effectiveness ofcrisis leadership, with a correlation coefficient of .270 atp < .05. This also implies that organization-orientedleadership behaviors partially mediate the relationshipbetween leadership traits and skills and effectiveness ofcollaborative crisis leadership.
The results indicated that leadership traits and skillshave a positive relationship with the perceived effec-tiveness of collaborative crisis leadership, which ismediated by leadership task, people, and organization-oriented behaviors.
H3: There is a positive relationship between task-oriented leadership behaviors and the perceivedeffectiveness of collaborative crisis leadership.
The results of analysis indicated that task-orientedleadership behavior was the strongest effective mediatingconstruct on endogenous variable, with a positive regres-sion coefficient .33 at p < 0.05. The finding showed thatthe effectiveness of crisis leadership level is positivelyinfluenced by task-oriented leadership behaviors. Inother words, the more the leaders in crisis situationsimplement their task-oriented behaviors as a leader’scompetency, the higher their effectiveness levels will be.
H4: There is a positive relationship between people-oriented leadership behaviors and the perceivedeffectiveness of collaborative crisis leadership.
The results of the revised SEM show that people-oriented leadership behaviors have a positive relationshipwith effectiveness of collaborative crisis leadership, with astandardized regression coefficient value of .262 at p < .05.The direction of the relationship is positive as expected,and the relationship between people-oriented leadershipbehaviors and the effectiveness of collaborative crisis lea-dership was found to be significant. The results of thisstudy indicate that people-oriented leadership behaviorsincrease the effectiveness of collaborative crisis leadership.
H5: There is a positive relationship between organi-zation-oriented leadership behaviors and the
INTERNATIONAL JOURNAL OF PUBLIC ADMINISTRATION 11
perceived effectiveness of collaborative crisisleadership.
The results of the SEM analysis also supported thelast hypothesis of this study. With the standardizedregression coefficient value of .27, there was a statisti-cally significant relationship at p ≤ .05 between organi-zation-oriented leadership behaviors, a mediating latentvariable, and the effectiveness of crisis leadership, anendogenous latent variable. The critical ratio score was2.146, which is higher than the suggested score of 1.96.Consequently, it is safe to claim that organization-oriented leadership behaviors positively influence effec-tiveness of collaborative crisis leadership.
The research hypotheses were supported by theempirical findings of the study, which means that theresults of the study are consistent with the propositionsof theoretical models and theories used in the study.The latent constructs have statistically significant rela-tionships with the perceived effectiveness of crisis lea-dership. Moreover, the results show that mediatingvariables are also vital for comprehending the relation-ship between core leadership competencies and theeffectiveness of crisis leadership. These hypothesesexpected that an increase in the exogenous variablewould produce an increase in mediator variables, andindirectly an increase in the endogenous variable. Thetheoretical framework that guided this study was builton the extant literature. The theoretical suppositions ofprevious studies that beyond other possible explanatoryfactors, the perceived effectiveness of crisis leadership isa function of the core leadership competencies utilizedby leaders during a crisis were supported and con-firmed by this study.
According to the results of this research, task-oriented leadership behaviors greatly increase the effec-tiveness of crisis leadership. However, the effects ofpeople-oriented and organization-oriented leadershipbehaviors on the effectiveness of crisis leadership arealso not low. Research results can be interpreted asprovincial and district governors should consider thepotential of their subordinates, develop and motivatethem to perform their tasks, and maintain good com-munication with them. Although the research resultsdo emphasize the importance of task-oriented behavior,it is not possible to say that focusing only on task-oriented behaviors alone will increase the efficiency ofcrisis management. In such a case, there will not be anychange or development either in the personnel or in thestructure of the organization.
Leaders generally need to coordinate effectively alldifferent governmental agencies, other sector represen-tatives, and even volunteer individuals. In such an
environment, leaders need different kinds of authoritysources, such as legal, contractual, and voluntary to dealwith issues in disseminating information, organizingfinancial–material allocations, and distributing respon-sibilities and authority. The results also confirm thefindings in previous studies in literature. Networkingand partnering behaviors of leaders are representedwith four indicators in the SEM revised model as theindicators of organization-oriented behaviors. Thosefour indicators had .50, .48, .72, and .77 factor loading,which are over the determined threshold (.40) andsignificant at p < .05 level. Based on the results of theSEM, networking and partnering behaviors of leadersduring a crisis have a significant impact on the per-ceived effectiveness of crisis leadership as mentioned inKapucu and Van Wart’s (2006) study.
Conclusion
This study analyzed the impact of leadership compe-tencies on perceived effectiveness of crisis leadership inthe crisis management context in Turkey. The vitalimportance of having an effective crisis managementsystem and adequate crisis leaderships can be under-stood in terms of continuity of government operations.A mismanaged crisis can damage the reputation of agovernment and erode the citizens’ trust of a govern-ment. Their core leadership competencies for crisissituations should be developed in training programsbefore appointing them to especially critical areas withcrisis potential such as frontier towns in southeast partof the country’s disaster-prone prareas.
The competency model that was developed in thisstudy can be utilized in several ways in the publicmanagement practice. The competency set is a goodsource to evaluate a candidate for district governors’adequate preparedness for an interview or examination.The more a candidate fits the requirements of thedistrict governor position, the more he/she will beperceived as an effective leader during a crisis.Therefore, a good match between the position andcandidates’ competencies will increase the potentialfor better crisis management. The general directorateof staff of the Turkish Interior Ministry may use thismodel as a criteria set when recruiting new districtgovernors.
Crisis management is one of the most importanttasks of provincial and district governors. A governorneeds to be adequately informed about his/her dutiesand responsibilities before being appointed to his/herworkplace. In this way, he/she recognizes and needs toincrease his/her necessary competencies for crisis lea-dership. Education and training programs are the main
12 N. KAPUCU AND Y. USTUN
tools to inform the governors properly. In light of thisor similar studies, an education and training programneeds to be arranged for governors in order to providethem with the required skills and competencies forcrisis leadership. These organizations may certify theattendants according to their achievement. Governorswith these certificates may be appointed to the pro-vinces and districts with high potential for crises.Such an implementation in personnel policy will leadto a surge of governors acquiring the expertise they willneed to improve their needed competencies.
Anyone who desires to be a governor in Turkey mustgraduate from some certain disciplines. In other words,formal education at undergraduate or graduate level inTurkey does not provide required competencies for crisisleadership by future governors. The Ministry of Interiorcan work closely with the universities to eliminate thisgap. Again, these competencies can be a framework forsuch a curriculum that aims to provide necessary trainingand preparedness for future governors.
This research has a cross-sectional design, which is atime-saving and efficient technique to examine researchhypotheses. The use of cross-sectional research posessome limitation due to a lack of a sequential timeline;in other words, a time-order sequence is not available inorder to infer causation (Shadish, Cook, & Campbell,2002). Utilization of multiple methods and sources, suchas agency records, interviews, and first-line managers’evaluations, could help researchers to collect and analyzeadditional data about the effects of leadership traits,skills, and behaviors on the perceived effectiveness ofcrisis leadership. Moreover, because SEM was used todiscover the relationships between variables, the study islimited to the quantitative method. Future researcherscan take the findings of this research as a starting point,and bring these results to a further point which they canuse for other qualitative or mixed methods and sources.With longitudinal research, the researchers can study thestability of and changes in predictors, and their impactson outcome variables linearly. The data from thisresearch project could be used as a basis for a long-itudinal study. This study uses a self-reported survey asthe primary data source, which makes it subject to themethod variance problem to some extent becauserespondents might have a tendency to not indicatetheir actual behaviors and views.
References
Amabile, T., & Amabile, T. M. (1983). The social psychologyof creativity (Vol. 11). New York, NY: Springer-Verlag.
Barrett, D. J. (2006). Strong communication skills a must fortoday’s leaders. Handbook of Business Strategy, 7(1), 385–390. doi:10.1108/10775730610619124
Bloisi, W., Cook, C. W., & Hunsaker, P. L. (2003).Management and organisational behaviour. Maidenhead,UK: McGraw-Hill.
Boin, A., T’ Hart, P., Stern, E., & Sundelius, B. (2005). Thepolitics of crisis management: Public leadership under pres-sure. New York, NY: Cambridge University Press.
Bolden, R., Gosling, J., Marturano, A., & Dennison, P. (2003).A review of leadership theory and competency frameworks.Report for Chase Consulting and the ManagementStandards Centre, Centre for Leadership Studies.University of Exeter, Exeter, United Kingdom.
Boomsma, A., &Hoogland, J. J. (2001). The robustness of LISRELmodeling revisited. Structural equation models: Present andfuture. A Festschrift in Honor of Karl Jöreskog (pp. 139–168).Chicago, IL: Scientific Software International.
Bryson, J. M. (2011). Strategic planning for public and non-profit organizations: A guide to strengthening and sustain-ing organizational achievement (Vol. 1). San Francisco,CA: John Wiley & Sons.
Byrne, B. M. (2010). Structural equation modeling withAMOS. New York, NY: LEA Publisher.
Carter, M. M. (2006). The importance of collaborative leader-ship in achieving effective criminal justice outcomes.Department of Justice National Institute of Corrections.Retrieved from http://nicic.gov/Library/021200
Daft, R. L. (2000).Management. Orlando, FL: Harcourt College.Dessler, G. (1997). Human resources management (7th ed.).
New York, NY: McGraw Hill Book Company.Drummond, H. (2000). Organizational behavior. New York,
NY: Oxford University Press.Evans, J. R. (1991). Creative thinking: In the decision and
Farazmand, A. (2007). Learning from the Katrina crisis: Aglobal and international perspective with implications forfuture crisis management. Public Administration Review,67(1), 149–160. doi:10.1111/j.1540-6210.2007.00824.x
Heller, N. A. (2012). Leadership in crisis: An exploration ofthe British Petroleum Case. International Journal ofBusiness and Social Science, 3(18), 21–32.
Hemansson, H. M. L. (2016). Disaster management colla-boration in Turkey: Assessing progress and challenges ofhybrid network governance. Public Administration, 94(2),333–349. doi:10.1111/padm.12203
Hoyle, R. (Ed) (1995). Structural equation modeling:Concepts, issues, and applications (pp. 100-117).Thousand Oaks, CA: Sage.
Hox, J. J., & Bechger, T.M. (1998). An introduction to structuralequation modeling. Family Science Review, 11, 354–373.
Ink, D. (2006). An analysis of the house select committee andWhite House reports on Hurricane Katrina. PublicAdministration Review, 66(6), 800–807. doi:10.1111/puar.2006.66.issue-6
Kapucu, N. (2010). Governance reforms: Comparative perspec-tives. Ankara, Turkey: International Strategic ResearchOrganization (ISRO) Publications.
Kapucu, N., & Garayev, V. (2011). Collaborative decision-making in emergency and crisis management.
International Journal of Public Administration, 34(6), 366–375. doi:10.1080/01900692.2011.561477
Kapucu, N., & Palabiyik, H. (2008). Turkish public adminis-tration: From tradition to modern age. Ankara, Turkey:International Strategic Research Organization (ISRO)Publications.
Kapucu, N., & Van Wart, M. (2006). The evolving role of thepublic sector in managing extreme events: Lessons learned.Administration & Society, 38(3), 279–308. doi:10.1177/0095399706289718
Kapucu, N., & Van Wart, M. (2008). Making matters worse:An anatomy of leadership failures in managing cata-strophic events. Administration and Society, 40(7), 711–740. doi:10.1177/0095399708323143
Katzenbach, J. R., & Smith, D. K. (1993). The wisdom ofteams: Creating the high-performance organization.Boston, MA: Harvard Business School Press.
Klann, G. (2003). Crisis leadership: Using military lessons,Organizational experiences, and the power of influence tolessen the impact of chaos on the people you lead.Greensboro, NC: Center for Creative Leadership Press.
Lussier, R. N. (2002). Human relations in organization. NewYork, NY: Mc Graw Hill.
McGuire, M., & Silvia, C. (2009). Does leadership in net-works matter? Examining the effect of leadership behaviorson managers’ perceptions of network effectiveness. PublicPerformance & Management Review, 33(1), 34–62.doi:10.2753/PMR1530-9576330102
Moore, L. L., & Rudd, R. D. (2004). Leadership skills andcompetencies for extension directors and administrators.Journal of Agricultural Education, 45(3), 22–33.doi:10.5032/jae.2004.03022
Murphy, P., & Dunn, P. (2012). Senior leadership in times ofcrisis. Noetic Notes, 3, 1.
Rhodes, R. A. (1997). Understanding governance: Policy net-works, governance, reflexivity and accountability.Maidenhead, UK: Open University Press.
Salas, E., Dickinson, T. L., Converse, S. A., & Tannenbaum, S. I.(1999). Toward an understanding of team performance andtraining. In R. W. Swezey, & E. Salas (Eds.), Teams: Theirtraining and performance (pp. 3–29). Norwood, NJ: Ablex.
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002).Experimental and quasi-experimental designs for general-ized causal inference. New York, NY: Houghton-Mifflin.
Spector, P. E. (2000). Industrial and organizational psychologyresearch and practice (2nd ed.). New York, NY: John Wiley& Sons, Inc.
Trochim, W. M. (2016). The research methods knowledge base,2nd Edition. Retrieved from http://www.socialresearchmethods.net/kb
Unlu, A., Kapucu, N., & Sahin, B. (2010). Disaster and crisismanagement in Turkey: A need for a unified crisis manage-ment system. Disaster Prevention and Management, 19(2),155–174. doi:10.1108/09653561011037977
Van Wart, M. (2004). A comprehensive model of organi-zational leadership: The leadership action cycle.International Journal of Organization Theory andBehavior, 7, 173–208.
Van Wart, M. (2011). Dynamics of leadership in public ser-vice: Theory and practice (2nd ed.). Armonk, NY: MESharpe Publishing.
Van Wart, M., & Kapucu, N. (2011). Crisis managementcompetencies: The case of emergency managers in theUSA. Public Management Review, 13(4), 489–511.doi:10.1080/14719037.2010.525034
Wan, T. T. H. (2002). Evidence-based health care manage-ment: Multivariate modeling approaches. Boston, MA:Kluwer Academic Publishers.
Williams, E. A., Pillai, R., Lowe, K. B., Jung, D., & Herst, D.(2009). Crisis, charisma, values, and voting behavior in the2004 presidential election. The Leadership Quarterly, 20(2),70–86. doi:10.1016/j.leaqua.2009.01.002
Woodman, R. W., Sawyer, J. E., & Griffin, R. W. (1993).Toward a theory of organizational creativity. Academy ofManagement Review, 18(2), 293–321.
Yukl, G. (2002). Leadership in organizations (5th ed.). UpperSaddle River, NJ: Prentice-Hall.
Zaccaro, S. J., Gilbert, J. A., Thor, K. K., & Mumford, M. D.(1992). Leadership and social intelligence: Linking socialperspectiveness and behavioral flexibility to leader effec-tiveness. The Leadership Quarterly, 2(4), 317–342.doi:10.1016/1048-9843(91)90018-W
