Aunque investigadores hayan conseguido identificando el conocimiento y las habilidades (Mumford, Zaccaro, Harding, Jacobs, & Fleishman, 2000) y rasgos y características personales (Stogdill, 1974; Zaccaro, Kemp, & más Malo, 2004) de líderes efectivos, ellos no han sido casi como exitosos en identificar ni definir esas dotes de mando evasivas que se caen en el dominio afectiva – lo que el Concilio Nacional para la Acreditación de la Educación de Maestro (NCATE) (2008) y el Concilio de Estado Principal Eduque a Oficiales (CCSSO) (1996; 2008) "disposiciones" de llamada. Por supuesto, que los profesores de la administración educativa lo encuentran difícil o definir o valorar disposiciones (Melton, Mallory, & Verde, 2010) no los ha aliviado de la responsabilidad para hacer así. NCATE requiere programas que preparan la escuela líderes para dirigir "disposiciones," así como "el conocimiento y las habilidades". Además, requiere que programas emplean justo, los métodos seguros y válidos para hacer así.

Burns had it right. The more we research leadership, the more we learn what we do not know. Should we focus on the leader’s personal qualities? Or, is it more important to examine what a leader accomplishes, long term? What matters most when choosing a leader? Do we give more weight to the experience and skill sets someone brings to the situation? Or, do we value more those elusive attributes variously included in concepts like “character” and “wisdom?” In other words, is it what the person has done, or can do? Or, is it what the person is like . . . or is likely to be? Most of us will answer “yes” to all of the above. Leadership is, indeed, an elusive concept.

Although researchers have succeeded in identifying knowledge and skills (Mumford, Zaccaro, Harding, Jacobs, & Fleishman, 2000) and personal traits and characteristics (Stogdill, 1974; Zaccaro, Kemp, & Bader, 2004) of effective leaders, they have not been nearly as successful in identifying or defining those elusive leadership qualities that fall into the affective domain – what the National Council for the Accreditation of Teacher Education (NCATE) (2008) and the Council of Chief State School Officers (CCSSO) (1996; 2008) call “dispositions.” Of course, that professors of educational administration are finding it difficult to either define or assess dispositions (Melton, Mallory, & Green, 2010) has not relieved them of the responsibility for doing so. NCATE requires programs that prepare school leaders to address “dispositions,” as well as “knowledge and skills.” Moreover, it requires that programs employ fair, reliable, and valid methods for doing so.

And herein lies the problem. The term “dispositions” poses a complex set of issues, in part due to the complex nature of constructs implied by the term. In this brief paper, the authors discuss the challenges posed by the need to assess leadership dispositions, and they review the current state of professional literature that informs attempts to do so. In addition, they offer two examples of promising practices designed to assist candidates in educational administration programs by providing them valid and reliable feedback on their leadership dispositions.

Many institutions (and individuals) shy away from assessing dispositions of administrative candidates and from using the results of those assessments for any summative (Scriven, 1991) purposes for fear of legal reprisal (Lindahl, 2009). However, courts are unlikely to intervene if the institution has taken appropriate measures to ensure that the assessment processes are valid, reliable, provide sufficient due process, and do not violate the civil rights of applicants or candidates (Ginzberg & Whaley, 2003; Lindahl, 2009).

The first condition courts would consider would be the extent to which the dispositions assessed are clearly related to job requirements (Griggs v Duke, 1971), which is known as predictive validity. Institutions of higher education may look to professional standards to determine those dispositions most essential for the profession. This is referred to as content validity. For educational administration programs, the most common source (Lindahl, 2009) of such standards is the 1996 Standards for School Leaders (Council of Chief State School Officers), promulgated by the Interstate School Leadership Licensure Consortium (ISLLC) under the aegis of the Council of Chief State School Officers, later adopted by the National Policy Board for Educational Administration and the Educational Leadership Constituent Consortium. These standards identified 43 specific dispositions that all school leaders should demonstrate. The revision of these standards in 2008 (Council of Chief State School Officers) maintained the same basic six standards with only a minor change in wording. Instead of explicating the specific knowledge, skills, and dispositions underlying each standard, the new standards substituted performance-based language. However, personal communication with the head of the revision committee, Joe Simpson, revealed that the revision committee in no way intended to suggest that the original skills, knowledge, and dispositions no longer applied (personal communication, April 20, 2008). The performance-based language was substituted to facilitate the assessment of school leaders via the standards (Sanders & Keaney, 2007). Professional organizations, states, and institutions were left to ensure that both potential and acting school leaders were equipped with the necessary knowledge, skills, and dispositions. Consequently the 43 dispositions of the original ISLLC Standards may still be considered valid. Some caution may need to be taken if an institution chooses to assess all 43 dispositions, for this places a considerable burden on the multiple assessors of these dispositions and may be superfluous, as there is considerable conceptual overlap among the dispositions and auto-correlation is likely. It may be prudent to categorize them and to reduce the number of specific dispositions assessed.

Other sources of essential dispositions may include the NCATE Unit Standards (2008) and each individual school or college of education’s conceptual model, as required by NCATE. Because these standards and conceptual models were developed through formal processes involving university faculty and highly respected P-12 practitioners, they may be considered to have face, or even expert, validity. This provides the foundation for further research by the institution on the extent to which the assessed dispositional levels of candidates later translates into predictive validity for their performance as school leaders.

Establishing the content validity of a set of standards is merely the first step in the process. Next, proper and varied instrumentation should be developed to ensure the reliability of the assessments. Because dispositions tend to be somewhat amorphous, assessors may hold differing perceptions on what they are assessing. Consequently, all instruments must be extremely clear as to their definition of each disposition. From one assessor to another, there should be very little need for interpretation; this should help to ensure the stability of the assessments over time, allowing for them to measure any true growth (or decline) relative to the disposition, but greatly reducing the “noise” in the measurement process.

Multiple assessment instruments are advised (Lang & Wilkerson, 2006). There is a plethora of types of instruments to be selected from, e.g., scenario-based, portfolio rubrics, reflective journaling, interview rubrics, observation rubrics, Thurstone Attitude Scales, Semantic Differential Scales, focus groups, projective techniques, qualitative text analyses, constructed response methods, perceptual evaluation protocols, human relations incidents and subsequent interviews, biographical and metaphorical self-assessment, and paragraph completion methods. This allows for the various instruments to be checked for concurrent validity and reliability. It also enhances the reliability and validity of the overall assessment process.

In addition to the use of multiple instruments, the assessment of dispositions should be done from multiple perspectives, by multiple assessors. For this reason, self-assessment and feedback from peers, instructors, and mentors are recommended. Clearly, for the assessment of dispositions to be a learning process, it should be heavily based on the candidate’s reflection and self-assessment (Wasicsko, 1977; 2007). However, self-assessment often does not match the assessments of others; consequentially, for a candidate to continue to develop dispositionally, external feedback at various stages of the preparation program is essential. Such external feedback can come from colleagues specifically tasked with providing such feedback on the candidate’s dispositions, full cohorts or classes of peers, professors, and field-based mentors. This feedback may come from the use of common instruments, which can facilitate the triangulation of the assessors’ perspectives, or from instruments uniquely designed to suit the circumstances of each group of assessors to the candidate.

Any time multiple assessors are involved, regardless of how valid and reliable the instruments may be, attention must be paid to ensuring inter-rater agreement (reliability). This is best accomplished by careful training of each assessor on the use of the instrument(s) and trial evaluations of that assessor’s scores compared to the scores of other assessors evaluating the same candidate’s performance. This can be done by using sample video or audio tapes of candidates, portfolio assessments, or joint examination of survey instrument results. Such training and evaluation of assessors must be done prior to any formal assessment by that individual and must be periodically re-checked over time in order to prevent an assessor deviating from the group norms.

Messick (1989) introduced the concept of consequential validity, or the concern for the proper use of assessment results. The assessment of dispositions may be done for formative or summative purposes (see Scriven, 1991). When used formatively, the assessments should guide the candidate for further development of specific dispositions which are not yet sufficient to meet the institution’s standards. This calls for individualized dispositional growth plans and mentoring along the preparation program. It would violate Messick’s concept of consequential validity to fail to develop and monitor the implementation of such growth plans.

However, violation of consequential validity is more likely to occur in the summative uses of the assessment of dispositions. For example, dispositions should be assessed as part of the admissions process. Because at least some dispositions can be developed over time with proper circumstances and feedback, applicants who do not display the required level of dispositional development in the admissions process should have their admissions deferred, rather than rejected. When used summatively at the conclusion of the candidate’s program of study, or prior to admission to the field-based internship, candidates failing to demonstrate the required levels of dispositions should be provided with remediation opportunities rather than being summarily dismissed.

These processes tie closely to the court’s likely next concern – procedural due process. Institutions carry such responsibilities as advising applicants and candidates of the institutional expectations regarding specific dispositions, as well as of the processes by which these will be periodically assessed, from the admission process until program completion. They must be informed of how assessment data will be used in decision-making regarding the candidate and the potential consequences of those decisions. Similarly, after admission, candidates must be provided planned opportunities to develop those dispositions deemed to be their personal strengths and weaknesses. Finally, appeals processes regarding the assessment of dispositions must be available, without penalty, to the applicants and candidates.

The court’s likely final, and overriding, concern would be that no individual’s or group’s civil rights be violated in the assessment of dispositions. In short, institutions must ensure that all assessment practices are equitably implemented and are without bias toward any individual or group. For example, institutions are obligated to ascertain if their assessment practices, including the specific dispositions identified, are gender, race, culture, handicap, or sexual preference-neutral. Confidentiality of assessment results must be maintained among the program faculty and administration.

Although these challenges may seem daunting, they are by no means insurmountable. As the next section of this article highlights, there is a growing research base to support the efforts of institutions in assessing dispositions. In addition, the Internet provides access to much of the work that many institutions of higher education are doing in this regard, which may not yet be part of the refereed knowledge base.

In two recent studies, Lindahl (2009) and Melton, Mallory, and Green (2010) described the current state of teaching and assessing leadership dispositions as “inconsistent” across educational leadership programs in the United States. From two exploratory studies, one of 43 programs using a cross sectional survey design (Melton, et al.), and the other study of 35 programs using a qualitative design (Lindahl), the researchers found that most educational leadership programs have a formal procedure in place for teaching and assessing leadership dispositions, but definitions, practices, and utility of assessments vary across programs. Although the two studies provided a glimpse into a relatively small sample of the more than 600 educational leadership programs in the United States (Levine, 2005), the literature seems to suggest that university-based principal preparation programs are in early stages of wrestling with definitions, assessors, instruments, and methods of teaching and assessing dispositions (Schulte & Kowal, 2005; Wasonga & Murphy, 2007; Williams, 2009).

So, specifically, what did the two studies reveal about the current state of teaching and assessing leadership dispositions? First, the vast majority of leadership programs rely on either NCATE or ISLLC for the definition of dispositions and subsequent list of dispositions to be taught and assessed (Lindahl, 2009; Melton et al., 2010). Many programs add to the ISLLC list based on research, professional practices, beliefs, and values identified by program faculty, standards, or other resources, such as state codes of ethics. Therefore, although the researchers observed considerable overlap and congruence with the original 1996 ISLLC list of dispositions (Lindahl; Melton et al.), a major theme that emerged was the apparent lack of consensus about what leadership dispositions needed to be taught and assessed. Lindahl observed that there appeared to be considerable confusion over dispositions, identified as values and beliefs, and candidate professional expectations of punctuality, attendance, class expectations, and academic honesty.

There was also variance in how to teach leadership dispositions, with the most common teaching method identified as modeling through professional interactions with candidates. Other methods of teaching dispositions involved case studies and scenarios. The norm was that faculty members spend about 10% of their time devoted to teaching and assessing dispositions, disproportionate to their focus on skills, behaviors, and knowledge (Lindahl, 2009). Very few programs require that leadership dispositions be embedded in all course syllabi, and faculty discussions about dispositions are minimal (Lindahl).

For accreditation purposes, many approaches to assessing candidate dispositions have been developed, ranging from checklists completed by professors, to self-reported descriptions of candidate beliefs, to qualitative and subjective means (Chandler, 1998; Melton et al., 2010; Schulte & Kowal, 2005; Stahlhut & Hawkes, 1994; Wasicsko, 2000). The most common assessment tool was the checklist-type of instrument, which often was incorporated into the candidate portfolio (Lindahl; Melton et al.). Several programs utilize rubrics designed to assess candidate dispositions once the self-assessment and/or portfolio assignments have been completed.

Intervals of assessment of dispositions varied, with most leadership candidates being assessed at the beginning and end of their programs (Lindahl, 2009; Melton et al., 2010). Another common practice was to assess leadership dispositions prior to or while candidates were enrolled in field-based internships. Very few programs use disposition assessment for admissions decisions, and rarer still was the decision made to dismiss a student based on disposition data (Lindahl). Students who were counseled out of programs based on poor “dispositions” were actually students who were often late, absent from class, or delinquent in assignments, demonstrating weak professional practices, often referred to as professional dispositions, by reporting universities.

In whatever assessment methods were employed, the most frequent assessors of dispositions were the candidate’s advisor or the candidate himself or herself (Lindahl, 2009; Melton et al., 2010). Although self-assessment abounds, it does not necessarily do so in conjunction with other assessors. Some programs use a panel of faculty to assess candidate dispositions, whereas other less frequent assessors were identified as intern supervisors or program coordinators. A few programs employed services of field-based mentors in assessing dispositions.

Major problems identified with assessing dispositions were concerns about assessment methods, such as inter-rater reliability, instrument validity, labor intensity, and skepticism about authentic assessment of dispositions (Lindahl, 2009; Melton et al., 2010). Another concern was that educators who completed a leadership degree with appropriate dispositions of the profession tend to be re-socialized in their school or district culture, thereby losing their capacity to influence as guided by values and beliefs of the profession (Lindahl). Only 19% of programs indicated that their disposition assessment system was very useful (Melton et al.).

Some interesting qualitative approaches to disposition assessment were described in literature. For example, in one southeastern university, Martin (2009) described the system of assessment of dispositions as one that involved interviews and conferences with candidates at the midpoint transition point of their program. Candidates participated in a mandatory 30-minute session with an educational leadership faculty member to discuss readiness for leadership during the first internship. Although disposition conferences were identified as time consuming, they provided a unique opportunity for faculty to discuss a student’s tendency to act in a certain way. The disposition data from the mid-point transition were used to decide if candidates are ready for field-based internship based on their assessment of the candidates’ dispositions.

Employing a quantitative design, several assessment methods in various stages of development, implementation, and utility also appear to be promising approaches. The University of Nebraska uses the Administrator Dispositions Index (ADI) developed by Schulte and Kowal (2005), which requires beginning candidates in educational leadership to assess themselves. As candidates progress through their courses, leadership dispositions are directly and indirectly addressed. Concerns about self-assessment led faculty to develop a follow-up survey so that supervisors of practicum candidates could assess candidates’ readiness to be school leaders. By comparing the ADI and results of the follow-up survey as matched pairs, educational leadership faculty are able to use the data to guide decisions about candidates’ readiness to be school leaders.

At Florida Gulf Coast University, the Educational Leader Candidate Belief Scale (ELCBS), developed by Carter, Rea, Valesky, Wilkerson, and Lang (2010), is undergoing validity studies after initial confidence from the first pilot study of the instrument. The ELCBS consists of a series of 53 statements (8 to10 per performance expectation), designed to measure candidate leadership dispositions using a systemic sampling process of ISLLC performance expectations dispositions. In addition, Northern Kentucky University uses an assessment system that employs 360° feedback for candidates, and Georgia Southern University is in the final stage of piloting the School Leader Dispositions Inventory (SLDI)^©, an instrument designed to assess dispositions of school leaders in situ. The Northern Kentucky University and Georgia Southern University systems are discussed in greater detail in the following sections.

Although it is true that most organizations are feedback-poor environments, empirical research and anecdotal evidence have supported the premise that 360° feedback leads to improved performance in the areas that are being evaluated (Lepsinger & Lucka, 2009). A 360° assessment helps leaders compare how their constituents perceive them with how they perceive themselves. In other words, a 360° assessment system is a reality check. More importantly, it allows leaders to reflect on the results, analyze the data, and develop a growth plan to enhance their leadership capacity.

Belief in a 360° process is based on four assumptions: (1) feedback is important for personal and professional growth; (2) most organizations are feedback-poor environments; (3) when we look in the mirror, we see who we think we should be more then who we really are; and (4) empirical research and anecdotal evidence has shown that 360° feedback leads to improved performance in the areas that are being evaluated (Lepsinger & Lucia, 2009).

The 360° process can improve team effectiveness, facilitate cultural change, inform the performance appraisal process, identify individual or organizational professional development needs, assist in making selection decisions, and allow supervisors to intervene through coaching. Equally as important, if many people participate in the 360° process within an organization, it can also be used to strengthen the collective leadership capacity of the organization. A tipping point can be reached to create a culture that values leadership more as a process than a position (Chirichello, 2010). Leadership then becomes the collective activities of the leader and followers to set direction, build commitment, and create alignment. The focus is on a process that is valued more than a person who holds a position (Martin, 2007).

Northern Kentucky University’s (NKU) College of Education and Human Services uses the 360° process to assess the traits, skills, characteristics, and dispositions of its senior administrative leaders as well as its learning associates, the term NKU uses to refer to candidates who are enrolled in the college’s Ed.D. program. The dispositions section of the Individualized Leadership Self-Assessment^© has also been used to screen applicants for the doctoral program, the principal preparation program, and teacher education programs. In addition, it is used to screen applicants for faculty positions throughout the college.

The Individual Leadership Self-Assessment^©, or ILSA, was designed to assess one’s capacity to lead. It can be used by aspiring and practicing leaders across professions. The instrument includes seven subsections: 1) traits, 2) skills, 3) characteristics, 4) perception of self, 5) perception of others, 6) perception of purpose, and 7) frame of reference. Subsections four through seven (i.e., perception of self, perception of others, perception of purpose, and frame of reference) are identified as dispositions. Respondents rate themselves and critical friends rate the respondents using a Likert scale from “1” to “7.” Each of the choices on the Likert scale is defined in each subsection using a common rubric.

During the introductory course in the Ed.D. program at NKU, conversations that focus on building leadership capacity begin to create a culture in which self-assessment is valued. Subsequently, the ILSA is completed by each of the learning associates at the end of the first semester, at the beginning of year two, and near the completion of the program. Each learning associate identifies no more than ten critical friends who are invited by the program director to complete the ILSA twice, once at the end of the first semester and once near the completion of the program. The subsections for the learning associates and critical friends are identical. During the first semester of the second year, each of the learning associates completes the ILSA for the other learning associates in her or his cohort.

The learning associates and the critical friends respond to the items on the ILSA using an on-line survey. The mean scores for each item from the critical friends and other cohort members are compared to the candidate’s own score for each item. After examining and discussing the results with faculty and other learning associates, each learning associate develops an Individualized Leadership Development Plan (ILDP).

During the fall semester, 2010, a reliability test was completed using Cronbach’s alpha to measure internal consistency of the ILSA. The results ranged from .899 to .944 for each of the subsections of the instrument. Faculty continue to review current research to ensure the validity of the elements in each subsection in the context of 21^st Century leadership traits, skills, characteristics, and dispositions.

The use of the 360° process at Northern Kentucky University’s College of Education and Human Services has assisted the faculty in the admissions process for the doctoral program, the principal preparation program, and teacher education programs. It has assisted search committees in narrowing the applicant pool for faculty positions throughout the college. In addition, faculty have reported that learning associates have shared how the system has given them new insights into their traits, skills, characteristics, and dispositions that went beyond their own self-perceptions.

Faculty in the educational leadership preparation program at Georgia Southern University (GSU) shared the concern that assessment of dispositions was problematic, especially when a system relied solely upon self-reported data or data reported by a single observer. More specifically, when professors of educational administration were surveyed, these concerns over assessing leadership dispositions focused on what dispositions should be assessed and whether such assessment systems could be valid (Melton, Mallory, & Green, 2010).

Attempting to address the issue of which dispositions to assess, a group of faculty members at Georgia Southern University constructed a list of 14 educational leadership dispositions based on the theoretical underpinnings of the work of Burns (1978) and McGregor (2006) – in other words, a set of dispositions aligned with transformational leadership theory. With the dispositions identified, they set about developing a method to assess those dispositions.

Finding fault with the single-observer or self-reporting instruments, the GSU faculty considered a number of other options, including a 360° system and a scenario-based instrument that utilized open-ended responses by the candidates. They decided that the 360° approach is limited by the kind and extent of experience that raters have with the administrator being observed. And, they also realized that analyzing reams of qualitative data from open-ended responses to scenarios was not practical as an on-going assessment system. In order to capture their candidates’ leadership dispositions in a way that would more closely emulate a field setting as opposed to a questionnaire, they stayed with the scenario-based system; but, they designed a set of responses for each of the scenarios, with each response designed to reflect an inclination toward one type of leadership approach. Hence, the creation of the School Leader Dispositions Inventory©, or SDLI(Reavis, Green, Mallory, & Melton, 2010), which is designed to assess the dispositions of school leaders in situ.

The instrument consists of 15 cases, with each case comprised of a situation and four possible responses a leader might make to each situation. The cases reflect real-life situations that have no single solution. After reading and reflecting upon a given situation, the participant indicates the degree of agreement or disagreement with each of the stated responses. Each of the four responses is aligned with either Theory Y, Theory X, Soft X, or Pseudo Y leadership constructs. For each of the four responses to each case, there is a scaled response ranging from “0” to “4,” with “0” indicating “not an option” and “4” indicating “I strongly agree with this course of action.” The average administration of the instrument takes 45 minutes, and it can be machined scored. A candidate’s responses to all of the items results in a leadership disposition profile as aligned with one of the four different leadership theory constructs.

The Theory Y profile is grounded in the work of McGregor (2006, original work published in 1960), which was later elaborated upon by “transformational leadership” theorists (Bass, 2008; Burns, 1978). The Theory Y school leader believes in the individual dignity and worth of each member of the organization. Thus, these leaders make wide use of shared decision making and distributive leadership; their approach to management emphasizes the development human resources within the school, person by person. A Theory X profile, on the other hand, is grounded in McGregor’s Theory X assumptions, and what Burns (1978) called “transactional leadership.” The Theory X school leader believes that most teachers and staff require constant and close direction because of they lack motivation, and they assume that most will avoid responsibility. The Soft-X school leader will demonstrate leadership behavior that is consistent with Theory X assumptions about the people in the school community. However, the Soft-X leader prefers to use persuasion and compromise rather than overt coercion to manage. Also know as Theory X, Pattern B (Argryis, 1971), basically, it is Theory X leadership with a soft touch. And finally, the Pseudo-Y school leader is a Theory-X school leader who attempts to look like a Theory-Y school leader. Also known as Pseudo-transformational leaders (Bass, 2008; Price, 2003), these school leaders, indeed, show the outward signs of being transformational or adhering to Theory Y assumptions. However, they use the illusion of a Theory Y approach as a means to achieving their self-interest.

In Phase I of the pilot study for the SDLI (Melton, Mallory, Tysinger, & Green, in press), the instrument was administered to 48 educational leadership candidates enrolled in five courses in an Ed.D. program. To assess the reliability of the instrument, the total scale was measured in the form of Cronbach's alpha, resulting in a coefficient of .85. Additionally, the follow-up items were clustered according to their alignment with Theory X, Theory Y, Soft-X, and Pseudo-Y for further reliability analyses. Results from the correlation matrix revealed no issues with multi-collinearity. For the item means, the average response showed more of a trend toward Theory Y and a low trend toward Pseudo Y. The overall item mean of Theory X items was 1.6. Theory Y items had a mean of 2.74. The mean for Soft-X items was measured at 1.95. Finally, the mean for Pseudo-Y items was 1.22.

In the second phase of the validation study, an additional 38 cases comprised of principals from California, Florida, Georgia, Missouri, North Carolina, South Carolina, Tennessee, and Texas, were added to the sample. These additional cases raised the Cronbach’s alpha for the SLDI to .90, with the probability of going higher with the deletion of some items. A third phase is planned in order to establish concurrent validity using the Multifactor Leadership Questionnaire^© (Avolio & Bass, 1995) and to measure test/re-test reliability.

Although the SDLI is still in the final stages of development, it promises to be a valid and reliable instrument for candidates in educational leadership programs to observe how their leadership dispositions align with current leadership theories. Moreover, it can be used in pre-test/post-test administrations for observing how leadership preparation programs influence the leadership dispositions of candidates.

The Greek historian Pausanias (trans. 2009) recorded that the maxim “Know thyself” was inscribed in the forecourt of the Temple of Apollo at Delphi. It is worth noting that Temple of Apollo at Delphi was the place where knowledge was venerated. Thus, it was at that very location that the ancient Greeks saw fit to remind themselves that the pursuit of knowledge begins with self-knowledge.

What the Greeks knew then holds true today. The development of leadership capacity for school administrators begins with self-knowledge. Certainly, a school leader’s self-knowledge must include an honest appraisal of the necessary technical skills. Moreover, self-knowledge requires an examination of one’s competence in the academic disciplines that form the basis for understanding leadership practice. But, leaders also need self-knowledge of their assumptions about the people they are leading. They need self-knowledge of their personal and professional values. They need self-knowledge of their attitudes, their beliefs, their worldviews. As elusive as these qualities might prove to be, professors of educational administration must acknowledge that assessing leadership dispositions is an incontestable part of preparing leaders of educational organizations. It follows that programs in educational leadership need to continue to refine their systems to ensure that the feedback given to candidates is valid, reliable, and relevant to their formation as school leaders.