This module has been peer-reviewed, accepted, and sanctioned by the National Council of the Professors of Educational Administration (NCPEA) as a scholarly contribution to the knowledge base in educational administration.This article is from the NCPEA Education Leadership Review, Volume 7, Number 2 August 2006 issue and authored by Jimmy Byrd, Celia Drews, and Jeanie Johnson.In an occupation enjoying very little security (Cooper, Fusarelli, & Carella, 2000), having fewer benefits than similar jobs in the private sector, and facing increased criticism in addition to greater complexities (Goodman & Zimmerman, 2000), the role of the school superintendent could be characterized as formidable. Due to the realities of the job, current superintendents are concerned about the future possibilities of quality candidates to fill superintendent vacancies (Cooper et al., 2000). In recent years, public school superintendents have faced increased demands from the No Child Left Behind Act of 2001 (U.S. Department of Education, 2002) and stringent state accountability standards. Yet, academic improvement does not happen by chance but rather through effective leaders having ample time to implement broad, sustainable reform (Fullan, 2002). By supporting campus-level leadership, the superintendent is ultimately responsible for the success or failure of student performance. Twenty-first century superintendents must have skills to augment instructional methods, in addition to interpret assessment data as well as explain their district’s achievement level compared to others in the state and nation (Hoyle, Bjork, Collier, & Glass, 2005). Hoyle and colleagues (2005) expressed that the success or failure of various superintendents (length of tenure) is a subject that is ambiguous and not thoroughly researched. Adding to the ambiguity, Gardner’s (1990) statement holds true, “Despite length of tenure, one thing is certain, for good or bad, the system will survive the superintendent” (p. 12). According to Cooper and coauthors (2000), the public perception of the superintendency is that of a job so daunting, few individuals desire to pursue the challenge. Given the challenges of the job, one pressing question is what are the factors that lead to superintendent turnover in the field? There has been much speculation by media about superintendent tenure and turnover; however, very little quantitative research exists detailing that the characteristics of superintendent tenure. Although the American Association of School Administrators (AASA) has conducted a national superintendent survey each decade, claiming to represent one in every five superintendents, in the most recent AASA survey only 11% of respondents represented the Southwest regions, with no identification of participants by state (Glass, Bjork, & Brunner, 2000). The current study attempts to add to the literature by revealing characteristics of the Texas superintendency and contributing factors to the length of tenure and rate of turnover.

Review of Superintendent Tenure Literature

Previous Studies on Superintendent Tenure Natkin, Cooper, Alborano, Padilla, and Ghosh (2002) have provided the most recent quantitative study on survival of the superintendency. Focused on the longevity of 292 superintendents from North Carolina and random districts across the U.S., these researchers found that superintendent turnover averaged 6 to 7 years, regardless of the district’s size or location. Factors significantly related to superintendents’ longevity in office were the extent of school board involvement in management, support for needed construction, consolidation of school systems, district poverty level, and superintendent’s post-graduate education. The research indicated that superintendent tenure had not markedly increased since 1975, and that superintendent turnover was not as serious an issue as once perceived. Despite this result, Natkin et al. revealed that when combined, high poverty of students enrolled in the district, minimal support for construction of new facilities, and micromanagement by school governance lead to shorter tenure. In 2003, the Council of Great City Schools (GCS) reported results of a survey conducted with member districts. Average tenure for urban superintendents was reported to be only 2.75 (up from 2.5 in 2001), but mean tenure for the immediate past GCS superintendents averaged over 4 years. Supporting GCS findings, the Council of Urban Board of Education (CUBE) reported the tenure of urban superintendents between 4 and 5 years (National School Board Association, 2002). Unequivocally, the most comprehensive study about superintendent characteristics is the AASA 2000 study of the American school superintendent (Glass et al., 2000). The 2000 AASA survey sampled 2,262 superintendents; average tenure of superintendents was estimated to be between 5 and 6 years, slightly lower than the previous survey in 1992. Glass and coauthors alleged the difference between the two surveys was due in part to new superintendents entering the field, maintaining that superintendent tenure has remained relatively static over the past 30 years. In fact, many superintendents have served more than half of their career in one district. Important to note is the difference between average tenure for urban superintendents and other superintendents. As revealed in the literature, urban superintendents have historically experienced shorter tenure than other superintendents. Because the role of the superintendent is so diverse for various reasons, geography and size being only two, length of tenure varies.

Superintendent as Instructional Leader Farkas, Foley, and Duffet (2001) found that more than one half of superintendents listed the most daunting task faced in the job is that of increasing student achievement. Additionally, 41% of school boards identified raising student achievement as a primary mission. Effective superintendents are recognized as vital to the success of a district’s improvement efforts (Forsyth, 2004). Although Byrd’s (2001) study revealed no correlation between superintendent leadership style and student achievement, the managerial role of superintendents had a significant effect. Byrd suggested that superintendents must increase lines of communication among stakeholders and allow for autonomy at the campus level in order to have a positive impact on student academic achievement. Superintendent as instructional leader differs from that of principal in that superintendents are responsible for regulating the overall capacity of the school system (Hoyle et al., 2005). Furthermore, superintendents of successful districts adopt a hands-on approach in regard to instructional matters (Cuban, 1984). These same superintendents utilize managerial influence over the behaviors of principals and teachers, in turn directly impacting student learning and achievement (Cuban, 1984; Hoyle et al., 2005). Managerial influence includes thoughtful staff selection and recruitment, clearly articulated mission and goals in regard to curricular issues, as well as financial planning that supports instruction (Hoyle et al., 2005).

Superintendent and School Board Relationships While instructional leadership is integral to the role of superintendent, the increasingly complex political aspects of the job must be handled as well (Education Writers Association, n.d.; Hoyle et al., 2005). Superintendent relationships with school boards were found to be a decisive element of superintendent tenure (Education Writers Association, n.d.). Often, conflict with the school board is cited as a common reason for superintendents leaving a district and hence their attrition (Rausch, 2001). Allen (1998) observed that superintendents listed the relationship with the board as a second reason for involuntary non-extension of a contract, while board members listed relationships with the superintendent as the major cause. Despite conflicts, Glass and co-researchers (2000) surmised that the school board and superintendent must work together to connect the school district with the needs of the community (Goodman & Zimmerman, 2000). While many school boards and superintendents described having mutually cooperative relationships (Foley, et al., 2001), Farkas, Johnson, Duffett, and Foleno (2001) reported that 65% of superintendents speculated that many school boards simply wanted leaders the board could control. Furthermore, over 80% of superintendents have reported feeling frustrated with politics and bureaucracy of the job (Farkas, et al., 2001; Farkas et al., 2003). A primary source of superintendent frustration stemmed from school boards micromanaging or interfering in superintendents’ administrative responsibilities (Harvey, 2003), with more than two-thirds of superintendents stating that their board meddled in issues not within the scope of its responsibility. According to Goodman and Zimmerman (2000), a quality working relationship between effective leaders and school boards is a “key cornerstone of the foundation for high student achievement” (p. 1). As local school boards are the sole evaluators of superintendent performance and renewal of contracts, a quality working relationship with members also directly influences the tenure of the superintendent. Superintendent Preparation and Education An assumption can be made that superintendents who are neither fully prepared nor well trained are prone to experience difficulties in their leadership role. A majority of superintendents reported that the average supervisory leadership programs in university schools of education were not aligned with the actualities of what is needed to effectively lead today’s public school systems (Farkas, et al., 2001; Farkas et al., 2003). Farkas, Johnson, and colleagues (2001) also stated that nearly one half of superintendents surveyed believed that revamping preparation programs would be very effective in improving school leadership. Superintendents identified the weaknesses of preparation programs as insufficient hands-on application, inadequate access to technology, and poor linkages of content to practice. Glass et al. (2000) further commented that the increased number of changes in school systems necessitated parallel changes in educational administration and supervisory leadership preparation programs. Thus, a focus on the increased challenges facing the superintendency today that include securing adequate financing to meet additional mandates, student assessment and analysis of school and community data, increased state and federal accountability, as well as instructional innovations for increasing student achievement would constitute a solid first step for preparation programs to take in endeavoring to reconnect with the field of educational administration (Hoyle et al., 2005). Educational level attained by superintendents is an additional factor impacting tenure and turnover. In the AASA 2000 survey, 45% of superintendents had obtained doctorates, 89% of which were in educational administration or supervision (Glass et al., 2000). Yet, contradictions are apparent in recent studies connecting superintendent tenure and level of education. In a study of Texas superintendents, Largent (2001) reported that short tenured superintendents were more likely to hold doctorates. In contrast, Natkin et al. (2002) presented evidence suggesting a strong correlation between tenure and level of education. The researchers established that median tenure was lengthened roughly one year for each level of education acquired.

Superintendent Pressures on Multiple Fronts Most superintendents agree that current issues schools face are similar to those confronted in years past, but not in size or complexity (Orr, 2002). In today’s American public schools, superintendents must guide challenging, dynamic education systems, while appropriately responding to social and political pressures (Rohland, 2002). In addition, Rohland speculated that the high standards and people-intensive nature of school districts are primary reasons the job of superintendent is so demanding. Similar to other professions, ascension on the career ladder in education is associated with increased exposure to criticism (Jazzar & Kimball, 2004). Fullan (1998) opined that due to the complex nature of executive leadership itself, there will always be dissatisfaction among constituents with respect to the leader’s performance. If the number of teachers were multiplied times the number of students, parents, and community members, the possibilities for conflict and outside pressures are endless (Parker, 1996). Success for the superintendent lies in gleaning wisdom from attacks and criticism, without being defeated in the process (Harvey, 2003).

Additional Factors Affecting the Superintendency Time, one of the superintendent’s most valuable resources, can quickly be exhausted by special interest groups’ demands and community pressures (Glass et al., 2000; Harvey, 2003). According to the Colorado Association of School Executives, (CASE) (2003) the role of superintendent is labor intensive, often requiring 80 or more hours a week. Glass and colleagues (2000) found evidence to support the widely-held belief that the job of superintendent has become increasingly complex, with salary and benefits insufficient for the level of responsibility and accountability demanded. However, superintendents polled by Cooper and his colleagues (2000) have surmised that improved pay and benefits would possibly attract and retain more qualified individuals in the superintendent profession. In regard to superintendent self-perception of effectiveness, lack of fiscal resources was cited as a major reason for inhibiting superintendent effectiveness (CASE, 2003) and for explaining why superintendents are leaving the profession (Glass et al., 2000). In the AASA survey (Glass et al.), superintendents described efforts to obtain sufficient fiscal resources as a never-ending struggle. Too many insignificant demands from various stakeholders and compliance with increased state-mandated reforms was also provided by superintendents as a key factor in hindering superintendent effectiveness. Reports of low superintendent tenure, some as low as 2.5 years (Natkin et al., 2002), have contributed to negativity and a sense of crisis (Cooper et al., 2000) surrounding the superintendency. Nevertheless, existing research does not definitively identify specific factors contributing to superintendent tenure and turnover. The success or failure of various superintendents in the field is a subject that is unclear (Hoyle et al., 2005). Therefore, the purpose of this study is to determine factors or combination of factors contributing to length of tenure and rate of turnover among public school superintendents. Method of Inquiry

Superintendents as Participants Participants in the study consisted of 141 Texas superintendents representing public school districts during the 2004-2005 academic year. Districts ranged in size from small rural districts, with under 200 students enrolled, to urban school districts with multiple campuses enrolling over 30,000 students. Of the superintendents surveyed, 130 were White (92.2%), 6 were African American (4.3%), and 5 Hispanic (3.5%); 124 (87.9%); participants were mostly male with 17 (12.1%) female. Of total participants, 15 (10.6%) held a master’s degree only, 71 (50.4%) held a master’s degree plus additional coursework, 42 (29.8%) held an Ed.D., 12 (8.5%) had earned a Ph.D., and 1 (.7%) held a Doctorate of Juris Prudence. Participants ranged from 32 to 69 years of age, with the average age being 53 (SD= 6.05). Further, salaries ranged from $60,000 to over $100,000, with just under one-half of participants earning over $100,000. Average tenure for all participants was 5.0 years (SD=3.75). Males had an average tenure of 4.8 years (SD=3.67), while 6.2 years (SD= 4.2) was the average for females. Tenure between superintendents who had changed districts within the past 5 years and superintendents who had been in the same district for 5 or more years varied greatly. For superintendents having changed districts within the past 5 years, tenure averaged 3.2 years (SD= 1.57), while superintendents remaining in the same district for 5 or more years, tenure averaged 10.2 years (SD=3.26). Nationally, superintendent tenure averaged between 5 and 6 years in one study (Glass et al., 2000), and between 6 and 7 years in another (Natkin et al., 2002). Development of the Research Instrument The Texas Superintendent Survey (TSS), consisting of 29 items, was developed by compiling various items from previously administered superintendent surveys (TSS can be accessed at http://www.troyisd.org/education/components/docmgr/default.php?sectiondetailid =178&fileitem=107&catfilter=67). Sources used in the development of the survey instrument included the CASE (2003) survey, which credited the AASA 2000 (Glass et al.) national survey instrument for many of its questions. Additionally, a survey developed by Beach and Reinhartz (1990) was used. Other items on the questionnaire were developed by the authors based upon a thorough review of the literature. Before survey completion, university professors and current, nonparticipating superintendents reviewed the survey and made suggestions that enhanced content validity. Variables Examined Dependent variable. The dependent variable for this study of superintendents was the length of tenure. Tenure was defined as the length of time in the position of superintendent in a single public school district. Superintendents reported the number of years and months superintendents had served in the current position at the time the survey was completed. Independent variables. Independent variables displayed in Table 1 included characteristics examined as possible influences of superintendent tenure, such as age, salary, highest educational level obtained, reported level of preparedness, school board and superintendent relationships, politics and bureaucracy impacting the role of superintendent, and fiscal resources for the district. Table 1 Definition, and Measures of Independent Variables Examined in the Study Variable Definition Measure GenderSuperintendent SalaryLevel of EducationLevel of PreparednessRelationship with the school board president Level of school board involvement in superintendent businessPolitics and Bureaucracy Fiscal Resources in the District The dollar value of superintendent compensation.Highest degree obtained from a college or university.Self-rating of preparedness for the superintendency.Described in degree of difficulty of working together.Described in degree of difficulty of working together and delineation of roles between the two.Described in contribution to instability in profession, time spent, and frustrations.Money allocated for all aspects of management in the district (Categorical Variable)1 = female 2 = maleListed in $5,000 increments(Categorical Variable)2 = $60,000 – $64,9993 = $65,000 – $69,9994 = $70,000 - $74,9995 = $75,000 - $79,9996 = $80,000 - $84,9997 = $85,000 - $89,9998 = $90,000 - $94,9999 = $95,000 - $99,00010 = over $100,000List of degrees(Categorical Variable)1 = Bachelor’s Plus2 = Master’s3 = Master’s Plus4 = Ed.D.5 = Ph.D(Categorical Variable)4 = Very Prepared3 = Prepared2 = Somewhat Prepared1 = Not Prepared(Categorical Variable)Likert scale from 1 to 5 with1 = least difficult through5 = most difficult(Categorical Variable)Likert scale from 1 to 5 with1 = least difficult through5 = most difficult(Categorical Variable)Ranked from 1 to 6, with 1 being the top contributor and 6 being the least.(Categorical Variable)(Categorical Variable)Likert scale from 1 to 5 with1 = least difficult through5 = most difficult

Procedure for Research To identify superintendents for this study, the authors accessed a state database listing superintendents’ positions with each district in which they had been employed for the past 5 years. Superintendent and the district for which a superintendent had been employed 5 years ago were located first. The second step involved determining if the superintendent was currently serving as superintendent in that same district or elsewhere within the state. To determine if an individual was currently a superintendent, a Texas Education Agency (TEA) database of current superintendents was searched. Superintendents still in the TEA superintendent database were contacted at the new district in which they were employed. Superintendents not found in the current TEA database were assumed to have retired, left the superintendent profession, or moved out of state. From the original database, a total of 176 superintendents were identified to have changed districts within the past 5 years. All 176 superintendents found to have changed districts while remaining in the superintendency were included in this study. A survey requesting information about the superintendent, and previous position and district, was mailed to each superintendent. Researchers used a similar process to identify superintendents that remained in the same district for 5 or more years. Of the 233 superintendents remaining in the same district for 5 or more years, 60 were randomly selected to participate in the study by using a table of random numbers (Patten, 2004). Each of the 60 superintendents was mailed a survey requesting information about the superintendent and current district. Of the 176 surveys mailed to superintendents changing districts within the 5-year timeframe, 105 completed surveys were returned, for a return rate of 60%. Of the 60 surveys mailed to superintendents employed in the same district for 5 years or more, 36 completed survey instruments were returned, resulting in a 60% return rate. In totality, 235 superintendents were selected to receive surveys. Of the 235 surveys mailed, 141 useable surveys were returned for an overall response rate of 60%.

Data Analysis Descriptive statistics and survival analysis modeling time-to-event status were utilized. Specifically, Cox regression was most applicable for modeling time-to-event status with the data collected on the TSS. The use of Cox regression allows covariates to be included in the prediction of superintendent tenure, while managing censored cases appropriately, in addition to providing coefficients for each covariate (SPSS, 2004). Analysis of data began by entering variables shown in Table 1 as independent variables into a Cox regression equation, with superintendent time in current position as the dependent variable. Measures of censored time-to-event data are not usually normally distributed. Censored data were therefore analyzed and reported via frequency distributions. Results of Research Descriptive Findings The TSS asked participants to rate difficulty in working with a variety of individuals on a scale of “1” for least difficult to “5” for most difficult. Of mobile superintendents, 32.7% rated working with school board presidents as somewhat difficult to most difficult, with only 13.9% of stationary superintendents responding similarly. Average tenure among participants was found to decrease as the level of difficulty working with the board president or board members increased. Other information provided by mobile superintendents indicated that difficulties with board members or characteristics of the board precipitated their decision to leave the district. The participants ranked six items in terms of contribution to instability in profession. While 38.5% of superintendents changing positions and 47.2% of superintendents remaining in the same district for 5 or more years rated superintendent/board relations and communication as the number one contributor to instability in the profession, increased politics had a statistically significant correlation with superintendent tenure. Fifty-five percent of tenured superintendents and 76% of superintendents changing districts during the timeframe of the study ranked increased politics in the profession as the number one or two contributing factor to instability in the profession. Average tenure among participants was found to have decreased as ratings of the role politics played in contributing to instability in the profession increased. Participants were asked to provide information about time spent in an average week. Statistically significant differences between groups on time spent in working with the school board (t = 1.99, p< .048) and community relations (t = 3.01, p< .003) were revealed. Superintendents new to the district spent more time establishing relationships with school board members and community groups. As these superintendents have not been in a district as long as stable superintendents, it seems logical that more time must be expended in building relationships. Of participants providing additional comments, extracurricular activities required many hours of attention each week from both groups. In regard to tenure, 26 (72.2%) superintendents who have remained in the same district for 5 or more years reported the current superintendency as their first. Additionally, 20% of superintendents changing jobs within 5 years reported to have left the previous district because of inadequate support from the board, while 62.5% left for better opportunities which became available elsewhere. Furthermore, the results indicated that after a superintendent reaches 5 years or more of service in a district, the average tenure increases significantly. Survival Analysis The results of the Cox regression reported in Table 2 revealed that working with the board president, not being able to get decisions made at the board level, and superintendent/board relations were statistically significant factors in determining the length of tenure among Texas public school superintendents. The results revealed that as the level of difficulty increased by a factor of one (scale 1 through 5) between superintendent and board president’s working relationship, the odds of a superintendent staying in the same district decreased by 22.2% (p = .003). Superintendents frustrated about not being able to work with the board to make decisions were 1.3 times more likely to leave the position when compared to those who maintained a cooperative relationship with the board (p = .019). Further, communication between board members and superintendents was a contributing factor that impacted tenure. As superintendents’ ratings of difficulty regarding superintendent-board relations and communication as contributing to instability in the profession (scale 1-6) increased, the odds of the superintendent staying in the same district decreased by approximately 10% (p = .048). Table 2 Results of Cox Regression Analysis Regarding Length of Tenure Among Texas Superintendents B SE Wald df Sig. Exp(B) Working with Board President -.251 .086 8.568 1 .003 .778 Not getting Decisions Made at the Board Level .171 .073 5.459 1 .019 1.187 Superintendent /Board Communication -.105 .072 2.089 1 .048 .901

Non-significant Factors Approximately one-half of superintendents changing districts within 5 years and superintendents maintaining their current position at least 5 years reported salaries of $100,000 or more. Educational attainment was asked of both categories of participants. Although level of educational attainment was a factor in other studies (Natikin et al., 2002), there was little difference between mobile and stationary superintendents in regard to highest degree obtained and tenure. Although finances were reported as a key factor impacting superintendent tenure in national studies (CASE, 2003; Glass et al., 2000), fiscal resources did not significantly impact superintendent tenure, even though results of the survey showed superintendents with less experience in the district spending more time each week on finances than superintendents who had been in the same district for at least 5 years. Discussion Results of this study targeting superintendent turnover revealed the importance of a good relationship between the superintendent and school board, primarily the school board president. As cited in the review of literature, superintendent relationships with school boards were found to be a vital factor in superintendent tenure (Education Writers Association, n.d.). Additionally, conflict with the board is often a reason superintendents leave a particular district (Rausch, 2001) and a top reason for involuntary non-extension of superintendent contracts (Allen, 1998). Akin to those reports, our study results reveal that the relationship between the superintendent and the school board president is statistically significantly in impacting superintendent tenure. Increased politics was revealed to have a statistically significant impact on superintendent tenure. Our findings support previous research reporting a majority of superintendents feeling frustrated with politics and the bureaucracy of the job (Farkas et al., 2001; Farkas et al., 2003;). Among the superintendent participants providing additional information in our study, all agreed that a primary contributing factor to instability in the profession stemmed from the legislature making educational decisions along party lines. Additionally, apathy in the legislative process was reported to be frustrating. The average age of participants was 53 years of age, similar to the national survey’s average age of 52 (Glass et al., 2000). Akin to Parker’s (1996) report, this study also revealed that many superintendents were seeking retirement after the currently held position. Paramount to the future of the superintendent profession in Texas is that based on responses to the TSS, nearly one-half (45%) of those participating will be retiring within the next 10 years. This is especially disquieting considering that during the last 5 years, among the 1029 Texas public school superintendents, 570 superintendents (55.4%) have left the profession while 459 (44.6%) are still working in it. Of those 459, only 283 (27.5%) have been in the same position during the past 5 years while 176 (17.1%) have changed jobs at least once during the timeframe of the current study. Our research reveals that superintendents who had been in a district for 5 or more years experienced an average tenure of over 10 years. Results of this study support the findings of others (e.g., Natkin et al. (2002); Glass et al., 2000), that superintendent tenure is much longer than the 2 to 3 years often reported by the media.

Conclusion In effect, this study contradicts the media hype regarding short superintendent tenure and rapid turnover. Overall, the results y support the majority of research on superintendent tenure, which has shown superintendent tenure to be roughly 5 to 6 years. Importantly, our research contributes an empirical study to the knowledge base. Research on district leadership and factors contributing to turnover have nationwide implications. Due to a paucity of research on superintendent tenure in general and superintendent tenure in the southwest in particular, a need to clarify any differences or similarities to nationwide studies was evident.

Future Research Because the relationship between the school board president and the superintendent is vital in determining superintendent tenure, studies of board presidents and other members are areas for future research. With the relationship between school boards and superintendents being a multifaceted concept, improved relationships between school board presidents and superintendents could lead to longer tenures and possibly a better focus on the importance of what really matters in schools—student academic achievement. Level of board involvement in managing the school district and the politics involved was found to be increasingly troubling to many superintendents. Reasons for board involvement in addition to the possible consequences on superintendent tenure are worthwhile to know. Additionally, the topics of fiscal resources and fiscal management in regard to superintendent tenure warrants further investigation. References Allen, P. R. (1998). Reasons for involuntary nonextensions of Missouri public school superintendent contracts in 1996. Dissertation Abstracts International, 59(9), 3292A. Beach, D., & Reinhartz, J. (1990, Spring). Endangered administrators. Texas Study of Secondary Schools, 46, 51-56. Austin, TX : Texas Association of Secondary School Principals. Byrd, J. K. (2001). Effective superintendent leadership strategies and management techniques for improving student performance as perceived by superintendents in selected school districts in Texas. Dissertation Abstracts International, 62(7), 2294A. Colorado Association of School Executives (CASE). (2003). The view from the inside: A candid look at today’s school superintendent. Engelwood, CO. Retrieved June 29, 2005, from www.co-case.org/displaycommon.cfm?an=&subarticlenbr=191%20. Cooper, B. S., Fusarelli, L. D., & Carella, V. A. (2000). Career crisis in the superintendency? The results of a national survey. Arlington, VA. American association of School Administrators. Retrieved July 5, 2005, from http://www.eric.ed.gov.zeus.tarleton. edu:81/ERICDocs/data/ericdocs2/content_storage_01_0000000b/80/21/e3/aa.pdf Council of Great City Schools. (2003, October). Superintendents: Characteristics, tenure, and salary. Urban Indicator, 7(1). Retrieved June 30, 2005, from http://www.cgcs.org/reports/ Cuban, L. (1984). Transforming the frog into a prince: Effective schools research and practice at the district level. Harvard Educational Review, 54(2), 129-151. Education Writers Association. (n.d.). Effective superintendents, effective boards: Finding the right fit. Washington, D.C. Retrieved June 29, 2005, from www.ewa.org/library/2004/leadership.pdf Farkas, S., Foley, P., & Duffett, A. (2001). Just waiting to be asked. New York: Public Agenda. Retrieved June 28, 2005, from http://www.publicagenda.org/research/pdfs/just_waiting_to_be_asked.pdf Farkas, S., Johnson, J., & Duffett, A. (2003). Rolling up their sleeves. New York: Public Agenda. Retrieved June 28, 2005, from http://www.publicagenda.org/research/pdfs/rolling_up_their-sleeves.pdf Farkas, S., Johnson, J., Duffett, A., & Foleno, T. (2001). Trying to stay ahead of the game. New York: Public Agenda. Retrieved June 28, 2005, from http://www.publicagenda.org/research/pdfs/ahead_of_the_game.pdf Forsyth, J. (2004, December). Does district leadership really matter? School Administrator. Retrieved July 5, 2005, from http://search.epnet.com.zeus.tarleton.edu:81/login.aspx?direct=true&db=tfh&an=15199836 Fullan, M. (1998). Leadership for the twenty-first century. In The Jossey-Bass Reader on Educational Leadership (pp. 156-163). San Francisco: Jossey-Bass. Fullan, M. (2002). The change leader. Educational Leadership, 59(8), 16-20. Gardner, J. W. (1990). On leadership. In The Jossey-Bass Reader on Educational Leadership (pp. 3-12). San Francisco: Jossey-Bass. Glass, T., Bjork, L., & Brunner, C. (2000). The study of the American school superintendency 2000. A look at the superintendent of education in the new millennium. American Association of School Administrators. Arlington, VA. Retrieved july 1, 2005, from http://www.eric.ed.gov.zeus.tarleton.edu:81/ERICDocs/data/ericdocs2/content_storage_01/0000000b/80/10/d6/e4.pdf. Goodman, R. H., & Zimmerman, W. G., Jr. (2000). Thinking differently: Recommendations for 21st Century school board/superintendent leadership, governance, and teamwork for high student achievement. Washington D.C: Education Commission of the States. Retrieved July 3, 2005, from http://www.nesdec.org Harvey, J. (2003, February). The urban superintendent: Creating great schools while surviving on the job. Orlando, FL: Council of Great City Schools,: Retrieved July 13, 2005, from www.cgcs.org/pdfs/UrbanSuptReport.pdf Hill, P. (2003, December). Smaller target, better job. Hoover Institute weekly essays. Retrieved July 6, 2005, from www-hoover.Stanford.edu/pubaffairs/we/2003/hill12.html Hoyle, J. (2002, September). The highest from of leadership. The School Administrator. Retrieved July 6, 2005, from http://staging.aasa.rd.net/publications/saarticledetail.cfm? ItemNumber=1785&snItemNumber=950&tnItemNumber=951 Hoyle, J., Bjork, L., Collier, V., & Glass, T. (2005). The superintendent as CEO. Thousand Oaks, Ca: Corwin. Jazzar, M., & Kimball, D. (2004, February). Lonely at the top. The School Administrator. Retrieved July 1, 2005 from http://staging.aasa.rd.net/publications/saarticledetail.cfm?ItemNumer=950tnItemNumber=951 Johnson, S. M. (1996). Leading to change. In The Jossey-Bass Reader on Educational Leadership (pp. 73-96). San Franciso: Jossey-Bass. Largent, J. (2001). Superintendent tenure: Characteristics of Texas superintendents, the school community, and their effect on tenure. Dissertation Abstracts International, 62(7), 2302A. National School Boards Association CUBE [Council of Urban Boards of Education]. (2002). Survey Report: Superintendent Tenure.Alexandria, VA; Retrieved June 29, 2005, from http://search.epnet.com.zeus.tarleton.edu:81/login.aspx?direct=truedb=erican=ED460201. Natkin, G., Cooper, B., Alborano, J., Padilla, A.,& Ghosh, S. (2002, April). Predicting and odeling superintendent turnover. Paper presented at the Annual Meeting of the American Educational Research Association, New Orleans, LA.Retrieved June 28, 2005, from http://www.eric.ed.gov.zeus.tarleton.edu:81/ERICDocs/data/ericdocs2/content_storage_01/0000000b/80/27/8e/5b.pdf. Orr, M. T. (2002, April). Learning the superintendency: Socialization, negotiation, determination. Paper presented at the meeting of the Annual American Educational Research Association, New Orleans, LA. Retrieved July 6, 2005, from www.superintendentresource.org/cmt/Resources.orr.pdf Parker, P. (1996). Superintendent vulnerability and mobility. Peabody Journal of Education,71(2), 64-77. Retrieved June 28, 2005, from http://search.epnet.com.zeus.tarleton.edu: 1/login.aspx?direct=true&db=aph&an=9605053348 Patten, M. L. (2004). Understanding research methods (4th ed.). Los Angeles: Pyrczak. Rausch, L. M. (2001). Factors contributing to exits from the superintendency in Indiana. Dissertation Abstracts International, 62(4), 1290A. Rohland, M. (Ed.). (2002, September). Understanding the evolving concept of leadership in education: Roles, expectations, and dilemmas. Laboratory for Student Success Review. Retrieved July 5, 2005, from http://search.epnet.com.zeus.tarleton.edu:81/login.aspx?direct=truedb=erican=ED474262SPSS (2004). Statistical Package for the Social Sciences: Version 12.0 SPSS Chicago, Ill. U.S. Department of Education. (2002, January 8). No Child Left Behind Act of 2001. Washington, DC: Office of Elementary and Secondary Education. [Online]. Available: http://www.ed.gov/policy/elsec/leg/esea02/107-110.pdf.