BACKGROUND AND CONTEXT
In my early teaching career, I had the opportunity to work in a high-achieving school district, where a newly appointed principal, despite holding a doctorate and being recognized as the state's Principal of the Year, struggled to adapt to the school's culture and community Over two years, it became clear that, although he was highly qualified, he was not the right fit for that particular school and ultimately left the position This situation highlighted a failure in the selection process, which did not adequately match the candidate's leadership style and qualifications with the specific needs of the school.
Transitioning from classroom teaching to an administrative role has prompted deep reflection on my career journey, now as an associate principal with aspirations for further advancement My thoughts frequently return to the qualities that should be prioritized in principal selection, including candidates' academic education, teaching and leadership experience, personal attributes, and leadership styles Additionally, I ponder how the backgrounds of selection committee members impact their decisions in recommending candidates for principal positions Through discussions with senior school leaders and extensive readings, I have gained insights into the factors that influence principal selection, yet I remain curious about the attributes valued by teachers on selection committees who interact with principals daily This dissertation research stems from my desire to explore these critical issues and understand the characteristics sought in effective school leaders.
Effective school leadership is crucial for student achievement, with principals playing a vital role in fostering a successful educational environment Research indicates that an effective principal is essential for a school's overall effectiveness, significantly influencing student outcomes Strong leadership not only impacts academic performance but also contributes to the overall success of the school.
Effective principals significantly influence student learning, second only to teachers (Doyle & Locke, 2014; Hattie, 2009; Leithwood et al., 2004) Their vital contributions to school success highlight the necessity of carefully selecting the most qualified candidates for principal positions.
This study explores the key attributes valued by teachers in principal-selection processes in New York State It aims to identify and evaluate the candidate qualities that teachers prioritize when recommending individuals for principal positions Additionally, the research investigates how teachers' characteristics and experiences influence their preferences and recommendations while serving on principal-selection committees.
This article begins with a historical overview of the principalship, highlighting its evolution over time It then examines the current responsibilities of modern principals, addressing significant trends and challenges faced by school leaders in New York State today The chapter wraps up by outlining the central problem, defining key terms for the dissertation, synthesizing the introductory content, and detailing the structure of the subsequent chapters.
The role of school leadership has evolved significantly since the 1600s, influenced by changing social norms, educational values, economic factors, and the onset of the digital age in the 21st century This article examines the development of the principalship, tracing its origins before the 20th century, highlighting key transformations throughout the 1900s, and detailing the current state of the role in the early 21st century.
Early public school principalships were influenced by the headmaster roles in private academies during the late 1700s and early 1800s, which were primarily focused on religious education These institutions were often led by ministers or individuals with theological training Similar to headmasters, public school principals managed a small number of teachers, performed basic administrative duties, and dedicated a significant part of their day to teaching.
In 2017, the roles of early school leaders, often called preceptors, head teachers, or principal teachers, highlighted their status as master teachers among peers (Kellough & Hill, 2015; Pierce, 2017; Rousmaniere, 2013) Operating with minimal local or state oversight, these leaders had the autonomy to manage their schools based on their personal values and beliefs (Rousmaniere, 2013).
Early in the 19th century, schools began to be divided into separate departments and sub- schools within a building, each of which was led by a different head or principal (Pierce, 2017)
In 1838, the Cincinnati School District became one of the first districts to place all departments and sub-schools within one building under a single individual (Pierce, 2017; Rousmaniere,
In 2013, a significant shift in the organizational model of schools emerged due to factors such as rising enrollment, standardized curricula, heightened emphasis on grading, and the formalization of departments As schools expanded, superintendents increasingly delegated supervisory responsibilities to building leaders, recognizing that they could no longer manage all tasks independently (Matthews & Crow, 2003; Pierce, 2017).
By the mid-1800s, the role of school principals had shifted primarily to management, leading most educational leaders to relinquish their teaching responsibilities as they took on increased supervisory tasks (Rousmaniere, 2013; Pierce, 2017) This transition resulted in a diminished focus on teaching within their duties, as highlighted by a review of board of education reports from that era.
1853 and 1900, Pierce (2017, pp 33-34) described the allocation of principals’ duties as comprising the following tasks:
Organization and general management (32 duties, 40.5%),
Building and grounds (6 duties, 7.6%), and
The shift towards greater supervisory and managerial roles for school principals has sparked controversy, with concerns about granting absolute administrative authority to individuals and the effectiveness of management-focused principals lacking proper training (Rousmaniere, 2013) This transition, coupled with minimal support and job security for principals, has led to significant principal turnover, negatively affecting the continuity of leadership essential for schools and their students By the late 19th century, as the responsibilities of principals expanded, some were finally allocated clerical assistants to help manage their increasing workload (Pierce).
2017) Theses assistants helped with such tasks as maintaining records, monitoring attendance, and performing other routine duties (Pierce, 2017)
By the early 1900s, principals had shifted their focus to curricula, requiring them to review teachers' lesson plans, create curriculum maps, and select appropriate teaching materials aligned with local community needs (Matthews & Crow, 2003; Pierce, 2017) This evolution in their role included holding regular staffing meetings to address school-related issues However, despite these curriculum responsibilities, principals did not return to direct instruction; instead, their roles expanded to encompass planning, designing, overseeing instruction, and managing teaching activities.
In the early 1900s, Chicago schools established the role of “extra-teacher” to alleviate the increasing managerial burdens on principals, who were moving away from direct teaching (Pierce, 2017) Other institutions introduced “general supervisor” roles to assist with administrative tasks that clerical staff could not manage (Matthews & Crow, 2003) These early administrative support positions laid the groundwork for modern assistant and associate principal roles, highlighting the respect and recognition that principals received as essential leaders in education (Beck & Murphy, 1993).
Principals play a crucial role in fostering community relations, as towns increasingly recognize the significant impact of schools and their leaders on local success Historically seen as a burden, schools are now viewed as vital contributors to civic initiatives, wartime efforts, and overall community development.
LITERATURE REVIEW
This chapter reviews research literature on factors influencing the selection of secondary school principals from three key perspectives First, it examines the characteristics of effective principals, highlighting various viewpoints Second, it analyzes the background and qualifications of principal candidates, focusing on the alternatives available to decision-makers Lastly, it explores how the personal biases and priorities of decision-makers, including teachers and administrators, impact the selection process These case-specific factors can interact with the values assigned to candidate characteristics, leading to variability in the selection outcomes.
This study excluded certain literature on the qualifications and selection criteria of school leaders due to its limited scope However, it did take into account related studies from regions outside of North America.
(Blackmore, Thomson, & Barty, 2006; Dinham, 2005; Goolamally & Ahmad, 2014; Gronn & Lacey, 2006; Kwan, 2012; Kwan & Walker, 2009; Parkes & Thomas, 2007; Walker & Kwan,
This review exclusively examines studies related to public secondary school principals in North America, excluding those focused on elementary schools or lacking cultural relevance The selection process for school principals is deeply influenced by the educational systems and communities they serve, and significant structural differences between elementary and secondary schools necessitate this focused approach.
This chapter's literature review is structured into three main sections: (a) the traits of effective secondary school principals, (b) the specific attributes assessed during the selection of candidates for these positions, and (c) the unique characteristics of decision-makers that influence their preferences and biases in recommending or selecting principals The chapter concludes by synthesizing the essential factors that impact the selection process for secondary school principals.
Characteristics of Effective Secondary School Principals
A considerable body of the relevant research literature has focused on the characteristics of effective principals and the salient qualities of other effective educational leaders (Ash,
Numerous studies have explored various educational topics, highlighting the contributions of researchers such as Hodge and Connell (2013), Awender (1978), and Bauck (1987) Significant findings from Blaise and Kirby (2009), as well as Blase and Blase (2000), further enrich the discourse on effective teaching strategies Brewer (1993) and Carlton (1987) provide foundational insights, while Clark, Martorell, and Rockoff (2009) examine the impact of teacher quality Additional research by Cotton (2003) and Crum and Sherman (2008) emphasizes the importance of professional development Daresh et al (2000) and Garza et al (2014) contribute to understanding leadership in education, supported by the work of Gerhart, Harris, and Mixon (2011), along with Griffing (2010) and Grissom and Loeb (2011).
Hallinger, 2011; Hauserman; Ivankova, & Stick, 2007; Hauserman & Stick, 2013; Herriot, 2012;
Numerous studies have highlighted the significance of effective leadership in education, emphasizing its impact on student achievement and school improvement Research by Hitt & Tucker (2016), Robinson et al (2008), and Marzano et al (2005) underscores the correlation between strong leadership practices and enhanced learning outcomes Additionally, the works of Leithwood & Sun (2012) and Waters et al (2003) further illustrate how leadership strategies can foster a positive school culture The contributions of various scholars, including Hudson & Rea (1996) and Hull (2012), collectively underscore the importance of continuous professional development for educational leaders to adapt to evolving challenges Overall, the evidence suggests that effective school leadership is crucial for driving educational success and fostering an environment conducive to learning.
Effective school leadership is crucial for a school's success, as demonstrated by research indicating that schools with capable principals achieve higher student performance compared to similar institutions with less effective leadership Understanding the key qualities of effective principals is essential, as their influence directly correlates with improved student outcomes.
Hitt and Tucker (2016) conducted a literature synthesis identifying 28 key leadership practices that significantly impact student achievement, categorized into five main themes: establishing and conveying the vision (ECV), facilitating a high-quality learning experience for students (FHQL), building professional capacity (BPC), creating a supportive organization for learning (CSO), and connecting with external partners (CEP).
Creating, articulating, and stewarding shared mission and vision [ECV]
Implementing vision by setting goals and performance expectations [ECV]
Modeling aspirational and ethical practices [ECV]
Communicating broadly the state of the vision [ECV]
Promoting use of data for continual improvement [ECV]
Tending to external accountability [ECV]
Maintaining safety and orderliness [FHQL]
Personalizing the environment to reflect students’ backgrounds [FHQL]
Developing and monitoring curricular program [FHQL]
Developing and monitoring instructional program [FHQL]
Developing and monitoring assessment program [FHQL]
Selecting for the right fit [BPC]
Providing opportunities to learn for whole faculty, including leader(s) [BPC]
Supporting, buffering, and recognizing staff [BPC]
Engendering responsibility for promoting learning [BPC]
Creating communities of practice [BPC]
Acquiring and allocating resources strategically for mission and vision [CSO]
Considering context to maximize organizational functioning [CSO]
Building collaborative processes for decision-making [CSO]
Sharing and distributing leadership [CSO]
Tending to and building on diversity [CSO]
Maintaining ambitious and high expectations and standards [CSO]
Strengthening and optimizing school culture [CSO]
Building productive relationships with families and external partners in the community [CEP]
Engaging families and community in collaborative processes to strengthen student learning [CEP], and
Anchoring schools in the community [CEP]
In their 2014 study, Garza, Drysdale, Gurr, Jacobson, and Merchant analyzed four successful school principals, identified by their schools' reputations, student outcomes, and peer recognition Their findings highlighted five key themes that characterize exceptional leadership: a commitment to social justice, a strong sense of care for students, a high level of ethical and moral responsibility, significant resiliency and persistence, and notable courage.
Parylo and Zepeda (2014) identified eight key characteristics of effective principals based on feedback from central office administrators, organizing these traits into four categories: documented characteristics, instructional skills, interpersonal skills, and perceptual characteristics.
Having a track record [Documented characteristics]
An instructional leader [Instructional skills]
A perfect fit to the school [Perceptual characteristics], and
In a study by Grissom and Loeb (2011), responses from principals and assistant principals in Miami-Dade County Public Schools were analyzed to identify essential skills for promoting school success Principals assessed their effectiveness in various tasks, while assistant principals evaluated their principals' effectiveness in the same areas The exploratory factor analysis of principals' responses revealed five key factors: instruction management, internal relations, organization management, administration, and external relations In contrast, the analysis of assistant principals' responses identified three overlapping factors: instruction management, internal relations, and organization management, all of which were also recognized by the principals.
A study by Schulte, Slate, and Onwuegbuzie (2010) explored college students' perceptions of effective principal qualities, involving participants from undergraduate education programs and graduate students with education degrees The researchers identified 29 themes categorized into five key factors: responsible and supportive leader (RSL), being impartial (BI), straightforward, task-oriented, and communicative (STC), professional and facilitator (PF), and collaborative, organized, and inclusive role model (COI) These findings provide valuable insights into the attributes that students associate with effective school leadership.
Works well with others [COI]
Crum and Sherman’s (2008) research on successful high school principals in Virginia revealed six key themes: developing personnel and fostering leadership, responsibly delegating tasks and empowering staff, acknowledging ultimate accountability, maintaining effective communication and rapport, facilitating instructional practices, and managing change effectively.
Robinson, Lloyd, and Rowe (2008) conducted meta-analyses to assess the impact of various leadership styles on student outcomes, analyzing 22 studies across over 3,800 schools Their findings indicated that transformational leadership had a minimal effect (d = 0.11), while instructional leadership was significantly more effective (d = 0.42), being nearly three times as impactful Other leadership styles collectively showed a mean effect size of (d = 0.30), surpassing transformational leadership In a second analysis focusing on leadership practices, 12 studies produced 199 indicators grouped by specific dimensions Key dimensions included establishing goals and expectations (d = 0.42), strategic resourcing (d = 0.31), planning and evaluating teaching (d = 0.42), promoting teacher development (d = 0.84), and ensuring a supportive environment (d = 0.27).
Waters, Marzano, and McNulty (2003) conducted a meta-analysis identifying 21 key leadership responsibilities that are significantly linked to student achievement Their research utilized teacher perceptions as the independent variable and quantitative measures of student achievement, such as standardized tests, as the dependent variable From over 5,000 studies reviewed, 70 were selected, encompassing data from 2,894 schools, approximately 1.1 million students, and 14,000 teachers.
Involvement with curriculum, instruction, and assessment (d = 0.16)
Knowledge of curriculum, instruction, and assessment (d = 0.24)
METHODOLOGY
Public-school principals are crucial for the success of students and schools, highlighting the importance of selecting the right individual for the role Given the extensive responsibilities and high expectations placed on principals, decision makers must carefully evaluate various candidate-specific attributes during the selection process This study examines how these attributes, referred to as alternative-specific attributes, affect the choices of decision makers, alongside the case-specific attributes that describe the decision makers themselves.
This study was guided by the following research questions:
This study investigates the influence of specific candidate attributes on public school teachers' recommendations for senior high school principal positions in New York State It aims to determine the extent to which these attributes impact teachers' choices in endorsing candidates for leadership roles.
The study investigates the relative prevalence of various preferences for candidates vying for senior high school principal positions, as indicated by latent classes identified through a discrete choice experiment.
The study investigates how the backgrounds and experiences of public school teachers in New York State shape their perceptions of the ideal candidate for a senior high school principalship It aims to explore the extent of these influences and identify any differences among various latent classes revealed through a discrete choice experiment.
This chapter provides a foundational understanding of the research approach and experimental design in this study by reviewing decision and choice theory, along with the associated research designs and analytic methods for studying judgments and choices A detailed examination of discrete choice theory and analysis is presented, focusing on the discrete choice experiment at the study's core, while highlighting the significance of multiple preference sets and latent classes in the results Following this theoretical framework, the chapter outlines the research design, including the blocked design, participant sampling plan, choice variables, optimal experimental design, data collection methods, and statistical analyses used to model discrete choices and identify latent preference classes Finally, the chapter concludes with a discussion on ethical considerations, methodological limitations, and a synthesis of the study's findings.
Decision and Choice Theory emphasizes that when making choices, decision makers must select one alternative while forgoing others, despite the attractive features of the rejected options This process of making discrete choices ultimately unveils the decision makers' preferences (Louviere, Hensher, & Swait, 2010; Raghavarao, Wiley, & Chitturi, 2011; Simon, 1997).
Decision and choice theory explores the processes behind decision-making, with a rich history of research dedicated to this field This overview highlights the evolution of key theories, including rational decision theory, administrative decision theory, expected utility theory, prospect theory, and random utility theory Additionally, it addresses social judgment theory and discrete choice theory, which serve as foundational elements for the guiding theories and methods utilized in this study.
Rational decision theory, or economic decision theory, posits that individuals operate under unbounded rationality when identifying and evaluating decision alternatives (Beach & Lipshitz, 1993; Simon, 1997) It suggests that people have comprehensive systems that allow them to consider all options and outcomes, aiming to select the optimal alternative in any situation (Edwards, 1954) This theory assumes individuals possess the cognitive ability to recognize all available alternatives, perform complex evaluations, and understand the probabilities of each option's outcomes (Edwards, 1954; Simon, 1997) Additionally, rational decision theory emphasizes the importance of correctly executing the decision-making process rather than solely focusing on achieving correct decisions (Beach & Lipshitz, 1993).
In rational decision theory, four key steps are hypothesized to be followed in decision making (Simon, 1997):
3 Determine all the consequences that follow from each of all those strategies, and
4 Select the strategy that optimizes achievement of the stated goal
Rational decision theory seeks to identify the optimal alternative by ranking each alternative-consequence relationship based on its means-ends rating, ultimately selecting the highest-rated option These ratings serve as estimations of the anticipated level of goal achievement.
Rational decision theory acknowledges that rationality varies depending on specific circumstances, as highlighted by Simon (1997), who identified six distinct perspectives on decision-making Objective rationality refers to choices that effectively maximize values in a given situation, while subjective rationality focuses on maximizing outcomes based on available information Consciously rational decisions involve a self-aware evaluation process, such as instinctively pulling away from a hot stove Deliberate rationality emphasizes purposeful decision-making, exemplified by a typist pressing a particular key Finally, organizational rationality aligns decisions with organizational goals, whereas personal rationality aligns them with individual objectives (Simon, 1997).
Rational decision theory assumes individuals engage in a comprehensive process when choosing among alternatives; however, they often lack the cognitive capacity and resources to evaluate all possible options and their consequences Additionally, this theory does not adequately explain how individuals formulate their goals (Beach & Lipshitz, 1993; Simon, 1953/1997) To address these limitations, Simon (1953/1997) introduced the theory of administrative decision making, which serves not as a replacement for rational decision theory but as a refinement that illustrates how rational decision-making principles manifest in real-world scenarios.
Administrative Decision-Making and the Theory of Bounded Rationality
Rational decision theory, a mainstay of economics, represented the predominant view of decision and choice theory until the 1940s, when it was challenged by Herbert Simon
In his work from 1945 to 1997, Simon critiqued the rational theory of decision making, arguing that its elegance fails to capture the realities of practical decision-making processes He posited that traditional economic theories represent an unattainable ideal due to the inherent limitations of human cognition and the practical constraints, such as time and resources, faced during decision-making Instead, Simon proposed a more realistic perspective through the "muddling" theory of administrative decision making, which reflects how decisions are actually made in practice.
The theory of rational decision-making is not entirely replaced but rather expanded to reflect how rationality is applied amidst cognitive constraints and limitations (Kalantari, 2010) Administrative decision theory recognizes that while decision-makers strive for optimal rationality as outlined in classical economic theory, real-world decisions occur under conditions of "bounded rationality" (Kalantari, 2010; Simon, 1945/1997) and are often better described as "muddling through" (Lindblom, 1959).
Administrative decision theory highlights that human rationality in decision-making is inherently limited by cognitive capacity and situational factors Individuals face constraints on the number of alternatives they can consider, influenced by their experiences and knowledge This theory also suggests that decision-makers possess only partial knowledge of the conditions affecting their choices As a result, decisions often require compromise, as no option can fully meet all desired outcomes; instead, individuals typically select the alternative that offers the best possible solution within their circumstances Ultimately, this theory posits that decision-makers must "satisfice," choosing the first alternative that is deemed "good enough."
Administrative decision-making theory contrasts with rational decision theory by acknowledging a less structured process that often leads to sub-optimal choices Decision makers limit their rationality by focusing on a narrow range of plausible alternatives and evaluating them based on their immediate availability Consequently, the set and order of alternatives considered can vary significantly over time, affecting the decision-making outcome.
RESULTS
This chapter presents the study's results, beginning with the unlabeled discrete choice experiment central to the research It starts by discussing the findings from the alternative-specific conditional logistic regression (asclogit), which operates under a traditional single-rational-actor framework Subsequently, it explores statistical analyses using the Akaike and Bayesian Information Criteria (AIC and BIC) to investigate unobserved heterogeneity and the possibility of multiple latent classes, challenging the assumptions of asclogit Finally, the chapter details the outcomes of a latent class logistic regression (lclogit) model, identified through a 3-latent class specification derived from AIC/BIC testing and alignment with established theoretical frameworks.
Part Two of this chapter details the findings from a systematic qualitative and mixed-methods analysis of responses to an open-ended narrative question It explores the application of traditional hermeneutic qualitative techniques alongside computer-aided quantitative content analysis, utilizing geospatial statistical modeling methods such as multidimensional scaling, co-occurrence network analysis, mapping, and hierarchical agglomerative cluster analysis The results reveal five emergent subthemes that align with two overarching major themes, ultimately connecting the qualitative and mixed-methods findings to the quantitative outcomes of the discrete choice analysis.
This chapter summarizes the study's findings, offering an evidence-based foundation for addressing the research questions that directed this investigation, as well as the conclusions and implications discussed in Chapter 5.
PART ONE RESULTS OF THE UNLABELED DISCRETE CHOICE EXPERIMENT
In this first part of the chapter, I discuss the discrete choice analyses that were conducted
This article presents the findings from an alternative-specific, conditional logistic regression, which operates under the assumption of a single, observed class, reflecting a single-rational-actor perspective Additionally, it details the outcomes of a series of latent class, alternative-specific, conditional logistic regression analyses, with a particular emphasis on the three-latent class model identified as the optimal statistical and theoretical fit for the experimental data.
Results from the Alternative-Specific, Conditional Logistic Regression
The alternative-specific, conditional logistic regression results, detailed in Tables 4.1a and 4.1b, reveal a highly significant single-class model (Wald X²(40) = 1142.74, p < 001; Tjur’s pseudo R² = 47; N = 212), with an AIC of 2665.10 and a BIC of 935.41 The analysis indicates that candidates without prior administrative experience have a 42% lower chance of being selected (p = 03) compared to those with administrative experience but no principal role The odds for candidates lacking administrative experience may be reduced by as much as 64% and no less than 7% when compared to those with administrative experience Conversely, candidates who have previously held a principalship are 82% more likely (p = 01) to be selected than those with administrative experience without a principal role With 95% confidence, candidates with prior principal experience have at least a 20% greater chance of being recommended, potentially increasing by up to 174% compared to candidates with non-principal-level experience.
Candidates with below-average instructional leadership abilities face a significant disadvantage, with a 62% lower chance of receiving recommendations compared to those with average skills (p < 001) Their likelihood of being selected may drop by as much as 77%, and is almost certainly (95% confidence) at least 36% lower than candidates with average skills Interestingly, candidates with above-average instructional leadership skills do not have any advantage over their average-skilled counterparts.
Candidates with below-average managerial skills face significantly lower odds of being recommended, with a reduction of 85% compared to their peers with average skills When controlling for other factors, these odds can drop by as much as 92% but are guaranteed to be at least 72% lower Conversely, candidates with above-average managerial skills enjoy a 49% higher likelihood of selection than those with average skills This selection advantage is statistically significant, with a minimum increase of 10% and potentially reaching up to 103%.
Results From the Alternative-Specific Conditional Regression b SE z p LL-b UL-b OR LL-OR UL-OR
Note Bayesian Information Criterion = 2935.41; Akaike Information Criterion = 2665.10 Wald X 2 (40) for the overall model 1142.74 p < 001 Tjur’s pseudo R 2 = 47 Upper and lower limits were calculated using confidence intervals of 95%
The analysis of case-specific effects from the alternative-specific conditional regression reveals several key insights The variable teachXtchexp shows a positive effect, though not statistically significant (p = 100) In contrast, hicredXinstrleadhi demonstrates a significant positive association (p = 013), indicating a potential impact on outcomes Other variables, such as hicredXadminexplo and hicredXintprleadlo, exhibit negative coefficients but lack statistical significance (p = 200 and p = 176, respectively) Overall, while some variables suggest trends, many do not reach conventional significance levels, highlighting the complexity of the relationships examined in this regression analysis.
Case-Specific Effects From the Alternative-Specific Conditional Regression b SE z p LL-b UL-b OR LL-OR UL-OR Wald test p needsXtchexp 0.003 0.014 0.23 816 -0.02 0.03 1.00 0.98 1.03 needsXadminexplo 0.188 0.200 0.94 347 -0.20 0.58 1.21 0.82 1.78 317 needsXadminexphi -0.117 0.195 -0.60 546 -0.50 0.26 0.89 0.61 1.30 317 needsXedlevel -0.021 0.158 -0.13 895 -0.33 0.29 0.98 0.72 1.34 needsXinstrleadlo -0.476 0.223 -2.13 033 -0.91 -0.04 0.62 0.40 0.96 044 needsXinstrleadhi 0.030 0.134 0.22 825 -0.23 0.29 1.03 0.79 1.34 044 needsXmgrleadlo -0.084 0.264 -0.32 750 -0.60 0.43 0.92 0.55 1.54 769 needsXmgrleadhi 0.086 0.146 0.58 559 -0.20 0.37 1.09 0.82 1.45 769 needsXintprleadlo 0.231 0.285 0.81 419 -0.33 0.79 1.26 0.72 2.20 697 needsXintprleadhi 0.046 0.135 0.34 733 -0.22 0.31 1.05 0.80 1.36 697
The study examines various case-specific and principal-candidate specific variables related to educational leadership Key variables include teaching experience (teach), advanced qualifications (hicred), and perceived school needs (needs) Additionally, the analysis considers candidates' previous administrative experience (adminexplo and adminexphi), instructional leadership abilities (instrleadlo and instrleadhi), and managerial and interpersonal leadership skills (mgrleadlo, mgrleadhi, intprleadlo, intprleadhi) Significant findings include Wald statistics of 6.56 for the interaction between advanced qualifications and instructional leadership, and 6.24 for the interaction between perceived needs and instructional leadership, with confidence intervals set at 95%.
Interpersonal skills play a crucial role in the candidate selection process, significantly impacting hiring outcomes Candidates with below-average interpersonal skills are 90% less likely to be chosen compared to those with average skills (p < 001) There is a 95% confidence level that this disadvantage could range up to 95%, but it is unlikely to fall below 79%.
The analysis identified two significant factors influencing teacher perceptions of candidates' instructional leadership skills in the single-class asclogit model Teachers with higher educational credentials, such as advanced certificates or doctorates, placed a greater emphasis on above-average instructional leadership skills (p = 01) Additionally, educators in high-needs schools expressed considerable concern (p = 03) regarding principal candidates with below-average instructional leadership abilities No other teacher-specific characteristics significantly impacted their preferences for principal candidates.
Results of the Latent Class Conditional Logistic Regression
To identify potential latent classes within the choice data from the experiment involving 219 participants, three latent class conditional logistic regression analyses were performed, specifying models with 2, 3, and 4 latent classes The Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) statistics were calculated for each model, as presented in Table 4.2 These metrics, along with the AIC and BIC from the asclogit, were utilized to determine the model specification that offers the best statistical fit.
Akaike and Bayesian Information Criteria for Latent Class Models in the Study
Number of Classes Source AIC BIC
Note asclogit = alternative-specific conditional logistic regression lclogit = latent class conditional logistic regression
In model selection, the best fit is indicated by the lowest AIC and BIC values, provided the model is theoretically sound In this instance, the AIC recommends a 4-latent class model, while the BIC favors a 3-latent class model When faced with differing suggestions from AIC and BIC, I prioritize the BIC due to its stronger penalty for model complexity, as supported by Dziak, Coffman, Lanza, & Li (2012).
The analysis revealed that the single-class model generated by asclogit fails to adequately represent the complexity and variety present in the choice data In contrast, the three-latent class model emerged as the best-fitting option, supported by robust statistical evidence, theoretical relevance, and the scientific principle of parsimony.
Analysis of the Best-Fitting Conditional Logistic Regression Model
The selected 3-latent class lclogit model demonstrated high statistical significance (p < 001, Tjur’s pseudo R² = 0.46) This article explores the preferences related to candidate factors and the potential impacts of teacher-related factors for each of the three latent class models.