Development,factor structure,and reliability of the Shared Instructional Leadership Scale in public secondary schools

Abstract

Shared instructional leadership may support informed decision making on matters of curriculum, instruction, and assessment. Given the various organizational processes and outcomes associated with this construct, it is important to be able to measure the degree to which it exists in schools. In this article we propose the Shared Instructional Leadership Scale and report its reliability and the validity of its factor structure. The scale was designed to assess the extent to which faculty perceive that principals, teachers, and school staff collaborate on instructional leadership practices. Drawing from a sample of 422 teachers nested in 107 schools, we generated four sub-samples to examine its psychometric properties. Next, using exploratory factor analysis techniques, we found the Shared Instructional Leadership Scale factor structure to be stable across all four sub-samples. Finally, we conducted confirmatory factor analysis on a second school-level sample (n = 103) and the results confirmed the Shared Instructional Leadership Scale had a unidimensional structure. We conclude with a discussion of the potential of the Shared Instructional Leadership Scale to inform practice and implications for future research, including directions for further scale validation.

Keywords

Shared instructional leadership scale development factor analysis

Introduction

Involving school staff in leadership responsibilities may facilitate informed decision making on instructional matters (Lee et al., 2012). There is growing evidence that principals collaborate with teachers and staff to create conditions to improve teaching and learning (City et al., 2009). Such work entails articulating a school vision (Finnigan and Stewart, 2009; Zepeda, 2013), establishing data teams (Goldring et al., 2009a; Urick and Bowers, 2014), and aligning school climate (Damanik and Aldridge, 2017), structure (Leithwood and Jantzi, 2000), curricula (Hallinger, 2005; Marks and Printy, 2003), professional learning experiences (Anthony et al., 2019), community partnerships (Goldring et al., 2009b), and teacher evaluations (Coldren and Spillaine, 2007) to improve instruction and student outcomes.

Shared instructional leadership (SIL) is an extension of instructional leadership, which focuses on the principal’s role in leading and improving instructional programs (Hallinger, 2003; Hallinger and Murphy, 1985). SIL incorporates collective contributions of principals and school community members in strengthening professional knowledge, teaching practice, and instructional coherence in order to improve student learning and academic outcomes (Lambert, 2002; Marks and Printy, 2003; Printy et al., 2009). Several studies have found a positive relationship between SIL and school achievement (Jackson and Marriott, 2012; Marks and Printy, 2003). Shared instructional leadership may indirectly impact achievement by strengthening teacher collaboration and professional community (Goddard et al., 2007; Johnson et al., 2018; Louis and Marks, 1998), teacher empowerment (Somech, 2005), teacher commitment (Bryk et al., 2010; Pounder et al., 1995), teacher collective efficacy (Goddard et al., 2015), and a school’s capacity for instructional improvement (Hallinger and Heck, 2010; Heck and Hallinger, 2009; King and Newmann, 2001).

Although a rich collection of empirical research on instructional leadership and SIL exists, there are few validated measures specifically assessing SIL. Therefore, we designed this study to expand upon existing measures to facilitate future research on the relationship between SIL and various school processes and outcomes. More specifically, the purpose of our study was two-fold: 1) to develop the Shared Instructional Leadership Scale (SILS), and 2) to evaluate the reliability and validity of the scale. We developed the SILS with a focus on collaboration between principals, teachers, and staff members. The scale highlights key instructional leadership practices, including developing and communicating an instructional vision, securing resources to ensure high-quality instruction, facilitating, participatory decision making, examining data, and continuously improving instruction. Our aim is to provide researchers and educators with an instrument to examine SIL in schools.

Literature review

This section discusses research and theory on instructional leadership and then turns to the ways in which it has evolved into a shared responsibility in schools.

Instructional leadership

Studies of instructional leadership originated in the early 1980s on schools identified as effective at fostering instructional quality and student achievement. The conventional conceptualization of instructional leadership emphasized the role of the principal as an inspector, director, and supervisor of educational issues (Hallinger, 2003; Hallinger and Murphy, 1985; Hoy et al., 2013). Indeed, according to Hallinger (2010), in the 1980s instructional leadership was thought to be delivered exclusively by the principal, and there was no consideration of it as a shared responsibility.

From this perspective, the principal defined and communicated the school’s goals and expectations (Heck et al., 1990), ensuring that its mission focuses on students’ academic progress (Hallinger, 2003) and on preparing students for success. This perspective envisions the principal as the leader of efforts intended to improve and coordinate curricula, promote teacher professional growth, and ensure systemic accountability (Goldring et al. 2009b; Hallinger and Murphy, 1985). Early instructional leadership scholars also envisioned the principal as responsible for fostering a positive climate and supportive environment for teaching and learning that aligned with the school’s mission (Dwyer, 1984; Hallinger and Murphy, 1985; Heck et al., 1990).

A key feature of instructional leadership, however, is that it supports teachers’ commitment to, involvement in, and capacity for professional development. Scholarly understanding of the construct thus grew to conceive of it as shared because when principals involve teachers and staff in instructional decision-making processes, they become key actors whose morale and capacity for instructional improvement are needed to increase student learning (Blase and Blase, 2000; Heck et al., 1990). In other words, school principals cannot lead in isolation; rather, they rely upon professional staff commitment to and involvement in the achievement of common school goals. From this perspective, to be effective, principal instructional leadership requires the coordinated and creative work of other professional educators including teachers, assistant principals, and instructional specialists, rendering its efficacious enactment shared.

Shared instructional leadership

In response to concerns that an exclusive focus on principal authority ignores the shared contributions of others, scholars proposed SIL as a revised conceptualization of instructional leadership (Lambert, 2002; Marks and Printy, 2003). SIL highlights principals’ and teachers’ collaborative efforts to improve teaching and learning (Goddard et al., 2004; Blase and Blase, 2000; Glickman, 1989; Kaplan and Owings, 1999; Lambert, 2002; Marks and Printy, 2003; Printy et al., 2009). SIL is not confined to a single role or position; rather it flows through networks in the school community and interactions among members who deploy personal resources to achieve common instructional goals (Bolman and Deal, 2013; Ogawa and Bossert, 1995).

In addition, Urick and Bowers (2014) note that SIL is associated with aspects of instructional, transformational, and distributed leadership. Extending instructional leadership to recognize teachers’ contributions to school improvement, SIL emphasizes the collaborative undertaking of principals and teachers to improve curriculum, instruction, and assessment (Hallinger, 2005; Marks and Printy, 2003). Similar to transformational leadership (Finnigan and Stewart, 2009; Leithwood and Jantzi, 1999), yet focusing on instruction and acknowledging the formal and informal leadership of teachers and others (Hallinger, 2003; Hallinger and Heck, 2010), SIL works to develop and communicate a coherent instructional vision. Related to shared leadership (Wahlstrom and Louis, 2008), yet with an emphasis on instruction, SIL provides teachers with opportunities to participate in making departmental or schoolwide decisions aimed at improving instruction within and beyond the classroom (Printy et al., 2009). Similar to distributed leadership, SIL practices are “stretched over the school’s social and situational contexts” (Spillane et al., 2001: 23) and occur as principals and members of the school community collaborate in the context of their school improvement efforts (Harris, 2004; Spillane, 2004, 2006). As an ideal type, SIL draws on aspects of instructional, transactional, and transformational leadership (Printy et al., 2009; Urick, 2016).

Although schools have historically struggled with enacting aspects of SIL such as broad collaboration on instructional decision making (Cohen, 1981; Little, 1990; Lortie, 1975), many education reforms from the early 1990s through the early 2000s in the U.S. required schools to restructure in ways that increased engagement with stakeholders such as teachers, comprehensive school reform providers, parents, and community members (Bryk et al., 2010; Hallinger, 2003; Louis and Marks, 1998). Research on nationally representative samples of principals indicates that a significant proportion of U.S. public schools regularly exercise aspects of SIL, such as principals and teachers working together to make instructional improvement decisions (Jackson and Marriott, 2012; Urick, 2016; Urick and Bowers, 2014). Indeed, in recognition of its importance to school improvement, multiple states have included aspects of SIL in their principal evaluation frameworks (Pennsylvania Department of Education, 2020; Virginia Department of Education, 2020). Yet, while state frameworks are used for evaluation purposes, they are rarely equipped with evidence of validation (Teh et al., 2014).

SIL enhances teachers’ professional status and uses their expertise to support their own professional growth as well as school improvement (Lambert, 2002; Printy and Marks, 2006). Principals who practice SIL work with teachers and staff to develop shared visions for school improvement, encourage teachers to collaboratively examine evidence of student learning, and help teachers implement strategies to improve instruction (Graczewski et al., 2009; Merrill and Daugherty, 2010). Collaboration and emerging teacher leadership, in turn, foster instructional coherence (Newmann et al., 2001) and teachers’ motivation, efficacy, and reflective teaching (Blase and Blase, 2000; Yost et al., 2009). Broad collaboration with an emphasis on instructional issues can lead to improvements in teaching practice and student achievement (Goddard et al., 2004, 2015; Newmann et al., 2001). In summary, elements of SIL include the following:

Shared visions: The principal works with teachers and staff to develop and communicate the visions of the school. All school members know why and how these visions influence the school.

Focus on instruction: The principal works with teachers, staff, and others to improve and align the instructional program, including school culture and climate, parent and community engagement, professional development, instructional quality, curricula, and assessments.

Monitoring of progress: The principal, teachers, and staff collaboratively monitor and evaluate the school’s progress towards achieving instructional improvement goals and student outcomes.

Broad collaboration: The principal values input from teachers, staff, and others and works collaboratively with them to inform instructional decisions. The school has participatory decision-making structures and processes.

Existing measures

Despite the importance of SIL in practice and research, there are few validated measures to assess it. To analyze existing measures, we began by conducting literature searches. Because researchers use various terms such as “leadership for learning” or “principal and teacher leadership for instructional improvement” to describe aspects of SIL, we searched the EBSCO research databases and the WorldCat collection of library content for peer-reviewed journal articles that included the words “shared,” “instruction(al),” and “leadership” in either the title or abstract. As a result of our search, we identified three existing measures of SIL (Marks and Printy, 2003; Urick, 2016; Urick and Bowers, 2014). We also added a measure of principal and teacher instructional influence (Jackson and Marriott, 2012) and a scale of collaborative leadership for learning (Hallinger and Heck, 2010) to our review of existing measures of SIL. Although each of the measures described in this section offers insight into how schools carry out aspects of SIL, these measures primarily operationalize SIL as teacher involvement in instructional decision making. Of the five measures, Hallinger and Heck’s (2010) is the only one that addresses shared vision-setting. Furthermore, most do not address progress monitoring. Three of the measures (Jackson and Marriott, 2012; Urick, 2016; Urick and Bowers, 2014) used data from the Schools and Staffing Survey (SASS), resulting in items that share a factor loading but that are not theoretically derived, which may explain why all four components of SIL are not reflected across the five measures.

Marks and Printy (2003: 383) created an index measuring SIL by summing nine items (α = .77). Three of the items were dummy coded (0 = No, 1 = Yes), assessing whether “there is evidence of significant instructional leadership in the school,” whether “from a principal or other school-based administrator” or “from a teacher or group of teachers.” Their measure also included six items with a three-point, Likert-type scale (low, medium, high), rating (1) the actual influence of teachers over curriculum, instruction, and assessment; and (2) the actual influence of principals over curriculum, instruction, and assessment. Both groups of items showed satisfactory internal consistency, but their construct validities were not reported. Marks and Printy’s (2003) measure of SIL partially addressed two of the four components of SIL, namely a focus on instruction and collaboration among teachers and administrators. Their measure did not address the influence of parents, staff, and students over instruction; neither did their measure address shared vision-setting or monitoring of progress.

Hallinger and Heck (2010) developed a measure of collaborative leadership for learning that included two subscales with five items each that loaded onto their collaborative leadership factor. One subscale focused on teacher perceptions of leadership within the school. It included five items on a five-point, Likert-type scale (α = 0.82). The second subscale focused on parent perceptions of school leadership, as well as parents’ involvement in improving the school’s instructional program. It included five items on a five-point, Likert-type scale (α = 0.88). Although Hallinger and Heck (2010) do not list the items for their measure, they described the items as addressing all four components of SIL, including stakeholders’ contribution to the school’s vision; a focus on instruction; monitoring progress; and broad collaboration of teachers, administrators, parents, and staff in making decisions concerning instructional improvement, as well as participation of students in school governance.

Drawing on principals’ survey responses to the 2003–04 administration of the SASS, Jackson and Marriott (2012) identified two factors—one factor that included five items on a four-point, Likert-type scale that measured teacher instructional influence, and a second factor that included five items on a four-point, Likert-type scale that measured principal instructional influence. Their measure addressed three of the four components of SIL. Of the three components addressed, they had the most items reflective of a focus on instruction (e.g., setting performance standards for students, establishing curriculum, determining the content of professional development, and hiring new teachers). They addressed monitoring of progress only in terms of evaluating teachers, as opposed to evaluating student outcomes. In terms of broad collaboration, they included teachers and principals as individuals who influence instructional decisions, but not staff, students, or other members of the school community. Their measure did not emphasize a shared vision for teaching and learning.

Urick and Bowers (2014) analyzed data from the 1999–2000 SASS survey and identified one factor with three items on a five-point, Likert-type scale that represented principals’ perceptions of the amount of instructional influence they shared with teachers around work such as setting performance for students, establishing curriculum, and planning in-service professional development for teachers. Of the four components of SIL, Urick and Bowers’ measure addressed two components, namely instruction and collaboration among the principal and teachers on matters of instruction; however, their measure did not address collaboration from a broad array of stakeholders (e.g., parents or community members), shared vision-setting, or monitoring of progress.

Urick (2016) also analyzed data from the 1999–2000 SASS survey and identified one factor to represent SIL. This factor included six items (α. = 0.836) on a five-point, Likert-type scale. Like Urick and Bowers (2014), data for this factor were based on principals’ perceptions of principals’ and teachers’ influence over setting performance standards for students, establishing curriculum, and planning in-service professional development for teachers. Of the four components of SIL, this measure addressed two of the components, namely a focus on instruction and collaboration among the principal and teachers on matters of instruction.

Given that SIL is an extension of other leadership ideal types, and even overlaps with them in practice (Printy et al., 2009; Urick, 2016; Urick and Bowers, 2014), we located existing measures of distributed (Heck and Hallinger, 2009; Hulpia et al., 2009), instructional (Goddard et al., 2015; Goldring et al., 2009b; Hallinger and Murphy, 1985), and transformational (Leithwood and Jantzi, 1999) leadership that focus on instruction and align with at least one of the three remaining aspects of SIL previously described: shared vision-setting, monitoring progress, or broad collaboration. Table 1 summarizes existing leadership measures in relation to a conceptual framework of SIL. The first column lists four components of SIL as an organizing framework. The second column summarizes the literature aligned with each component. The third column summarizes existing measures of SIL. The fourth column summarizes existing measures of distributed leadership (DL), instructional leadership (IL), and transformational leadership (TL) in relation to the four components of SIL.

Table 1.

Alignment between components of shared instructional leadership and existing leadership measures.

Component of SIL	SIL conceptual frameworks or qualitative studies	SIL quantitative measures	DL, IL, and TL quantitative measures
Shared vision	Hallinger, 2003; Lambert, 2002; Printy and Marks, 2006; Printy et al., 2009	Hallinger and Heck, 2010	Hallinger and Murphy, 1985; Heck and Hallinger, 2009; Leithwood and Jantzi, 1999
Focus on instruction	Hallinger, 2003; Lambert, 2002; Printy and Marks, 2006; Printy et al., 2009	Hallinger and Heck, 2010; Jackson and Marriott, 2012; Marks and Printy, 2003; Urick, 2016; Urick and Bowers, 2014	Goddard et al., 2015; Goldring et al., 2009b; Hallinger and Murphy, 1985; Heck and Hallinger, 2009
Monitoring progress	Hallinger, 2003; Lambert, 2002; Printy and Marks, 2006; Printy et al., 2009	Hallinger and Heck, 2010; Jackson and Marriott, 2012	Goddard et al., 2015; Goldring et al., 2009b; Hallinger and Murphy, 1985; Heck and Hallinger, 2009
Broad collaboration	Hallinger, 2003; Lambert, 2002; Printy and Marks, 2006; Printy et al., 2009	Hallinger and Heck, 2010; Jackson and Marriott, 2012; Marks and Printy, 2003; Urick, 2016; Urick and Bowers, 2014	Goddard et al., 2015; Goldring et al., 2009b; Heck and Hallinger, 2009; Hulpia et al., 2009; Leithwood and Jantzi, 1999

As Table 1 demonstrates, existing measures of SIL emphasize principals’ and teachers’ collaboration around matters of instruction; however, there is less emphasis on vision-setting and monitoring of instruction. Of the existing SIL measures, Hallinger and Heck’s (2010) measure of collaborative leadership for learning is the only one that addresses all four aspects of SIL reflected in the conceptual and qualitative literature, which includes a focus on instruction, vision-setting, broad collaboration, and monitoring instruction. The next section describes how we drew on the conceptual and qualitative literature concerning SIL, as well as existing measures of SIL and existing measures of distributed leadership, instructional leadership, and transformational leadership to inform the development of the SILS.

Method

This section describes the initial development of the SILS, revisions to the instrument, survey administration, and a description of methods used to analyze the psychometric properties of the instrument. In developing the SILS, we aimed to capture key aspects of SIL outlined in Table 1. The primary goal of this study was to develop the SILS, and to evaluate the validity of the factor structure and the reliability of the SILS.

Instrument development

To address the content validity of the SILS, we needed to make sure that the items making up the instrument were relevant to this form of leadership (DeVellis, 2012). To form the theoretical basis for the instrument, we consulted the aforementioned scholarly literature on SIL and items from existing measures of instructional, distributed, and transformational leadership that address various aspects of SIL. We began instrument development by drafting items that addressed behaviors reflective of the four components of SIL, namely (1) developing and communicating a shared instructional vision, (2) an emphasis on aligning and improving the instructional program, (3) monitoring and evaluating the school’s progress toward improving instructional practices and student outcomes, and (4) collaborating broadly with stakeholders within and beyond the school to make instructional decisions (Hallinger, 2003; Lambert, 2002; Printy and Marks, 2006; Printy et al., 2009). Using these four components of SIL as an organizing framework, we referenced items included in existing measures to draft and revise items for this instrument.

Existing SIL measures primarily focus on principals’ and teachers’ collaboration around matters of instruction (Jackson and Marriott, 2012; Marks and Printy, 2003; Urick, 2016; Urick and Bowers, 2014); however, we broadened collaboration to also include staff (Hulpia et al., 2009). In addition to a focus on instruction, we considered collaborative efforts around framing and communicating a common vision (Hallinger, 2003; Hallinger and Heck, 2010; Hallinger and Murphy, 1985; Leithwood and Jantzi, 1999; Printy and Marks, 2006) and monitoring the instructional program (Goldring et al., 2009b; Hallinger and Heck, 2010; Hallinger and Murphy, 1985).

Building upon existing SIL measures that focus on collaboration among internal members of the school (Goddard et al., 2015; Jackson and Marriott, 2012; Marks and Printy, 2003; Urick, 2016; Urick and Bowers, 2014), we incorporated their collective efforts to coordinate with the broader community to improve instructional programs (Goldring et al., 2009b; Hallinger and Heck, 2010; Leithwood and Jantzi, 1999; Newmann et al., 2001). Principals, as opposed to teachers, have historically led efforts to forge relationships with community members (Epstein, 2001; Sanders and Harvey, 2002). However, educational policies and initiatives in the U.S. since the early 2000s have increasingly called for schools’ collaboration with external stakeholders to develop and deliver instructional programming (e.g., culturally relevant instruction, problem-based learning, career and technical education, STEM education, early college high schools). Although building and district administrators remain central figures in forging community partnerships, teachers who contribute to instructional leadership have a heightened role in recommending and aligning community resources with instructional improvement efforts.

The items were designed to represent SIL as a unidimensional construct comprising seven items that cut across the collaborative instructional leadership behaviors of principals, teachers, and other school staff. The seven questions in the scale asked teachers to describe the extent to which they agreed that the principal, teachers, and staff in their building work collaboratively to perform each particular practice, choosing their response from a six-point, Likert-type scale: “strongly disagree” (1) to “strongly agree” (6). We used a six-point scale to increase response variability and to discourage “neutral” responses.

Following initial instrument development and preliminary revisions, we sought input from educational administration faculty with K-12 leadership experience. To guide our discussions, we asked these reviewers if the instrument seemed to reasonably reflect SIL. We asked if any items were ambiguous or difficult to understand. We also asked if they recommended any edits to the instructions, survey items, or Likert scale. Reviewers affirmed item content and provided feedback that guided revisions to item wording. All feedback informed the revision process. As described in the next section, we administered two rounds of the SILS and analyzed the internal reliability and construct validity of the underlying factor structure. The final version of the instrument is displayed in Table 2.

Table 2.

The shared instructional leadership scale.

Shared instructional leadership scale
Instructions Please respond to each of the items on this questionnaire by selecting the response that best reflects practices at your school. Read each statement carefully. Then, select the response that indicates your level of agreement regarding feelings about whether each specific behavior was demonstrated at your school during the past school year. For each behavior, a 6 represents “Strongly agree” and a 1 represents “Strongly disagree.”
The principal, teachers, and other staff work collaboratively to…
	Strongly disagree	Disagree	Somewhat disagree	Somewhat agree	Agree	Strongly agree
	1	2	3	4	5	6
1. develop an instructional vision.
2. communicate an instructional vision.
3. identify potential community partnerships that align with school’s goals.
4. ensure resources for high-quality instruction.
5. make instructional decisions.
6. examine student achievement data.
7. improve the school’s instructional program.

Survey administration and analysis of psychometric properties

The SILS was used to collect data on teachers’ perceptions of SIL in their building from two samples—the first sample was used to create four sub-samples for exploratory factor analysis (EFA), and the second sample was used for confirmatory factor analysis (CFA). For the first sample, a link to the online survey was distributed by email to 6,200 principals and teachers at 150 public secondary schools in a Midwestern U.S. state. We emailed two reminders to request that recruited individuals respond to the survey. We received 422 total valid responses from teachers. Two principals and zero school staff members responded to the survey. Since we received only two total responses from principals and other staff members, an insufficient number for any inferential statistics analysis, only the teachers’ responses were used in this study.

To conduct the EFAs for the first sample, we generated four sub-samples: sub-sample 1a: an individual-level sample which included all teacher responses (n = 422); sub-sample 1b: a school-level sample including the responses from the schools which had a response rate at or above 10% of their total teaching staff (n = 60); sub-sample 1c: a school-level sample including the responses from all the schools where at least one teacher provided a response (n = 117); and sub-sample 1d: a school-level sample including the responses from the schools had a response rate below 10% of their total teaching staff (n = 57). The school-level responses were generated by aggregating the teachers’ responses. We then conducted an EFA on each sub-sample. The design of the sub-samples allowed us to consider whether the observed factor structures appeared to be influenced by either average school-level response rates or aggregation bias.

For the second sample, a second survey was distributed by email to 5,269 teachers at 141 public secondary schools in a Midwestern state. We received 658 valid teacher responses in total. We used a school-level sample (n = 103), including the responses from the schools that had a response rate at or above 10% of their total teaching staff (n = 587) for the CFA. The CFA sample includes 587 teachers from 103 public secondary schools. Once again, the school-level response was generated by aggregating the teachers’ responses by school.

Results

This section discusses the EFA, the CFA, and the reliability tests on this survey.

Exploratory factor analysis

Sub-sample 1a. Individual level sample

Missing value imputation. We used SAS 9.4 to examine all the responses, revealing a small percentage of missing item data. Only one respondent had a missing value under question three. We used the strategy of multiple imputations to replace the missing values with a set of acceptable values which represented a distribution of the possible values, generating several data sets. For a data set with 10% missing data, five imputations have above 98% efficiency in increasing the variance estimates (Yuan, 2000). Although we had relatively little missing data, we still used the expectation maximization algorithm and the Markov Chain Monte Carlo method to impute missing values, since the deletion of the remainder would bias the inferential analysis results (Fichman and Cummings, 2003). Five imputations were thus conducted to estimate missing values of the section responses. The maximum mean difference for any of the variables across the five imputed data sets was 0.002, the maximum variance difference was 0.012, the maximum correlation difference was 0.001, and the average error was smaller than 0.001. Given that only trivial differences existed between the five data sets, we combined them into a single data set by using medians. Table 3 reports the descriptive statistics for sub-sample 1a, the individual-level sample.

Table 3.

Descriptive statistics of sub-sample 1a, individual-level sample (n = 422 teachers).

	SIL1	SIL2	SIL3	SIL4	SIL5	SIL6	SIL7
Mean	4.47	4.39	4.09	4.60	4.33	4.62	4.60
Median	5.00	5.00	4.00	5.00	5.00	5.00	5.00
SD	1.28	1.27	1.38	1.25	1.34	1.26	1.27
Variance	1.64	1.61	1.91	1.56	1.80	1.59	1.61
Skewness	−.92	−.76	−.51	−.96	−.85	−.96	−1.06
Kurtosis	.46	.14	−.50	.41	.07	.53	.87

Polychoric correlations

We conducted an EFA based on the polychoric correlations. We calculated polychoric correlations instead of Spearman’s rank correlations or Pearson correlations. Given that an ordinal scale was used to measure the seven items, it was not appropriate to use either raw data or a Pearson correlation matrix to perform an EFA. The choice between Spearman rank and polychoric correlations hinges upon whether the underlying distribution is multivariate normal. Since a multivariate normal distribution is likely to underlie the Likert scale of extent (from “strongly agree” to “strongly disagree”), we decided to use the polychoric correlation matrix (Choi et al., 2010). Table 4 reports the polychoric correlations for the seven variables.

Table 4.

Polychoric correlations of sub-sample 1a (n = 422 teachers).

	SIL1	SIL2	SIL3	SIL4	SIL5	SIL6	SIL7
SIL1	1.00
SIL2	.94^*	1.00
SIL3	.77^*	.77^*	1.00
SIL4	.84^*	.82^*	.73^*	1.00
SIL5	.84^*	.85^*	.70^*	.82^*	1.00
SIL6	.70^*	.71^*	.57^*	.71^*	.74^*	1.00
SIL7	.85^*	.85^*	.74^*	.86^*	.85^*	.82^*	1.00

* Correlation is significant at the 0.01 level.

Parallel analysis, EFA, and reliability

We conducted an EFA based on a polychoric correlation matrix. EFA was used instead of principal component analysis (PCA). PCA is appropriate when the nature of the data is event-typed, while EFA is appropriate when the measured items are within the control of the respondents. The practices measured by the questionnaire are not events and are best characterized as the manifestation of latent structures in the respondents’ minds. For example, the item “the principal, teachers, and other staff work collaboratively to develop an instructional vision” does not indicate a specific event. Based on the above considerations, an EFA using polychoric correlations was appropriate for the data analysis.

We used SAS 9.4 to conduct the parallel analysis and the EFA. Before conducting an EFA, a parallel analysis was conducted to determine the number of factors to retain. The pervasive methods of determining the number of factors such as Kaiser criterion and the screen test are accurate 22% and 57% of the time, according to simulation studies (Hayton et al., 2004). Parallel analysis retains the factors that have eigenvalues greater than those derived from randomly generated data having the same number of variables and sample size. In order to avoid a bias, only the factors having eigenvalues greater than the 95th percentile of those from the simulated data are retained.

The results of the parallel analysis recommended that one factor should be retained. Thus, we conducted an EFA retaining one factor. We employed Varimax as the pre-rotation method. There was no rotation because a one-factor solution was suggested. All items loaded above .78 on the factor and were retained because they exceeded the minimum advised retention value of .30 (Tabachnick and Fidell, 2013). Table 5 reports the factor structure and the eigenvalue for each item. The factor accounted for 78.80% of the combined item variance. The Cronbach’s alpha was .95 and the Guttman Split-Half Coefficient was .91, indicating good internal consistency of the items in the scale.

Table 5.

Factor pattern and eigenvalues of sub-sample 1a (n = 422 teachers).

	Factor loading	Eigenvalue
SIL7	0.94	−.10
SIL1	0.94	5.51
SIL2	0.94	.16
SIL5	0.91	−.04
SIL4	0.90	.01
SIL3	0.79	.03
SIL6	0.79	−.06

Sub-sample 1b. Schools with a response rate at or above 10% of total staff

Pearson correlations

There were no missing values in this sample. We aggregated the responses from school members for each school that had a response rate at or above 10% of their total staff, generating a sample of schools (n = 60). We transformed the ordinal data to be continuous after aggregation. Thus, we calculated Pearson correlations as the basis of the EFA. Table 6 reports the descriptive statistics of sub-sample 1b, while Table 7 represents the Pearson correlation matrix.

Table 6.

Descriptive statistics of sub-sample 1b, school-level sample with response rate 10% and above (n = 60).

	SIL1	SIL2	SIL3	SIL4	SIL5	SIL6	SIL7
Mean	4.57	4.45	4.15	4.65	4.35	4.73	4.67
Median	4.78	4.63	4.00	4.78	4.53	4.83	4.92
SD	.86	.83	.93	.81	.86	.88	.86
Variance	.73	.69	.87	.65	.73	.78	.74
Skewness	−.85	−.69	−.23	−.99	−.67	−1.31	−.88
Kurtosis	.44	.12	−.78	.92	−.12	2.49	.36

Table 7.

Pearson correlations of sub-sample 1b (n = 60).

	SIL1	SIL2	SIL3	SIL4	SIL5	SIL6	SIL7
SIL1	1.00
SIL2	.96^*	1.00
SIL3	.84^*	.87^*	1.00
SIL4	.91^*	.92^*	.81^*	1.00
SIL5	.92^*	.93^*	.82^*	.93^*	1.00
SIL6	.74^*	.75^*	.62^*	.77^*	.76^*	1.00
SIL7	.90^*	.92^*	.83^*	.93^*	.91^*	.82^*	1.00

* Correlation is significant at the 0.01 level (2-tailed).

Parallel analysis, EFA, and reliability

A one-factor solution was employed according to the results of the parallel analysis. We conducted an EFA using the same process as sub-sample 1a. All items loaded above .30 on the factor and were therefore retained. Table 8 reports the factor structure and the eigenvalue for each item. The factor accounted for 85.51% variance. The Cronbach’s alpha was .97 and the Guttman Split-Half Coefficient was .94, indicating good internal consistency of the items in the scale.

Table 8.

Factor pattern and eigenvalues of sub-sample 1b.

	Factor loading	Eigenvalue
SIL2	.97	.13
SIL7	.96	−.09
SIL4	.96	−.01
SIL1	.96	5.99
SIL5	.96	−.02
SIL3	.86	.02
SIL6	.79	−.03

We also conducted supplementary school-level EFA analyses on sub-sample 1c (all schools where at least one teacher responded) and sub-sample 1d (schools with a response rate less than 10%). The results of EFA and reliability tests for sub-samples 1c and 1d point to one factor retaining all seven items. Specifically, the single factor extracted for sub-sample 1c explained 71.54% of the common item variance, Cronbach’s alpha was .94, and the Guttman Split-Half Coefficient was .87. For sub-sample 1d, the single extracted factor accounted for 78.81% of the shared item variance, Cronbach’s alpha was .96, and the Guttman Split-Half Coefficient was .91. Since the items focused on the overall practices of a school rather than school staff’s individual behaviors, we decided to conduct our CFA for sample 2 at the school level. In addition, to represent a higher proportion of teachers per school on average and because sub-sample1b (10% or more response rate) explained a greater proportion of total item variance and had higher reliability coefficients than sub-sample 1c, we conducted the CFA for sample 2 with only those schools having a response rate of 10% or more.

Confirmatory factor analysis

We used the same method to impute missing values as we used for the EFA. Of the total 587 respondents, six had missing data. In the CFA we used the schools with a response rate at or above 10% of total staff to better represent each school. We aggregated the responses from staff for each school, generating a continuous data set for the school sample (n = 103). Most of the schools included in this sample had a response rate between 10% and 20%. Table 9 reports the response rates of schools in this sample, and Table 10 reports the descriptive statistics for the imputed data set. Table 11 represents the Pearson correlations between the items. All the values of skewness and kurtosis were between −2 and 2, indicating that the data did not violate the assumption of normality. Thus, the CFA used maximum likelihood as the estimation method.

Table 9.

Response rates.

Number of schools	Response rates	Percentage	Accumulative percentage
50	10–14.99%**	48.54%	48.54%
31	15–19.99%	30.10%	78.64%
13	20–24.99%	12.62%	91.26%
4	25–29.99%	3.88%	95.15%
4	30–44.99%	3.88%	99.03%
1	60%	.97%	100%

** Six schools’ response rates were rounded up to 10%, among which the lowest response rate was 9.52%.

Table 10.

Descriptive statistics for CFA (n = 103).

	SIL1	SIL2	SIL3	SIL4	SIL5	SIL6	SIL7
Mean	4.07	4.02	3.71	4.33	4.18	4.65	4.33
Median	4.14	4.00	3.67	4.40	4.33	4.67	4.38
SD	.67	.69	.73	.76	.76	.66	.73
Variance	.45	.47	.54	.58	.58	.43	.54
Skewness	.77	.71	−.07	.55	.69	.39	.55
Kurtosis	.83	.93	.57	.30	.81	.38	.34

Table 11.

Pearson correlations for CFA (n = 103).

	SIL1	SIL2	SIL3	SIL4	SIL5	SIL6	SIL7
SIL1	1.00
SIL2	.88^*	1.00
SIL3	.67^*	.68^*	1.00
SIL4	.71^*	.71^*	.62^*	1.00
SIL5	.81^*	.79^*	.59^*	.80^*	1.00
SIL6	.65^*	.65^*	.42^*	.62^*	.77^*	1.00
SIL7	.82^*	.82^*	.67^*	.77^*	.80^*	.66^*	1.00

* Correlation is significant at the 0.01 level (2-tailed).

We used LISREL 9.2 to conduct the CFA. Figure 1 represents the final model. Drawing on the literature, we modeled error covariance between questions 1 and 2, 4 and 5, and 5 and 6 to take other common influences for the items into the calculation of the model fit indices. We did this because the three pairs of items were closely related to each other in terms of the school practices that they represented. The error covariance between the three pairs was added one by one. The whole model was reanalyzed each time an error covariance was freed. These changes were made for several reasons. First, collaboratively developing and communicating an instructional vision (items 1 and 2) might covary because of the connection between constructing and articulating school goals (Harmon and Schafft, 2009). Next, collaboratively ensuring resources for high-quality instruction and making instructional decisions (items 4 and 5) might covary due to the community’s value and demand related to some instructional programs (Major, 2013). Finally, collaboratively making instructional decisions and examining student achievement data (items 5 and 6) might covary due to the degree of the school’s autonomy in instructional issues and the support from the local accountability system (Wohlstetter et al., 2008). In the final model, all the indices indicated a good model fit. All the structure coefficients and factor loadings were significant (p < .001). The value of Chi-Square was non-significant, indicating a good model fit (p = .33). The value of RMSEA was .04, which was smaller than .05. The value of SRMR was .03, which was smaller than .05. The value of GFI was .97. The model is over-identified (df = 11) (Hooper et al., 2008).

Figure 1.

Final model of shared instructional leadership.

Discussion

These results are based on teachers’ perceptions of SIL. In the schools that value collective contributions from school members in leadership practices, the principal, teachers, and other staff tend to collaborate on the following activities: developing and communicating an instructional vision, improving instructional programs, securing resources for instruction, examining student achievement data, and identifying community partnerships that support instructional improvement efforts. Our study findings suggest principals, teachers, and other staff seem to have less collaborative input in identifying potential community partnerships, but more input in the practices involving instruction, resources, and examining student achievement data. Even so, as our measurement model results suggest, the degree to which teachers experience any of these types of SIL in their schools tends to covary positively with the other indicators of SIL found in the measure developed in this study.

Question 3 had the lowest mean across the four sub-samples in the EFA, indicating that while it covaried positively with the other six items, respondents tended to report that “the principal, teachers, and other staff work collaboratively to identify potential community partnerships that align with school’s goals” to a somewhat lesser degree than the other types of SIL found in the survey, perhaps because this item was the most directly focused on sharing instructional leadership with actors outside the school. In contrast, the items asking questions about collaborative efforts in instructional vision, instruction, resources, and student achievement data received more agreement from the respondents. Thus, while all of the collaborative practices reflected in the SILS (e.g., defining and communicating the school’s mission, resource management, managing curriculum and assessment, and promoting school-community partnerships) have already received rich support from the existing literature (Blase and Blase, 2000; Hallinger, 2003; Leithwood, 1994), educators may tend to focus less on opportunities to share leadership outside the school than within. In order to change this, principals may need to help staff members consider how community partnerships can strengthen the school’s instructional program.

Although the items’ factor loadings were stable across the four EFA sub-samples, the school-level data from the schools with higher response rates seemed to best serve our examination of the survey’s psychometric properties. This is because the factor derived from this sample explained the highest proportion of shared variance among the items compared with the factors from other samples. The values of kurtosis and skewness tended to increase when the analysis switched from the individual-level data and the schools with a response rate at or above 10%, to the schools with a lower response rate and the sample of all schools. One possible explanation is related to the low response rates of some schools. Many of these schools had only 1 or 2 responses, which meant that the two latter samples literally mixed the individual-level data and school-level data. This mixture might have caused weaker normality. Although none of these samples violated the assumption of normality, we suggest keeping the data consistent in level, either individual level or school level. However, considering that the survey aims to assess a school-level construct, we suggest using school-level data to represent the school more accurately.

The modifications of the model in CFA indicated a correlation between three pairs of items: developing and communicating an instructional vision, ensuring resources for high-quality instruction and making instructional decisions, and making instructional decisions and examining student achievement data. The principal, teachers, and other school staff tended to contribute more collaboratively to these paired practices. If they collaboratively developed an instructional vision (item 1), they also tended to communicate the vision in a shared way (item 2). Also, if the principal, teachers, and other staff worked collaboratively to ensure resources for high-quality instruction (item 4), they also tended to make shared contributions on other instructional decisions (item 5). In schools with SIL, it may be that principals empower school staff to use their knowledge in data-driven decision-making processes because this approach improves data relevance (item 5 and item 6) (Heck and Hallinger, 2009; Park and Datnow, 2009; Wayman, 2005). Overall, the results support that the responses of teachers to the items were driven by the underlying unidimensional construct of SIL theorized based on our literature review.

Limitations and future research

Although the concept of SIL involves the principal, teachers, and other staff, the sample used in this study included only secondary teachers, which means the factor structure is based on secondary teachers’ perspectives. Future studies should attempt to explore the perspectives of principals and other staff, to permit the generalization of the measure to other populations or to explore differences of factor structure across these populations. Furthermore, the sample schools were characterized as public secondary schools. The factor structure and internal consistency should be re-examined in samples with different characteristics, such as elementary schools and private schools.

Three rounds of reminders were conducted before the completion of survey response collection. The overall individual-level response rate of the first sample was 6.80%, and of the second was 12.50%. Most of the schools sampled in this study had response rates ranging from 10% to 20%. The individual-level response rate of the final sample (CFA sample) was 15.60%, and the average organizational-level response rate of the final sample (response rate above 10%) was 16.42%, which was slightly beyond one standard deviation (18.20) of the average (35.00%) derived from an overall examination of response rates of organizational research (Baruch and Holtom, 2008). Our data collection was conducted a decade after that meta-analysis. Response rates for organizational studies might have declined due to increasing survey saturation (Rogelberg and Stanton, 2007). Additionally, it is possible that non-respondents may have lower satisfaction with their organizations (Rogelberg et al., 2000). The distribution of response scores that did not cluster at the lower or higher level of SIL provided evidence that respondents may fairly represent the population to some extent. However, we note that the rates are low compared with several studies using survey in the field of education. It may have limited the representation of teachers’ opinions.

We believe that the SILS represents a valid measure of the construct of SIL and that it can be used by future researchers; however, we suggest that future researchers examine the psychometric properties of the SILS in a sample of schools with higher mean response rates. Face-to-face instead of online data collection, as well as incentives, may help future researchers reach a higher response rate.

The SILS instrument reflects important facets of SIL, namely shared instructional vision-setting, a focus on instruction, monitoring of progress, and broad collaboration. We found the factor structure quite stable across four sub-samples of teachers and schools with different response rates. However, future researchers may wish to attend to the relatively lower loadings on some items such as teacher participation in community partnerships, as this work may primarily involve principals.

Finally, future researchers should also conduct concurrent and predictive validity tests to further understand the usefulness of the survey and the processes and outcomes with which it may be associated. Researchers can use this instrument to assess the SIL of schools and then examine the relationship between SIL and other outcomes, such as the quality of instructional practice, student and teacher social and emotional well-being, and student learning. Future research might also examine teacher efficacy beliefs as a mediator between SIL and student achievement, to examine the survey’s predictive validity (Comrey, 1988).

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Xi Zhan

Anika B Anthony

Karen Stansberry Beard

References

Anthony

Gimbert

Luke

Hurt

(2019) Distributed leadership in context: Teacher leaders’ contributions to novice teacher induction. Journal of School Leadership 29(1): 54–83.

Baruch

Holtom

(2008) Survey response rate levels and trends in organizational research. Human Relations 61(8): 1139–1160.

Blase

(2000) Effective instructional leadership: teachers’ perspectives on how principals promote teaching and learning in schools. Journal of Educational Administration 38(2): 130–141.

Bolman

Deal

(2013) Reframing Organizations: Artistry, Choice and Leadership. San Francisco: Jossey-Bass.

Bryk

Sebring

Allensworth

Luppescu

Easton

(2010) Organizing Schools for Improvement: Lessons from Chicago. Chicago: University of Chicago Press.

Choi

Peters

Mueller

(2010) Correlational analysis of ordinal data from Pearson’s r to Bayesian polychoric correlation. Asia Pacific Education Review 11(4): 459–466.

City

Elmore

Fiarman

Teitel

(2009) Instructional Rounds in Education. Cambridge, MA: Harvard Education Press, 30.

Cohen

(1981) Sociology looks at team teaching. Research in Sociology of Education and Socialization 2: 163–193.

Coldren

Spillaine

(2007) Making connections to teaching practice: the role of boundary practices in instructional leadership. Educational Policy 21(2): 369–396.

10.

Comrey

(1988) Factor-analytic methods of scale development in personality and clinical psychology. Journal of Counseling and Clinical Psychology 56(5): 754–761.

11.

Damanik

Aldridge

(2017) Transformational leadership and its impact on school climate and teachers’ self-efficacy in Indonesian high schools. Journal of School Leadership 27(2): 269–296.

12.

DeVellis

(2012) Scale Development: Theory and Applications. Thousand Oaks, CA: Sage.

13.

Dwyer

(1984) The search for instructional leadership: routines and subtleties in the principal’s role. Educational Leadership 41(5): 32–37.

14.

Epstein

(2001) School, Family, and Community Partnerships: Preparing Educators and Improving Schools. Boulder: Westview Press.

15.

Fichman

Cummings

(2003) Multiple imputation for missing data: making the most of what you know. Organizational Research Methods 6(3): 282–308.

16.

Finnigan

Stewart

(2009) Leading change under pressure: an examination of principal leadership in low-performing schools. Journal of School Leadership 19(5): 586–618.

17.

Glickman

(1989) Has Sam and Samantha’s time come at last? Educational Leadership 46(8): 4–9.

18.

Goddard

Kim

Miller

(2015) A theoretical and empirical analysis of the roles of instructional leadership, teacher collaboration, and collective efficacy beliefs in support of student learning. American Journal of Education 121(4): 501–530.

19.

Goddard

Hoy

(2004) Collective efficacy beliefs: Theoretical developments, empirical evidence, and future directions. Educational Researcher 33(3): 3–13.

20.

Goldring

Cravens

Murphy

Porter

Elliott

Carson

(2009a) The evaluation of principals: what and how do states and urban districts assess leadership? The Elementary School Journal 110(1): 19–39.

21.

Goldring

Porter

Murphy

Elliott

Cravens

(2009b) Assessing learning-centered leadership: connections to research, professional standards, and current practices. Leadership and Policy in Schools 8(1): 1–36.

22.

Graczewski

Knudson

Holtzman

(2009) Instructional leadership in practice: what does it look like, and what influence does it have? Journal for Students Placed at Risk 14(1): 72–96.

23.

Hallinger

(2003) Leading educational change: reflections on the practice of instructional and transformational leadership. Cambridge Journal of Education 33(3): 329–352.

24.

Hallinger

(2005) Instructional leadership and the school principal: a passing fancy that refuses to fade away. Leadership and Policy in Schools 4: 221–239.

25.

Hallinger

(2010) Developing instructional leadership. In: Brent

(ed.) Developing Successful Leadership. Dordrecht: Springer, 61–76.

26.

Hallinger

Heck

(2010) Leadership for learning: does collaborative leadership make a difference in school improvement? Educational Management Administration & Leadership 38(6): 654–678.

27.

Hallinger

Murphy

(1985) Assessing the instructional management behavior of principals. The Elementary School Journal 86(2): 217–247.

28.

Harmon

Schafft

(2009) Rural school leadership for collaborative community development. The Rural Educator 30(3): 4–9.

29.

Harris

(2004) Distributed leadership and school improvement: leading or misleading? Educational Management Administration & Leadership 32: 11–24.

30.

Hayton

Allen

Scarpello

(2004) Factor retention decisions in exploratory factor analysis: a tutorial on parallel analysis. Organizational Research Methods 7(2): 191–205.

31.

Heck

Hallinger

(2009) Assessing the contribution of distributed leadership to school improvement and growth in math achievement. American Educational Research Journal 46(3): 659–689.

32.

Heck

Larsen

Marcoulides

(1990) Instructional leadership and school achievement: validation of a causal model. Educational Administration Quarterly 26(2): 94–125.

33.

Hooper

Coughlan

Mullen

(2008) Structural equation modelling: guidelines for determining model fit. Electronic Journal of Business Research Methods 6(1): 53–60.

34.

Hoy

Miskel

Tarter

(2013) Educational Administration: Theory, Research and Practice. New York: McGraw-Hill.

35.

Hulpia

Devos

Van Keer

(2009) The influence of distributed leadership on teachers’ organizational commitment: a multilevel approach. The Journal of Educational Research 103(1): 40–52.

36.

Jackson

Marriott

(2012) The interaction of principal and teacher instructional influence as a measure of leadership as an organizational quality. Educational Administration Quarterly 48(2): 230–258.

37.

Johnson

Reinhorn

Simon

(2018) Ending isolation: the payoff of teacher teams in successful high-poverty urban schools. Teachers College Record 120(5): 1–46.

38.

Kaplan

Owings

(1999) Assistant principals: the case for shared instructional leadership. NASSP Bulletin 83(610): 80–94.

39.

King

Newmann

(2001) Building school capacity through professional development: conceptual and empirical considerations. International Journal of Educational Management 15(2): 86–93.

40.

Lambert

(2002) A framework for shared leadership. Educational Leadership 59(8): 37–40.

41.

Lee

Hallinger

Walker

(2012) A distributed perspective on instructional leadership in International Baccalaureate (IB) schools. Educational Administration Quarterly 48(4): 664–698.

42.

Leithwood

(1994) Leadership for school restructuring. Educational Administration Quarterly 30(4): 498–518.

43.

Leithwood

Jantzi

(1999) Transformational school leadership effects: a replication. School Effectiveness and School Improvement 10(4): 451–479.

44.

Leithwood

Jantzi

(2000) Principal and teacher leadership effects: a replication. School Leadership and Management 20(4): 415–434.

45.

Little

(1990) Teachers and colleagues. In: Lieberman

(ed.) Schools as Collaborative Cultures: Creating the Future Now. Bristol, PA: The Falmer Press; Taylor and Francis, 164–193.

46.

Lortie

(1975) Schoolteacher: A Sociological Study. Chicago: University of Chicago Press.

47.

Louis

Marks

(1998) Does professional community affect the classroom? Teachers’ work and student experiences in restructuring schools. American Journal of Education 106(4): 532–575.

48.

Major

(2013) How they decide: a case study examining the decision-making process for keeping or cutting music in a K–12 public school district. Journal of Research in Music Education 61(1): 5–25.

49.

Marks

Printy

(2003) Principal leadership and school performance: an integration of transformational and instructional leadership. Educational Administration Quarterly 39(3): 370–397.

50.

Merrill

Daugherty

(2010) STEM education and leadership: a mathematics and science partnership approach. Journal of Technology Education 21(2): 21–34.

51.

Newmann

Smith

Allensworth

Bryk

(2001) Instructional program coherence: what it is and why it should guide school improvement policy. Educational Evaluation and Policy Analysis 23(4): 297–321.

52.

Ogawa

Bossert

(1995) Leadership as an organizational property. Educational Administration Quarterly 31: 224–243.

53.

Park

Datnow

(2009) Co-constructing distributed leadership: district and school connections in data-driven decision-making. School Leadership and Management 29(5): 477–494.

54.

Pennsylvania Department of Education (2020) Teacher and principal evaluation survey. Available at: www.education.pa.gov/Teachers%20-%20Administrators/Teacher%20and%20Principal%20Evaluation%20Survey/Pages/default.aspx (accessed 16 April 2020).

55.

Pounder

Ogawa

Adams

(1995) Leadership as an organization-wide phenomena: its impact on school performance. Educational Administration Quarterly 31(4): 564–588.

56.

Printy

Marks

(2006) Shared leadership for teacher and student learning. Theory into Practice 45(2): 125–132.

57.

Printy

Marks

Bowers

(2009) Integrated leadership: how principals and teachers share instructional influence. Journal of School Leadership 19(5): 504–532.

58.

Rogelberg

Stanton

(2007) Understanding and dealing with organizational survey non-response. Organizational Research Methods 10: 195–209.

59.

Rogelberg

Luong

Sederburg

Cristol

(2000) Employee attitude surveys: examining the attitudes of noncompliant employees. Journal of Applied Psychology 85(2): 284–293.

60.

Sanders

Harvey

(2002) Beyond the school walls: a case study of principal leadership for school-community collaboration. Teachers College Record 104(7): 1345–1368.

61.

Somech

(2005) Directive versus participative leadership: two complementary approaches to managing school effectiveness. Educational Administration Quarterly 41(5): 777–800.

62.

Spillane

(2004) Educational leadership. Educational Evaluation and Policy Analysis 26(2): 169–172.

63.

Spillane

(2006) Distributed Leadership. San Francisco: Jossey-Bass.

64.

Spillane

Halverson

Diamond

(2001) Investigating school leadership practice: a distributed perspective. Educational Researcher 30(3): 23–28.

65.

Tabachnick

Fidell

(2013) Using Multivariate Statistics. Upper Saddle River, NJ: Pearson.

66.

Teh

Chiang

Lipscomb

Gill

(2014) Measuring school leaders’ effectiveness: An interim report from a multiyear pilot of Pennsylvania’s framework for leadership Available at: https://ies.ed.gov/ncee/edlabs/regions/midatlantic/pdf/REL_2015058.pdf. (accessed 16 April 2020).

67.

Urick

(2016) Examining US principal perception of multiple leadership styles used to practice shared instructional leadership. Journal of Educational Administration 54(2): 152–172.

68.

Urick

Bowers

(2014) What are the different types of principals across the United States? A latent class analysis of principal perception of leadership. Educational Administration Quarterly 50(1): 96–134.

69.

Virginia Department of Education (2020) Guidelines for uniform performance standards and evaluation criteria for principals. Available at: www.doe.virginia.gov/teaching/performance_evaluation/principal/index.shtml (accessed 16 April 2020).

70.

Wahlstrom

Louis

(2008) How teachers experience principal leadership: the roles of professional community, trust, efficacy, and shared responsibility. Educational Administration Quarterly 44(4): 458–495.

71.

Wayman

(2005) Involving teachers in data-driven decision making: using computer data systems to support teacher inquiry and reflection. Journal of Education for Students Placed at Risk 10(3): 295–308.

72.

Wohlstetter

Datnow

Park

(2008) Creating a system for data-driven decision-making: applying the principal-agent framework. School Effectiveness and School Improvement 19(3): 239–259.

73.

Yost

Vogel

Rosenberg

(2009) Transitioning from teacher to instructional leader. Middle School Journal 40(3): 20–27.

74.

Yuan

(2000) Multiple imputation for missing data: concepts and new development. In: Proceedings of the Twenty-Fifth Annual SAS User’s Group International Conference, SUGI, April 2000, paper 267-25. Rockville, MD: SAS Institute Inc. Available at: https://support.sas.com/rnd/app/stat/papers/multipleimputation.pdf. (accessed 16 April 2020).

75.

Zepeda

(2013) The Principal as Instructional Leader: A Practical Handbook. Larchmont, NY: Eye on Education.