Content uploaded by Helen M. Marks
Author content
All content in this area was uploaded by Helen M. Marks on Feb 11, 2015
Content may be subject to copyright.
10.1177/0013161X03253412 ARTICLEEducational Administration QuarterlyMarks, Printy / SCHOOL LEADERSHIP RELATIONS
Principal Leadership and School Performance:
An Integration of Transformational and
Instructional Leadership
Helen M. Marks
Susan M. Printy
Focusing on school leadership relations between principals and teachers, this study ex-
amines the potential of their active collaboration around instructional matters to en-
hance the quality of teaching and student performance. The analysis is grounded in two
conceptions of leadership—transformational and instructional. The sample comprises
24 nationally selected restructured schools—8elementary, 8 middle, and 8 high schools.
In keeping with the multilevel structure of the data, the primary analytic technique is hi-
erarchical linear modeling (HLM). The study finds that transformational leadership is a
necessary but insufficient condition for instructional leadership. When transformational
and shared instructional leadership coexistin an integrated form of leadership, the influ-
ence on school performance, measured by the quality of its pedagogy and the achieve-
ment of its students, is substantial.
Keywords: school performance; school leadership; shared instructional leadership
Schools depend on leadership throughout the organization to shape pro-
ductive futures through a process of self-renewal (Senge et al., 1999, 2000).
To enlargetheleadershipcapacityof schools attempting toimprove theiraca-
demic performance, some principals involve teachers in sustained dialogue
and decision making about educational matters. While remaining central
370
Educational Administration Quarterly
Vol. 39, No. 3 (August 2003) 370-397
Authors’ Note: The project from which this study draws its data was conducted under the aus-
pices of the Wisconsin Center for Education Research and the Center on Organization of
Schools and supported by the U.S. Department of Education, Office of Educational Research
and Improvement (Grant No. R117Q000005-95). An earlier version of the article was presented
at the 2001 Annual Meeting of the American Educational Research Association held in Seattle,
Washington.The opinions expressed are those of the authors and do not necessarily reflect those
of the supporting agencies.
DOI: 10.1177/0013161X03253412
© 2003 The University Council for Educational Administration
agents for change, these principals recognize teachers as equal partners in
this process, acknowledging their professionalism and capitalizing on their
knowledge and skills (Darling-Hammond, 1988; Rowan, 1990).
Focusing on school leadership relations between principals and teachers,
this study examines the potential of their active collaboration around instruc-
tional matters to enhance the quality of teaching and student performance.
The analysis is grounded in a comparison of two conceptions of leadership—
transformational and instructional. Functioning as leaders, principals can
serve to transform school cultures or to maintain them (Firestone & Louis,
1999; Leithwood & Jantzi, 1999). Transformational leadership, put briefly,
provides intellectual direction and aims at innovating within the organiza-
tion, while empowering and supporting teachers as partners in decision mak-
ing (Conley & Goldman, 1994; Leithwood, 1994). Instructional leadership,
as we reconceptualize it, replaces a hierarchical and procedural notion with a
model of “shared instructional leadership.”
Sharedinstructional leadership involvesthe activecollaboration ofprinci-
pal and teachers on curriculum, instruction, and assessment. Within this
model, the principal seeks out the ideas, insights, and expertise of teachers in
these areas and works with teachers for school improvement. The principal
and teachers share responsibility for staff development, curricular develop-
ment,and supervision ofinstructional tasks. Thus,theprincipal is notthe sole
instructional leader but the “leader of instructional leaders” (Glickman,
1989, p. 6).
We investigate these conceptions of leadership and their relationship to
school performance, measured as pedagogical quality and student achieve-
ment, within a sample of 24 nationally selected restructuring elementary,
middle, and high schools (Newmann & Associates, 1996). At the time of the
study, all of these schools were decentralized and practiced a form of site-
based management. In almost all of the schools, teachers reportedly exer-
cised substantial influence on school practice in matters of curriculum,
instruction, and assessment (Marks & Louis, 1997). However, because
administrators and teachers varied in their commitment to mutual collabora-
tion and continuous improvement through learning, we are able to study the
relationship of transformational and shared instructional leadership to the
quality of teaching and learning.
BACKGROUND
Two primary images of school principalship have prevailed in recent
decades—instructional leadership and transformational leadership (Hallinger,
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 371
1992). Instructional leadership, developed during the effective schools movement
ofthe 1980s, viewedthe principal as theprimary source ofeducationalexper-
tise. Aimed at standardizing the practice of effective teaching, the principal’s
role was to maintain high expectations for teachers and students, supervise
classroom instruction, coordinate the school’s curriculum, and monitor stu-
dent progress (Barth, 1986). For principals who lacked the skills to accom-
plish these tasks, coaching and on-site assistance were in short supply.
Instructional leadership in practice fell far short of the ideal (Cuban, 1984;
Murphy & Hallinger, 1987).
Moreover, the hierarchical orientation of instructional leadership con-
flicted with the democratic and participative organization of schools that
emerged in the late 1980s with school restructuring and the movement to
empowerteachers asprofessionaleducators (Marks &Louis, 1997). Because
critics had attributed to the educational bureaucracy schools’failure to edu-
cate effectively (Carnegie Forum on Education and the Economy, 1986), a
fundamental restructuring initiative entailed decentralizing to schools authority
over such matters as budgets, hiring, curriculum, and instruction. When prin-
cipals adopted this model fully, they shared management decisions with
teachers and other constituents (Malen, Ogawa, & Kranz, 1990).
Because teachers possessed critical information about their students and
how they learn, teachers needed discretionary authority to make their own
curricular and instructional decisions (Hallinger, 1992; Sykes, 1990). The
latitude to make such decisions would improve both teachers’work life and
student achievement (Darling-Hammond & Goodwin, 1993; Maeroff, 1988;
Schlechty, 1990). Moreover, educational reform had a greater chance of suc-
cesswhen teachers were involved(Blase &Kirby,2000;Conley & Goldman,
1994). Functioning in leadership capacities (e.g., site council chairs, staff
developers, or lead teachers), teachers could shape the goals and cultures of
their schools while retaining their ties to the classroom (Conley & Goldman,
1994). In so doing, teachers gained greater legitimacy as leaders (Little,
1988; Smylie & Denny, 1990).
To accomplishthe reforms centraltoschool restructuring, scholarsof edu-
cation espoused a model of transformational leadership. Transformational
leadership focuses on problem finding, problem solving, and collaboration
with stakeholders with the goal of improving organizational performance
(Hallinger, 1992). To develop the collective capacity of the organization and
its members to achieve these results, transformational leadership seeks to
raise participants’ level of commitment (Burns, 1978), to encourage them in
reaching their fullest potential (Bass & Avolio, 1993), and to support them in
transcending their own self-interest for a larger good (Bass & Avolio, 1993;
372 Educational Administration Quarterly
Leithwood, Tomlinson, & Genge, 1996; Sagor & Barnett, 1994; Silins,
Muford, Zarins, & Bishop, 2000).
Transformational leadership affirmed the centrality of the principal’s
reform role, particularly in introducing innovation and shaping organiza-
tional culture (Conley & Goldman, 1994; Leithwood, 1994). While concen-
trating on renewing the organization and its personnel, however, trans-
formationalleadership lacked an explicitfocus on curriculumandinstruction
(Hallinger & Leithwood, 1998). Transformational and shared instructional
leadership are complementary, in our view, but neither conceptualization
embraces the other. When they operate in tandem, however, the leadership
approaches are integrated. Few studies have examined these relationships
empirically, the objective of this study.
THEORETICAL FRAMEWORK
Instructional Leadership Conventionally Understood
Instructional leadership, narrowly defined, focuses on leadership func-
tions directly related to teaching and learning (Murphy, 1988). In a broader
view, instructional leadership also refers to all other functions that contribute
to student learning, including managerial behaviors (Donmoyer & Wagstaff,
1990; Murphy, 1988). Such an action orientation theoretically encompasses
everything a principal does during the day to support the achievement of stu-
dents and the ability of teachers to teach (Sebring & Bryk, 2000).
In a review of the literature on instructional leadership, Murphy (1990)
noted that principals in productive schools—that is, schools where the qual-
ity of teaching and learning were strong—demonstrated instructional leader-
ship both directly and indirectly. Although these principals practiced a con-
ventional rather than a shared form of instructional leadership, they
emphasized four sets of activities with implications for instruction: (a) devel-
oping the school mission and goals; (b) coordinating, monitoring, and evalu-
ating curriculum, instruction, and assessment; (c) promoting a climate for
learning; and (d) creating a supportive work environment (Murphy, 1990).
Focused on learning, they infused management decisions and regular school
routines with educational meaning (Dwyer, 1984).
In the context of teacher professionalization, however, critics regarded the
existing models of instructional leadership as paternalistic, archaic, and
dependent on docile followers (Burlingame, 1987; Poplin, 1992; Sheppard,
1996). If teachers are committed and competent, they argued, traditional
forms of instructional leadership are not appropriate (Sergiovanni, 1991).
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 373
Rather, principals should be concerned with facilitating teachers’exercise of
initiative and responsibility in instructional matters (Glanz & Neville, 1997;
Senge et al., 2000). Such an approach is consistent with educational reforms
in the professionalization of teaching that equip teachers to play informed
and active roles in improving schooling (Little, 1993).
Shared Instructional Leadership
Unlike the conventional notion of instructional leadership, shared instruc-
tional leadership is an inclusive concept, compatible with competent and
empowered teachers. The principal invests teachers with resources and
instructionalsupport(Rosenblum, Louis, & Rossmiller,1994) and maintains
congruence and consistency of the educational program (Conley & Goldman,
1994). Teachers’ participation in shared instructional leadership occurs
informally as well as being manifest in formal roles (Prestine & Bowen,
1993). Teachers assume leadership responsibility when they interact with
other adults in the school community around school reform efforts, encour-
age others to improve their professional practice, or learn together with their
school colleagues (Moller & Katzenmeyer, 1996).
Several models of shared instructional leadership recast the process of
instructionalsupervision. In these models, teachersassume responsibility for
their professional growth and for instructional improvement. The principal
becomes less an inspector of teacher competence and more a facilitator of
teacher growth (Poole, 1995). Whereas the principal remains the educational
leader of the school, teachers, who have requisite expertise or information,
exercise leadership collaboratively with the principal. Collaborative inquiry
supplants principal-centered supervisory practices (Reitzug, 1997). As teachers
inquire together, they encourage each other toward answers for instructional
problems.Leadershipfor instruction emerges from boththeprincipal and the
teachers. Principals and teachers discuss alternatives rather than directives or
criticisms and work together as “communities of learners” in service to stu-
dents (Blase & Blase, 1999). Principals contribute importantly to these com-
munities when they promote teacher reflection and professional growth.
When teachers interact with principals as they engage in these activities, the
teachers report positive changes in their pedagogical practices, including
usingvarious and innovativetechniques and being willingto take risks(Blase
& Blase, 1999).
Principals and teachers both play a part in forging an effective leadership
relationship. Principals must provide opportunities for teacher growth, but
teachers are also responsible for seizing these opportunities (Blase & Kirby,
2000). Strong leadership on the part of the principal, however, often affirms
374 Educational Administration Quarterly
teachers’ responsibility and accountability for change (Louis, 1994). The
relationship is a reciprocal one, where those in formal roles step aside to let
others step into leadership roles (Prestine & Bowen, 1993). This phenome-
non is often subtle and might not be readily apparent except in certain critical
incidents that threaten change efforts (Prestine & Bowen).
Shared instructional leadership, therefore, is not dependent on role or
position. Its currency lies in the personal resources of participants and is
deployed through interaction (Ogawa & Bossert, 1995). Such leadership
extends throughout the organization with revised structures permitting coor-
dinatedaction (Ogawa& Bossert, 1995;Pounder, Ogawa,& Adams,1995).
Transformational Leadership
Transformational leadership has been the subject of systematic inquiry in
nonschool organizations for several decades. Supplying conceptual ground-
ingfortransformational leadership, Burns (1978) focusedon the relationship
between the leader and the “followers.” When the relationship focuses on the
continuing pursuit of higher purposes, change for the better occurs both in
the purposes and resources of those involved and in the relationship itself.
Whereas the transformational leader plays a pivotal role in precipitating change,
followers and leaders are bound together in the transformation process.
The importance of developing followers to their fullest potential extended
the concept of transformational leadership (Bass, 1985; Bass & Avolio,
1993). Transformational leaders motivate followers by raising their con-
sciousness about the importance of organizational goals and by inspiring
them to transcend their own self-interest for the sake of the organization. In
theirrelationships with followers,this theory posits, transformationalleaders
exhibit at least one of these leadership factors: idealized influence, inspira-
tionalmotivation,intellectualstimulation, and individualizedconsideration.
Leithwood and colleagues have described and assessed the effectiveness
oftransformationalleadership in schools (Leithwood, 1994,1995; Leithwood,
Dart, Jantzi, & Steinbach, 1993; Leithwood et al., 1996; Leithwood & Jantzi,
1990; Leithwood, Jantzi, & Fernandez, 1994; Leithwood, Jantzi, & Steinbach,
1999). They have distinguished nine functions of transformational leader-
ship clustering in three areas—those that are (a) mission centered (develop-
ing a widely shared vision for the school, building consensus about school
goals and priorities), (b) performance centered (holding high performance
expectations, providing individualized support, supplying intellectual stimu-
lation),and(c) culture centered (modeling organizationalvalues,strengthen-
ing productive school culture, building collaborative cultures, and creating
structures for participation in school decisions).
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 375
By seeking to foster collaboration and to activate a process of continuous
inquiry into teaching and learning, transformational leaders attempt to shape
a positive organizational culture and contribute to organizational effective-
ness (Fullan, 1991; Leithwood et al., 1996). But even in collaborative cul-
tures where principals’ transformational efforts encourage teachers to con-
tribute leadership and expertise in teaching and learning, principals have a
central and explicit role in instruction (Sebring & Bryk, 2000; Sheppard,
1996). When principals who are transformational leaders accept their
instructional role and exercise it in collaboration with teachers, they practice
an integrated form of leadership.
The Leadership Models and School
Performance: Research Findings
Investigating three domains of principal instructional leadership, Heck,
Larsen, and Marcoulides (1990) demonstrated both direct and indirect
effectson studentachievementfortheir measures of principalinfluence oper-
ating through school governance, instructional organization, and school cli-
mate. Specifically, an inclusive approach to governance worked to promote
an effective system of instructional organization and a school climate sup-
portive of teaching and learning. Their study found direct effects on achieve-
mentfor instructional organization andclimate and indirect effectsfor gover-
nancethrough its positiveinfluence on thesetwo domains. Effortsthat builda
sense of teamwork in the school proved particularly important, such as clari-
fying, coordinating, and communicating a unified educational mission to
teachers, students, and community. Important instructional leadership vari-
ables that influence achievement are not those tied to close supervision of
instruction(Hallinger& Murphy,1985; Heck et al., 1990).Todate, studies of
instructional leadership have not evaluated its relationship to pedagogical
quality.
To improve organizational performance (Hallinger, 1992), transforma-
tional school leaders focus on the individual and collective understandings,
skills, and commitments of teachers. Transformational leaders may chal-
lenge teachers to examine their assumptions about their work and to rethink
instructional processes; they may establish expectations for quality peda-
gogy and support teachers’ professional growth (Leithwood, Jantzi, &
Steinbach, 1998; Leithwood, Leonard, & Sharratt, 1998). To our knowledge,
no study has investigated these relationships empirically. Moreover, although
transformational principals can enhance student engagement in learning,
studies have not shown any direct effects on student achievement (Leithwood,
1994; Silins et al., 2000).
376 Educational Administration Quarterly
Linking Transformational and Shared
Instructional Leadership: Theory of Action
Although the importance transformational leadership places on vision
building can create a fundamental and enduring sense of purpose in the orga-
nization, the model lacks an explicit focus on teaching and learning. Instruc-
tional leadership, emphasizing the technical core of instruction, curriculum,
and assessment, provides direction and affects the day-to-day activities of
teachers and students in the school. The action orientation of shared instruc-
tional leadership moves a school staff forward to accomplish each goal and,
in so doing, to enact the vision. Transformational leadership builds organiza-
tional capacity whereas instructional leadership builds individual and collec-
tive competence. Instructional leadership is shared, in that specific leader-
ship functions are carried out by many people working in collaboration
(Firestone, 1996).
The theory of action underlying this model holds that the efficacious prin-
cipal works simultaneously at transformational and instructional tasks. As a
transformational leader, the principal seeks to elicit higher levels of commit-
ment from all school personnel and to develop organizational capacity for
school improvement. As an instructional leader, the principal collaborates
with teachers to accomplish organizational goals for teaching and learning.
Whereas these leadership dimensions are analytically distinct, they may
coherein practice in an integratedmodel of leadership. Integratedleadership,
then, reflects the transformational influence of the principal and the shared
leadership actions of the principal and teachers.
Hypothesis and Research Questions
Building on the premise outlined above, we hypothesize that while trans-
formational leadership is necessary for reform-oriented school improve-
ment, it is insufficient to achieve high-quality teaching and learning. Shared
instructional leadership, its essential complement, describes the dynamic
collaboration between the principal and teachers on curricular, instructional,
and assessment matters to further the core technology of schools—teaching
and learning.
Thus, we inquire into the relationship of transformational and shared in-
structional leadership to the pedagogical practice of teachers and to student
performance on authentic measures of achievement. Recognizing that
schools provide a context for teaching and learning that is shaped by the ages
or grade levels of the students enrolled and, as well, by compositional or de-
mographic factors, we pose three research questions:
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 377
1. What is the relationship between transformational and shared instructional
leadership in restructuring elementary, middle, and high schools?
2. How do schools with varying approaches to leadership differ according to
their demographics, organization, and performance?
3. What is the effect of transformational and shared instructional leadership on
school performance as measured by the quality of pedagogy and the achieve-
ment of students?
METHOD
Sample and Data
To studyschool restructuring inthe United States,the Center on Organiza-
tion and Restructuring of Schools undertook a national search for public
schools that had made substantial progress in their reform efforts. Out of a
nationally nominated pool of 300 schools, the center selected 24 elementary,
middle, and high schools, 8 at each grade level, to participate in its School
Restructuring Study (SRS). Despite the selection criteria for nomination and
inclusion in the study, the schools in the SRS sample varied substantially in
their goals, their capacity for reform, and their success in restructuring. (See
Berends & King, 1992, and Newmann & Associates, 1996, for additional
detailson sample selectionand for profilesof the SRSschools.) Representing
16 states and 22 school districts, most of the SRS schools are urban, enrolling
substantial proportions of economically disadvantaged and minority
students.
Comparedwith public schools nationally,schools inthis sample are larger
(enrolling, on average, 777 students compared with a national average of 522
students). In the sample elementary and middle schools, NAEP achievement
levels in reading and mathematics are at or above the national average. In the
high schools, NAEP achievement is below the national average (a result that
may be attributable to the high school sample—mostly 9th- and 10th-grade
students taking a NAEP test normed for 12th-grade students). (See also
Marks and Louis, 1997.)
Our study employs several of the quantitative and qualitative instruments
that were part of the SRS design. Teachers responded to a survey querying
them about their instructional practices, professional activities, and percep-
tions of their school and its organization. Over 80% (910) of them turned in
surveys, completing 95% of the items. During each participating school’s
study year, teams of three researchers spent a week in the fall and a week in
the spring on site. As part of the data collection process during each visit, the
researchers conducted interviews with 25-30 staff members at each school as
378 Educational Administration Quarterly
well as with school and district administrators. Researchers also observed
governance and professional meetings at each school, and they collected and
analyzed written documentation pertaining to the school’s restructuring
efforts.
The instruction and assessment practices of 144 core-class teachers (3
mathematics and 3 social studies teachers from each school) received special
scrutiny. Trainedto evaluateinstruction according tostandards of intellectual
quality, the researchers rated the instruction in each core class at least four
times, with two researchers observing at least half the classes. The interrater
reliabilityfor the joint observationswas .78. To evaluatethe qualityof assess-
ment,the SRS asked allcore teachers to providetwo written assessment tasks
thatwere representativeof howthey typically assessedlearning. Subject mat-
ter specialists from the center and trained teacher practitioners rated the
assessment tasks on standards of intellectual quality. Teams of two raters
scored the tasks independently, resolving any differences in their judgments
through discussion until they arrived at a consensus score.
The center also collected from the teachers the work that students com-
pleted in response to the assessment tasks, totaling over 5,000 assignments.
Trained researchers and practitioners rated the student work according to
standards for authentic achievement. At least one third of these papers were
evaluatedby teams oftwo raters. The interrater reliabiltieswere.77 for social
studies, .70 for mathematics. (For more information about the instruments
andprocedures for observingteachers,collecting and ratingassessment tasks
and student work, see Newmann, Secada, & Wehlage, 1995, Newmann,
Marks, & Gamoran, 1996, and Newmann & Associates, 1996.)
Dependent Measures
Pedagogical quality.The measureof pedagogical quality isconstructedas
an index that sums teachers’scores on two components of pedagogy: class-
room instruction and assessment tasks. Classroom instruction scores are the
summed ratings for observed instruction on four standards of authenticity:
(a) higher order thinking (students manipulate information and ideas, rather
than merely reproduce them), (b) substantive conversation (students enter
intodiscussionabout subject matter with their teacherand/or with peers, thus
enhancing their understanding of concepts and ideas), (c) depth of knowl-
edge (students focus on disciplinary ideas or concepts to produce an under-
standing of complex relationships), and (d) connections to the world beyond
the classroom (students work on issues and problems to apply their knowl-
edge). The measure of classroom instruction is standardized (M=0,SD = 1).
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 379
Its reliability (i.e., internal consistency) as indicated by Cronbach’s alpha (α)
is .85.
Assessment task. Assessment task scores are the summed ratings on seven
standardsforauthentic assessment: (a) organizationof information (asks stu-
dents to organize, synthesize, interpret, explain, or evaluate complex infor-
mation), (b) consideration of alternatives (asks students to consider alterna-
tive solutions, strategies, perspectives, or points of view), (c) disciplinary
content (asks students to show an understanding of disciplinary ideas, theo-
ries, or perspectives), (d) disciplinary process (asks students to use the meth-
odologicalapproach of the discipline),(e) elaborated written communication
(asks students to express their understanding, explanations, or conclusions
through extended writing), (f) problem connected to the real world (asks stu-
dents to address an issue, problem, or concept external to the school), and (g)
audience beyond school (asks students to communicate with an audience
other than their teacher and class- or schoolmates). The measure of assess-
ment tasks is standardized (M=0,SD = 1) (Cronbach’s α= .79). The peda-
gogicalquality composite measureis standardized (M=0,SD =1,α= .79).
Academic achievement. Academic achievement is a measure of authentic
student performance, specifically, the sum of averaged student scores in
mathematics and social studies on three standards of intellectual quality:
(a) analysis, (b) disciplinary concepts, and (c) elaborated written communi-
cation. Analysis rates students’ work as it reflects higher order thinking
throughsuch processes asorganizing,synthesizing, interpreting,hypothesiz-
ing, and evaluating. Disciplinary concepts rates students’work as it reflects
understandingand the ability to workwith and manipulate disciplinaryideas,
concepts, and theories. Elaborated written communication rates students’
work on its clarity, coherence, quality of articulation, and richness of argu-
ment. The measure of academic achievement is standardized (M=0,SD =1,
α= .72).
Additional information on the conceptual framework of authentic intel-
lectual quality that underlies the pedagogical quality and academic achieve-
ment measures or on their component variables and construction may be
found in Newmann and Associates (1996).
Independent Measures
Leadership. Although governance rather than school administrative lead-
ershipitself was a majorarea of inquiry intheSRS, the researchers spentcon-
380 Educational Administration Quarterly
siderableamounts of time withthe school principals. Duringeach of their site
visits, the researchers conducted a formal interview lasting 60-90 minutes
with the principal or, in the case of three schools (one at each grade level) that
elected to abandon the principalship as conventionally understood, a princi-
pal surrogate, typically a designated teacher or a coordinating team (cf. Ta-
ble1). Additionally,while at eachof the schools,the researchers observedthe
principal’s interactions formally and informally with teachers, staff mem-
bers, and other professionals of the school community. The researchers
viewed the principals in action at such gatherings as curriculum committees,
school improvement committees, administrative councils, and faculty meet-
ings. Interviews with many teachers at each school also attested to the nature
of principals’ leadership. Based on these data, the SRS researchers produced
the case studies and coding reports.
A systematic and thorough process ensured the validity of both these sets
of documents. At the conclusion of the study years, each research team col-
laborated to write a case study summarizing and synthesizing the interview,
observation, and documentation data collected at the school the team visited.
The 24 case studies, typically about 150 single-spaced pages in length, fol-
lowed an identical topic outline. As part of a rigorous peer review, other cen-
ter staff members reviewed and critiqued the drafts of the case studies. Based
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 381
TABLE 1
Demographic and Performance Characteristics by School Leadership Compared
Low Limited Integrated
Leadership Leadership Leadership
(N= 9) (N= 6) (N= 7)
School demographics
Number of elementary 3 2 2
Number of middle 3 2 2
Number of high 3 2 3
Size 656 977 1,008
Percentage free/reduced lunch 51.0* 31.0 24.0
Percentage African American 26.0 18.0 21.0
Percent Hispanic 29.0 11.0 17.0
NAEP achievement –.36 .13 .36
School leadership
Number of schools with principal surrogate 3 0 0
School performance
Pedagogical quality –.67 .00 .86**
Authentic achievement –.83 .21 .85***
NOTE: NAEP = National Assessment of Educational Progress.
*p≤.05. **p≤.01. ***p≤.001.
on the reviews, the research team revised the drafts. To facilitate systematic
retrieval of case study data, the full center research team developed a stan-
dardized list of more than 100 items for coding the case study data. Two
researchers from the team that had visited the school coded the case sepa-
rately. The researchers resolved disagreements through discussion until they
reached consensus. Codes were later converted into variables. Several were
included in the component indices of shared instructional leadership. Taken
together with the survey data, they provide the basis for the construction of
the leadership measures. For the three schools that elected to function with a
principal surrogate, respondents to the surveys and the researchers complet-
ingthe coding reportsapplied to thesurrogate items referringto the principal.
Because leadership was not a primary focus of the SRS study, as we have
indicated, construct validity in the measurement of transformational and
shared instructional leadership was of paramount concern. If the constructs
were to measure what they purported to measure, they needed to meet two
tests: (a) to reflect the conceptual basis for each of the two leadership
approaches and (b) as composite measures, to be internally consistent or reli-
able. In the variable descriptions that follow, we describe how each of the
leadership constructs meets these tests.
Transformational leadership, constructed from teacher survey items and
coding reports, maps onto the transformational leadership domains of ideal-
ized influence, intellectual stimulation, individualized consideration, and
inspirational motivation (Bass & Avolio, 1993) and, largely, onto the three
clusters of leadership functions discussed earlier—mission centered, perfor-
mance centered, and culture centered (Leithwood et al., 1999). Table 2 con-
tains a listing of the SRS coding and survey items used in this study to mea-
sure transformational leadership compared with the dimensions of the
concept specified by Bass and Avolio (1993) and Leithwood and colleagues
(1999).
The SRS measure of transformational leadership includes five items.
Items 1 and 2 are from the coding reports: (a) There is evidence of significant
intellectualleadership from the principal orother school-based administrator
and(b) principal shares powerwith teachers. Item 1 isdummy coded (0 =No,
1 = Yes). Item 2 is measured on a three-point scale of low, medium, and high.
Items 3-5 are Likert-type scale items from the teacher survey: (c) The school
administration’s behavior toward the staff is supportive and encouraging, (d)
the principal is interested in innovation and new ideas, and (e) the principal
influenced restructuring (α= .85). The measure is constructed as a factor and
standardized for the analysis (M= 0, SD = 1).
Shared instructional leadership, constructed from the coding reports, cap-
tures raters’ assessment of principal instructional leadership, teacher
382 Educational Administration Quarterly
instructional leadership, and the extent that principal and teachers interacted
on matters of instruction, curriculum, and assessment. Tapping the elements
of shared instructional leadership discussed earlier (e.g., Blase & Blase,
1999; Ogawa & Bossert, 1995; Prestine & Bowen, 1993), these items reflect
principal focus on instruction, teachers exercising instructional leadership
roles beyond the classroom, and the mutual engagement of principal and
teachers as leaders in the core areas of instruction, curriculum, and
assessment.
The nine component items all come from the coding. Items 1-3 are
dummy coded (0 = No, 1 = yes): (a) There is evidence of significant instruc-
tional leadership in the school, (b) significant instructional leadership comes
from a principal or other school-based administrator, and (c) significant
instructional leadership comes from a teacher or group of teachers. Items 4-9
are ratings on a 3-point scale of low, medium, and high: (d) the actual influ-
ence of teachers over curriculum, (e) the actual influence of teachers over
instruction, (f) the actual influence of teachers over student assessment, (g)
the actual influence of principals over curriculum, (h) the actual influence of
principals over instruction, and (i) the actual influence of principals over stu-
dent assessment. The measure is constructed as an index summing the items
(α= .77) and standardized (M= 0, SD = 1).
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 383
TABLE 2
Transformational Leadership: SRS Indicators and Theoretical Elements Compared
Leithwood, Jantzi, &
SRS Bass & Avolio (1993) Steinbach (1999)
There is evidence of significant
intellectual leadership from
the principal or other school-
based administrators.
Intellectual stimulation Holds high expectations
Provides intellectual
stimulation
Models organizational values
The school administration’s
behavior toward the staff is
supportive and encouraging.
Individualized
consideration Provides individualized
support
Builds collaborative culture
The principal is interested in
innovation and new ideas. Inspirational motivation Strengthens productive school
culture
The principal influenced
restructuring. Inspirational motivation
Intellectual stimulation Develops widely shared vision
The principal shares power
with teachers. Idealized influence Creates structures for
participation in decisions
Builds consensus about school
goals
NOTE: SRS = School Restructuring Study.
School demographics. Grade-level indicator variables for elementary,
middle, and high school—if Yes, coded 1, all others, 0; school size, number
of students enrolled; school socioeconomic status (SES), the proportion of
students receiving federal lunch subsidy; percentage African American, pro-
portion of African American students; percentage Hispanic, proportion of
Hispanic students; average NAEP achievement, aggregated student score on
a baseline test of basic knowledge and skills in mathematics and reading/
writing.
Control variables. When pedagogical quality is the dependent variable,
the control variables include classroom compositional measures: percentage
female, proportion of girls enrolled in the class; percentage African Ameri-
can, proportion of African American students enrolled in class; percentage
Hispanic, proportion of Hispanic students enrolled in class; average SES,
student score on the SES scale (tapping parental education and household
possessions) aggregated to the classroom level; average NAEP achievement,
individual student scores on the baseline test aggregated to the classroom
level.
When student achievement is the dependent variable, the controls account
for student background characteristics: Female, student gender dummy vari-
able, Yes coded 1, No coded 0; African American race, Yes coded 1, No
coded 0; Hispanic ethnicity, Yes coded 1, No coded 0; SES—student SES;
NAEP achievement, student baseline test score.
Analytic Approach
To examine the relationship between shared instructional leadership and
transformational leadership in the schools, we use a scatterplot analysis
(Research Question 1). The scatterplot displays the distribution of schools
according to their comparative ranking on these two leadership dimensions.
The transformational leadership and shared instructional leadership mea-
sures are standardized so that the average score for a school in the study sam-
ple is 0 and the standard deviation is 1. We overlay a quadrant on the
scatterplot, with the axes placed at 0 on each leadership measure. In this way,
we situate schools relative to the other study schools as either low or high on
both shared instructional leadership and transformational leadership or low
on one dimension and high on the other.
Based on this distribution, we construct a categorical variable to parallel
theschools’quadrant positions, forinstance, low onboth forms ofleadership,
high on both forms, low on one form and high on the other. Using one-way
analysis of variance (ANOVA), we compare means for the schools on their
384 Educational Administration Quarterly
demographic, organizational, and performance characteristics according to
the categorical measure of school leadership (Research Question 2).
Because of the multilevel nature of the data (teachers in schools, class-
rooms in schools, and students in classrooms in schools), we use hierarchical
linear modeling (HLM) in both its 2-level and 3-level applications to investi-
gate the effect of school approach to leadership on the two dependent vari-
ables—pedagogical quality and student achievement (Research Question 3).
The first step in the HLM analysis is to formulate an unconditional model,
one with no predictors at the individual or group levels (Bryk & Raudenbush,
1992). The unconditional model yields the data needed to estimate the
intraclass correlation or the proportion of the variance in the particular out-
come that is between groups. In this study, we are interested in the effect of
leadership as a school-level predictor in accounting for that variance, while
controlling at the individual level for teacher or student characteristics that
could affect the outcome independently of the contribution of leadership.
Because we have described the technical details of these analyses elsewhere,
we will not repeat them here (See, for example, Marks & Louis, 1999.)
RESULTS
Our first analysis addresses the relationship between transformational
leadershipand shared instructionalleadershipin the sampleschools. Accord-
ing to their position on the scatterplot, the schools cluster into three of the
four quadrants (Figure 1). Nine schools, three at each grade level, scored low
on both forms of leadership; six schools, two at each grade level, scored high
ontransformationalleadership, low on sharedinstructional leadership; seven
schools—two elementary, two middle, and three high schools—scored high
on both transformational and shared instructional leadership. (Two schools
were dropped from the analyses because of missing data on the leadership
measures.)
The empty quadrant at the upper left of the scatterplot, representing low
transformational and high shared instructional leadership, suggests that
transformational leadership with the behaviors it implies are a necessary,
although insufficient, condition for shared instructional leadership. Put
another way, if a principal demonstrates no capacity for transformational
leadership—for example, articulating an intellectual vision, providing struc-
tures for participatory decision making, building consensus toward a produc-
tive school culture, and promoting collaboration, the principal will be ill-
disposed to share responsibility with teachers in matters of instruction, cur-
riculum, and assessment in a shared instructional leadership model.
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 385
Thenineschools that scored low onboth leadership measures did not ben-
efit from principals’ leadership influence in either a transformational or
instructional sense. For the purposes of this analysis, we refer to the leader-
shipinthese schools as low(that is, below the samplemean on both measures
of leadership). According to the findings of the field researchers as reported
386 Educational Administration Quarterly
-3 -2 -1 0 1 2 3
-3
-2
-1
0
1
2
3
BI
J
QR
C
DE
L
M
S
T
U
F
G
H
N
0P
V
W
X
Elementary Schools Middle Schools High Schools
B Sumpter I Morris Q Fremont
C Ashley J Selway R Wallingford
D Eldorado L Red Lake S Flinders
E Winema M Shining Rock T Huron
F Careen N Copan U South Glen
G Falls River O Okanagan V Cibola
H Lamar P Ottawa W Island
X Marble Canyon
Shared Instructional Leadership
Low High
Low High
Transformational Leadership
Figure 1. Transformational Leadership and Shared Instructional Leadership
in the case studies of the schools, where both transformational and shared
instructional leadership were low, the schools were in one of three situations:
(a) They deliberately operated without a principal, choosing instead an
administrative team or teacher-in-charge approach, (b) they were in transi-
tion with either an interim or a new principal, or (c) they had an established
but ineffective principal. Any instructional leadership present in these nine
schools came from teachers.
In the six schools where principals demonstrated strong transformational
but little instructional leadership, neither did teachers, in any but one school,
provide evidence of instructional leadership. As these principals sought to
furtherreform, they focusedon change in areasother than instruction, suchas
coordinated social services, structural innovations, or development of link-
ages with school reform networks. They engaged teachers in these efforts but
not in the areas of instruction, curriculum, and assessment. In our analysis,
accordingly, we refer to the leadership in these schools as limited.
In the remaining seven schools, both transformational leadership by the
principal and shared instructional leadership involving both principals and
teachers were above average. The principals provided strong instructional
leadership in all these schools while they facilitated leadership by the teach-
ers, whom they regarded as professionals and full partners in furthering high-
quality teaching and learning. In all but one of the schools, teachers, viewing
their responsibilities as extending beyond their classrooms, also functioned
as instructional leaders. We refer to the leadership in these schools as
integrated.
School Characteristics and Leadership
Patterns: Observed Differences
Thelow leadership schools,enrolling an averageof just over650 students,
tended to be smaller than the other schools by 300-350 students (Table 1).
They enrolled the largest proportion of poor students—51%, whereas the
integrated leadership schools enrolled the smallest—24% (p≤.05). About
one fourth the enrollment at the low leadership schools is African American,
compared with 21% at the integrated leadership schools and 18% at the lim-
ited leadership schools. The proportion of Hispanic students—close to 30%
at the low leadership schools—is 17% at the integrated leadership schools
and11%at the schools where leadershipislimited. Student baseline achieve-
ment measured by NAEP performance averaged –0.36 SD at the low leader-
ship schools, 0.13 SD at the limited leadership schools, and 0.36 SD at the
integrated leadership schools. All three schools, noted earlier, as operating
with a principal surrogate are low leadership schools.
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 387
The school performance measures reflect distinct group differences. Low
leadership schools averaged –0.67 SD on pedagogical quality, compared
withthelimited leadership schools scoring at themean and integrated leader-
ship schools scoring well above at 0.86 SD (p≤.01). Authentic achievement
scores in the low leadership schools averaged –0.83 SD; in the limited lead-
ership schools, 0.21 SD; and in the integrated leadership schools, 0.85 SD
(p≤.001).
Thus, the comparison of observed means for the school groups by type of
leadership indicated significant and systematically patterned differences
among them in the students they enroll. Low leadership tended to be found in
smaller schools where students were poor, minority, and lower achieving.
Integrated leadership, in contrast, typified larger schools enrolling the lowest
proportions of poor, minority, and lower achieving students, whereas limited
leadership schools occupied a middle ground in relation to these school and
student characteristics. Recognizing these systematic differences among the
SRS schools, in subsequent analyses we will adjust statistically for their
potentially confounding influence.
Why the pattern of relationships we have described should exist is not
altogether clear. We recognize that the pattern reflects an empirical observa-
tion, not an inevitability (see, for example, Bryk, Lee, & Holland, 1993; Edu-
cation Trust, 2002). Strong school performance, as we have argued, depends
on integrated leadership mobilizing the collective action of individuals to
producehigh-quality teaching and learning.Where leadership is low, bydefi-
nition, schools lack the collaborative effort of principal and teachers around
matters of curriculum, instruction, and assessment. Without such mutual
engagement to challenge and excite students about learning, particularly in
poor urban schools that may have become accustomed to failure, weak stu-
dent performance is likely to be the norm (Haberman, 2002). We have
encountered a similar occurrence in an earlier study that compared the SRS
schools with a sample of elementary, middle, and high schools in a midwest-
ern urban school district struggling to meet state-imposed accountability
standards (Marks & Printy, 2002). In the lowest achieving schools, adminis-
trators were the most likely to centralize authority and control. These adminis-
trators feared that broadening decision making would threaten the control
they needed to maintain to keep their schools from even greater failure.
Integrated Leadership and School
Performance: Multilevel (HLM) Analyses
Based on the observed relationship between transformational and shared
instructional leadership, we modified our research strategy somewhat to
388 Educational Administration Quarterly
focus on integrated leadership—the coexistence at high levels of transforma-
tional and shared instructional leadership. To capture the effect of integrated
leadership, we constructed an indicator variable to represent schools where
shared instructional leadership and transformational leadership coexist as
compared with leadership in all other schools—that is, those where
transformational and instructional leadership were generally low and those
where transformational leadership was high but lacked principal and teacher
collaboration around curriculum, instruction, and assessment.
Psychometric properties of the school performance measures. Based on
theunconditional HLM analyses, wecomputedintraclass correlations for the
two dependent variables (Table 3). Substantial variation exists between
schools on these outcomes. Twenty-five percent of the variance in pedagogi-
cal quality is between schools compared with 40% of the variance in student
achievement. Table 3 also presents the HLM reliabilities for the dependent
variables. The HLM reliability is a measure estimating the ratio of observed
to true variance in the outcomes (Bryk & Raudenbush, 1992). The reliability
forauthentic achievementis .93, whereas thereliability for pedagogical qual-
ity is comparatively low, at .64. (Because HLM reliabilities are sensitive to
sample size, the lower reliability probably reflects the smaller sample of
teachers in this analysis, that is, the subsample of 6 teachers in each school
whose classes were observed and assessment tasks evaluated.)
Pedagogical quality. The analysis focusing on the quality of pedagogy as
anoutcome is a two-levelHLM analysis(Table 4).Because we treat teachers’
pedagogy as a classroom rather than an individual characteristic, the analysis
adjusts for classroom compositional characteristics, based on students’ per-
sonal and academic backgrounds, that have the potential to influence peda-
gogy—the proportion of female and minority students, the average SES of
the students, and their average NAEP achievement score. In so doing, we fol-
low the precedent of Newmann and Associates (1996), noting that we have
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 389
TABLE 3
Psychometric Properties of the Dependent Variables
Pedagogical Quality Authentic Achievement
Intraclass correlation 25.0% 40.0%
Hierarchical linear modeling reliability .64 .93
Cronbach’s α.79 .72
not found systematic variation in the quality of pedagogy that reflects differ-
ences in teachers’ social and professional backgrounds.
In schools with integrated leadership, average pedagogical quality is 0.6
SD higher than in other schools, a difference that very likely reflects the
sharedengagement of both administratorand teachers around mattersof ped-
agogy (p≤.05). The backgrounds of the students in these teachers’ class-
rooms are not influential for school average pedagogy, with the exception of
baseline achievement as measured on the NAEP assessment. In schools
where classroom average prior achievement is higher, pedagogical quality
tends to be higher by 0.4 SD (p≤.001). The model explains 26% of the
between-school difference in pedagogical quality.
Authentic achievement. The student performance analysis entails a three-
level HLM model, although the model contains no predictors at Level 2, the
classroom level (Table 5). The model takes into account student background
characteristics that have the potential to affect their achievement beyond
school effects (Newmann et al. 1996).
Schoolswithintegrated leadership are higher achievingby close to 0.6 SD
(p≤.01). As was the case with pedagogical quality, strong student perfor-
mance probably reflects the concerted work of administrator and teachers
focused on curriculum, instruction, and assessment. Student background
characteristics are somewhat influential. Girls achieve at higher levels than
boysby 0.1 SD (p≤.001), whereasminority students, both African American
and Hispanic, perform at lower levels than their peers (p≤.01). Although
390 Educational Administration Quarterly
TABLE 4
Integrated Leadership and Pedagogical Quality: A Two-Level Hierarchical
Linear Modeling Model Controlling for Classroom Composition
Dependent Variable
Pedagogical Quality
Fixed effects
Intercept –.25
Integrated leadership .59*
% femalea–.01
% Blacka.14
% Hispanica.27
Average SESa.13
Average NAEP achievementa.39***
Between-school variance in pedagogical quality explained: (%) 26.0
NOTE: SES = socioeconomic status; NAEP = National Assessment of Educational Progress.
a. Standardized variable (M= 0, SD = 1).
*p≤.05. ***p≤.001.
student SES is not a significant factor, prior achievement is. High scores on
the NAEP assessment will add close to 0.3 SD to students’achievement. The
model accounts for 57% of the between-school variance in authentic
achievement.
DISCUSSION
The starting point for the study was a recognition both of the importance
ofinstructionalleadership if schools are to improveand of its evolvingnature
in the context of teacher professionalism. Early conceptions of instructional
leadership had focused on the principal’s role in managing school processes
and procedures related to instruction and supervision. As the challenge of
school reform demanded the principal to become an agent of change, the
managerial role of instructional leader lost its centrality. Transformational
leadership emerged as the model needed by principals to lead schools
through reform. Transformational leadership emphasized the ingredients of
change—ideas, innovation, influence, and consideration for the individual in
the process.
Asthe school reform movementevolved,however,principalsfelt pressure
to be accountable for school improvement and the achievement of students.
Dealing with accountability in the context of systemic change, principals
also had to face the implications of the standards movement, curriculum
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 391
TABLE 5
Integrated Leadership and Authentic Student Achievement: A Three-Level
Hierarchical Linear Modeling Analysis Controlling for Student Characteristicsa
Dependent Variable
Authentic Achievement
Fixed effects
Intercept .29
Integrated leadership .56**
Female .11***
Black –.11**
Hispanic –.11**
SESb.03
NAEP Achievementb.26***
Between-school variance in authentic achievement explained: (%) 57.0
NOTE: SES = socioeconomic status; NAEP = National Assessment of Educational Progress.
a. The model contains no predictors at Level 2.
b. Standardized variable (M= 0, SD = 1).
**p≤.01. ***p≤.001.
frameworks, and new forms of assessment. Responding to these demands
with an outmoded conception of instructional leadership was senseless, but
engaging teachers in a collaborative dialogue about these issues and their
implications for teaching and learning was essential. Thus, the conception of
shared instructional leadership that we proposed to evaluate emphasized the
principal’s interactive role with teachers in the central areas of curriculum,
instruction, and assessment.
Conducting the analyses based on data from restructuring schools in this
unique sample had a major advantage and a major disadvantage. Because of
the leadership for change inevitably entailed in school restructuring, on the
positive side, we believed the schools in this study would be promising sites
for finding transformational leadership. Moreover, because a central goal of
school restructuring is improved student achievement, we also expected
instructional leadership to be prevalent. These expectations were largely
borne out by the data. Yet, even in this rather select sample of schools, as the
analysesdemonstrated,considerable variation existedin these two approaches
to leadership. Although the variation made the study possible, it serves as a
reminder that effective school leadership, as distinct from management, is a
relatively rare commodity.
The disadvantage in focusing on this sample, however, is that we are
unable to generalize our findings. We view our findings as suggestive, there-
fore, posing a pattern that, ideally, we want to investigate in a random sample
of U.S. elementary, middle, and high schools. Nonetheless, the data permit-
ted us to examine the relationship between transformational and shared
instructional leadership and to investigate the implications of these leader-
ship models for school performance measured as pedagogical quality and
student achievement. The absence of shared instructional leadership in
schools that lacked transformational leadership is an important finding.
Whereas transformational leadership is its prerequisite, moreover, shared
instructional leadership will not develop unless it is intentionally sought and
fostered. This latter finding supports the observation of Hallinger and
Leithwood (1998) that transformational leadership does not imply instruc-
tional leadership.
Our second set of findings establishes the importance of what we termed
integrated leadership—transformational leadership coupled with shared
instructional leadership. Where integrated leadership was normative, teach-
ers provided evidence of high-quality pedagogy and students performed at
high levels on authentic measures of achievement. Although this study does
not provide details on how principals and teachers shared instructional lead-
ership, a follow-up investigation will present an analysis from the case study
datato showhow shared instructionalleadership workedin the SRSschools.
392 Educational Administration Quarterly
This study suggests that strong transformational leadership by the princi-
pal is essential in supporting the commitment of teachers. Because teachers
themselves can be barriers to the development of teacher leadership (Smylie
& Denny, 1990), transformational principals are needed to invite teachers to
share leadership functions. When teachers perceive principals’instructional
leadership behaviors to be appropriate, they grow in commitment, profes-
sional involvement, and willingness to innovate (Sheppard, 1996). Thus,
instructional leadership can itself be transformational.
Ourfindings suggest that teachershave both thedesire and the expertiseto
lead. We disagree with the view of Leithwood and Jantzi (1999), therefore,
that adding the notion of leadership to teaching does a disservice to both
teachers and leaders. We argue instead that our findings demonstrate the
importanceof cultivatingteacher leadership forenhanced school performance.
In summary, the integrated view of leadership we propose highlights the
synergistic power of leadership shared by individuals throughout the school
organization. We agree with the argument of Donaldson (2001) that past
understandings of school leadership have failed to meet two functional tests:
that leadership promote organizational improvement and be sustainable for
the leaders themselves. The study demonstrates the effectiveness of inte-
grated leadership—both transformational and instructional—in eliciting the
instructional leadership of teachers for improving school performance.
Arguably, principals who share leadership responsibilities with others would
be less subject to burnout than principal “heroes” who attempt the challenges
and complexities of leadership alone. When the principal elicits high levels
of commitment and professionalism from teachers and works interactively
with teachers in a shared instructional leadership capacity, schools have the
benefitof integrated leadership;they are organizationsthat learn and perform
at high levels.
REFERENCES
Barth, R. (1986). On sheep and goats and school reform. Phi Delta Kappan,68(4), 293-296.
Bass,B. M. (1985).Leadership and performance beyond expectations. New York:Free Press.
Bass, B. M., & Avolio, B. J. (1993). Transformational leadership: A response to critiques. In M.
M. Chemers & R. Ayman (Eds.), Leadership theory and research: Perspectives and direc-
tions (pp. 49-80). San Diego: Academic Press.
Berends, M., & King, M. B. (1994). A description of restructuring in nationally nominated
schools: Legacy of the iron cage? Educational Policy,8, 28-50.
Blase, J., & Blase, J. (1999). Principals’ instructional leadership and teacher development:
Teacher perspectives. Educational Administration Quarterly,35, 349-378.
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 393
Blase, J., & Kirby, P. (2000). Bringing out the best in teachers: What effective principals do.
Thousand Oaks, CA: Corwin Press.
Bryk, A. S., Lee, V. E., & Holland, P. B. (1993). Catholic schools and the common good. Cam-
bridge, MA: Harvard University Press.
Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models: Applications and data
analysis methods. Newbury Park, CA: Sage.
Burlingame, M. (1987). Images of leadership in effective schools literature. In W. Greenfield
(Ed.), Instructional leadership: Concepts, issues, and controversies (pp. 3-16). Toronto:
Allyn & Bacon.
Burns, J. M. (1978). Leadership. New York: Harper & Row.
Carnegie Forum on Education and the Economy. (1986). A nation prepared: Teachers for the
21st century. Washington, DC: Carnegie Foundation.
Conley,D. T.,&Goldman, P.(1994). Tenpropositions for facilitative leadership. In J. Murphy&
K. S. Louis (Eds.), Reshaping the principalship: Insights from transformational reform
efforts (pp. 237-262). Thousand Oaks, CA: Corwin Press.
Cuban, L. (1984). Transforming the frog into a prince: Effective schools research, policy and
practice at the district level. Harvard Educational Review,54(2), 129-151.
Darling-Hammond, L. (1988). Policy and professionalism. In A. Lieberman (Ed.), Building a
professional culture in schools (pp. 55-77). New York: Teachers College Press.
Darling-Hammond,L., & Goodwin, A. L.(1993). Progress towardsprofessionalism in teaching.
In G. Cawelti (Ed.), Challenges and achievements of American education: The 1993 ASCD
yearbook (pp. 19-52). Alexandria, VA: Association for Supervision and Curriculum
Development.
Donaldson, G. A. (2001). Cultivating leadership in schools: Connecting people, purpose, and
practice. New York: Teachers College Press.
Donmoyer, R., & Wagstaff, J. G. (1990). Principals can be effective managers and instructional
leaders. NASSP Bulletin,74(525), 20-29.
Dwyer, D. C. (1984). The search for instructional leadership: Routines and subtleties in the prin-
cipal’s role. Educational Leadership,41(5), 32-37.
Education Trust. (2002, May). Dispelling the myth . . . Over time. Washington, DC: Author.
Firestone, W. A. (1996). Leadership: Roles or functions? In K. Leithwood et al. (Eds.), Interna-
tional handbook of educational leadership and administration (pp. 395-418). Netherlands:
Kluwer Academic.
Firestone, W. A., & Louis, K. S. (1999). Schools as cultures. In J. Murphy & K. S. Louis (Eds.),
Handbook of researchon educational administration (2nd ed., pp. 297-322). San Francisco:
Jossey-Bass.
Fullan,M. (1991). The new meaning ofeducational change. New York:TeachersCollegePress.
Glanz, J., & Neville, R. F. (1997). Educational supervision: Perspectives, issues and controver-
sies. Norwood, MA: Christopher-Gordon.
Glickman, C. (1989). Has Sam and Samantha’s time come at last? Educational Leadership,
46(8), 4-9.
Haberman, M. (2002). The pedagogy of poverty versus good teaching. Retrieved from http://
www.enc.org/topics/equity/articles/document.shtm?input=ACQ-111376-1376
Hallinger,P. (1992). The evolving role of American principals: From managerial to instructional
to transformational leaders. Journal of Educational Administration,30(3), 35-48.
Hallinger, P., & Leithwood, K. (1998). Unseen forces: The impact of social culture on school
leadership. Peabody Journal of Education,73(2), 126-151.
Hallinger, P., & Murphy, J. (1985). Assessing the instructional management behavior of princi-
pals. Elementary School Journal,86, 217-247.
394 Educational Administration Quarterly
Heck, R. H., Larsen, T. J., Marcoulides, G. A. (1990). Instructional leadership and school
achievement: Validation of a causal model. Educational Administration Quarterly,26, 94-
125.
Leithwood, K. (1994). Leadership for school restructuring. Educational Administration Quar-
terly,30, 498-518.
Leithwood, K. (1995). Cognitive perspectives on leadership. Journal of School Leadership,5,
115-135.
Leithwood, K., Dart, B., Jantzi, D., & Steinbach, R. (1993). Building commitment to change and
organizational learning (Phase 4 final report). Victoria, BC: British Columbia Ministry of
Education.
Leithwood, K., & Jantzi, D. (1990). Transformational leadership: How principals can help
reform school cultures. School effectiveness and school improvement,1(4), 249-280.
Leithwood, K., & Jantzi, D. (1999). The relative effects of principal and teacher sources of lead-
ership on student engagement with school. Educational Administration Quarterly,35
(Suppl.), 679-706.
Leithwood, K., Jantzi, D., & Fernandez, A. (1994). Transformational leadership and teachers’
commitment to change. In J. Murphy & K. S. Louis (Eds.), Reshaping the principalship (pp.
77-98). Thousand Oaks, CA: Corwin Press.
Leithwood, K., Jantzi, D., & Steinbach, R. (1998). Leadership and other conditions which foster
organizational learning in schools. In K. Leithwood & K. S. Louis (Eds.), Organizational
learning in schools (pp. 67-92). Lisse: Swets & Zeitlinger.
Leithwood, K., Jantzi, D., & Steinbach, R. (1999). Changing leadership for changing times.
Philadelphia: Open University Press.
Leithwood, K., Leonard, L., & Sharratt, L. (1998). Conditions fostering organizational learning
in schools. Educational Administration Quarterly,34(2), 243-276.
Leithwood, K., Tomlinson, D., & Genge, M. (1996). Transformational school leadership. In K.
Leithwood et al. (Eds.), International handbook of educational administration (pp. 785-
840). Netherlands: Kluwer Academic.
Little, J. W. (1988). Assessing the prospects for teacher leadership. In A. Lieberman (Ed.),
Buildinga professional culture in schools(pp. 78-108). NewYork: TeachersCollege Press
Little,J. W. (1993). Teachers’professional development ina climate of educational reform. Edu-
cational Evaluation and Policy Analysis,15(2), 129-151.
Louis, K. S. (1994). Beyond “managed change:” Rethinking how schools improve. School
Effectiveness and School Improvement,5, 2-24.
Maeroff, G. I. (1988). A blueprint for empowering teachers. Phi Delta Kappan,69, 472-477.
Malen, B., Ogawa, R. T., & Kranz, J. (1990). What do we know about school-based manage-
ment? A case study of the literature—A call for research. In W. H. Clune (Ed.), Choice and
control in American education: Vol. 2. The practice of choice, decentralization and school
restructuring (pp. 289-342). London: Falmer.
Marks, H. M., & Louis, K. S. (1997). Does teacher empowerment affect the classroom? The
implications of teacher empowerment for instructional practice and student academic per-
formance. Educational Evaluation and Policy Analysis,19(2), 245-275.
Marks, H. M., & Louis, K. S. (1999). Teacher empowerment and the capacity for organizational
learning. Educational Administration Quarterly,35(5), 707-750.
Marks, H. M., & Printy, S. M. (2002). Organizational learning in a high-stakes accountability
environment: Lessons from an urban school district. Research and Theory in Educational
Administration,1(1), 1-35.
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 395
Moller, G., & Katzenmeyer, M. (1996). The promise of teacher leadership. In R. H. Ackerman,
G. Moller, & M. Katzenmeyer (Eds.), Every teacher as a leader: Realizing the potential of
teacher leadership (pp. 1-18). San Francisco: Jossey-Bass.
Murphy, J. (1988). Methodological, measurement, and conceptual problems in the study of
instructional leadership. Educational Evaluation and Policy Analysis,10(2), 117-139.
Murphy, J. (1990). Principal instructional leadership. In R. S. Lotto & P. W. Thurston (Eds.),
Advances in educational administration: Changing perspectives on the school (Vol. 1, Pt. B,
pp. 163-200). Greenwich, CT: JAI.
Murphy, J., & Hallinger, P. (1987). New directions in the professional development of school
administrators: A synthesis and suggestions for improvement. In J. Murphy & P. Hallinger
(Eds.), Approaches to administrative training in education (pp. 245-282). Albany, NY:
SUNY Press.
Newmann, F. & Associates. (1996). Authentic achievement: Restructuring schools for intellec-
tual quality. San Francisco: Jossey-Bass.
Newmann,F.M., Marks, H. M., &Gamoran, A.(1996). Authenticpedagogy andstudent perfor-
mance. American Journal of Education,104(4), 280-312.
Newmann, F. M., Secada, W. G., & Wehlage, G. G. (1995). A guide to authentic instruction and
assessment: Vision, standards and scoring. Madison: Wisconsin Center for Educational
Research.
Ogawa, R. T., & Bossert, S. T. (1995). Leadership as an organizational quality. Educational
Administration Quarterly,31, 224-243.
Poole, W. (1995). Reconstructing the teacher-administrator relationship to achieve systemic
change. Journal of School Leadership,5, 565-596.
Poplin, M. (1992). The leader’s new role: Looking to the growth of teachers. Educational Lead-
ership,49(5), 10-11.
Pounder, D. G., Ogawa, R. T., & Adams. (1995). Leadership as an organization-wide phenom-
ena:Its impact onschool performance. Educational Administration Quarterly,31, 564-588.
Prestine,N. A. & Bowen, C.(1993). Benchmarks forchange: Assessingessential school restruc-
turing efforts. Educational Evaluation and Policy Analysis,15(3), 298-319.
Reitzug,U. C. (1997). Images ofprincipal instructional leadership:From supervisionto collabo-
rative inquiry. Journal of Curriculum Supervision,12, 356-366.
Rosenblum, S., Louis, K. S., & Rossmiller, R. A. (1994). School leadership and teacher quality
of work life in restructuring schools. In J. Murphy & K. S. Louis (Eds.), Reshaping the
principalship: Insights from transformational reform efforts (pp. 99-122). Thousand Oaks,
CA: Corwin Press.
Rowan, B. (1990). Commitment and control: Alternative strategiesfor the organizational design
of schools. In C. B. Cazden (Ed.), Review of research in education: Vol. 16 (pp. 353-389).
Washington, DC: American Educational Research Association.
Sagor, R., & Barnett, B. (1994). The TQE Principal: A transformed leader. Thousand Oaks, CA:
Corwin Press.
Schlechty, P. C. (1990). Schools for the 21st century: Leadership imperatives for educational
reform. San Francisco: Jossey-Bass.
Sebring,P.B.,& Bryk, A. S. (2000).School leadershipand the bottom line inChicago. Phi Delta
Kappan,81, 440-443.
Senge, P., Cambron-McCabe, N., Lucas, T., Smith, B., Dutton, J., & Kleiner, A. (2000). Schools
that learn: A fifth discipline fieldbook for educators, parents, and everyone who cares about
education. New York: Doubleday/Currency.
396 Educational Administration Quarterly
Senge, P., Kleiner, A., Roberts, C., Ross, R., Roth, G., & Smith, B. (1999). The dance of change:
The challenges to sustaining momentum in learning organizations. New York: Doubleday/
Currency.
Sergiovanni, T. (1991). The principalship: A reflective practice perspective. Boston: Allyn &
Bacon.
Sheppard, B. (1996). Exploring the transformational nature of instructional leadership. Alberta
Journal of Educational Research,52, 325-344.
Silins, H., Mulford, B., Zarins, S., & Bishop, P. (2000). Leadership for organizational learning.
In K. Leithwood (Ed.), Understanding schools as intelligent systems (pp. 267-292). Stam-
ford, CT: JAI.
Smylie, M. A., & Denny, J. W. (1990). Teacher leadership: Tensions and ambiguities in organi-
zational perspectives. Educational Administration Quarterly,26, 235-259.
Sykes, G. (1990). Fostering teacher professionalism in schools. In R. F. Elmore and Associates
(Eds.), Restructuring schools: The next generation of school reform (pp. 59-96). San Fran-
cisco: Jossey-Bass.
Helen M. Marks is an associate professor in the School of Educational Policy and Leadership at
The Ohio State University. Research interests include school organization and improvement,
educational policy issues, and social and political context of education. Her most recent publi-
cation is “Community Service in the Transition: Shifts and Continuities in Participation From
High School to College,” with S. R. Jones (in press with the Journal of Higher Education).
Susan M. Printy is an assistant professor in the Department of Educational Administration at
Michigan State University where she researches teachers’communities of practice, principal
and teacher leadership, and organizational learning.Her most recent publication is “Organiza-
tional Learning in High-stakes Accountability Environments: Lessons from an Urban School
District” co-authored with H. M. Marks (2002, Theory and Research in Educational
Adminstration).
Marks, Printy / SCHOOL LEADERSHIP RELATIONS 397
