Science mapping the knowledge base in educational leadership and management: A longitudinal bibliometric analysis,1960 to 2018

Abstract

This review employed science mapping methods to analyze the evolution of the knowledge base in educational leadership and management from 1960 to 2018. Descriptive trend analysis, citation analysis, co-citation analysis, and visualization of similarities were used to document growth and change in the ‘intellectual structure’ of the educational leadership and management knowledge base as it evolved through the decades. The review analyzed a database comprised of 22,492 articles published in 21 Scopus-indexed journals over six decades. The authors found that contributions to the knowledge base have evolved from primarily Anglo-American male scholars up until 2000 to increasing gender and geographic diversity in the past 20 years. The review identified several ‘schools of thought’ that emerged across four generations of EDLM scholarship. These include: Leadership for Learning, Leading Change, Leading Teachers, and School Effectiveness and School Improvement. The review also documented a broader evolution in the field’s intellectual structure from a focus on ‘administration’ during the 1960s and 1970s to the embrace of ‘leadership for learning’ as the dominant theme during recent generations. This paradigm shift has not only reshaped the focus of research but also the identity of educational leadership and management as a field of study.

Keywords

Review of research educational administration educational leadership science mapping bibliometric review knowledge production

The knowledge base in educational leadership and management (EDLM) has been of interest to scholars for the past 60 years (Eidell and Kitchell, 1968; Oplatka, 2009). Scholars have reviewed the evolution of EDLM scholarship through both qualitative (e.g. Bossert et al., 1982; Oplatka, 2009; Riehl, 2000) and quantitative approaches (e.g. Gumus et al., 2018; Hallinger and Kovačević, 2019; Murphy et al., 2007; Wang, Bowers and Fikis, 2017). Collectively, these reviews have traced the lineage of scholarship in EDLM as a field of applied practice (Hallinger, 2014).

In a recent companion review, the authors used science mapping to examine the accumulated knowledge base in EDLM as a single entity (Hallinger and Kovačević, 2019). While this approach enabled the identification of overall topical and conceptual trends in the EDLM knowledge base, it did not illuminate the progression of knowledge over time. Thus, the purpose of this review is to build upon the prior effort by conducting a quantitative, longitudinal review of the evolution of the EDLM knowledge base over time (see also Oplatka, 2009; Wang et al., 2017). More specifically, the review addressed the following research questions.

RQ1: How has the growth trajectory of EDLM journal publications changed across the decades between 1960 and 2018?

RQ2: What authors and articles have evidenced the greatest citation impact during different generations of EDLM scholarship between 1960 and 2018?

RQ3: How has the intellectual structure of the EDLM knowledge base evolved between 1960 and 2018?

In order to address these research questions, this review analyzed bibliographic data associated with 22,492 articles published in 21 Scopus-indexed EDLM journals over the past six decades. Bibliometric analyses included descriptive trend analysis, citation analysis, co-citation analysis and visualization of similarities. Quantitative data analyses were organized in order to highlight the longitudinal evolution of the EDLM knowledge base since 1960.

The value-added from this paper derives from the ability to document empirically how the field has changed through the decades. As we will elaborate, prior efforts at analyzing the evolution of the EDLM knowledge base were based on smaller samples and less varied methods of bibliometric analysis (e.g. Murphy et al., 2007; Oplatka, 2009; Wang et al. 2017). Thus, we argue that the current review offers a more comprehensive picture of the field’s evolution.

Four generations of EDLM scholarship

The authors began by consulting prior publications that had reviewed the evolution of the knowledge base in EDLM (e.g. Gumus et al., 2018; Murphy et al., 2007; Oplatka, 2009; Wang et al., 2017). These scholars divided the literature into a variety of different time periods (e.g. decades, generations, mixed-duration time periods), engaged different foci (e.g. methods, concepts, topics), and used different review methods (e.g. context analysis, text mining, simple bibliometric analysis). After reviewing these approaches, the authors decided to divide this review into four time periods, or ‘generations’: 1960 to 1979; 1980 to 1999; 2000 to 2009; 2010 to 2018. This approach took into account space constraints that would have made it impractical to include data analyses for each of six decades as opposed to four generations.

Formal scholarship in EDLM emerged during the late 1950s with the advent of the Theory Movement (Bridges, 1982; Erickson, 1979; Griffiths, 1957; Oplatka, 2009; Wang et al., 2017). This ‘1st generation’ of EDLM scholarship was marked by the launch of the first ‘research journals’ specializing in educational administration (see Table 1). First generation EDLM scholars devoted considerable energy towards exploring epistemological issues concerned with the knowledge base in educational administration (e.g. Eidell and Kitchel, 1968; Greenfield, 1979; Griffiths, 1979). Although these pioneering scholars addressed a proliferation of topics, both the scope and quality of empirical research were characterized as ‘limited’ by previous reviewers of 1st generation research (Bridges, 1982; Campbell, 1979; Erickson, 1979).

Table 1.

Characteristics of the 21 SCOPUS-indexed EA journals included in the review rank-ordered by total articles included in the review.

#	Journal name	Scopus inclusion	Total articles	Scopus h-index	Scopus quartile
1	NASSP Bulletin (NASSP)	1960–2018	5941	19	Q3
2	Educational Leadership (EdLd)	1994–2018	2587	47	Q3
3	Economics of Education Review (EER)	1981–2018	1601	65	Q1
4	Education & Urban Society (EUS)	1968–2018	1364	31	Q2
5	Journal of Educational Administration (JEA)	1963–2018	1319	46	Q1
6	International Journal of Ed. Management (IJEM)	2000–2018	1188	40	Q2
7	Educational Administration Quarterly (EAQ)	1965–2018	1163	60	Q1
8	Urban Education (UrbEd)	1965–2018	1158	43	Q1
9	Ed. Management Admin. & Leadership (EMAL)	1972–2018	1071	31	Q1
10	School Leadership & Management (SO/SLAM)	1981–2018^a	788	35	Q2
11	Management in Education (MiE)	1987–2018	637	11	Q2
12	School Effect. and School Improvement (SESI)	1990–2018	612	45	Q1
13	Journal of Ed. Adm. and History (JEAH)	1968–2018^b	521	13	Q2
14	Improving Schools (ImpSch)	1998–2018	414	15	Q2
15	International Journal of Leadership in Ed. (IJLE)	1998–2018	395	29	Q2
16	Journal of Personnel Evaluation in Ed. (JPEE)	1988–2007	299	n/a	n/a
17	Journal of Educational Change (JEC)	2005–2018	296	26	Q1
18	Advances in Educational Administration (AEA)	2003–2017	284	n/a	n/a
19	International Journal of Management in Ed. (IJME)	2012–2018	283	9	Q3
20	Education as Change (EAC)	2008–2018	275	9	Q3
21	Leadership and Policy in Schools (LPS)	2010–2018	164	12	Q1

^aPublished as School Organisation from 1981 to 1996

^bNot included in Scopus from 2001 to 2009

The 1st generation of EDLM scholarship lasted until the late 1970s when the advent of research on ‘effective schools’ (Edmonds, 1979) heralded a significant shift in the direction of the field and the launch of a 2nd generation in EDLM research. Interest in effective schools research quickly morphed into new lines of inquiry into how school leadership impacted teaching and learning (e.g. Bossert et al., 1982; Hallinger and Murphy, 1985; Hallinger et al., 1996; Heck et al., 1990; Leithwood, 1994). This accelerating interest in ‘leadership for learning’ was stimulated by policy emphases on ‘education reform’, ‘school change’, and ‘school improvement’ that emerged in the 1980s and continued throughout the 1990s (Edmonds, 1982; Leithwood and Montgomery, 1982; Louis and Miles, 1990; Miles, 1993). The 2nd generation also witnessed the emergence of new lines of inquiry into equity, gender, leadership preparation, and teacher effectiveness (see also Oplatka, 2009; Wang et al., 2017).

Transition to a 3rd generation began in the late 1990s with publication of companion research reviews authored by Hallinger and Heck (1996, 1998). Whereas earlier research reviews (e.g. Bridges, 1982; Erickson, 1979) had emphasized a lack of meaningful findings concerning the effects of school leadership, these updated reviews found that improved quality of empirical research had yielded substantive progress. These reviews also proposed a more refined understanding of how school leadership ‘makes a difference’ in student learning, and offered explicit conceptual and methodological recommendations for moving this line of inquiry forward in the future (see also Gumus et al., 2018).

Third generation EDLM scholars took up these challenges and continued to produce robust empirical evidence (e.g. Geijsel et al., 2003; Hallinger and Heck, 2010; Marks and Printy, 2003; Spillane et al., 2004; Supovitz et al. 2010; Witziers et al., 2003). In contrast to the 2nd generation’s sole focus on ‘principal leadership’, the 3rd generation scholarship began to adopt conceptualizations of shared (Marks and Printy, 2003), teacher (York-Barr and Duke, 2004), distributed (Gronn, 2002; Spillane et al., 2004), and collaborative leadership (Hallinger and Heck, 2010). This generation also witnessed the consolidation of earlier concerns for ‘equity’ and ‘gender’ into a more comprehensive line of inquiry under the rubric of ‘social justice’ (Oplatka, 2009; Shields, 2004; Theoharis, 2007; Wang et al., 2017).

The 4th generation, which emerged around 2010, has been characterized by dual themes of internationalization and consolidation in EDLM scholarship. This is reflected in the rapid expansion of geographic sources of EDLM scholarship to include Latin America, Asia, and Africa (Hallinger, 2019 ). Concepts such as accountability, instructional leadership, leadership for learning, social justice and distributed leadership which originated in North America and Europe have continued to be the focus of empirical research.

Method

Science mapping reviews analyze bibliographic data associated with published studies in order to illuminate structural and relational features of a knowledge base (Hallinger and Kovačević, 2019; Small, 1999b; Van Eck and Waltman, 2017; White and McCain, 1998). The ability to work with a large number of studies enables science mapping reviews to discern patterns that might not emerge in research reviews that employ critical synthesis or meta-analysis to analyze smaller sets of studies (Hallinger, 2014). It should be noted, however, that science mapping relies on the analysis of bibliographic data (e.g. citations, author affiliation, keywords) associated with documents in order to reveal features of the knowledge base. Unlike traditional research synthesis and meta-analysis, science mapping does not examine and synthesize substantive findings of studies. Consequently, questions concerning the quality and findings of studies within the knowledge base are not addressed in this review. In sum, science mapping aims to document and synthesize features of the EDLM knowledge base, thereby complementing rather than replacing other review methods.

Identification of sources

Criteria used to determine the eligibility of documents for this review included type and source of documents, time period of publication, and topical focus. We focused on peer-reviewed journal articles in the belief that they provide a more consistent level of quality control (Hallinger, 2014). The authors relied on the Scopus index for the identification of relevant articles based on analyses which confirm that Scopus offers broader coverage of EDLM sources than the Web of Science (see Hallinger and Kovačević, 2019; Falagas et al., 2008). The review encompassed the period from 1960 through August 2018.

We used a journal-based, rather than keyword-based, strategy for the identification of review documents. This reduced ambiguity in making inclusion/exclusion decisions since we could reasonably assume that all articles published in EDLM specialization journals would be relevant for this review. Many other EDLM reviews conducted over the past 40 years have adopted this approach (Bridges, 1982; e.g. Campbell, 1979; Hallinger, 2014; Oplatka, 2009; Wang et al., 2017).

We used a three-step process to identify relevant EDLM specialization journals. First we used the ‘source’ search engine in Scopus to identify titles of journals that focused specifically on EDLM topics (e.g. Educational Management Administration & Leadership). Next, we conducted a ‘content-based’ search for EDLM-related journal articles in Scopus. Using the search terms ‘educational leadership and management’ we identified a large number of articles. We scanned the names of journals associated with these articles in order to identify relevant journals that had been missed in step one.

In the final step, we compared the list of journals accumulated through these steps with journals used in prior reviews of EDLM scholarship (e.g. Bridges, 1982; Cherkowski et al., 2012; Gumus et al., 2018; Hallinger and Bryant, 2013; Oplatka, 2009). Our list of journals overlapped substantially with these other reviews. At the same time, we also excluded some Scopus-indexed journals that had been included in other reviews of research either on EDLM or lines of inquiry within EDLM. For example, although the Journal of School Leadership (JSL) had been included in several reviews of research conducted on North American research in EDLM (e.g. Bridges, 1982; Leithwood and Montgomery, 1982; Riehl, 2000), it is not indexed in Scopus and was, therefore, omitted.

A trickier decision concerned how to manage the sometimes ambiguous boundary between specialization journals that focus on EDLM and journals with related or overlapping but broader missions. Again, it should be noted that the authors were not ‘searching within’ the included journals for relevant articles, but treating the entire published corpus of a journal as relevant to EDLM. For example, although the Journal of Personnel Evaluation in Education (JPEE) was included in the review, its rebranded successor, Educational Assessment Evaluation and Accountability (EAEA), was not. This was due to a change in the espoused mission of the journal from personnel evaluation in educational administration to a broader focus that often goes beyond EDLM issues.

A similar issue arose with respect to education policy and higher education journals. For example, the authors considered including education policy journals (e.g. Educational Evaluation and Policy Analysis, Journal of Educational Policy, Educational Policy, Policy Futures in Education) that feature some relevant EDLM articles. However, we opted not to include ‘policy journals’ because they included many more policy articles that are not centrally concerned with school leadership and management. A similar rationale applied to a set of higher education journals (e.g. Journal of Higher Education, Teaching in Higher Education, Higher Education in Europe, Studies in Higher Education). Through this process, we identified 21 Scopus-indexed EDLM journals that contained 22,633 articles published between 1960 and 2018 (see Table 1).

The authors followed PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for the identification of documents (Moher et al., 2009). We used Scopus filters to exclude 141 documents comprised of editorials, comment, letters, and book reviews (see Figure 1). This left a database of 22,492 articles. Scopus’ ‘year’ filters were then used to disaggregate the full database into separate databases corresponding to each of the four generations.

Figure 1.

Data analysis

We exported bibliographic data on the 22,492 articles from Scopus into separate Excel files corresponding to each of the four generations. The stored data included author name, author affiliation, article title, keywords, abstracts, and various citation data. We conducted descriptive analyses of the full database (e.g. size, growth trajectory) using Scopus analytical tools as well as Excel. Then we conducted bibliometric analyses of the four databases.

Citation analysis examines the frequency with which a document (or author) has been cited within a specific document index (e.g. Scopus) Thus, our citation analyses yielded total ‘Scopus citations’, a common measure of scholarly influence, for authors and documents in each generation (Small, 1999b). Over time, scholars have come to accept that a document (or author) which is frequently cited has made significant contributions to the advancement of knowledge. Although citation analysis offers a useful means of identifying prominent authors, documents and journals, the findings are limited by the scope of the index from which the documents were obtained (e.g. Scopus).

Consequently, co-citation has evolved as a complementary approach to analyzing scholarly impact (Zupic and Čater, 2015). Co-citation is defined as the frequency with which two units (e.g. authors) are cited together in other documents (Small, 1999b). Co-citation analysis assumes that when two authors are frequently ‘cited together’ they share an intellectual affinity (Small, 1999b; Zupic and Čater, 2015).

In order to further clarify this concept, we provide an example of co-citation. In Figure 2 empirical studies on school leadership effects authored by Heck et al. (1990) and Hallinger (2003) both appear in the reference lists of three other documents. Thus, they are considered ‘co-cited’ documents and each gain three co-citations.

Figure 2.

Example of document co-citation in educational administration.

It should be emphasized that co-citation analysis checks for co-citations of a document (e.g. Heck et al. 1990) in the ‘reference lists’ of documents included in the reviewer’s dataset. Therefore, unlike citation analysis, co-citation analysis is not limited to documents residing in a particular database (e.g. Scopus). Thus, for example, in Figure 2 the Hallinger (2003) document was neither included in our database, nor even in Scopus. Since co-citation analysis is based on documents cited in the reference lists of the review documents, it captures a much broader literature. This capability to include co-cited documents located outside the review database enables co-citation analysis to provide a broader perspective on scholarly impact than basic citation analysis (White and McCain, 1998; Zupic and Čater, 2015).

In this review, author co-citation analysis conducted in VOSviewer software was also used to create visual representations or ‘network maps’ of the EDLM knowledge base (Van Eck and Waltman, 2017). These maps show relationships among key authors based on their patterns of co-citation. Author co-citation maps were used to identify the ‘schools of thought’ that comprised the ‘intellectual structure’ of the EDLM knowledge base for each of the generations.

It should be emphasized that, unlike keyword analysis and text mining (e.g. Wang et al., 2017), co-citation analysis is not based on the direct inspection of the content of papers. Nonetheless, Strotman and Zhao (2012) argued that the analysis of ‘authors as the unit of bibliometric studies present[s] a good compromise between the granularity of the individual paper and that of an entire journal’ (p. 1821). With this in mind, the authors triangulated patterns derived from document citation analysis, author citation analysis, and author co-citation analysis to determine thematic trends.

Results

Presentation of results is organized in terms of the four generations of EDLM scholarship.

1st Generation of EDLM Scholarship: 1960–1979

For the sake of parsimony, we have included cross-generation citation tables for documents (Table 2) and authors (Table 3). The rank order in these tables is based on total ‘Scopus citations’.

The volume of EDLM scholarship witnessed slow growth during the 1st generation with 3867 articles published in seven of the 21 journals (see Table 1). In total, 98.8% of the 1st generation EDLM corpus was authored in the USA (80%), Canada, United Kingdom (UK), and Australia. These data locate the birth of formal EDLM scholarship in these Anglo-American nations.

Table 2.

Top five most highly cited EA documents by generation, 1960 to 2018 (n=22,492).

Time Period	Rank	Documents	Journal	Type	Topic	Scopus Cites
1st Gen. 1960–1979	1	Sergiovanni (1967)	EAQ	Emp	Teacher satisfaction	60
	2	Miskel et al. (1979)	EAQ	Emp	School effectiveness	50
	3	Griffiths (1979)	EAQ	Com	Theory in ed. admin.	49
	4	Mohrman et al. (1978)	EAQ	Emp	Participation in D-M	44
	5	Bridges (1967)	JEA	Con	Instr’l leadership	42
	5	Holdaway (1978)	EAQ	Emp	Teacher satisfaction	42
2nd Gen.1980–1999	1	Hallinger and Heck (1996)	EAQ	Rev	Leadership effects	492
	2	Hallinger and Heck (1998)	SESI	Rev	Leadership effects	417
	3	Sanders and Wright (1997)	JPEE	Emp	Teacher effects	332
	4	Monk (1994)	EER	Emp	Teacher effects	295
	5	Bossert et al. (1982)	EAQ	Rev	Instr’l leadership	292
3rd Gen.2000–2009	1	Robinson et al. (2008)	EAQ	Rev	Leadership effects	518
	2	Jeynes (2007)	UE	Rev	Parent involvement	396
	3	Hartog (2000)	EER	Rev	Teacher market	386
	4	Mazzarol and Soutar (2002)	IJEM	Emp	Higher ed. student	352
	5	Marks and Printy (2003)	EAQ	Emp	Leadership effects	338
4th Gen.2010–2018	1	Nores and Barnett (2010)	EER	Rev	Early childhood ed.	157
	2	Hanushek (2011)	EER	Rev	Teacher quality	139
	3	Hallinger and Heck (2010)	SLAM	Emp	Leadership effects	138
	4	Supovitz et al. (2010)	EAQ	Emp	Instr’l leadership	138
	5	Leithwood et al. (2010)	EAQ	Emp	Leadership effects	128
All Gens.1960–2018	12345	Robinson et al. (2008) Hallinger and Heck (1996) Hallinger and Heck (1998) Jeynes (2005) Jeynes (2007)	EAQEAQSESIUEUE	RevRevRevRevRev	Leadership effectsLeadership effectsLeadership effectsParent involvementParent involvement	36482830233220381808

Table 3.

Overview of top five cited educational administration authors by generation, 1960-2018.

Time period	Top-cited authors	Nation	Affiliation	Articles^a	Scopus Citations	CPD^b
Gen. 11960–1979	Hoy W	USA	Ohio State U.	11	120	10.9
	Bridges E	USA	UCSB/Stanford U.	8	120	15.0
	Griffiths D	USA	New York U.	5	90	18.0
	Willower D	USA	Penn. State U.	15	89	6.0
	Greenfield T	USA	U. of Toronto	4	64	16
Gen. 21980–1999	Hallinger P	USA	Vanderblit U.	22	1353	61.5
	Heck R	USA	U. of Hawaii	19	1290	67.9
	Leithwood K	Can	OISE	16	1020	63.8
	Louis KS	USA	U. of Minnesota	10	593	59.3
	Hoy W	USA	Ohio State U.	17	500	29.4
Gen. 32000–2009	Leithwood K	Can	OISE/U. of Toronto	19	1662	87.5
	Harris A	UK	IOE-U. of London	32	1298	40.6
	Jeynes W	USA	Cal. State-Long Beach	9	1218	135.3
	Robinson V	NZ	U. of Auckland	6	614	102.3
	Hoy W	USA	Ohio State U.	16	609	38.1
Gen. 42010–2018	Hallinger P	HK/Thai	HKIEd/Mahidol U.	50	784	15.7
	Heck R	USA	U. of Hawaii	10	412	41.2
	Daly A	USA	U. of Cal-San Diego	18	362	20.1
	Leithwood K	Can	U. of Toronto	11	273	24.8
	Walker A	HK	HKIEd	22	268	12.2
All	Leithwood K	Can	OISE/U. of Toronto	46	2955	64.2
Gens.	Hallinger P	USA	Vanderbilt//Mahidol	77	2339	30.4
1960–2018	Heck R	USA	U. of Hawaii	35	1966	56.2
	Harris A	UK	IOE/U.Malaya/U.Bath	52	1632	31.4
	Hoy W	USA	Ohio State U.	50	1289	25.8

^a Minimum of four documents.

^b CPD = citations per document.

Citation analysis identified the most highly cited 1st generation articles (see Table 2). These included articles published by Sergiovanni (1967), Miskel, Fevurly and Stewart (1979) and Holdaway (1978) on administrator behavior and teacher job satisfaction, Bridges (1967) on instructional leadership, Mohrman, Cooke and Mohrman (1978) on participation in decision-making, and Griffiths (1979) on theory in educational administration. These papers provide a representative picture of topical foci among the 25 most-cited 1st generation articles (not tabled). Although ‘leadership’ was not a common topic in this generation, the Bridges (1967) article in JEA on ‘instructional leadership’ presaged a future trend.

Scopus citation data in Table 3 reveals the most highly cited 1st generation scholars: Wayne Hoy, Edwin Bridges, Daniel Griffiths, Donald Willower, Thomas Greenfield. Analysis of the 100 top-cited scholars of this generation (not tabled) found only two female scholars (Pat Schmuck and Marilyn Gittell) and two scholars from outside of the USA (Peter Ribbins of the UK and William Walker of Australia). The relatively small volume of documents (see Table 1) and Scopus citations per scholar observed (see Table 3) were indicative of this period. The pool of EDLM scholars was small, there were few research-oriented journals, and knowledge production was highly labour intensive.

Next we employed author co-citation analysis (ACA) to examine broader scholarly impact on EDLM research (not tabled).¹ The 10 most highly co-cited 1st generation authors were Donald Willower (104 co-citations), James Coleman (95), Andrew Halpin (69), Daniel Griffiths (66), Herbert Simon (63), James March (61), Talcott Parsons (61), Wayne Hoy (57), Neal Gross (57), and Marilyn Gittell (49). Notably, only five of these scholars were located in departments of educational administration (not tabled). The others were management and social science scholars whose publications were frequently cited by educational administration scholars.

Author co-citation analysis (ACA) was also used to identify ‘schools of thought’ that comprised the intellectual structure of 1st generation EDLM scholarship (see Figure 3). On the ACA map, node size reflects co-citation frequency and ‘links’ between nodes indicate the extent to which two scholars have been co-cited. Proximity of scholars on the map reveals their relative degree of intellectual affinity. Scholars within the same ‘cluster’ (denoted by shade of the nodes) are considered to be members of the same ‘school of thought’. As noted earlier, within science mapping, a ‘school of thought’ refers to a common scholarly tradition or line of inquiry within a field of study.

Figure 3.

Author co-citation map of 1st generation EDLM scholarship, 1960–1979 (threshold 20 citations, display 69 authors).

The author co-citation map for the 1st generation reveals five clusters, each comprised of scholars associated with a particular ‘school of thought’ (see Figure 3).

Administrative Behaviour in Schools: Scholars led by Donald Willower, Wayne Hoy and Cecil Miskel began to apply social science methods to understand how administrative behavior was associated with teacher satisfaction and other features of schools.

Theory in Educational Administration: Scholars led by Roald Campbell, Daniel Griffiths, and Thomas Greenfield theorized if and how social science theories could inform the development of educational administration as a field of study.

Organizational Change in Schools: Scholars led by Matt Miles, Richard Schmuck, Richard Runkel, and Richard Carlson undertook studies of change and innovation which heralded the launch of a ‘school of thought’ that has persisted to the present.

Organizational Theory and Behaviour: Scholars led by James March, Charles Bidwell, Edwin Bridges, and Karl Weick applied theories of organizational behaviour to the administration of schools.

Social and Economic Effects of Schooling: Social science scholars led by James Coleman, Christopher Jencks, Henry Levin, Patrick Moynihan, Marilyn Gittell, David Easton and Ray Rist anticipated another long-term theme in this literature with sociological, econometric and anthropological studies of ‘educational effectiveness’.

It is notable that on this map only three of the schools of thought on the 1st generation map consisted primarily of scholars located in ‘departments of educational administration’. The others were comprised of scholars primarily associated with management and social science disciplines. Also note the central positioning on the co-citation map of scholars associated with management, sociology and social psychology (e.g. March, Simon, Bidwell, Stogdill, Katz, Kahn). This reflects the centrality of their scholarship to the emergence of EDLM scholarship and reprises observations of other reviewers that 1st generation scholars were searching for an intellectual purpose for EDLM as a field of study (Bridges, 1982; Campbell, 1979; Oplatka, 2009).

2nd Generation of EDLM Scholarship: 1980–1999

During the 2nd generation, publication volume doubled (i.e. 7794 articles) as eight additional EDLM specialization journals were launched (see Table 1) and more scholars associated with EDLM entered the field. Publication output continued to be led by American scholars (72%), with Anglo-American scholars contributing 94.7% of the 2nd generation corpus. Nonetheless, during the 2nd generation, scholars from additional nations began to publish in EDLM journals with greater frequency (e.g. Israel, 51; Hong Kong, 48; Netherlands, 44; New Zealand, 39).

Document citation patterns in Table 2 reflect the increasing coherence of 2nd generation scholarship. Three of the five most highly cited articles (Bossert et al., 1982; Hallinger and Heck, 1996, 1998) were reviews of research on school leadership effects on teaching and learning. The other two articles focused on ‘teacher effectiveness’. A similar focus on ‘teacher and leader effectiveness’ was clearly discernable in 18 of the 20 top-cited 2nd generation articles (not tabled). The most highly cited authors of the 2nd generation were all associated with interwoven themes of school leadership, school improvement, and change (see Table 3). Indeed, this same pattern applied to 18 of the 20 top-cited articles (not tabled).

Author co-citation analysis (not tabled) revealed additional influential authors of this generation: Joseph Murphy (537 co-citations), Michael Fullan (491), Ken Leithwood (435), Philip Hallinger (420), Matt Miles (365), Tom Sergiovanni (344), Milbrey McLaughlin (341), Brian Rowan (323), Peter Mortimore (311), and Wayne Hoy (292). This list shows some continuity from the 1st generation (e.g. Hoy, Miles, Sergiovanni, McLaughlin) as well as the entry of new scholars into the field.

The author co-citation map in Figure 4 offers a visual validation of change in the intellectual structure of the field during this generation. The clusters on the 2nd generation map are more coherent, thereby suggesting greater distinctiveness of the ‘schools of thought’ (see Figure 4). Notably, and in contrast with the 1st generation map, all three 2nd generation ‘schools of thought’ focused on the improvement of education for students.

Figure 4.

Author co-citation map of 2nd generation EDLM scholarship, 1980–1999 (threshold 20 citations, display 80 authors).

Leadership for Learning: Scholars led by Joseph Murphy, Ken Leithwood, Philip Hallinger, Tom Sergiovanni, Brian Rowan, and Wayne Hoy reoriented EDLM scholarship through their dual focus on leadership – as opposed to administration – and student learning.

Leading Change: Scholars led by Michael Fullan, Matt Miles, Karen S. Louis, Milbrey McLaughlin, Richard Elmore, Linda Darling-Hammond, and Susan Rosenholtz built on the legacy of 1st generation scholarship through empirical studies of change in teacher practices and education reform.

School Effectiveness and School Improvement: Scholars led by Peter Mortimore, Ron Edmonds, Wilbur Brookover, David Reynolds, David Hopkins, Pamela Sammons, Charles Teddlie and Sam Stringfield built on the legacy of 1st generation ‘effectiveness scholarship’. This built momentum for a long-term program of research that began with a focus on school effectiveness and later extended into school improvement.

In addition, we note that all three 2nd generation ‘schools’ were led by scholars located in ‘departments of educational administration’. Indeed, the central region on the 2nd generation author co-citation map was now claimed by scholars specializing in ‘educational leadership’ rather than ’management’ and ‘social sciences’. The field was beginning to find distinctive ‘voices’. Based upon his central position on the map and density of links to scholars in all three clusters, Joseph Murphy emerged as the key ‘boundary spanning scholar’ of this generation. Boundary-spanning scholars play a key role by interpreting and integrating ideas from different ‘schools of thought’ (White and McCain, 1998).

The author co-citation map also highlights two other trends that emerged during this generation. First, the map portrays the field’s transition from its prior intellectual identity as ‘educational administration’ to a new identity as a field of ‘educational leadership and management’. This transition became codified over time as academic departments gradually changed their names to reflect this paradigm shift. A second trend was the gradual emergence of female scholars specialising in EDM research. In the 2nd generation, 11% of the scholars on the co-citation map were female. While this still represented a small percentage of the EDLM professoriate, the measurable increase from the 1st generation foreshadowed a trend that would continue in subsequent decades.

3rd Generation of EDLM Scholarship: 2000–2009

Between 2000 and 2009, 5099 articles were published in 17 EDLM journals.² Publications from scholars in the same four Anglo-American nations (81%) still dominated the knowledge base. Nonetheless, critiques of the EDLM knowledge base published in the late 1990s had set the stage for an increase in ‘emerging regions’ scholarship (see Hallinger, 2019). Indeed, the percentage of emerging regions publications rose from 5.3% in the 2nd generation to 19% in the 3rd generation. This was reflected by a rise in contributions from scholars in the Netherlands (89), Israel (71), Hong Kong (69), South Africa (57), New Zealand (57), Cyprus (33), Spain (31), Taiwan (28), Sweden (27), Belgium (26), Turkey (24), and China (23). This signalled the first wave of a coming sea change in sources of EDLM scholarship in the 4th generation.

Document citation analysis identified a strengthening focus on ‘students’ in 3rd generation EDLM scholarship (see Table 2). More specifically, the top-cited articles included two on leadership and student learning (Leithwood et al., 2008; Robinson et al. 2008), one that examined the effects of parental involvement on student achievement (Jeynes, 2007), one on student choices in international higher education (Mazzarol and Soutar, 2002), and one on education and student earnings after graduation (Hartog, 2000). More broadly, the top 20 highly cited papers evidenced a continuing interest in leadership and learning, and related factors that impact student achievement (not tabled).

Citation analysis of authors found that the top five and 12 of the 20 top-cited scholars were associated with Leadership and Learning (see also Megan Tschannen-Moran, Peter Sleegers, Allan Walker, Chris Day, Daniel Muijs). Author co-citation analysis for the 3rd generation found that the most highly co-cited scholars were noted for their publications on Leadership for Learning and Leading Change. These included Ken Leithwood (962 co-citations), Michael Fullan (879), Wayne Hoy (573), Alma Harris (562), Andrew Hargreaves (518), Philip Hallinger (459), Joseph Murphy (442), Karen S Louis (438), Linda Darling-Hammond (414), and David Reynolds (395). This suggests trends of continuity and growth as additional scholars joined 2nd generation scholars in a long-term programs of research (Leithwood, 2005).

The 3rd generation author co-citation map offers visual representation of these patterns of continuity and differentiation. A new ‘school of thought’, Leading Teachers, emerged even as the three 2nd generation ‘schools’ continued to evolve.

Leadership for Learning: Scholars led by Ken Leithwood, Philip Hallinger, Alma Harris, Ron Heck, and Chris Day continued to explore the effects of school leadership on student learning through empirical research and review.

Shared Leadership for Change: Scholars led by Michael Fullan, Andrew Hargeaves, Joseph Murphy, Karen S. Louis, and Linda Darling-Hammond expanded the scope of research on Leading Change by incorporating a more explicit focus on the contributions of shared leadership.

Leading Teachers: This ‘school’ could be traced back to the 1st generation ‘school’ of Administrator Behaviour led by Wayne Hoy, Megan Tschannen-Moran and Anthony Bryk, scholars in this school of thought focused on examined how school leadership influences the attitudes and practices of teachers.

School Effectiveness and School Improvement (SESI): This ‘school’, led by Peter Mortimore, David Reynolds, Pamela Sammons, David Hopkins, Louise Stoll, Charles Teddlie, Sam Stringfield, Jaap Scheerens, and John MacBeath expanded 2nd generation SESI scholarship into more international contexts and increased the sophistication and scope of empirical investigation into school improvement.

Although scholars associated with EDLM continued to dominate this co-citation map, the contributions of management and social science scholars are also visible as part of the evolving warp and woof of the field’s intellectual structure. Note, for example, management scholars such as Schein, Argyris, Senge and Bass located in the left-central region of the map. This reflects co-citations of their ‘managment research’ in the publications of EDLM scholars.

As indicated in Figure 5, highly productive 2nd generation scholars were joined by a new cohort of scholars whose impact became ‘visible’ during this era (e.g. Spillane, Tschannen-Moran, Harris, Bush, Hargreaves, Walker). The ACA map also reveals an increasing representation of highly productive female EDLM scholars (e.g. Louis, Sammons, Darling-Hammond, Robinson, Harris, Marshall, McLaughlin, Marks, Stoll, Tschannen-Moran, Gunter, Jantzi, Steinbach, Lieberman, Little, Kruse). Based upon the size of their nodes, locations on the map, and density of links across clusters, Ken Leithwood and Michael Fullan were the key boundary-spanning scholars during this generation.

Figure 5.

Author co-citation map of 3rd generation EDLM scholarship, 2000–2009 (threshold 20 citations, display 80 authors).

4th Generation of EDLM Scholarship: 2010–2018

A total of 5732 articles were published in 19 EDLM journals between 2010 and August 2018. This represented a larger volume of articles published in a shorter period of time than during any of the preceding generations. This reflects the increasing publication output of EDLM journals (see Table 1), as well as the expansion of EDLM into a global field of study.

Although publication statistics confirm a continued dominance of ‘Anglo-American scholarship’ during this generation (i.e. 66%), they also highlight growing contributions from scholars located in ‘emerging regions’. Thus, for example, scholars from South Africa (257), Israel (161), Germany (155), Netherlands (127), Hong Kong (112), New Zealand (80), China (79), Malaysia (77), Sweden (75), Spain (74), Norway (69), Belgium (64), Finland (54) and Thailand (53) all contributed more than 50 articles to the 4th generation corpus (not tabled). This offers empirical evidence of the increasing geographical diversification of the EDLM knowledge base (see also Hallinger, 2019).

Highly-cited 4th generation articles indicate a continuing emphasis on issues related to the efficiency and effectiveness of schools. Three papers assessed leadership ‘effects’ on teaching and learning (Hallinger and Heck, 2010; Leithwood et al., 2010; Supovitz et al., 2010). The other two examined the ‘effects’ of early childhood interventions on students (Nores and Barnett, 2010) and the ‘effects’ of teacher quality (Hanushek, 2011). Examination of the 20 most highly cited articles found that 12 focused on Leadership and Learning, and five on School Change and Improvement.

Author citation analysis found a singular focus on Leadership for Learning and Change. Indeed, we were shocked to find that this theme described, without exception, all 50 of the most highly cited scholars. The most influential authors of this generation by co-citation impact were Ken Leithwood (2791 co-citations), Philip Hallinger (1988), Alma Harris (1538), James Spillane (1277), Michael Fullan (1270), Wayne Hoy (1005), Karen S. Louis (991), Ron Heck (971), Eric Hanushek (868), and Joseph Murphy (865). The earlier trend of increasing gender diversity continued in this generation with female scholars comprising 22 of the top 100 co-cited authors.

The author co-citation map revealed four closely related ‘schools of thought’ (see Figure 6).

Figure 6.

Author co-citation map of 4th generation EDLM scholarship, 2010–2018 (threshold 20 citations, display 90 authors).

Leadership for Learning: During this decade, scholars led by Ken Leithwood, Philip Hallinger, Doris Jantzi, Ron Heck, Allan Walker, and Clive Dimmock began to employ integrated models of Leadership for Learning with a concurrent growth in relevant empirical studies from Asia.

Shared Leadership for Change: Scholars led by Alma Harris, Michael Fullan, Andrew Hargeaves, Chris Day, Peter Gronn, Tony Bush and Steven Ball extended prior research with a focus on how shared leadership can bring about sustainable systemic change.

Leading Teachers: Scholars led by Wayne Hoy, James Spillane, Karen S Louis, Joseph Murphy, Megan Tschannen-Moran, Anthony Bryk, Susan Loeb, and Linda Darling-Hammond examined how school leaders shape relationships with teachers that have positive effects for school quality and improvement.

School Effectiveness and School Improvement: Fourth generation research led by Pamela Sammons, David Hopkins, David Reynolds, Bert Creemers, Louise Stoll, Sam Stringfield and Charles Teddlie has further shifted the focus of this ‘school’ towards studies of school improvement in a broader range of international settings.

Evidence of integration within the knowledge base is suggested by the intermingling of scholar nodes and density of links connecting the ‘schools’. For example, scholars in the Leadership for Learning and Shared Leadership for Change are comingled on the map. These patterns reflect the fact that these authors have frequently drawn on concepts associated with other ‘schools of thought’. Leithwood was the key boundary-spanning scholar in this generation.

Discussion

This research review sought to extend earlier reviews of the knowledge base in educational leadership and management (Gumus et al., 2018; Hallinger, 2019; Hallinger and Kovačević, 2019; Murphy et al., 2007; Oplatka, 2009; Wang et al., 2017). Using bibliometric review methods, the authors analyzed 22,492 articles published in 21 Scopus-indexed EDLM journals between 1960 and 2018. A signal feature of this review lies in the use of a longitudinal approach to analyzing trends within and across four generations and 60 years of EDLM scholarship. This concluding section highlights limitations of the review, offers our synthesis of the findings, and discusses implications for future research.

Limitations

Earlier we highlighted methodological limitations of this review. Here we wish to add one additional limitation that concerns the generalizability of our findings to the field of EDLM as a whole. Our review examined a very large database of journal articles drawn from a large proportion of all international, peer-reviewed research-oriented journals specializing in EDLM. Moreover, the use of author and document co-citation analysis enabled the identification of documents beyond those contained in our dataset of Scopus-indexed journal articles. Nonetheless, it is possible that a different method (e.g. topical analysis via text mining) or scope (e.g. inclusion of books, dissertations, handbook chapters, conference papers) of review could yield slightly different results. Thus, the results of this review should be interpreted as tentative benchmarks that can be reinterpreted through reviews that use alternative methods and against new trends as they develop in the future.

Interpretation of the findings

The science mapping method employed in this review offers a unique perspective on the long-term historical development of the knowledge base in EDLM. The EDLM knowledge base encompassed in Scopus-indexed EDLM journals grew steadily from 1682 articles published during the 1960s into a corpus of 22,462 articles by mid-2018. Moreover, this growth trend has gained momentum over time with 25% of the full corpus published between 2010 and 2018. This finding concerning the size of the EDLM knowledge base takes on even greater significance when we consider that this review excluded EDLM-related articles published in general education, policy and higher education journals.

We observed broadly similar growth trends with respect to gender and geographical diversity of EDLM scholarship across the four generations. For example, during the 1st generation only 2% of the 100 top cited scholars were female. In contrast, female scholars represented 21% of the 100 top cited authors in the 4th generation. Similarly analysis of the geographical sources of publications revealed a trend towards increasing diversity beginning in the early 2000s and continuing until today. This bodes well for the development of a truly global knowledge base in EDLM (Hallinger, 2019).

One of the hallmarks of science mapping is its capacity to surface ‘canonical’ scholars and documents that have made significant contributions sustained over a period of several decades (White and McCain, 1998). Our review identified two canonical scholars in the EDLM literature: Ken Leithwood and Wayne Hoy. The influence of their scholarship can be traced clearly across three and four generations of the field's development respectively (see Tables 2 and 3). More broadly, all of the highly cited scholars and documents included in Tables 2 and 3 have had significant impact on the development of the EDLM knowledge base.

A series of author co-citation analyses yielded ‘schools of thought’ associated with each of the four generations of EDLM scholarship. Our analyses reaffirmed Oplatka’s (2009) observation that in its first generation EDLM was a field searching for an intellectual purpose. At the same time, with the benefit of hindsight, we can see that seeds planted by 1st generation scholars grew into the ‘schools of thought’ that comprise the intellectual structure of the EDLM knowledge base in 2018.

For example, take the 1st generation ‘school’, Organizational Change in Schools, led by Matt Miles, Richard Schmuck and Philip Runkel. As more scholars joined this line of inquiry in the 1980s and 1990s (e.g. Fullan, Louis, McLaughlin, Little, Firestone, Stoll, Hargreaves), this grew into a robust ‘school’ comprised of theory and research on Leading Change in Schools. Similarly, 1st generation seeds planted by social scientists studying the effects of education on children and society (e.g. Coleman, Jencks, Levin, Hanushek) evolved into a ‘school of thought’ populated by scholars associated with School Effectiveness and School Improvement (e.g. Edmonds, Mortimore, Brookover, Lezotte, Reynolds, Sammons, Teddlie, Stringfield, Hopkins, Sammons). Finally, 1st generation scholarship on Administrative Behaviour in Schools associated with Don Willower, Wayne Hoy, and Cecil Miskel evolved into a ‘school of thought’ associated with research on Leading Teachers (e.g. Louis, Hoy, Tschannen-Moran, Murphy, Spillane, Goddard, Tarter, Bryk, Goldring, Wahlstrom).

While our analyses portray the emergence of these ‘schools of thought’ as ‘evolutionary’, a hallmark finding of this review lies in the empirical documentation of an undeniable ‘paradigm shift’ which took place during the 2nd generation of EDLM scholarship. Our data revealed a sea change during which the ‘intellectual center’ of the field shifted from the ‘Administration of Schools’ to ‘Leadership for Learning’ in the space of a decade. Unlike other ‘schools of thought’ that emerged over time, there were no clear precursors from which we can trace the ‘school’ of Leadership for Learning. Indeed, the only EDLM scholars identified with ‘leadership’ on the 1st generation co-citation map (see Figure 3) were Tom Sergiovanni, Edwin Bridges, and Neil Gross. But, tellingly, they were not even located in the same cluster. Comparison of the 1st and 2nd generation author co-citation maps affirms that this paradigm shift took place over the course of a single generation. Further inspection of the 3rd and 4th generation maps indicates that a dual focus on ‘leadership’ and ‘learning’ has cohered into the ‘intellectual purpose’ of the field (see Figures 5 and 6).

Results of the citation analyses fill out the picture of the field’s evolution portrayed by the co-citation maps. The highly cited documents included articles that focused on other related themes directly linked to what Murphy (1990) termed the ‘technical core’ of schooling: teaching and learning. These highly cited articles documented the ‘effectiveness’ of leaders (Hallinger and Heck, 1996, 1998; Leithwood et al., 2008; Robinson et al., 2008), teachers (Hanushek, 2011; Monk, 1994; Sanders and Wright, 1997), parent involvement (Jeynes, 2003, 2005, 2007), and early childhood education (Nores and Barnett, 2010) in relation to student learning. These complementary themes illustrate how ‘student learning’ has become the conceptual anchor of the EDLM knowledge base.

These cross-generational trends offer a complementary perspective on findings reported in other reviews of the EDLM knowledge base (see Table 4). In most cases, the ‘schools of thought’ identified in our review for each generation are also reflected in the ‘topics’ identified in prior reviews conducted by Oplatka (2009) and Wang et al. (2017). The main feature that distinguishes their findings is the broader scope of topics identified for each of the generations (see Table 4).

Table 4.

Comparison of findings from reviews mapping the evolution of the EDLM knowledge base.

Review	Design	1st Generation 1960–1979	2nd Generation 1980–1999	3rd Generation 2000–2009	4th Generation 2010–2018
Opaltka (2009)	Journals: JEA, EAQ, EMALDuration: 1960–2007Time periods: 4Focus: Topical analysisMethod: context analysis	epistemologypolitics and policyfinancedecision-makingattitudes and traitsorg./admin. behavior	epistemologyadmin. beliefsvalues and attitudesequityorg. behaviorleader preparationeducation reformpolicy and governanceschool changeeconomicshigher education	Ed. leadershipleader preparationprincipalrole/careerinnovation/changesocial justiceeducation reformaccountabilityteacher career	- - - - - - - - - - - - - - - -
Wang et al. (2017)	Journals: EAQDuration: 1965–2014Time periods: 5Focus: Topical analysisMethod: text mining	epistemologypolicyfinanceorg. behaviorlegal issueshigher education	epistemologyequitycollective bargainingorg. behaviorschool effectivenessinternational contextteacher recruitment	social justiceorganizational behaviorteaching and instructional leadershiplaw	Teaching and instructional leadershipleader preparation and developmentschool effectivenesstrust
This review	Journals: 21Duration: 1960–2018Time periods: 4Focus: MultipleMethod: co-citation analysis	epistemologysocial-economic effects of schoolsorg. theoryadmin. behaviororg. change in schools	leadership for learningleading changeSESI¹	leadership for learningshared leadership for changeSESIleading teachers	leadership for learningshared leadership for changeSESIleading teachers

¹SESI = school effectiveness and improvement.

For example, although ‘social justice’ surfaced as a prominent 3rd generation ‘topic’ in the Oplatka and Wang reviews (see Table 4), it did not cohere into a ‘school of thought’ in ours. Notably, however, the Oplatka and Wang reviews both employed review methodologies that yielded measures of ‘topic frequency’. Our review relied on author co-citation analysis to identify ‘dominant themes’ (i.e. ‘schools of thought’). In order to gain insight into the impact of the different methodologies, we re-analyzed our data to identify the most ‘productive’ 3rd generation authors (i.e. documents per author) rather than the most ‘highly cited’ authors. We found that seven of the 20 most ‘productive’ 3rd generation scholars were associated with ‘social justice’ topics (e.g. Holloway, Schlozman, Hunter, Wrigley, Calabrese, Woods, Oplatka). This illustrates how different methods can yield somewhat different results.

We also considered the possibility that differences in the composition of the review databases could account for differences in our results. Whereas our review examined 21 journals, the Opaltka and Wang et al. reviews examined one and three journals respectively (see Table 4). Thus, for example, Wang et al.’s conclusion concerning the social justice emphasis was identified solely based on articles published in Educational Administration Quarterly (EAQ). This result might not necessarily be representative of other 20 EDLM journals included in this review. This emphasizes the importance of paying close attention to the composition of the database being examined in any review of research.

Implications of the findings

The findings presented in this review suggest several implications. First, we found the intellectual structure of the EDLM knowledge base in 2018 comprised of several distinctive, coherent, and reasonably stable ‘schools of thought’. These intellectual pillars of the EDLM knowledge base have evolved through programmatic research sustained over a period of several decades. This should be acknowledged and celebrated as a victory in the development of EDLM as a field of study.

Second, we noted that the ‘schools of thought’ that comprise the present intellectual structure of EDLM reflect a common theme. This centers on how leaders foster student learning, productive change, and improvement in schools. Forty years ago scholars who argued for this emphasis in ‘educational administration’ could not have imagined the extent to which the field has ‘re-formed’ around a distinctive moral purpose (e.g. Bossert et al., 1982; Bridges, 1982; Erickson, 1979; Murphy, 1990).

This self-organization of the field around a set of distinctive ‘schools of thought’ seems, on the whole, positive. At the same time, we wonder if a mature knowledge base should feature a wider variety of themes and how new ‘schools’ can ‘break out’ from the existing intellectual structure of the field. New knowledge residing outside the existing ‘schools of thought’ can encounter obstacles in ‘finding or reaching an audience’ (Birnbaum, 1981; Small 1999a; Upham et al., 2010). In that sense, the existing ‘schools of thought’ influence how creators of novel ideas position themselves within the field (Upham et al., 2010). These ‘invisible colleges’ (Culnan, 1986) create social, intellectual and normative mechanisms that shape the creation of new knowledge (March and Sutton, 1997; Upham et al., 2010). Nonetheless, this review has documented multiple instances where new perspectives broke free from the existing structure of the knowledge base (e.g., instructional leadership, transformational leadership, distributed leadership).

The trend of conceptual integration that started during the 3rd generation and has continued into 4th generation research suggests an ongoing challenge for young scholars entering the field today. This is to avoid the comfort of choosing a belief system that resides in one particular ‘school of thought’. Individual development and collective progress require us to leverage insights from multiple ‘schools of thought’ (Bell et al., 2002). In that sense, finding a balance between ‘atypical knowledge’ and ‘conventional knowledge’ represents a possible key to transform innovation into impact (Uzzi et al., 2013).

Further insight into this process can be gained by following the scholarship of boundary-spanning scholars such as Murphy in the second generation, Fullan and Leithwood in the third, and Leithwood in the fourth (Uzzi et al., 2013). These scholars integrated concepts from conventional theories with novel ideas from outside of their ‘home school of thought’ (Uzzi et al., 2013). It is interesting to note that, while boundary-spanning scholars tend to be centrally located on a co-citation map, they often ‘reside’ at the periphery of their own ‘school of thought’ (Upham et al., 2010). Thus, generative strategies for young scholars could include collaboration with scholars located in another ‘school of thought’. Similarly, they can draw upon natural differences in perspective that flow from geographical diversity and team up with colleagues located in elsewhere in the world (Uzzi et al., 2013).

Finally, we wish to reflect on the strengths of science mapping as a method of systematic review. It was the longitudinal-quantitative perspective offered by science mapping that enabled us to reveal the contributions of educational administration’s pioneers and portray the dynamic evolution of the EDLM knowledge base from a focus on educational administration 60 years ago to educational leadership today. Notably, this review only employed a limited set of the tools used in science mapping. Thus, we encourage other EDLM scholars to explore this approach to reviewing knowledge accumulation of the field as a whole or for specific lines of inquiry (see Hallinger, 2019; Hallinger and Kovačević, 2019; Wang et al., 2017).

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

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

ORCID iD

Philip Hallinger

Notes

References

Bell

Whitwell

Lukas

(2002) Schools of thought in organizational learning. Journal of the Academy of Marketing Science 30(1): 70–86.

Birnbaum

(1981) Integration and specialization in academic research. Academy of Management Journal 24(3): 487–503.

Bossert

Dwyer

Rowan

Lee

(1982) The instructional management role of the principal. Educational Administration Quarterly 18(3): 34–64.

Bridges

(1967) Instructional leadership: A concept re-examined. Journal of Educational Administration 5(2): 136–147.

Bridges

(1982) Research on the school administrator: The state-of-the-art, 1967–1980. Educational Administration Quarterly 18(3): 12–33.

Campbell

(1979) A critique of the Educational Administration Quarterly. Educational Administration Quarterly 15(3): 1–19.

Cherkowski

Currie

Hilton

(2012) Who should rank our journals…and based on what? Journal of Educational Administration 50(2): 206–230.

Clarke

O’Donoghue

(2017) Educational leadership and context: A rendering of an inseparable relationship. British Journal of Educational Studies 65(2): 167–182.

Culnan

(1986) The intellectual development of management information systems, 1972 – 1982: A co-citation analysis. Management Science 32(2): 156–172.

10.

Edmonds

(1979) Effective schools for the urban poor. Educational Leadership 37(1): 15–24.

11.

Edmonds

(1982) Programs of School Improvement: An Overview. Educational Leadership 40(3): 4–11.

12.

Eidell

Kitchel

(eds) (1968) Knowledge Production and Utilization in Educational Administration. Eugene, OR: ERIC.

13.

Erickson

(1979) Research on educational administration: The state-of-the-art. Educational Researcher 8(3): 9–14.

14.

Falagas

Pitsouni

Malietzis

Pappas

(2008) Comparison of PubMed, Scopus, Web of Science and Googlescholar: Strengths and weaknesses. The FASEB Journal 22(2): 338–342.

15.

Geijsel

Sleegers

Leithwood

Jantzi

. (2003) Transformational leadership effects on teachers’ commitment and effort toward school reform. Journal of Educational Administration 41(3): 228–256.

16.

Greenfield

(1979) Research in educational administration in the United States and Canada: An overview and critique. Educational Administration 8(1): 207–245.

17.

Griffiths

(1979) Intellectual turmoil in educational administration. Educational Administration Quarterly 15(3): 43–65.

18.

Griffiths

(1957) Toward a theory of administrative behavior. In: Campbell

Gregg

(eds) Administrative Behavior in Education. New York: Harper, pp. 43–65.

19.

Gronn

(2002) Distributed leadership as a unit of analysis. The Leadership Quarterly 13(4): 423–451.

20.

Gumus

Bellibas

Esen

Gumus

(2018) A systematic review of studies on leadership models in educational research from 1980 to 2014. Educational Management Administration and Leadership 46(1): 25–48.

21.

Hallinger

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

22.

Hallinger

(2014) Reviewing reviews of research in educational leadership: An empirical assessment. Educational Administration Quarterly 50(4): 539–576.

23.

Hallinger

(2019) Science mapping the knowledge base on educational leadership and management from the emerging regions of Asia, Africa and Latin America, 1965–2018. Educational Management Administration & Leadership 1741143218822772.

24.

Hallinger

Bryant

(2013) Mapping the terrain of educational leadership and management in East Asia. Journal of Educational Administration 51(5): 618–637.

25.

Hallinger

Heck

(2010) Collaborative leadership and school improvement: Understanding the impact on school capacity and student learning. School Leadership and Management 30(2): 95–110.

26.

Hallinger

Heck

(1998) Exploring the principal’s contribution to school effectiveness: 1980-1995. School Effectiveness and School Improvement 9(2): 157–191.

27.

Hallinger

Heck

(1996) Reassessing the principal’s role in school effectiveness: A review of empirical research, 1980–1995. Educational Administration Quarterly 32(1): 5–44.

28.

Hallinger

Kovačević

(2019) A bibliometric review of research on educational administration: Science mapping the literature, 1960 to 2018. Review of Educational Research 89(3): 335–369.

29.

Hallinger

Murphy

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

30.

Hanushek

(2011) The economic value of higher teacher quality. Economics of Education Review 30(3): 466–479.

31.

Hartog

(2000) Over-education and earnings: Where are we, where should we go? Economics of Education Review 19(2): 131–147.

32.

Heck

Larsen

Marcoulides

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

33.

Holdaway

(1978) Facet and overall satisfaction of teachers. Educational Administration Quarterly 14(1): 30–47.

34.

Jeynes

(2005) A meta-analysis of the relation of parental involvement to urban elementary school student academic achievement. Urban Education 40(3): 237–269.

35.

Jeynes

(2003) A meta-analysis: The effects of parental involvement on minority children’s academic achievement. Education and Urban Society 35(2): 202–218.

36.

Jeynes

(2007) The relationship between parental involvement and urban secondary school student academic achievement: A meta-analysis. Urban Education 42(1): 82–110.

37.

Leithwood

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

38.

Leithwood

(2005) Understanding successful principal leadership: Progress on a broken front. Journal of Educational Administration 43(6): 619–629.

39.

Leithwood

Harris

Hopkins

(2008) Seven strong claims about successful school leadership. School Leadership and Management 28(1): 27–42.

40.

Leithwood

Montgomery

(1982) The role of the elementary school principal in program improvement. Review of Educational Research 52(3): 309–339.

41.

Leithwood

Patten

Jantzi

(2010) Testing a conception of how school leadership influences student learning. Educational Administration Quarterly 46(5): 671–706.

42.

Louis

Miles

(1990) Improving the Urban High School: What Works and Why. New York: Teachers College Press.

43.

March

Sutton

(1997) Crossroads-organizational performance as a dependent variable. Organization Science 8(6): 698–706.

44.

Marks

Printy

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

45.

Mazzarol

Soutar

(2002) ‘Push-pull’ factors influencing international student destination choice. International Journal of Educational Management 16(2): 82–90.

46.

Miles

(1993) 40 years of change in schools: some personal reflections. Educational Administration Quarterly 29(2): 213–248.

47.

Miskel

Fevurly

Stewart

(1979) Organizational structures and processes, perceived school effectiveness, loyalty and job satisfaction. Educational Administration Quarterly 15(3): 97–118.

48.

Moher

Liberati

Tetzlaff

Altman

(2009) Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Annals of Internal Medicine 151(4): 264–269.

49.

Mohrman

Jr Cooke

Mohrman

(1978) Participation in decision making: A multidimensional perspective. Educational Administration Quarterly 14(1): 13–29.

50.

Monk

(1994) Subject area preparation of secondary mathematics and science teachers and student achievement. Economics of Education Review 13(2): 125–145.

51.

Murphy

(1990) Restructuring the technical core of preparation programs in educational administration. UCEA Review 31(3): 4–5.

52.

Murphy

Vriesenga

Storey

(2007) Educational Administration Quarterly, 1979–2003: An analysis of types of work, methods of investigation and influences. Educational Administration Quarterly 43(5): 612–628.

53.

Nores

Barnett

(2010) Benefits of early childhood interventions across the world: (Under)investing in the very young. Economics of Education Review 29(2): 271–282.

54.

Oplatka

(2009) The field of educational administration: A historical overview of scholarly attempts to recognize epistemological identities, meanings and boundaries from the 1960s onwards. Journal of Educational Administration 47(1): 8–35.

55.

Riehl

(2000) The principal’s role in creating inclusive schools for diverse students: A review of normative, empirical and critical literature on the practice of educational administration. Review of Educational Research 70(1): 55–81.

56.

Robinson

Lloyd

Rowe

(2008) The impact of leadership on student outcomes: An analysis of the differential effects of leadership types. Educational Administration Quarterly 44(5): 635–674.

57.

Sanders

Wright

Horn

(1997) Teacher and classroom context effects on student achievement: Implications for teacher evaluation. Journal of Personnel Evaluation in Education 11(1): 57–67.

58.

Sergiovanni

(1967) Factors which affect satisfaction and dissatisfaction of teachers. Journal of Educational Administration 5(1): 66–82.

59.

Shields

(2004) Dialogic leadership for social justice: Overcoming pathologies of silence. Educational Administration Quarterly 40(1): 109–132.

60.

Small

(1999a) A passage through science: Crossing disciplinary boundaries. Library Trends 48(1): 72–108.

61.

Small

(1999b) Visualizing science by citation mapping. Journal of the American Society for Information Science 50(9): 799–813.

62.

Spillane

Halverson

Diamond

(2004) Towards a theory of leadership practice: A distributed perspective. Journal of Curriculum Studies 36(1): 3–34.

63.

Strotmann

Zhao

(2012) Author name disambiguation: What difference does it make in author-based citation analysis? Journal of the American Society for Information Science and Technology 63(9): 1820–1833.

64.

Supovitz

Sirinides

May

(2010) How principals and peers influence teaching and learning. Educational Administration Quarterly 46(1): 31–56.

65.

Theoharis

(2007) Social justice educational leaders and resistance: Toward a theory of social justice leadership. Educational Administration Quarterly 43(2): 221–258.

66.

Upham

Rosenkopf

Ungar

(2010) Positioning knowledge: Schools of thought and new knowledge creation. Scientometrics 83(2): 555–581.

67.

Uzzi

Mukherjee

Stringer

Jones

(2013) Atypical combinations and scientific impact. Science 342(6157): 468–472.

68.

Van Eck

Waltman

(2017) Citation-based clustering of publications using CitNetExplorer and VOSviewer. Scientometrics 111(2): 1053–1070.

69.

Wang

Bowers

Fikis

(2017) Automated text data mining analysis of five decades of educational leadership research literature: Probabilistic topic modeling of EAQ articles from 1965 to 2014. Educational Administration Quarterly 53(2): 289–323.

70.

White

McCain

(1998) Visualizing a discipline: An author co-citation analysis of information science, 1972–1995. Journal of the American Society for Information Science 49(4): 327–355.

71.

Witziers

Bosker

Kruger

(2003) Educational leadership and student achievement: The elusive search for an association. Educational Administration Quarterly, 34(3): 398–425.

72.

York-Barr

Duke

(2004) What do we know about teacher leadership? Findings from two decades of scholarship. Review of Educational Research 74(3): 255–316.

73.

Zupic

Čater

(2015) Bibliometric methods in management and organization. Organizational Research Methods 18(3): 429–472.