A Systematic Review of Methodological Approaches in Educational Leadership Research from 2016 to 2019 ()
1. Introduction
A review undertaken by Heck and Hallinger (2005) on the state of research in educational leadership reveals that prior to 1950 the knowledge base in educational leadership was drawn from theoretical claims. Practice within the field focused on stories told by former educational leaders, and their prescriptions for practice were based on their ideological beliefs and personal experiences in life.
A more recent literature review of research on the development of school leadership conducted by Jensen (2016) covered the period from the 1950s to 2016, with the aim of identifying the primary research methodologies, strengths and weaknesses of the existing research. The review was based on the assumption that “what we know is the result of how we come to know it.” To support the analysis, Jensen (2016) applies Gunter’ (2006) conceptualisation of knowledge domains as a framework to analyse the underlying concerns that might have affected the research within three eras, from the 1950s, the 1980s onward, and from 2000 to 2016. According to Gunter (2006) , “knowledge provinces [domains] mean what is being asserted as constituting the truth underpinning the intention behind any leadership activity” (2006, p. 263). In other words, the method by which we research educational leadership is consequential to our knowledge about the educational leadership phenomenon. Table 1 explains the types of knowledge production, containing six knowledge domains developed through a set of different approaches to purpose.
Similarly, in this systematic review, I have used Gunter (2006) knowledge domains as a comprehensive tool for identifying the main methodological approaches used to research educational leadership in the period from 2016 to 2019.
Based on Jensen’s (2016) interpretation it appears that in the 1950s educational leadership research had instrumental and descriptive purposes. Researchers used scientific approaches such as quantitative research methods, formulating hypotheses and testing collected data through surveys to develop effective strategies in order for school leaders and school staff to achieve organisational goals. Moreover, educational leadership programmes were focused on the behaviour science of school leaders, describing their characteristics and testing their actions. Research describes the period from 1947 to 1985 as “the behavioural science era”, and Burns (1997) refers to the model of educational research in this period as the empirical “objective scientific model,” using quantitative research methods and surveys and questionnaires as the main sources for collecting data.
However, in the mid-1970s scholars in the field of social sciences encountered a variety of issues which could not be quantified, and these required more intellectual work to understand. For instance, responding to philosophical questions of morality and life. Therefore, qualitative approaches were introduced in response to these types of issues. The diversity of research designs was especially evident from the 1980s onward, when the role of school leadership changed from being a manager (managing stability) to a leader, to influence others and complete the process of change (Connolly et al., 2019) . Hence, to understand the process of change, scholars and practitioners found qualitative research methods more suitable for dealing with more naturalistic inquiries, as well as what Gunter (2006) called conceptual clarification. Observations and interviews were the most common methods for collecting information, while researchers using their own experiences and others involved in the research to understand the real nature of the research and find out what constitutes a fact. Burns (1997) describes the model of educational research as “qualitative, naturalistic and subjective” in nature; and Hallinger (2003) refers to this period as the “era of globalisation.”
Subsequently, as declared by Jensen (2016) , from 2000 to 2016, all six types of knowledge domains were apparent in educational leadership research, although studies had more evaluative purposes for measuring the effectiveness of school leadership and identifying the conditions to improve. Nevertheless, in terms of data collection methods, surveys and interviews were popular. The field began to adapt a variety of research designs and a wide range of data collection techniques such as case studies, document analysis, focus groups and video recording.
Gradually, significant increase in empirical research and diversity of methods in the educational leadership field made it difficult, especially for novice researchers and scholars, to accumulate knowledge and integrate the findings into single evidence to use as a guide. Accordingly, this review attempts to identify the main methodological approaches used for research leadership in the field of education. Comparable to the literature review carried out by Jensen (2016) , the present review uses Gunter’ (2006) knowledge domains as a comprehensive tool to identify the main methodological approaches in the field. In so doing, I have selected 23 methodologically diverse empirical studies on educational leadership, conducted between 2016 and 2019. Additionally, I have taken a quantitative approach to analyse the selected studies, particularly focusing on three sets of variables: 1) research designs, 2) sample size, and 3) techniques of data analysis employed in the selected studies.
The following questions guide the review.
2. Review Questions
1) What types of research designs were employed?
2) How many participants took part?
3) Which data analysis techniques were employed?
3. Methods
This section of the review presents the methods employed to identify and collect data. This consists of the search strategy (the procedures for data collection), criteria for including/excluding sources and data extraction.
3.1. Search Strategy
To ensure scientific quality, only peer-reviewed journals were considered in this review. The review was sourced from eleven well-recognised journals, including Educational Management Administration & Leadership (EMAL); Journal of Higher Education Policy and Management; Educational Administration Quarterly (EAQ); School Leadership and Management; Journal of Educational Administration (JEA); International Journal of Leadership in Education; Leadership and Policy in Schools; Africa Education Review; National Education Policy Centre; Universal Journal of Educational Research; and Eurasian Journal of Educational Research, which were frequently used in earlier reviews of educational leadership and management researches (Oplatka & Arar, 2017; Hammad & Hallinger, 2017) . Additionally, scholarly databases incorporated in the search included the Educational Resources Information Centre (ERIC) and The British Educational Leadership, Management and Administration Society (BELMAS) to increase the reliability of the selected sources.
This review aimed at selecting a representative sample of high-quality empirical research studies in educational leadership. The review employed search criteria that provided a bounded set of sources. The rationale behind choosing the period from 2016 to 2019 was that the time span was considered both substantial and sufficient for the purpose of the review, in order to include the most recent papers. A Boolean search string was used to select sources; using main keywords inspired by a comprehensive review of the research studies in the field (such as Leithwood & Jantzi’s, 2005 review). I also tried several search strings to provide the most representative dataset such as educational leadership and school improvement and UK; “educational leadership” and “school improvement” and UK; education and leadership and “primary and secondary schools” and “staff development” and UK. Scanning the abstracts and the keywords provided (for example, methods, research methodology, and data analysis methods) with each paper confirmed the findings were empirical studies. This search produced 252 research and review articles. All the articles were carefully screened to eliminate those which did not fit the inclusion criteria in Table 2. For example, articles written in other languages and articles that were not peer-reviewed were identified and excluded from the dataset. Consequently, the number of articles drawn from the above dataset was reduced to 16.
To increase the certainty of identifying all relevant studies, I also went through bibliographic database searches of ERIC from the EBSCOhost platform within the same timeline (2016-2019), using keywords used by (Robinson, 2007) such as “DE Leadership Styles” OR DE “Transformational Leadership”, and found 685 papers. I then reviewed the titles and abstracts of each paper and found only seven papers aligned with the inclusion criteria (Table 2). The combined search methods provided me with a set of 23 studies.
3.2. Eligibility Criteria
Inclusion and exclusion criteria are summarised in Table 2 below.
3.3. Data Extraction
Data extraction included gathering information relevant to my review questions from each of the 23 sample studies. The method I used for this purpose was as follows: initially, data were coded to facilitate quantitative analysis (Gough, 2007) , and extracted into an excel spreadsheet (for example, author, year, type of journal and the studies’ locus). Following my judgements about the nature of the studies, I then entered information such as keywords, methods of data analysis, and sample size into a table (see Table 3 below). Finally, as will be discussed in the following section on data evaluation and analysis, descriptive statistics were employed to examine aspects related to variables such as research designs, study sample sizes and techniques of data analysis.
![]()
Table 2. The inclusion & exclusion criteria.
![]()
Table 3. Characteristics of sample size studies in educational leadership in 11 selected journals.
3.4. Findings
An empirical analysis of academic journals resulted in identification of twenty-three sample studies (see Table 4 below). Of the 23 studies included in the review, 10 (44%) used mixed methods (a combination of qualitative and quantitative data), employing interviews, observations, focus groups and document analysis as their main methods to collecting data; 5 (22%) used case studies; 4 (18%) used qualitative methods; and equally 4 (18%) used quantitative methods, using surveys and questionnaires as their data collection methods.
The bulk of the evidence drawn from the sample studies carried out in 20 countries including the UK. As the table below shows, the majority of the evidence included in this review (22 out of 23 - 96%) was conducted outside of the UK, even though the term “UK” was in the search process, only one UK-based study was found (1 out of 23 - 4%).
3.5. Data Evaluation and Analysis
What types of research designs were employed? (first review question)
To answer my first research question, I read the abstracts, research methods and data collection sections of each sample studies. The results indicated that the majority of studies (44%) adopted mixed methods, 22% adopted case study designs, 18% included qualitative and 18% included quantitative research designs. Furthermore, in terms of data collection methods, as Table 5 presents, the most common techniques employed in both qualitative and quantitative research studies are interviews at 61%, with observations at 35%, focus groups 18%, document analysis 18%, surveys 35% and questionnaires at 31% (see Table 5 below).
![]()
Table 4. Sample studies in educational leadership (2016-2019).
(PIMRS: Principal Instructional Management Rating Scale).
How many participants took part?
To answer the second review question, I thoroughly reviewed all the sample studies to investigate the number of participants engaged in each individual study. Sadly, the sample section of some studies was blurred, for instance study (1) by Mitchell and Sackney (2016) and study (20) by Marfan and Pascula (2018) only referred to the number of selected schools/case studies rather than the actual number of participants. Furthermore, some studies with mixed methods designs such as study (6) by Day et al. (2016) , study (11) by Timor (2017) , study (13) by Yakavets et al. (2017) , and study (15) by Yakavets (2017) were not specific regarding the number of people involved at either the qualitative or quantitative stage of their studies (see Table 6 below).
At the end, I used the sample mean statistic strategy (mean of sample values collected) to determine a representative number of the population participated in the entire 23 sample studies. As a result, 30 is the average number of participants taking part in the qualitative research studies, and 400 is the average number of sample size for quantitative studies.
Which data analysis techniques were employed?
As noted by Briggs and Coleman (2007) , “the types of analysis you are engaged in will always depend very heavily on the nature of the project, the people involved and the focus of the investigation.” (2007: p. 353)
In the same vein, to explore the most common techniques of data analysis in the field, I reviewed all of the 23 sample studies in detail, attempting to associate the nature and the focus of each study with the particular technique(s) employed to analyse both qualitative and quantitative data. Table 7 below presents the number and percentages of the techniques used in these sample studies to analysis and interpret both qualitative and quantitative data.
4. Results
The findings reveal that since 2016 mixed methods (44%) and case studies (22%) have increasingly been used in educational leadership research. With regard to the data collection methods, interviews (61%), observations (35%), focus groups (18%) and document analysis (18%) were the most common methods of collecting qualitative data; while surveys (35%) and questionnaires (31%) were the dominant methods of collecting quantitative data.
In terms of data analysis techniques, content analysis (56%), thematic analysis (45%) and grounded theory (33%) were the top three techniques researchers used to make sense of qualitative data. On the other hand, descriptive statistics analysis (70%) and factor analysis (30%) were the most commonly used techniques to analyse quantitative data. Finally, referring to the population of people participated in the entire 23 sample studies, the average sample size for collecting qualitative information could be 30 participants, and 400 could be a good sample size for quantitative studies.
5. Discussion and Conclusion
The present review examined the methodological features of 23 empirical studies conducted between 2016 and 2019 in the field of educational leadership. The focus of the analysis centred on three sets of variables: 1) research designs, 2) sample size, and 3) techniques of analysis employed in the 23 selected studies. Following Jensen’s (2016) literature review on the development of educational leadership, I used Gunter’ (2006) knowledge domains, firstly to provide a brief background of the research in the field of educational leadership from 1950s to 2016. Secondly, I used the knowledge domains as a comprehensive tool to identify the main methodological approaches used to research educational leadership from the period 2016 to 2019. The results showed that in the earlier decades (from the 1950s onward) social science researchers had more instrumental and descriptive perspectives and were attempting to prescribe effective strategies related to leaders and leadership. Scientific or quantitative methods were the most relevant model of educational leadership research in those periods, and surveys and questionnaires were the common sources for collecting data in the field.
Gradually, with increased issues and changes in the field, scholars and practitioners used qualitative approaches for those issues to understand the processes involved, while using quantitative approaches for measuring the impact, effectiveness of leadership, and the relation between variables. As time passed, more complex research designs (mixed methods and case studies) were developed, and researchers became more flexible in applying the six knowledge domains to different extents according to the purpose of the research they aimed to explore. Surveys and interviews were still popular; however, a wide range of data collection methods began to emerge in the field (case study, document analysis, focus group and video recording).
Consequently, within this review’s time frame (from 2016 to 2019), practically all of the six knowledge domains are still observable to different extents within the field, although most research has been concerned with humanistic and conceptual purposes, interpreting human experiences with the aim of gaining a deep understanding of the nature and the process involved for conceptual clarity. Table 8 below portrays the research trajectory in the field of educational leadership and management from 1950s to 2019.
![]()
Table 8. Educational leadership and management research trajectory (1950-2019).
Overall, the findings indicate that since 2016 there has been an increase in using mixed methods approaches and case studies in researching educational leadership and management. This could be explained in terms of Mackenzie and Knipe’s (2006) claim, where they assert that most educational leadership researchers are showing interest in using qualitative data collection methods and analysis or a combination of both qualitative and quantitative (mixed) methods to observe the cause of events.
Additionally, excessive use of mixed methods can be described in relation to their advantages over using any single method (qualitative or quantitative), which limits the depth and richness of a research study. As specified by Briggs and Coleman (2007) , mixed method approaches allow researchers to be creative when integrating data, and triangulate findings which can enhance the validity and utility of the work (ibid. 2007). However, mixed methods research studies have been criticised by educationists such as Zhang and Creswell (2013) since their narrative frequently lacks descriptions of how the mixing or integration of the qualitative and quantitative elements was achieved. In addition, as demonstrated by O’Cathain (2010) , the use of mixed methods does not necessarily make a study robust or rigorous; the quality of individual qualitative and quantitative components in a mixed method study may affect the quality of mixed methods researches if that component is under-resourced, under-developed or under-analysed in the whole study (ibid. 2010).
Furthermore, case studies were the second most popular methodology used in the sample studies included in this review. This is consistent with Burns’ (2000) statement that using a case study assists a researcher in gaining an in-depth understanding with meaning for the subject, focusing on process rather than products, and on discovery of facts rather than confirmation.
Regarding possible and adequate data collection methods in the field of education and leadership since 2016, the review identified the main methods, such as interviews, observations, focus groups, document analysis, surveys and questionnaires, to collect both qualitative and quantitative data. Consequently, more than half of the qualitative data drawn from interviews explored the in-depth experiences of research participants and the meanings they attribute to these experiences. Lastly, the review highlighted content analysis, thematic analysis and grounded theory as the best fit for interpreting qualitative data. According to Bengtsson (2016) , the purpose of content analysis is to organise, elicit meaning from the data collected, and draw realistic conclusions from it. Similarly, grounded theory analysis enables researchers to focus on the main concern, explaining how the concern is resolved and finally generating a theory. In agreement with Thorne (2000) , data analysis is the most complicated stage of qualitative research that unfortunately receives the least constructive discussion in the educational leadership field, since qualitative researchers often eliminate a detailed description of how analysis was conducted within published research. This issue is one of the current review’s limitations because two sample studies only explained the coding and developing themes instead of mentioning the intellectual process involved in generating results. As advised by Thorne (2000) , researchers should be clearer about what they are doing, why they are doing it, and include a clear description of analysis methods.
Moreover, in respect to the techniques used to analyse quantitative data, the biggest portion of the data were analysed through using descriptive statistical analysis. It seems that this technique helps researchers to describe and simplify large amounts of data in a manageable and sensible way.
The second limitation of the review, as already discussed elsewhere, concerned the sample size recruited in the sample studies. For example, study (1), (6), (11), (15) and (20) did not mention the number of people engaged in their studies. It appears that the issue needs greater attention by researchers in order to enhance the credibility of research. The diverse context of the review not only supported the strength of the conclusion, but it also recognised the lack of educational and leadership research in the UK during the period of 2016 to 2019. Nevertheless, this systematic review represents diverse and rich resources for current and future scholarship. Studying these resources provides a better appreciation of how methodological approaches have been developed in the field of educational leadership and management over time.