The origin of comparative education as an academic field goes back to the 19th century when people attempted to create exceptional models of educational practices which can be implemented elsewhere in the world (Bereday, 1964).
Since then, actors in education, such as scholars, government and educational organizations, have conducted a wide range of studies to learn from each other, trying to improve the existed systems and pursuing educational equity globally (Bray, 2007). As comparative education has emerged as a distinct discipline in the world, some scholars (Bray, 2007. Fairbrother, 2007. Arnove, 2012) start to pay attention to the inner elements of research such as the fundamental features in methodologies.
Methods play a vital role in the accomplishment of research in the field of comparative education. In the book Comparative education research: Approaches and methods by Bray, Adamson and Manson (2007 a), they summarize three phases of comparative education from the direction of methods based on the work of Bereday (1964). The first stage, which covered the whole 19th century, was seen as a “period of borrowing” (Bray et al., 2007 a, P. 2).
In this period, people emphasized the collection and analyses of descriptive data so that people could identify the “best practice” from one system and then copy them in various settings. At the beginning of 20th century, the second phase “period of prediction” (Bray et al., 2007 a, P. 2) appeared as people began to connect education system with conditions of the society. After 1950s, scholars paid greater attention to the social environment which may lead to certain educational phenomena.
This last stage was the “period of analysis” (Bray et al., 2007 a, P. 2), which focused on unfolding theories by conducting research, as well as constructing clear frameworks or methods to examine the possibility of implementing the “best practice” (Bray et al., 2007 a, P. 3). The third phase was not a total jump from the previous two states; instead, it filled the blank of the area between predicting and borrowing by proposing a systematized tool of comparison to examine the validity (Bray et al., 2007.Bereday, 1964).
This paper focuses on various aspects of methodology used and discussed in comparative education literature, which could ideally offer insights into the methods and the implementations in empirical studies. It is divided into five sections. Section one starts with the general background of the field of comparative education.
The paper then provides a brief description of historical debates on approaches within the field of comparative education in section two. Chapter three elaborates on the common way of categorizing methodologies, whereas many concrete examples of methodologies from empirical works are discussed in section four. At the end, section five entails a conclusion of some highlights from the literature review.
Historical Debate on Methods
As the field of comparative education garnered more and more attention in the second half of the twentieth century, some scholars (Bray & Thomas, 1995) argued that the traditional research in this field was limited by the units which were compared because they only concentrated on the differences and similarities among countries and neglected the influence of other comparable parts such as states, schools, and even on an individual level (p. 472).
To provide more comprehensive understanding on choosing the effective model of comparison, Bray and Thomas (1995) proposed a framework to classify multilevel studies. This three-dimensional framework, which is also known as the “cube” model, has three categories of comparative units. By showing this framework, Bray and Thomas encourage people to identify the units of comparison from different directions more specifically.
Although this model does cover most of the foci of comparative education research, not everyone agrees with this category. Bray and his colleges (Bray, et al. 2007b) claim that research with epistemological approaches (this type requires scholars to combine methods as well as purposes and contexts of their studies), sometimes does not encourage people to use this multi-level model. According to them (Bray, et al. 2007b), the “cube” is more likely to be a model with which one can describe the structure of the existing studies than a tool guiding people to conduct certain kinds of comparison (p. 371).
That is, although the cube model allows researchers to investigate educational issues by simply building a comparative framework without understanding the backgrounds of each unit, research methods, like the epistemological approach, emphasize good knowledge on the countries or units which have been under-examined (Bray et al. 2007b). Such an advantage benefits largely on setting hypotheses about the causal factors prior to the actual research (Bray et al. 2007b).
Besides the context of comparison, scholars have also prompted rubrics of “good methods” (Noah,1985). According to Noah (1985), a good method should allow researchers:
- to collect descriptive evidence of an educational system and its achievement or weakness;
- to assist government and institution in developing educational practices;
- to connect educational issues and social background;
- to generalize some frameworks or models which could be valid in many settings.
To accomplish these goals, scholars need to not only depend on concepts in other disciplines (namely, social science, psychology, and history), but also on solid foundation within the field of education (Noah, 1985).
Yet, many problems in research methods have been discovered at the mean time. Noah (1985) also deemed some of them lacking the abilities and funding to collect suitable data, uncertainties on reliability and validity of some data collected previously for other purpose, difficulty of finding the scales working for all type of domestic structure, and bias on ethnography throughout the whole procedure.
However, not all scholars believe the above-mentioned “good method” could fit all studies. Such methods highlight the objectiveness of research, but some studies, which involve researchers as a main means of collecting and analyzing data about participants’ perspective to certain issues, do not have to be totally objective (Fairbrother, 2007). Moreover, these approaches value large amount of data as the main source of evidence, which will not fit in the approaches focusing on process of certain development or policy (Gorur, 2017Fischman et al., 2018).
Common Categories of Methods
Among all the fundamental features of certain educational researches in the field of comparative education, the decisions on choosing the right approach largely lie on their aims and actors (Bray, 2007).
According to Bray (2007), main actors in comparative education include schools who are interested in the comparison of schools and systems in which their children could benefit more; practitioners intending to develop a better operational mechanism; policy makers who want to identify tools for achievement in various levels; international agencies which generalize patterns within and among countries in order to provide feasible suggestions; academics with the goal of seeking for better understanding of determinants and procedures which modify educational systems, as well as the influence of education on society in different settings (p. 16).
Based on these purposes of all actors in education, studies on comparison in diverse directions are usually conducted with the most suitable and economical approaches. This paper mainly focuses on research concerning policymakers, international agencies and academics. These are the three common ways to categorize methods in comparative education. The first one depends on the unit of comparison and it see the methods as either single-level or multi-level methodology (Bray & Thomas, 1995, Resnik, 2012).
The second one is based on the way of dealing with data and evidence, and under this situation, there are quantitative and qualitative analysis (Fairbrother, 2007). The third type of category is built on the second one, but it adds an alternative approach of quantitative and qualitative method to offset the limitations of them (Klees, 2008). As the rationale for fist classification has been discussed in previous chapter (see the discussion about historical debate on methods), the following paragraphs will elaborate the second and third models.
Gregory P. Fairbrother (2007), provides an explicit summary on how quantitative and qualitative methodology differ in these fundamental aspects in his journal Quantitative and Qualitative Approaches to Comparative Education.
According to Fairbrother (2007), quantitative and qualitative approaches vary in purpose, data source and collection process, and analyzing tools. Firstly, unlike quantitative research, which is tightly related to large numerical data and statistical analysis, qualitative study rests on exploring and interpreting essential meanings of certain settings (Fairbrother, 2007, p. 40).
Secondly, quantitative research uses a systematic procedure of data collecting and analyzing, and its absence of researchers during this process contributes to an objective point of view (Fairbrother, 2007). In contrast, qualitative study sees “researchers themselves as instruments of data collections”, taking into consideration of individuals’ (both participants and researchers) perspective (Fairbrother, 2007, p. 42). Thirdly, these two tools do not share a same strategy of doing research. Quantitative one intends to have a clear hypothesis, while the other one more likely to build its depth during the entire research (Fairbrother, 2007).
Another way of classifying the research methods in the field of education is discussed by Klees (2008). He agrees that quantitative approach alone, is insufficient to explore every aspects of a certain educational issues, although it has been a favored method for many actors in education such as government and international organizations (i.e., UN agents, World Bank, and OECD) (Klees, 2008).
Then people started to seek various alternative methods besides quantitative ones. Klees (2008) introduces a new way of grouping the methods and names them with meaningful titles highlighting the purpose of analysis: there are three categories which are “quantitative/positivist methods, qualitative/ interpretive methods, and critical/transformative methods” (Klees, 2008, p. 316).
In his theory, qualitative methods, to some extent, also tries to generate objective and generalized conclusion, which is quite different with Fairbrother’s (2007) interpretation of qualitative approach (Klees, 2008). In contrast, critical/transformative perspective casts doubts on the fundamental debate on the objectiveness of quantitative/positivist research and qualitative/interpretive research; it affirms that none of these two methods could be neutral and comprehensive in all aspects, therefore, studies should tie closely with specific interests (Klees, 2008).
Because of this nature of critical/transformative studies, they usually have the focus on marginalized groups and involving labels like “participatory, action, feminist, indigenous, critical, critical ethnography, and critical race” (Klees, 2008, p. 316). By using this critical approach, researchers could investigate and compare certain issues based on the local culture, even if such studies sometimes lack the ability of generalization in other situations (Klees, 2008). The critical method, in a way, is the best reflection of Noah’s (1985) idea that good method should tie closely with the indigenous/local cultural and social background.