SciELO - Scientific Electronic Library Online

vol.42 número3Apresentação da Seção Temática - Métodos de Educação ComparadaMétodos Quantitativos na Educação Comparada e em Outros Cursos: são válidos? índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados




Links relacionados


Educação & Realidade

versão impressa ISSN 0100-3143versão On-line ISSN 2175-6236

Educ. Real. vol.42 no.3 Porto Alegre jul./set. 2017 


Comparative Studies: historical, epistemological and methodological notes

Juan Ignacio PiovaniI 

Nora KrawczykII 

IUniversidad Nacional De La Plata (UNLP), La Plata/Bs. As - Argentina

IIUniversidade Estadual de Campinas (UNICAMP), Campinas/SP - Brazil


In this article some historical, epistemological and methodological issues related to comparative studies in the social sciences are addressed, with specific reference to the field of education. The starting point is a discussion of the meaning of comparison, its logical structure and its presence in science and in everyday life. It follows the presentation and critical appraisal of the perspectives regarding comparison as a scientific method. It is argued that, even rejecting this restrictive meaning of comparison as a method, there is some consensus on the specificity of comparative studies within the social sciences. And in relation to them, the article address in more detail those studies that can be defined as trans-contextual (cross-national and cross-cultural), with emphasis on the main methodological and technical challenges they face. The socio-historical comparative perspective, which has gained importance in recent years in the field of education, is also discussed.

Keywords: Comparative Studies; Comparative Method; Social Sciences; Methodology


En este artículo se abordan algunas cuestiones históricas, epistemológicas y metodológicas relacionadas con los estudios comparativos en las ciencias sociales, con referencias específicas al campo de la educación. A partir de una consideración general de la comparación, su estructura lógica y su presencia en la ciencia y en la vida cotidiana, se avanza hacia la presentación y valoración crítica de las posturas que la definen como método científico. Por otra parte, se sostiene que, aún cuando se rechace este significado restrictivo de la comparación como método, existe cierto consenso en torno de la especificidad de los estudios comparativos en las ciencias sociales. Y en relación con ellos, en el artículo se discuten con mayor detalle los de carácter trans-contextual (cross-national y cross-cultural), con énfasis en los principales desafíos metodológicos y técnicos que afrontan. Se analiza también la perspectiva comparativa socio histórica, que ha adquirido importancia en los últimos años en el campo de la educación.

Palabras-clave: Estudios Comparativos; Metodo Comparado; Ciencias Sociales; Metodologia


Comparative studies occupy a prominent place in the social sciences, not just because of the value of the descriptions, explanations or interpretations of reality that can be made from them, but also, and especially in the last decades, for being constructed as an input for the diagnosis of social problems and for the design of public policies and, at the same time, as a reference parameter and a source of legitimation.

Although in many thematic areas this centrality is relatively recent, rather remote antecedents of comparative analysis are recorded, especially in the field of political studies. But political science is not only distinguished by the pioneering development of empirical comparative studies; it has also been one of the most fertile academic fields for the epistemological and methodological discussion of the comparison, and in which it has been proposed, perhaps more than any other, the thesis that assigns to the comparison the status of scientific method.

However, this formulation that considers comparison as a method has been the object of severe criticism, even by political scientists. Nevertheless, the challenge of this position has not meant to ignore the importance that the comparison has - and that it has historically - in the production of scientific knowledge. Either does it imply a denial of the entity or specificity of what we usually call comparative studies.

With this problematic framework as a background, in this article we will address some historical questions, epistemological and methodological aspects related to the comparison, the comparative method and comparative studies in the social sciences, making specific references to the field of education.

We begin with some general considerations about the comparison, understood as cognitive activity, since in addition to being a central feature of scientific activity, it is significantly present in our daily lives. In fact, we resort to the comparison on a recurring way - even tacit - to organize the knowledge and relate ourselves to others and to the world.

Otherwise, the history of science shows that the practice of comparison has been and continues to be an indispensable resource to respond to problems of natural and social knowledge. But we must not forget the important differences between comparison as a way of thinking and as a scientific procedure in the social sciences. The first one compares simple operations; the second one compares complex operations (Krawczyk, 2013), although the difference is not in the complexity of the logical structure of comparisons, that as pointed out by Marradi (1991) does not present significant contrasts in science and everyday life, but in the selection and definition of objects and the properties that are compared, as well as in the care and systematicity of the procedures of production and analysis of the data from which the comparisons are made.

The fact that the comparison is so naturalized in our ways of knowing, and that is not exclusive matter of the science, may be one of the reasons that explain the tendency to overlook their analysis, even in many of the specialized texts about comparative research.

Anyway, it is clear that in this work we are interested in focusing more specific senses and applications of comparison in the scientific field, in which, especially from the 18th century, this became a part of the methodological procedures which contributed to the consolidation of different disciplines such as modern sciences. That is why our text will continue with a presentation and critical assessment of the positions that define comparison as a method. Also, trying to show that even when this restrictive meaning is rejected, there is some consensus in the social sciences around the specificity of the comparative studies. And in relation to them, we will dwell in more detail on those of a trans-contextual nature (cross-national and cross-cultural), with emphasis on the main methodological and technical problems faced. Finally, we will also dedicate some paragraphs to the presentation of the comparative socio-historical perspective that, at least in the field of education, has gained importance in recent years and which centrally recovers the historicity of social processes which are the subject of comparative analysis.

Comparison, Comparative Method, Comparative Studies

What do we mean when we speak about an act of comparison? In everyday speech, following a classic dictionary definition19, comparison is presented as the act to observe two or more things to discover their relationships or to estimate their differences and similarities. In epistemological language, it is defined as an intellectual operation through which the states of one (or more) objects are checked on the basis of at least one common property (Fideli, 1998; Marradi, 1991). Therefore, from the logical point of view, an act of comparison implies: objects (which can obviously be subjects, groups, institutions, cities, countries etc.), at least one property of objects, the state of objects in this property20 and the point of time in which they were relieved21.

In science, comparison can be used for various purposes. The traditional epistemological postures, also standard calls (Mulkay, 1979; Outhwaite, 1987), emphasize the role of comparison in the formulation and testing of hypotheses and, in a broader sense, in the production of nomothetic knowledge. In this context, the emphasis of the comparison falls on the properties (variables), which is not surprising, considering the centrality that for these positions acquires the Galilean assumption according to which science must establish relations (mathematics) between the properties of the objects (Marradi; Archenti; Piovani, 2007).

The use of comparison for nomothetic purposes is widespread in the natural sciences, especially in the design, practice and analysis of experiments. But it also has importance in social sciences, mainly in the framework of those orientations that have promoted the development of sophisticated statistical techniques that allow determining the empirical relations between observable properties. In historical perspective, these methodological and technical developments could be thinking, in some way, as an attempt to safeguard the Galilean assumption regarding the cognitive ends of science even in those situations in which the method that best performs it - the experiment - becomes impracticable22 (Piovani, 2006)23.

In the case of social sciences, the impossibility of accepting, or rejecting, the assumptions of traditional conceptions of science has given rise to a series of perspectives in which comparison is used for rather ideographic or interpretive purposes. The emphasis then laid the objects of comparison themselves, and not so much on the properties.

A certain parallelism could be established between these two forms of comparison - centered on properties or objects - with the classic distinction between quantitative and qualitative methods of social research: the comparative approaches that emphasize the properties are generally inscribed in a quantitative orientation and those that put the emphasis on the objects in a qualitative orientation. However, it should also be noted that, in line with the current calls for the integration with the investigative styles and methodological triangulation, Ragin24 proposes a comparative research strategy that establishes a dialogue between variables [properties] and cases [objects].

Beyond the differences that the comparison can acquire in scientific work, from the epistemological point of view (nomothetic or ideographic orientation; input for explanation or interpretation) the methodological (use in the framework of quantitative or qualitative approaches) so far we have referred to the comparison in a relatively general sense. If we think instead in methodical and procedural terms, the preceding considerations are insufficient, especially if it is intended to use the comparison as a criterion to define and delimit a particular methodical strategy that has some autonomy in the set of methods recognized and practiced by the scientific community.

This would imply transcending the comparison as a single act of knowledge and recognize that an investigation comprises a complex and articulated set of decisions and practices that are conducted in a reasonably systematized and organized way. In this sense, proposals that emphasize the comparison with the status of scientific method charge relevance, specifically in the context of the social sciences.

The expression comparative method was diffused with force from the 1960s. One of the clearest formulations of this position is made by the Dutch political scientist Arend Lijphart (1971)25. Its starting point is a conception according to which scientific activity seeks to establish “[…] general empirical relations between two or more variables, controlling, keeping constant, all others” (Lijphart, 1971, p. 70). Following Smelser (1967), Lijphart considers the comparison as a method of empirical control of hypothesis that distinguishes it from three other scientific methods: experimental, statistics and case studies. Their classification involves a hierarchical organization according to the degree of relative effectiveness of each method for the empirical control of the hypotheses (Fideli, 1998). In the proposed scheme, the comparative method occupies the third place, after the statistical method and before the case studies.

Before proceeding with the discussion on the comparative method, to which we will return shortly, it should be noted that the classification proposed by Lijphart presents serious limitations. In the first place, from an epistemological point of view it seems too restrictive to determine to the science a single role, defined in terms of the establishment of general empirical relations between two or more variables. The Galilean assumption about the ends of science, to which reference has already been made, operates in a remarkably explicit way26.

In second place, if what is involved is to establish relationships between variables, then it must be admitted that the comparison plays a fundamental role in any procedure used for this purpose. In order to establish empirically relations between variables, it is necessary to determine the states of the object studied in these variables, and this cannot be achieved without recourse to comparison, regardless of the determination of the status is made within the framework of a procedure that allows identifying the causal direction of the relationship - as the experiment - or that is limited to determine the correlation or association from the statistical analysis of a data matrix.

Statistical analysis, meanwhile, which seems a bit excessive to present per se as one of the methods of science, is based on what Marradi (1977) denominates as atomistic assumption, which among other things involves consideration of the data regardless of the object to which it refers. This enables to construct frequency distributions of the values of a given variable and, from this, the establishment of association of relations between two or more variables. In this operation, the comparison occupies a preponderant place, so it does in the experiment, in which one variable is manipulated and others are monitored in order to evaluate the real effect of that considered independent from the verification of the states at two time points (before and after a manipulation, for example), or the differences between an experimental group and a control group. Whatever the experimental design, resorting to comparison is inevitable. Finally, it seems legitimate to say that a case study cannot be developed without comparisons. In the worst of situations, the study of the changes of the observed case over time will involve some form of comparison. Probably the only important difference is that, most of the time, the study of a case does not imply comparisons tending to the construction of nomothetic knowledge, as in the experiment or in the statistical analyses (at least in its more classic variant). Based on these reviews, many authors argue that it makes no sense to propose a comparative method:

Comparison has a fundamental role in the experimentation of the physical sciences as in the almost-experimental designs used in the human sciences. Statistical analysis is primarily done on the basis of comparisons; is inevitably even compared in the case studies (Fideli, 1998, p. 11-12).

But criticism has implications further broad, if we resort so often to comparison and if as cognitive activity is present in all forms of research: What sense could it have to speak not only about comparative method, but even about comparative research? In fact, the plurality of forms that can be compared and the purposes by which it can be compared has led to the assertion that all scientific activity should be considered as comparative (Swanson, 1971).

However, comparison does not have the same place in all investigations, even if their presence is unavoidable anyway. Indeed, it is possible to identify studies in which the cognitive objective implies an explicit and conscious confrontation, according to Sartori27. In these cases, the research design is traversed in all its dimensions and facets by the central objective of comparing, and this requires a theoretical support that endows meaning and guides the systematic comparison of some units about some properties, as well as methodological decisions and technical instruments that make it operational. We are not talking about a method in the strict sense, but of a type of research.

In this same line Fideli (1998, p. 12) states that “[…] criticize the proposal of Lijphart is not to deny the relevance of the research practices that are usually labeled with the comparative method label”, especially when it refers to some specific styles of social research considered globally, in which complex structures or systems are confronted (linguistic, cultural, institutional, social, political, educational etc.) on the basis of their states in global properties (for example: the degree of structural differentiation, the degree of political stability etc.), or that use data referring to different societies, cultures or nations, or even to sub-national units.

Moreover, this type of research has reached a high degree of specificity and institutionalization in many social sciences. As an example, comparative education, whose history goes back to the beginning of the 20th century in the central countries, mainly in the United States of America, where already in 1899 and 1900 was offered a university course pioneer on this subject. This process of institutionalization of comparative studies in education was intensified at the end of the Second World War, when emerged the policies of international cooperation and, with them, the first world’s education agencies (ONU, international office of Education in Geneva) (Goergen, 1991).

During the second half of the 20th century, in the framework of the process of international reconstruction of democratic capitalism, conceived based on development theory, comparative education experienced an unprecedented moment, focusing on comparisons between the characteristics of education systems in different countries, with the objective of providing elements for the formulation and implementation of educational development policies in peripheral countries28.

This remarkable degree of institutionalization of comparative research has also fueled debates about its status, far beyond the methodological issues that, as previously noted, constitute a relatively classical demarcation criterion, and from which the idea of the comparative method has been developed. In this sense, and as in other social sciences, it is usual to find in comparative research in education positions that define it as a discipline (for example Heath (1958)), while others, as Phillips and Schweisfurth (2014), argue that it would not be a discipline in the strict sense, but an almost-discipline. Chávez Rodríguez (2008), instead, considers it a science. Arnove (1980), Cowen (1996), Rust et al. (1999) and Bray, Adamson and Mason (2014), among many others, consider it a field of study.

It is beyond the scope of this article to further these discussions in relation to the status of comparative studies, whether they constitute sub-disciplines, fields, sub-fields or specialties. But it is evident, and cannot fail to be mentioned, the presence of some elements that literature usually emphasizes when characterizing an area or field of knowledge in the forged terms: a group of researchers who identify themselves as specialists in the area; subjects to grade level, seminars of undergraduate and also graduate programs; exchange networks and academic associations and professionals; national and international congresses, specialized academic journals etc.

If one considers exclusively the Ibero-American educational field, it is possible to point the existence of national associations (for example: Sociedad Argentina de Estudios Comparados en Educación (SAECE), Sociedad Española de Educación Comparada (SEEC), Sociedad Mexicana de Educación Comparada (SOMEC) Sociedade Brasileira de Educação Comparada (SBEC), among others), which in turn are nucleated in the World Council of Comparative Education Societies. Likewise, national and international congresses are regularly organized and magazines such as the Revista Latinoamericana de Educación Comparada or the Revista Española de Educación Comparada. To all this must be added the proliferation of specialized literature, impossible to cite exhaustively, which covers issues such as the same definition of the field of comparative education and the discussion of theoretical and methodological approaches, through the thousands of studies that report or analyze results of empirical research. On the other hand, the presence of comparative education in official institutions is also significant (for example, Ministries of Education), multilateral organizations such as UNESCO, which have been concerned with promoting it, the thinks tanks of very diverse political and ideological orientations.

Many of these investigations, principally those of the World Bank, seek to legitimize the uniformity of educational agendas in Latin America through the homogeneity of the diagnosis of the respective educational systems. These same agendas have even been exported to other regions, such as Africa. They are studies that return to the functionalist perspective of Comparative Education of the 1950s, mainly in the United States of America29 (Rosar; Krawczyk, 2001). Thus, international organizations were important inducers of the education reforms carried out in this period in Latin America, but also in the production of knowledge as a regulatory and governance device (Nóvoa, 1995).

In methodological terms, it should be noted that comparative studies are not limited to a particular strategy. In general they are more frequent, or at least have more diffusion (probably because they are promoted by international organizations) research focusing on the secondary analysis of statistical data. But these statistical comparisons are also made from primary data, in the framework of studies that include the design of survey instruments and samples, as well as the field work and the systematization and analysis of data30. However, this greater visibility of comparisons based on national statistics does not imply that there is no comparative research using different methodologies, even qualitative, as life stories or case studies. In the field of comparative education, for example, in 1979, Stenhouse raised the importance of using case studies to contribute to the understanding of the educational phenomenon (Crossley; Vulliamy, 1984).

About the research that is part of the Latin American critical tradition in social sciences, it can be pointed out that it is distanced from the comparative approach by its strong association with the positivist paradigm and with the imperialist policy of development. In consequence, the academic productions of the critical tradition with comparative analysis were practically nonexistent, at least until the dawn of the 21st century.

Beyond the methodological solutions adopted in each situation, it has been argued that the existence of a dissimilarity of environments could be considered a necessary condition for an investigation to be defined as comparative. In this sense, it becomes relevant what Fideli (1998) denominates cross-contextual comparison31, which encompasses two forms of a comparative research more canonic and disseminated in the social sciences: the studies cross-national and the studies cross-cultural, which will be further discussed.

The Cross-Contextual Comparative Research

Kohn defines the cross-national research such as that in which “[…] comparable data from two or more nations are systematically used” (Kohn, 1987, p. 714)32. Its origins are in the middle of the 17th century, in the context of the decline of the Holy Roman-Germanic Empire after the Peace of Westafalia (1648), which resulted in the emergence of more than 300 relatively small autonomous states. In this historical framework, especially in the Germanic intellectual environments there was interest in comparing the situation of the states, and in particular the disintegration of the Germanic states in contrast to the consolidation of some of the modern European national states (as France, for example).

Only two years after the Peace of Westphalia, in 1850, Hermann Conring would take over the chair of Politics at the University of Helmstedt, in which he gave his famous lessons entitled Notitia Rerum Publicarum y Notitia Statuum Germania (John, 1884), and would begin to write a comparative, unfinished work, in which emphasizes the historical evolution of the states (Jori, 2006). According to Westergaard (1932), Conring was a pioneer in the development of a tradition of comparative analysis of states, which he defined as Staatenkunde.

This tradition would be continued almost a century later in the notorious University of Göttingen, founded in 1737. In this institutional environment, the proposal of Conring acquired the status of autonomous science by the work of Achenwall, who called it Statistik (Cullen, 1975). In Göttingen was developed a school of international fame in which new methodological instruments were designed for the comparative study of the states, among them, the data matrix (even today used in research cross-national). In this matrix, the information about the states (that were located in the horizontal dimension: lines) y the relevant properties for comparison (which were located in the vertical dimension: columns) (Piovani, 2013).

Although it never completely lost its validity, this line of investigation had a significant rebirth from the mid-twentieth century, in the framework of the Comparative Policy, with the publication of very important works as The Civic Culture: political attitudes and democracy in five nations, by Almond and Verba (1963); Social Origins of Dictatorship and Democracy: lord and peasant in the making of the modern world, by Barrington Moore Jr. (1966) and States and Social Revolutions: a comparative analysis of France, Russia, and China, by Skocpol (1979).

But this type of analysis was not limited to the field of political studies. As is known the constitution of national education systems and educational policies have also resorted, at different times, to comparative studies between states. Initially these studies coexisted with the travel literature, which became important as a description of the mechanisms and efforts from different states to organize their national education systems. The project of a science of comparative education took greater strength inasmuch as the belief in the linear relationship between formal education and progress was consolidated, and in education of the population as the best indicator of the development of a country (Goergen, 1991).

Kohn (1987) distinguishes four types of cross-national research:

a. the one in which the nation is the object of study;

b. that which the nation is the context of study;

c. that which the nation is the unit of analysis;

d. that which is transnational.

In the first type, the primary interest is in the country or countries under investigation. In the second one, the focus is on the evaluation of generalizations concerning the way in which some institutions operate or the ways in which specific social structures impact the subjects. In the third type the researcher seeks to establish relations between the characteristics of nations. In this sense, nations are classified from one or more dimensions. In the last type, nations are analyzed as components of the international system.

The first variant - already mentioned in this article in the preceding section - is one of the most widespread, between other reasons, because it is systematically promoted by international agencies. Most part of the time it acquires the form of comparison of statistical information of different countries with respect to the same subject. However, the semantic capacity of these works, understood in terms of the adjustment between the narrative (the report) and the referent of what is narrated, has often been questioned. When considering the problems of equivalence, it will be possible to understand the reason for these criticisms.

The cross-cultural research, for its part, also has a long tradition. Urry (1984), of agreed with Rowe, suggests that the basis on which interest in other cultures began to take shape (exotic cultures), it lies in the rediscovery of European classical antiquity during the Renaissance. In this context, the conquest and colonization of America raised the need to explain other cultures. However, the systematic study of such differences, initially through what is known as ethnology, was not established as a recognized field of study but until mid-nineteenth century, in Europe and North America. This occurred around institutions that promoted the collection of information (through mediators) and publications about other cultures. But toward the end of the nineteenth century it began to become evident that the experts should become data collectors if they wanted to reach a deeper understanding of the cultural other (Urry, 1984; Burgess, 1995). In the beginning of the 20th century Malinowski conducted this idea to its highest point, creating what Leach qualifies as a “[…] theory of ethnographic fieldwork” (Leach apud Urry, 1984, p. 49). However, this kind of research, in its most classic forms, is not interested in the comparison between cultures or cultural systems, but by the detailed description of a specific no-Occidental culture. It can be considered comparative only in the sense acquired by the terms cross-national and cross-cultural in some North American works, which expand the extension of these concepts to include the case of studies of a single country or a single culture, provided that is not their own. The investigation cross-cultural in the strict sense is actually a much more recent phenomenon.

In the case of research cross-national as the cross-cultural, the central problem, from the methodological point of view, is that of equivalence (what is common). Osgood captures in a simple way, and at the same time overwhelming, the core of this problem: when is the same really the same?, when is the same really different?, when is the different really different?33 This question, which crosses vertically all the instances of the research process, affects decisions and practices such as the definition of the unit of analysis and the selection of samples, the use of concepts and terms, the measurement and construction of indicators, and the collection and analysis of data (Piovani, 1998).

A first problematic question refers to the actual identification of the objects to be compared. Fideli (1998, p. 9) asks: “How do you declare that two objects are comparable? The answer is simple: two objects can be compared only when they have at least one property in common. It is difficult to imagine, for example, a property on which to compare John with the United States of America”.

But even before the evaluation of the comparability of two objects, is the question of how they are conceptualized, on the basis of which criteria they are conceived. This problem is evident in the case of research cross-national and cross-cultural: “Zelditch sets the example of Italy, making a list of ten reasons to consider it a unique society. And so many reasons to consider it composed of different societies” (Marradi 1991, p. 73).

The difficulties of conceptualization are not simply limited to the objects of comparison. They are not minor in the case of the identification of relevant properties and their respective states. Smelser (1976) argues that the problem of comparing the same property in different social and cultural systems is presented in two planes: the conceptual defining and the operative.

In the definition and operationalization of properties, the risk of ethnocentrism is always latent: “How to compare the crime rates de of societies that define criminal acts by radically different ways?” (Marradi, 1991, p. 74).

The collection of information is not unrelated to the problems of equivalence. There are two central aspects associated with the collection and measurement 34 of the data: a) What information needs to be collected? , and b) How will it be collected and evaluated? The answer to the first question should be framed in the logical derivation of the way the research problem has been formulated, identifying the central aspects and variables involved in the study and the indicators that will be used to verify the states of the observation of units in the variables. But this is not a simple task. If the indicators change in meaning in different contexts, it follows that the constructs they represent may likewise lose comparability from one context to another (Bynner; Chisholm, 1998).

The second aspect, the collection and analysis of information, also finds serious difficulties in comparative research. According to Marradi (1991, p. 74)

[...] the operational survey of states on the properties presents the most inaccessible difficulties when, for example, to neutralize the difference between verbal expressions in different cultures, the phrases of a questionnaire was replaced by images, it has been found that even the three-dimensional interpretation of perspective and chiaroscuro were not universal in the human mind, but a cultural convention of the Occidental civilization.

In other words, even if an agreement could be reached on the relevant indicators, operational definitions for their survey should be constructed paying close attention to the possible biases that could lead to.

There is no an algorithm that solves the problems of equivalence. But the most recommended strategy currently involves the collaborative work of groups from different countries or areas subject of comparison, experts on themes, dimensions and properties from which it is expected to establish the comparisons. Although it is generally a hard and slow job, with a strong critical component and a craftsmanship, it is assumed that the consensus that can be reached in the definition of objects, the properties, the indicators and their respective operational definitions will give to the comparisons much more solid bases than those made from the universalized imposition of a model whose construction has had a strong local cultural and social bias.

On the other hand, in the current phase of globalization of capitalism, in which the uniformity of national policies is produced through, among others, the supranational constraints, comparative research may acquire a very rich interpretative potential, considering the national historicities for the analysis of the convergences and of the specificities in the concretion of global policies. However, this approach, which will be discussed below in its guidelines and general applications in the field of education, still requires further theoretical and methodological development.

The Socio-Historical Approach in the Comparative Research in Education

The aforementioned perspectives go through the redefinition of the field of comparative education and comparative research in education. But it is also interesting to dwell on the analysis of another approach - the socio-historical or sociological historical35 -, on which was based one of our investigations: the study of educational reform during the last decade of the 20th century in four Latin American countries - Argentina, Brazil, Chile and Mexico - (Krawczyk; Vieira, 2012).

Through this approach, the reformulation of the comparison proposal was sought, from the mere description and analysis of the facts to the analysis of the historical sense of the facts (Nóvoa, 1995). This also implies the recovery of historicity in the concretion of the phenomena and current processes in order to understand its uniqueness. The complexity of reality can no longer be treated from perspectives that seek a single explanation, objective, neutral (Pereyra, 1990; Madeira, 2008; Krawczyk; Vieira, 2012).

The analysis of the results of comparative research, quite clearly, the importance of the historicity of countries in understanding national processes, as well as the scientific production generated in each of them.

The process of external induction imposed uniformity in educational policy in the region, as a result of the growing importance of international agencies and the leadership of the Banco Mundial in the design and implementation of the Reform. Meanwhile, when reflecting on the aspects related to the particular conditions of development of each of the countries it is observed that this reform is processed in societies with different histories that, somehow, contributes in the concretion of the changes, or which give it national specificities. In this sense, in the homogeneity discussed in the regional debate, certain diversity emerges in the new logic of social regulation and in the new role of the State in the countries studied. This does not invalidate that the educational reform has produced institutional ruptures which weakened historically constituted social practices36.

In recent years, although it is still incipient, has increased the concern to incorporate the historical dimension in comparative research in education. However, it is still common the naturalization of concepts and categories of analysis for not understanding and / or not explaining their historicity. This historicity influences the ideas of the researchers and is reflected, although many times it is not explicit, in the choice of topics and in the definition of categories and of the assumptions from which the transformations are analyzed, among other things. The institutional spaces where the investigations are carried out and the political environment is an important dimensions that stand out in the research experiences. The wording of the investigations suggests different approaches in different countries, which are manifested in the character of the knowledge produced and operates in the educational debate. Beyond the relationship between intellectual production and the social base in which they arise and with which they interact, there is also a historically shaped link between national and international trends that configures different impacts of the reform in the different countries.


This article has addressed some relevant aspects related to comparative research in social sciences. Starting from the conceptualization of the comparison and the analysis of its logical structure, it has advanced towards the examination of the different forms that it can acquire in the scientific practice. On the other hand, the position that considers the comparison with the scientific method has been characterized as well as its limits. However, despite accepting the widespread criticism of this position it has been tried to attempt to make to reserve some specificity for comparative studies.

For this, the idea of an alleged method has been renounced, as a demarcation criterion, and has instead to replace the kind of investigation. Although in all scientific research the comparison, as a singular act of knowledge, have an unleashed role, it has been argued that comparative studies constitute a particular type, in which the main cognitive objective involves the systematic and deliberate comparison of complex objects (institutions, cities, countries, educational systems, political systems etc.). In addition, a widespread argument has been taken on comparative research: the need to consider a diversity of contexts or environments in which are placed the aforementioned complex objects.

When analyzing the different approaches in comparative research it is possible to see, even if only synthetically, the theoretical references that support each of them and that provide the coordinates of the comparison. This has allowed cutting, within the broad spectrum of comparative studies, a specific variant: the cross-contextual studies. In relation to them has been characterized its two typical forms: the cross-national and cross-cultural studies. On the other hand, some attention has also been given to comparative socio-historical approaches.

On the other hand, the main technical and methodological challenges are faced by empirical comparative studies. But it is evident that the problems transcend this merely procedural dimension, especially in the case of cross-cultural studies. In fact, in relation to them, literature has also raised complex philosophical dilemmas: is there a space of common meaning between different cultures, without which the cross-cultural comparison would be practically impossible? The case of the cross-national investigation, although not exempt of philosophical and methodological problems, results in this sense a little different. The modern idea of the Nation-State, exported and imposed from Western Europe, has been stabilized and globally consolidated in some way. Beyond the critical judgments that can be made about the circumstances in which these historical processes of stabilization and consolidation occurred, it seems obvious that there are certain properties that make States comparable to each other. However, these comparisons cannot be thought linearly, especially in the current context of globalization in which the sense of national states is in question and new configurations arise between the national, and validity of the comparisons.

If we specifically consider the field of study in which we have focused most of our references - the education -, it is clear that the new phase of globalization also entails new research questions in educational policy. Undoubtedly, transnational, economic, cultural, social and political phenomena, as well as transnational processes of educational regulation (Barroso, 2006), not only those promoted by international organizations, but also by the different blocks (Nafta, European Union, and, perhaps especially educational Mercosur) face researchers in comparative educational policy with a series of specific theoretical and methodological challenges. Many of these processes are silent, and for this reason it is the task of researchers to identify them, not because of their relative invisibility are less influential.

In this same sense, it is also interesting the statement made by the Dutch researcher Saskia Sassen (2007), who affirms that the global not only transcends the framework of the nation-state and global institutions (WTO), but partly inhabits national or sub-national institutions. Therefore, according to her, we have a challenge ahead that is the construction of new conceptual frameworks and analytical categories that allow a more complex study of globalization, detecting also the globalizing dynamics within the institutions and of national social relations, in which national and non-national elements are intersected. In other words, these processes do not necessarily belong to the global scale, but nevertheless they are part of globalization.

The profound transformations of the contemporary world pose new issues and new problems to the definition of educational policy and, although they are not recognized by the current policy, the challenge is to be sensitive to them in order to go beyond the questions and issues surrounding the political debate stage and produce a knowledge that allows to elaborate proposals to the public school recovers a better place in the society and generates a democratic dynamic in the distribution of knowledge and culture. In line with these proposals, it is worth remembering that comparative studies compare not only by the procedure, but because, as an analytical and interpretive resource, the comparison allows this type of analysis an adequate exploration of their fields of work and the achievement of the objectives proposed. In this sense, the comparative study requires, first and foremost, a research question that justifies the need for comparison.

It is clear that comparative research involves many difficulties. But despite everything, as stated by Geertz (1992, p. 152), “[…] the risks are not the most important thing, because running them leads to a thorough review of our understanding about what it means to open (a little) the conscience of a group (part of) the way of life of another, and by this way (part of) the own conscience”.

Translated from Spanish by Tikinet Edição Ltda.


ALMOND, Gabriel; VERBA, Sidney. The Civic Culture: political attitudes and democracy in five nations. Princeton: Princeton University Press, 1963. [ Links ]

ARNOVE, Robert. Comparative Education and World-Systems Analysis. Comparative Education Review, Chicago, v. 24, n. 1, p. 48-62, 1980. [ Links ]

BARROSO, João (Org.). A Regulação das Políticas Públicas de Educação. Lisboa: Educa, 2006. [ Links ]

BRAY, Mark; ADAMSON, Bob; MASON, Mark (Ed.). Comparative Education Research: approaches and methods. Hong Kong: Springer, 2014. [ Links ]

BURGESS, Robert. In the Field: an introduction to field research. Londres: Routledge, 1995. [ Links ]

BYNNER, John; CHISHOLM, Lynne. Comparative Youth Transition Research: methods, meanings and research relations. European Sociological Review, Oxford, v. 14, n. 2, p. 131-150, 1998. [ Links ]

COWEN, Robert. Last Past the Post: comparative education, modernity and perhaps post-modernity. Comparative Education, London, v. 32, n. 2, p. 151-170, jun. 1996. [ Links ]

CROSSLEY, Michael; VULLIAMY, Graham. Case-Study Research Methods and Comparative Education. Comparative Education , London, v. 20, n. 2, p. 193-207, 1984. [ Links ]

CULLEN, Michael. The Statistical Movement in Early Victorian Britain. Brighton: Harvester Press, 1975. [ Links ]

FIDELI, Roberto. La Comparazione. Milano: Angeli, 1998. [ Links ]

GEERTZ, Clifford. La Interpretación de las Culturas. Barcelona: Gedisa, 1992. [ Links ]

GOERGEN, Pedro. Educação Comparada: uma disciplina atual ou obsoleta? Revista Pro-Posições, Campinas, v. 2, n. 3, dez. 1991. [ Links ]

HEATH, Kathryn. Is Comparative Education a Discipline? Comparative Education Review , Chicago, v. 2, n. 2, p. 31-32, 1958. [ Links ]

JOHN, Vincenz. Geschichte der Statistik. Stuttgart: Ferdinand Encke, 1884. [ Links ]

JORI, Alberto. Hermann Conring (1606-1681): der begründer der deutschen rechtsgeschichte. Tübingen: MVK, 2006. [ Links ]

KOHN, Melvin. Cross-National Research as an Analytic Strategy. American Sociological Review, New York, v. 52, n. 6, p. 713-731, 1987. [ Links ]

KRAWCZYK, Nora Rut. Pesquisa Comparada em Educação na América Latina: situações e perspectiva. Revista Educação Unisinos, São Leopoldo, v. 17, n. 3, p. 199-204, 2013. [ Links ]

KRAWCZYK, Nora Rut; VIEIRA, Vera Lúcia. Uma Perspectiva Histórico-Sociológica da Reforma Educacional na América Latina: Argentina, Brasil, Chile e México nos anos 1990. Brasília: Liberlivro, 2012. [ Links ]

LIJPHART, Arend. Il Metodo della Comparazione. Rivista Italiana di Scienza Politica, Bologna, v. 1, n. 1, p. 67-92, 1971. [ Links ]

MADEIRA, Ana Isabela. Perspectivas actuáis da Investigação em Educação Comparada: um olhar luso-brasileiro. In: ENCONTRO INTERNACIONAL DA SOCIEDADE BRASILEIRA DE EDUCAÇÃO COMPARADA PUCRS, 4., 2008, Porto Alegre. Anais... Porto Alegre: 2008. P. 1-27. [ Links ]

MARRADI, Alberto. Comparación. In: REYES, Róman (Ed.). Terminología Científico-Social. Barcelona: Antropos, 1991. P. 65-84. [ Links ]

MARRADI, Alberto. Esperimento, Associazione e Insieme non Standard. In: BETTIN, Gianfranco. Politica e Società: studi in onore di Luciano Cavalli. Padova: Cedam, 1977. P. 675-689. [ Links ]

MARRADI, Alberto; ARCHENTI, Nélida; PIOVANI, Juan Ignacio. Metodología de las Ciencias Sociales. Buenos Aires: Emecé, 2007. [ Links ]

MOORE JUNIOR, Barrington. Social Origins of Dictatorship and Democracy: lord and peasant in the making of the modern world. Boston: Beacon Press, 1966. [ Links ]

MULKAY, Mike. Science and the Sociology of Knowledge. Londres: Allen & Unwin, 1979. [ Links ]

NÓVOA, António. Modelos de Análise em Educação Comparada: o campo e a carta. In: NÓVOA, António. Modèles d’Analyse en Éducation Comparée: le cham et la carte. Les Sciences de l’Éducation pour l’ère Nouvelle, Paris, v. 2, n. 3, 1995. P. 9-61. [ Links ]

OUTHWAITE, William. New Philosophies of Social Science. Basingstoke: Macmillan, 1987. [ Links ]

PEREYRA, Miguel Angel. La Comparación, una Empresa Razonada de Análisis: por otros usos de la comparación. Revista de Educación, Madrid, n. 1, p. 23-76, 1990. [ Links ]

PHILLIPS, David; SCHWEISFURTH, Michele. Comparative and International Education: an introduction to theory, method and practice. London: Bloomsbury, 2014. [ Links ]

PIOVANI, Juan Ignacio. Alle Origini della Statistica Moderna. Milano: Franco Angeli, 2006. [ Links ]

PIOVANI, Juan Ignacio. Comparative Research Design: life courses impact on the disengagement with the political functions of citizenship in contemporary Western Europe. [S.l.: s.n.], 1998. [ Links ]

PIOVANI, Juan Ignacio. De Objeto a Método: notas históricas sobre estatística e pesquisa social. Sociologia & Antropologia, São Paulo, v. 3, n. 5, p. 245-270, 2013. [ Links ]

RODRÍGUEZ, Justo Chávez. Aproximación a la Educación Comparada como Ciencia. Educación, Havana, n. 123, p. 21-26, 2008. [ Links ]

ROSAR, Maria de Fatima Felix; KRAWCZYK, Nora Rut. Diferenças da Homogeneidade: elementos para o estudo da política educacional na América Latina. Educação e Sociedade, Campinas, v. 22, n. 75, p. 33-43, 2001. [ Links ]

RUST, Val; SOUMARÉ, Aminata; PESCADOR, Octavio; SHIBUYA, Megumi. Research Strategies in Comparative Education. Comparative Education Review , Chicago, v. 43, n. 1, p. 86-109, fev. 1999. [ Links ]

SALVAT, Juan; MARCO, Joaquín; VICENS, Francisco. Diccionario Enciclopédico Salvat. Barcelona: Salvat Editores, 1986. [ Links ]

SASSEN, Saskia. Una Sociología de la Globalización. Buenos Aires: Katz, 2007. [ Links ]

SKOCPOL, Theda. States and Social Revolutions: a comparative analysis of France, Russia, and China. New York; Cambridge: Cambridge University Press, 1979. [ Links ]

SMELSER, Neil. Comparative Methods in the Social Sciences. Englewood Cliffs: Prentice-Hall, 1976. [ Links ]

SMELSER, Neil. Notes on the Methodology of Comparative Analysis of Economic Activity. Social Science Information, Paris, v. 6, n. 2-3, p. 101-117, 1967. [ Links ]

SWANSON, Guy. Frameworks for Comparative Research: structural anthropology and the theory of action. In: VALLIER, Ivan (Ed.). Comparative Methods in Sociology: essays on trends and applications. Berkeley: University of California Press, 1971. P. 141-202. [ Links ]

URRY, James. A History of Field Methods. In: ELLEN, Roy (Ed.). Ethnographic Research: a guide to general conduct. Londres: Academic Press, 1984. P. 35-61. [ Links ]

WESTERGAARD, Harald. Contributions to the History of Statistics. Londres: P. S. King, 1932. [ Links ]

Received: September 04, 2016; Accepted: January 18, 2017

Juan Ignacio Piovani is a Magister in Advanced Methods of Social Research and Statistics (City University London) and Doctor in Methodology of the Social Sciences (University of Rome). Full Professor and Director of the Doctorate in Social Sciences, Faculty of Humanities, Universidad Nacional de La Plata. Researcher of the National Council for Scientific and Technical Research in the IdIHCS (FaHCE/UNLP-CONICET). E-mail:

Nora Krawczyk is Magister in Education and Society from the Latin American Faculty of Social Sciences - FLACSO/Argentina and a Doctorate in Education from the State University of Campinas - Unicamp/Brazil. She is currently a professor and researcher in the Department of Social Sciences and Education of the School of Education/University of Campinas - UNICAMP and researcher of the National Research Council - CNPq. She has experience in the area of politics and sociology of Education in Brazil and in Latin America. E-mail:

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License