CC BY
62
ВОПРОСЫ ИСТОРИИ, ТЕОРИИ И МЕТОДОЛОГИИ
COMPARATIVE ANALYSIS AS A BASIC RESEARCH ORIENTATION: KEY METHODOLOGICAL PROBLEMS*
N.P. Narbut, I.V. Trotsuk
Sociology Chair Peoples' Friendship University of Russia Miklukho-Maklaya str., 10/2, Moscow, Russia, 117198
To date, the Sociological Laboratory of the Peoples' Friendship University of Russia has accumulated a vast experience in the field of cross-cultural studies reflected in the publications based on the results of mass surveys conducted in Moscow, Maikop, Beijing, Guangzhou, Prague, Belgrade, and Pristina. However, these publications mainly focus on the comparisons of the empirical data rather than methodological and technical issues, that is why the aim of this article is to identify key problems of the comparative analysis in cross-cultural studies that become evident only if you conduct an empirical research yourself — from the first step of setting the problem and approving it by all the sides (countries) involved to the last step of interpreting and comparing the data obtained. The authors are sure that no sociologist would ever doubt the necessity and importance of comparative analysis in the broadest sense of the word, but at the same time very few are ready to discuss its key methodological challenges and prefer to ignore them completely. We summarize problems of the comparative analysis in sociology as follows: (1) applying research techniques to the sample in another country — both in translating and adapting them to different social realities and worldview (in particular, the problematic status of standardization and qualitative approach); (2) choosing "right" respondents to question and relevant cases (cultures) to study; (3) designing the research scheme, i.e. justifying the sequence of steps (what should go first — methodology or techniques); (4) accepting the procedures that are correct within one country for cross-cultural work (whether or not that is an appropriate choice).
Key words: comparative analysis; cross-cultural study; quantitative and qualitative approach; methodological principles and challenges; technical decisions; comparability; linguistic, functional and pragmatic equivalence.
It is unlikely that anyone today would dare to dispute the fact that the contemporary sociology, at least for its empirical part, is fond of comparative analysis in the broadest sense of the word, i.e. sociologists conducting empirical studies (frankly speaking, mainly opinion polls) seek to compare either different groups by their values, aims, priorities, life trajectories, etc., or the same socio-demographic, professional or generational groups
* The research was supported by the Russian Foundation for Humanities. Grant № 15-03-00573.
in different time periods or socio-cultural milieus. "Comparison in sociology is inescapable. The importance and utility of comparative research are as old as the discipline itself. In a now famous quote, E. Durkheim insisted that «Comparative sociology is not a particular branch of sociology; it is sociology itself, in so far as it ceases to be purely descriptive and aspires to account for facts»". Although comparative research flourishes within this discipline, persistent methodological problems remain" [26. P. 619]. One of them is the "status" of the comparative research orientation in our discipline, which is ambiguous for at least two reasons: on the one hand, the concept "comparative studies" (or "comparative analysis") is not among the clearly defined in the sociological discourse (for instance, there is no way to draw a demarcation line between "comparative studies" and "cross-cultural studies" regardless numerous attempts to do so). On the other hand, the comparative research orientation is often an implicit part of sociological work not even mentioned in the title of the project. There are references to the comparative analysis when researchers focus on the study of a number of countries, but there are no guarantees that under a comparatively-sounding label one will find any comparisons instead of "thick" descriptions of the empirical data for every case under consideration. At the same time, it is a generally accepted unspoken and unwritten rule that there is no need to mention comparative orientation in the title of one-country studies because there are to be comparisons to identify gender, generational, economic or regional differences.
One may argue that the concept "cross-cultural studies" is ambiguous, however, unfortunately, that is not true for substantive (what we do within the research) rather than formal (how we name our work) reasons well formulated by E.K. Scheuch long time ago with reference to the study of ethnic stereotypes, but this applies to the general problem of ethnocentrism as a biased analytical perception of any other culture and society by both ordinary people and social scientists. "Cross-cultural research in the more ambitious meaning of the term is not necessarily the same as research conducted in more than one culture (or country or society). A survey of cross-cultural research shows that researchers tend implicitly to treat their own culture as the norm and all others as variations thereof... A discipline that began as a science of societies has become a discipline of the host society in which it is practiced, oriented towards the particular social problems of this country and reflecting the distinctive pre-scientific intellectual traditions of a particular culture... The bias in sociological thinking brought about by the structural conditions under which sociology is practiced tends to be reflected in research results. Thus, the seemingly objective figures which result from cross-cultural studies serve to support ethnocentric thinking... Culture-bound thinking is also reflected in the phrasing and especially in the concepts of a research problem" [32. P. 7]. This is true even for such seemingly culture-free concepts as professionalization [e.g.: 29], or achievement [e.g.: 25], or companionship [e.g.: 5]. Moreover, the problem of such ethnocentrism "should not be taken to mean that the fault is peculiar to American sociology... The concepts Gemeinschaft and Gesellschaft sometimes used in cross-cultural explanations are no more culture-free (and were ideologically contaminated to begin with)... In principle, there is nothing new in the realization that sociological explanations, concepts,
and theories are affected by a failure to reflect critically upon the social conditions in which they were developed. However, the character and the extent of this bias in cross-cultural research is unexpected..." [32. P. 9] (1). Furthermore, "comparative research has sometimes been branded as «safari» research... often being initiated by social science entrepreneurs who have wanted to go on safari in virgin lands. A more serious accusation is that of «scientific colonialism» of which there are many shades and nuances... In many comparative studies the meaning of the strategic concepts have been simply ascribed by a national team in a central country or by national teams with money and research ideas... In the most devastating critical examples... in a comparative study of values the imposed theoretical framework resulted in «seeking and examining non-existing phenomena»" [2. P. 171].
Actually, that is why most comparative studies are conducted within the framework of the quantitative approach: firstly, it is very hard to keep up one's ethnocentric stereotypes when you conduct long semi- or non-formalized individual or (focus-) group interviews, for the realities expressed in the words of informants usually destroy the researchers' myths of the everyday thinking: secondly, numbers (frequencies, average meanings, crosstabs, indexes, etc.) provided by the quantitative approach, by definition, are considered more accurate, objective and reliable than any analytical generalizations not supported by some percentages. As E.K. Scheuch aptly notices, quantitative data is harmless within cross-cultural studies for it is allegedly ascertaining rather than evaluative: there if nothing to dispute about or to get angry with when you read that, say, "Southern Europeans spend more time eating, resting and in outdoor activities generally than other Europeans; ...Americans spend more time in front of the TV set than people of any other country. All these statements are empirically true and yet false if slightly rephrased, e.g.: it is characteristic of culture X that A is more frequent; or, to go one step further: it is a peculiarity of culture X that makes people do A more frequently. The error lies in equating a somewhat unusual frequency observed in one country with a global characteristic presumed peculiar to it" [32. P. 10]. Still another reason that the examples of the qualitative approach in cross-cultural studies are so rare is that it is not designed (or presupposed) for rigorous testing of pre-formulated hypotheses — rather for assessing their applicability in the given socio-cultural context [eg.: 22].
Nevertheless, the tradition of comparative studies emerged in sociology (as well as the empirical tradition in general) within the qualitative approach, which developed in the early Chicago school with its distinct cross-cultural orientation based on practical considerations and the expectation that the immigrant city Chicago can be a substitute for a variety of cultures [e.g.: 24]. As some researchers describe the situation, "at that time, Chicago contained such a multitude of unassimilated minorities that it could be argued that all the variability necessary for a truly cross-cultural study could be obtained right at the doorstep of the University of Chicago" [32. P. 12]. At the same time, the applied orientation as it was implemented in the Chicago school research is partly responsible for the terminological confusion in definitions of comparative studies: the interest of the Chicago school representatives very fast shifted to the study of intra-societal (within-culture) rather than cross-cultural differences, i.e. peculiarities of Italians,
Poles, and other ethnic groups, and gangsters or other marginal groups were understood as resulting mainly from a set of deviant social norms and traditions not the specific national and cultural features determining behavior patterns and social attitudes.
The successors of the Chicago school refused the interpretation of intra-societal studies as cross-cultural in the more general sense of the term also due to the cooperation with the Federal Government of the United States since the early 1930's [e.g.: 32]. It is a generally known fact that cooperation with national or regional administrative bodies for the aims of social planning and helping specialized agencies to carry out their tasks usually leads to the outside (of the disciplinary borders) determination of the research problems and methodological and technical ways of solving them. Thus, interest in cross-cultural comparisons and deep detailed qualitative analysis as a rule decreases when sociological research becomes policy-oriented, because governments seek to obtain "numbers" (i. e. quantitative data) necessary to make macro-structural decisions. And it must be said that the renewal of the interest in cross-cultural research after the World War II did not change the situation fundamentally: regardless of the shifts in research goals — from the emphasis on obtaining information, which would permit to predict different nations' opinions and actions, to the general question on the reasons why some nations produce aggressive authoritarian regimes and others do not at the same stage of industrial, social, political, economic and cultural development — the comparative analysis focused mainly on describing quantitatively expressed differences between societies measured with the help of formalized questionnaires in mass surveys.
"Cross-cultural researchers, by definition, can never reject the core axioms of a scientific paradigm: that cultures exist, that they consist of elements, and that these elements can be qualitatively as well as quantitatively compared across cultures" [27. P. 16], but obviously too often prefer quantitative data and generalizations. That is why the most famous comparative studies today are World Values Survey [35] (based on the technique developed by R. Inglehart [e.g., 18]), and European Social Survey [11] (based on the technique proposed by S. Schwartz [e.g., 33; 34]), although both projects are criticized for not always equivalent samples and, thus, not valid cross-national comparisons. In fact European Social Survey was designed as a continuation of International Social Survey Programme [20] that would eliminate all its errors and biases [e.g.: 30] and help "to measure and interpret the changes in attitudes, values, perceptions and behaviors of the population of Europe, show the dependence of these changes on the social, economic and political conditions" [3. P. 136].
Considering the Russian tradition of cross-cultural studies, we must admit its quantitative preferences too, although it started as qualitative cross-cultural research (for instance, as "the civilizations approach"): "most scholars understand cross-cultural comparison as the comparison of a social phenomenon in different societies, and perhaps at different historical times, with the aim of establishing the common «causal» basis of shared features..., or the unique features of a particular culture or society, ...which does not imply that cross-cultural research should be quantitative" [22. P. 6]. Moreover, many of the first comparative studies, especially within the anthropological framework, were qualitative and brought very interesting results that were later supported by quan-
titative tests on worldwide samples [e.g.: 23]. Thus, "quantitative worldwide cross-cultural research... represents only one type of cross-cultural research. In addition, there is also a purely qualitative tradition of worldwide research that has made significant contributions to the development of new social theories; this tradition, by definition, is entirely unable to offer rigorous procedures for testing hypotheses. Qualitative approaches can only provide illustrations of validity or invalidity of those hypotheses/ theories. However, we cannot just ignore this approach because many qualitative findings have led to the formulation and testing of new hypotheses through quantitative methods. In other words, qualitative cross-cultural research has long constituted a basis for the development of quantitative research" [22. P. 7; see also: 4]. For instance, L.B. Alaev at first conducted a series of qualitative cross-cultural studies of village communities in class societies, and after that constructed a questionnaire for the systematic collection of data on the social and political organization of complex preindu-strial societies [ 1 ] and introduced a method of coded data collection to the quantitative cross-cultural research. Of course, this example represents rather anthropological than sociological tradition of comparative studies, however, (a) the former constitute the methodological basis of the latter and from the historic perspective is the first research strategy to study cultures as different "objects"; (b) "the topic of comparative research transcends subject matter, time, space and methodological affiliation; it relates to the international audience of sociologists across all regions of the world studying a wide range of subject matter and operating from diverse methodological standpoints", including anthropological [26. P. 619].
As we can conclude from all the above, the cross-national or comparative orientation in sociology as a discipline that relies basically on the observational and quantitative data obtained in different settings (represented by nation-states, for example) offers both great possibilities and serious problems in measurement and interpretation. The first type of challenges is terminological, thus, easily acceptable for omitting: the variety of terms — cross-cultural, cross-societal, cross-national, comparative — is a mere expression of the fact that in generalizing the results we focus on comparing different states/societies/cultures as determining differences. The other two types of problems are interconnected, because interpretation of data depends on the ways of obtaining it and on the conceptual grounds for choosing the method and techniques. Even a very brief overview of the relevant literature shows that the sociological discourse for the last few decades has been concerned rather with research administration and technology than with methodological issues such as the nature of the object and parameters available for either quantitative or qualitative measurement. That is a serious problem for comparative studies — the focus on certain and separate actions rather than methodological grounds for choosing them, but we cannot change the situation. Let us accept it and summarize the comparative orientation's difficulties, challenges and technical aspects in general as determining methodological ones and not vice versa (as it should be in the ideal case), relying on the long-term experience of the Sociological Laboratory of the Peoples' Friendship University of Russia in comparative studies due to the cooperation with China Youth and Children Research Centre (Beijing), Belgrade University (Serbia) and Charles University of Prague (Czech Republic) (2).
First, in any cross-cultural work one always (and rightly) expects great difficulties in applying research techniques to another country — both in translating and adapting them to different social realities and worldview. In general there is a belief that the more standardized tools (structured interviews or tests instead of group discussions, because the wording of the former can be controlled by translation-retranslation method) we use, the more easily the research technology can be exported (for example, we do not expect serious cultural differences in answering formalized questions about marriage status, education or profession). This belief creates a false sense of security, because it is based on the suggestion that identical stimuli are necessarily functionally equivalent in different countries or cultures; however, allegedly identical and absolutely neutral characteristics, such as age, sex, education or occupation, do not always have the same functional importance (or even meaning) in different cultures (for instance, being self-employed when a person retires or in his working age in an Eastern and Western country does not mirror the life conditions — rather the economic organization of the country).
There is nothing wrong with standardization in cross-cultural studies per se, but there are two problems connected with the standardization belief. (1) "In cross-national comparative research, standardized instruments or indices are available for only a small number of variables" [15. P. 5]: for occupation [e.g.: 19], education [e.g.: 21], status [e.g.: 12], income [e.g.: 16], ethnicity [e.g.: 14], and some other socio-demographic variables. (2) There are no guarantees that the standardization of the research tool will provide comparable data and help to overcome language barriers through the translation process. To achieve the comparability, we "require the syntactic, the semantic and the pragmatic understanding and analytical processing of the source language text... Researchers involved in cross-national research soon became familiar with the concept of functional equivalence, which stresses the importance of transferring meaning as opposed to translating literally and which emphasizes the fact that an iterative process of back-translation enhances face validity in the intercultural use of measurements. Face validity is achieved when a test appears valid to examinees who take it, personnel who administer it and other untrained observers... Translation problems stem mainly from lack of knowledge or awareness of cultural differences. A translating team which is bilingual but not bicultural cannot completely understand cultural differences. In such a case, functional equivalence is difficult to achieve. Ideally, therefore, all roles in the translation team (translators, reviewers and adjudicator) should be filled by persons with a bicultural background so that they can competently discuss the correct wording of a question" [15. P. 6, 7].
Frankly speaking, this condition is very hard to fulfill in most sociological projects, because practice is more complex than theory, which, for instance, provides the following set of rules to standardize and harmonize socio-demographic or socio-economic variables in cross-cultural research [15; 17]: "1) find a common definition of what is to be measured; 2) make sure that this common definition works in each of the countries surveyed; 3) analyze the national concepts and structures behind the variable of interest; 4) identify the similarities between the national concepts and structures underlying that
variable; 5) find a valid indicator or a set of indicators (depending on the variable of interest as well as on national specifics); 6) decide whether the variable of interest should be measured by the same instrument in every country or culture (input harmonization) or whether it should be which are harmonized after data collection (output harmonization); 7) test whether the chosen instrument reflects the empirical structures found in the different countries or cultures and whether the chosen instrument is logically related to the common definition; 8) make sure that the chosen measurement instrument can be understood by the average layperson in a particular culture and can be answered correctly by all respondents regardless of national and cultural contexts". According to these rules, the finally chosen survey instrument should reflect the empirical structures found in various countries or cultures and be logically related to the common definitions of the phenomena under study.
Second, while conducting comparative analysis one faces not only the problem of making the "same" research technology, but also the challenge of choosing "right" respondents to question. Sometimes, when you read methodological literature on comparative studies, you may suddenly think that research technology is the only problem cross-cultural study should overcome, and once the project groups agree on the questionnaire, you can happily travel from your country to other research sites with the ready-made concepts and instruments, and afterwards organize data into a report alone. This is not true: such an approach reduces (or confuses) experience in research administration with the rules of comparative methodology (or good research design). For instance, it is believed that the best strategy of the comparative work is to reduce with-in-group variability of the groups to be compared. Our experience disproves this rule for the artifacts of formal organization that may negatively affect the results of the research are easy to overcome if heterogeneous groups are studied. When we decided to compare representative student samples in Moscow and Beijing the formal criteria would not let us claim for comparativity: we do not possess the same administrative resources in Moscow as our Chinese colleagues in Beijing, who constructed a classical stratified random sample of student population, while we used the 'profession' criteria (natural sciences, engineering, social sciences and humanities, etc.) and "snowball" sampling to have the right to compare worldviews and value orientations of the student youth in two countries.
While generally recognizing the challenges of choosing "right" respondents to question, researchers often miss or deliberately ignore the problem of choosing cases (cultures), which usually depends on the type of intent that guides a study and determines the choice of an optimal design among the variety of possible ones. Within comparative studies there are two widespread strategies to select cultures: (a) in such a way that they differ simultaneously in several factors and also the one that interests the researchers the most, which guarantees the minimum level of internal variability (for instance, this approach helped sociologists to prove that nuclear family remains the key form of family organization in all contemporary societies and, thus, can be considered universal); (b) in such a way that cases are as similar as possible in several respects except for the phenomena to be studied (for example, such an approach is an op-
timal research solution to identify factors that determine different level of political and social activities in the same economic and political situation within one regional framework).
However, the challenges of choosing "objects" do not exhaust the second type of problems the comparative approach faces if we step over the disciplinary borders of sociology. We must confess that we are a bit luckier than, let us say, anthropologists, who (a) deny the right to compare traditional non-industrialized communities and industrialized nation-states; (b) question the right to treat a modern nation-state as a culture in the anthropological sense of the word (i. e. as a unit of both observation and analysis); (c) question the very interpretation of the nation-state as an integration of sub-institutional behavior forms, because such an assumption leads to ignoring social organization at all; (d) even consider scientific comparative analysis impossible due to the problem of defining cultures as wholes and constructing cultural units for comparison, because cultures are not "partible", "they are neither wholes nor complexes of traits, but domain constructs rather than perceivable objects with any obvious boundaries... thus, there are no principled reason for selecting one threshold over another" [13. P. 294]. This critique denies the use of theory for constructing cultural units and suggests to focus on methods of obtaining empirical data as guarding us against magnifying subjective biases by using systematic data collection and analytical techniques [28. P. 12].
Undoubtedly, the above considerations are not typical or necessary for comparative studies in sociology: we take the internal heterogeneity of complex societies that we compare in cross-cultural research for granted and believe that characteristics of individuals are mediated in different ways by the network of higher-order social structures and institutions, Thus, in mass surveys we attribute between-country differences in the distribution of characteristics to one of several factors (or to some combination of them): national culture; processes of mediation by higher-order institutions; different population subgroups; differentiated conditions that explain dissimilarities of behavior; or to much more prosaic explanations through the traditions of everyday life including working-place and family interactions, etc. That is, many of the differences will probably disappear if we control structural factors to be constant, especially demographic variations — within-culture differences may be greater than between-culture ones. "The tendency immediately to invoke culture or society as explanatory factors capable of accounting for differences between two bodies of data is an aggravated version of the ecological fallacy. The likelihood of such a fallacy is reduced, and the quality of the explanatory scheme dramatically improved if we no longer try to link observations on one level (mostly responses by individuals) directly to high-order constructs such as societies. The collection and interrelation of data on more than one level is a most promising development in research methodology" [32. P.23]. Nevertheless, as a rule, we still have to decide on the scale of analysis: "The choice lies between a small and a relatively large N (i.e., sample size), which each pose specific problems. In the case where the researcher chooses to include a large number of units (e.g. countries) with only scant, more general comparative variables she or he runs the risk of producing superficial,
though potentially statistically sound results. On the other hand, if the researcher chooses to include only a few units of analysis with numerous variables she or he takes the risk of having too many variables and too few cases to effectively test causal models" [26. P. 621; see also: 10].
Third, there is a typical rule in many sociological textbooks and in our biased scientific perception: at first, one should choose the basic concept and its theoretical definition, regardless its cultural and political implications, and make an attempt to reach consensus in the definition, and only then proceed to observables and descriptors expressed in a number of empirical indicators. Unfortunately, this rule, although being right in the essence, does not always work in comparative studies, that is why one should accept the possibility that after completing field work each research team has the right to classify the indicators measured in the way so as to suit their particular way of fitting the world's diversity into some theoretical conceptual "boxes". Furthermore, in some cases this approach is more preferable for it does not let us ignore the fact that cross-cultural comparisons logically are no more than observations under differing conditions of recording/obtaining data: even if the same questions are used (in different countries), respondents may understand them differently, and their use of the scale to answer the value questions might also be dependent on the temporal or cultural context [e.g.: 7].
Fourth, there is obviously not enough sociological works focusing on methodological rather than technical aspects of comparative analysis. "Most writing refers to the technology of research: to difficulties of developing and enforcing comparable data collection routines, or to often unhappy experiences in administering cross-cultural work" [32. P. 9], although the critical analysis long ago showed that rules of procedure that are correct within one country can be detrimental in cross-cultural work [e.g.: 31]. Thus, the best and the most correct strategy to conduct a comparative study is to begin with methodology, and then go further to particular techniques as determined, valid and justified by the chosen methodological framework. For instance, while planning and designing the comparative study of patriotism among Russian and Czech students at the beginning of 2000s, we discovered that the clear and universally recognized and accepted by the Russian youth and population in general notion of patriotism was absolutely irrelevant for the Czech society. The Russians distinguish citizenship and nationality due to the diverse ethnic composition of the country, but this does not affect their clear identification in terms of patriotism; the Czechs do not understand the meaning of the word "patriotism", because they distinguish their nationality and citizenship in the self-identification (although there is a clear regional differentiation in the country — the Czech Republic (the capital is Prague), Moravia (Brno) and Silesia (Opava), the country is considered a one-nation state, so patriotism here acts as a self-identification as a "Czech"). Unlike the multi-ethnic Russian state, where the concept of patriotism is of a supra-ethnic nature, in the Czech Republic it merges with different ethnic and territorial issues. Thus, the correct conceptualization of patriotism and the focus on the methodology of the survey helped us to make a functionally and pragmatically comparable questionnaire that measured all three components of patriotism as a youth value
orientation — cognitive, affective and conative — and revealed that in both countries none of them was dominant or more pronounced than the other two.
Summing up all the above, we can confidently state, that the happy days of the cross-cultural research have come: comparative studies have become a common type of research rather than an exception causing doubts and suspicions in validity, objectivity, reliability and neutrality in estimates. Nevertheless, there are still serious methodological (and, thereafter, technical) challenges to overcome to have the right to consider how different "settings" (nations, cultures, societies) become crucial in producing different "data". Unfortunately, quite often cross-cultural research, by the implications inherent in its thematic, methodological, and organizational design, is still largely a within-country research simply duplicated in a number of countries. This is partly determined by the fact that our socio-culturally biased thinking is often (frankly speaking, always and inevitably) reflected in cross-cultural research, since the latter is less held in check by researchers' self-awareness than in any intra-cultural research. That is the only reason why some rules of good sociological research procedure may not always hold for cross-cultural work.
Probably, there are no final decisions for the problems of comparative analysis in sociology summarized above, except for the relationship between quantitative and qualitative approaches. Today it is generally accepted that there are good reasons for combining both in comparative research, and no good reasons to eliminate quantitative tradition from the field, because without efforts to carry out quantitative comparisons a set of questions (what nations/cultures really are, how to compare the same variables measured through different empirical indicators, etc.) can never be solved or even addressed. Obviously, without at least attempts to discuss and overcome the problems summarized in the article the comparative analysis in sociology will be theoretically unsound and will not have an adequate research design. We hope that our article does contribute to the methodological reflections on the "use" of culture as an explanatory factor in the correct sociological way and to further discussions of the comparative analysis challenges not confined to the description of empirical data.
NOTES
(1) A reader may wonder why the authors refer to such old articles and data. The answer is simple: such examples show, on the one hand, that not much has changed in the cross-cultural studies considering their key methodological problems; on the other hand, that the sociological community is ungrateful and does not pay attention to the problems already identified — ignores them at all, or the ways of conceptualizing them in the past.
(2) We intentionally do not refer to our numerous publications based on the results of mass surveys conducted in Moscow, Maikop, Beijing, Guangzhou, Prague, Belgrade, and Pristina, because they mainly focus on the comparisons of the empirical data rather than methodological and technical issues. The aim of this article is to identify key problems of the comparative analysis in cross-cultural studies that become evident only if you conduct an empirical research yourself — from the first step of setting the problem and approving it by all the sides (countries) involved to the last step of interpreting and comparing the data obtained.
REFERENCES
[1] Alaev L.B. Opyt tipologii srednevekovyh obshchestv Azii [A typology of medieval societies of Asia] // L.B. Alaev (Ed.). Tipy obshchestvennyh otnoshenij na Vostoke v srednie veka. M., 1982.
[2] AllardtE. Challenges for comparative social research // Acta Sociologica. 1990. Vol. 33. № 183.
[3] Andreenkova A.V. Mezhstranovye sravnitel'nye issledovanija v social'nyh naukah: metodologija, etapy razvitija, sovremennoe sostojanie [Cross-country comparative analysis in social sciences: Methodology, development, and the current state] // Mir Rossii. 2011. № 3.
[4] Bondarenko D.M., Korotayev A. V. "Early state" in cross-cultural perspective: A statistical reana-lysis of Henri J.M. Claessen's database // Cross-Cultural Research. 2003. Vol. 37.
[5] Burgess E.W., Wallin P. Engagement and Marriage. Chicago, 1953.
[6] Burton M.L., White D.R. Cross-cultural surveys today // Annual Review of Anthropology. 1987. № 16.
[7] Davidov E. Theoretical implications of comparative survey research: Why the wheel of cross-cultural methodology keeps on being reinvented // International Journal of Comparative Sociology. 2010. Vol. 51.
[8] Davidov E., Meuleman B., Cieciuch J., Schmidt P., Billiet J. Measurement equivalence in cross-national research // Annual Review of Sociology. 2014. Vol. 40.
[9] Davidov E., Schmidt P., Schwartz S. Bringing values back in: Testing the adequacy of the European Social Survey to measure values in 20 countries // Public Opinion Quarterly. 2008a. Vol. 72.
[10] Ebbinghaus B. When less is more: Selection problems in large-N and small-N cross-national comparisons // International Sociology. 2005. Vol. 20.
[11] European Social Survey // URL: http://www.ess-ru.ru.
[12] Ganzeboom H. BG., Treiman D.J. Three Internationally Standardised Measures for Comparative Research on Occupational Status // J.H.P. Hoffmeyer-Zlotnik, C. Wolf (Eds.). Advances in Cross-National Comparison. A European Working Book for Demographic and Socio-Economic Variables. N.Y., 2003.
[13] Gatewood J.B. Distributional instability and the units of culture // Ethnology. 2000. Vol. 38.
[14] Hoffmeyer-Zlotnik J.H.P. How to measure race and ethnicity // J.H.P. Hoffmeyer-Zlotnik, C. Wolf (Eds.). Advances in Cross-National Comparison. A European Working Book for Demographic and Socio-Economic Variables. N.Y., 2003.
[15] Hoffmeyer-Zlotnik J.H.P. Harmonisation of demographic and socio-economic variables in cross-national survey research // Bulletin de Méthodologie Sociologique. 2008. Vol. 98. № 5.
[16] Hoffmeyer-Zlotnik J.H.P., Warner U. Methodological discussion of the income measure in the European Social Survey. Round 1 // URL: http://mrvar.fdv.uni-lj.si/pub/mz/mz3.1/hoffmeyer.pdf.
[17] Hoffmeyer-Zlotnik J.H.P., Wolf C. Comparing demographic and socio-economic variables across nations. Synthesis and Recommendations // J.H.P. Hoffmeyer-Zlotnik, C. Wolf (Eds.). Advances in Cross-National Comparison. A European Working Book for Demographic and Socio-Economic Variables. N.Y., 2003.
[18] Inglehart R. Changing values among Western publics from 1970 to 2006 // West European Politics. 2008. Vol. 31.
[19] International Standard Classification of Occupations: ISCO-88. Geneva, 1990.
[20] International Social Survey Programme // URL: http://www.issp.org.
[21] International Standard Classification of Education // J.H.P. Hoffmeyer-Zlotnik, C. Wolf (Eds.). Advances in Cross-National Comparison. A European Working Book for Demographic and Socio-Economic Variables. N.Y., 2003.
[22] Khizrieva A. G., de Munck V. C., Bondarenko D.M. The Moscow School of quantitative cross-cultural research // Cross-Cultural Research. 2003. Vol. 37. № 5.
[23] Korotayev A. V. Unilineal descent organization and deep Christianization: A cross-cultural comparison // Cross-Cultural Research. 2003. Vol. 37.
[24] Lazarsfeld P.F., Leeds R. International sociology as a sociological problem // American Sociological Review. 1962. Vol. 27.
[25] McClellandD.C. The Achieving Society. Princeton, N.J., 1961.
[26] Mills M., van de Bunt G.G., de Bruijn J. Comparative research: Persistent problems and promising solutions // International Sociology. 2006. Vol. 21.
[27] de Munck V.C. Contemporary issues and challenges for comparativists: An appraisal // Anthropological Theory. 2002. Vol. 2. № 5.
[28] de Munck V., Korotayev A. Cultural units in cross-cultural research // Ethnology. 2000. Vol. 38.
[29] Parsons T. Essays in Sociological Theory. Glencoe, 1954.
[30] Rokkan S. The development of cross-national comparative Research: A review of current problems and possibilities // Social Science Information. 1962. Vol. 1. № 21.
[31] Scheuch E.K. Comparative sample surveys: Progress in the cross-cultural use of surveys // S. Rokkan (Ed.). Comparative Research across Cultures and Nations. Paris-The Hague, 1968.
[32] Scheuch E.K. Society as context in cross-cultural comparisons // Social Science Information. 1967. Vol. 6. № 7.
[33] Schwartz S.H. Universals in the content and structure of values: Theoretical advances and 20 countries // Advances in Experimental Social Psychology. Vol. 25 / Ed. by M. Zanna. N.Y., 1992.
[34] Schwartz S.H., Melech G., Lehmann A., Burgess S., Harris M., Owens V. Extending the cross-cultural validity of the theory of basic human values with a different method of measurement // Journal of Cross-Cultural Psychology. 2001. Vol. 32. № 5.
[35] World Values Survey // URL: http://www.worldvaluessurvey.org.
СРАВНИТЕЛЬНЫЙ ПОДХОД КАК БАЗОВАЯ ИССЛЕДОВАТЕЛЬСКАЯ ОРИЕНТАЦИЯ: КЛЮЧЕВЫЕ МЕТОДОЛОГИЧЕСКИЕ ПРОБЛЕМЫ*
Н.П. Нарбут, И.В. Троцук
Кафедра социологии Российский университет дружбы народов ул. Миклухо-Маклая, 10/2, Москва, Россия, 117198
На сегодняшний день Социологическая лаборатория Российского университета дружбы народов обладает огромным опытом в сфере проведения сравнительных исследований, частично отраженным в публикациях авторов по результатам массовых опросов, реализованных в Москве, Майкопе, Пекине, Гуанчжоу, Праге, Белграде и Приштине. Однако в большинстве своем эти работы фокусируются на сопоставительном анализе эмпирических данных, а не на методологических и методических аспектах кросс-культурных исследований. Цель статьи — обозначить базовые проблемы сравнительных исследований, которые зачастую становятся очевидны только в том случае, если вы сами реализуете социологический проект — начиная с формулировки исследовательской проблемы и согласования ее теоретической и эмпирической интерпретации со всеми сторонами (участву-
* Исследование выполнено при поддержке РГНФ. Грант № 15-03-00573.
ющими в исследовании странами) и заканчивая интерпретацией и сравнением полученных данных. Авторы статьи уверены, что сегодня никто из социологов не будет оспаривать важность и нужность сравнительных исследований, однако мало кто заинтересован в обсуждении методологических проблем сопоставительного анализа. На наш взгляд, ключевые проблемы сравнительных исследований можно объединить в следующие группы: (1) использование разработанного для одной страны инструментария в ином социокультурном контексте (трудности перевода и адаптации вопросов к иным социальным реалиям и мировоззренческим схемам; проблемный статус стандартизации и качественного подхода); 2) отбор «правильных» респондентов и релевантных кейсов (культур); 3) разработка дизайна исследования, или обоснование последовательности шагов (что следует делать первым — выбирать методологию или методику); 4) применение апробированных во внут-ристрановом контексте методик для кросс-культурных проектов (насколько это возможно в принципе и для конкретных проектов в частности).
Ключевые слова: сравнительный анализ; кросс-культурное исследование; количественный и качественный подходы; методологические принципы и проблемы; методические решения; сопоставимость данных; лингвистическая, функциональная и прагматическая эквивалентность.
ЛИТЕРАТУРА
[1] Алаев Л.Б. Опыт типологии средневековых обществ Азии // Типы общественных отношений на Востоке в средние века / Под ред. Л.Б. Алаева. М., 1982.
[2] Андреенкова А.В. Межстрановые сравнительные исследования в социальных науках: методология, этапы развития, современное состояние // Мир России. 2011. № 3.
