Contextualizing the Religious Survey

Possibilities and Limitations

in Religion and Society
Author:
Astrid Krabbe Trolle Assistant Professor, University of Copenhagen, Denmark a.trolle@hum.ku.dk

Search for other papers by Astrid Krabbe Trolle in
Current site
Google Scholar
PubMed
Close

Abstract

Censuses and surveys represent a two-edged sword. They are both a technology of governance for national and former colonial administrations and a tool of recognition for the minoritized. In this article, I discuss the history of censuses and surveys in a Danish context, arguing that the regional and local history of registration is crucial for understanding how and why religious identity becomes visible and important as a measure for the population. Applying the case of a national survey on religiosity in relation to the Danish Evangelical Lutheran Church in 2020, I ask how religion comes alive through the strategic use of artificial ideal types aimed at mapping a religious mainstream. Surveys introduce a distance to messy religious reality, thereby reducing complexity and richness. Yet this distance also allows the researcher to ask new questions that go beyond the immediate religious experience.

This Symposium asks how our scholarly engagement with religions and religiosity is shaped by both individual and disciplinary genealogies. As Bandak and Stjernholm point out in their introduction, these genealogies can have epistemological, tactical, and personal qualities, yet the recurring theme for our engagement remains religion, with its inherent tension and opaqueness. Religion has a multifaceted presence that sieves through social fabrics and in the process is molded to mean very different things. In James Beckford's (2003: 20) words, the best strategy for scholars of religion would be to “map the varieties of meaning attributed to religion in social settings, to discern the relative frequency of the prevailing meanings and to monitor changes over time.” This view of religion as a social activity shaped by use, frequency, and time is helpful in portraying how sociologists approach the subject of religion. As might be evident from the quote, religion is understood as inherently social and measurable from a sociological perspective. The emphasis on measurability places some of my work as a sociologist of religion in the quantitative tradition, where the translation of individual religiosity into numbers for statistical analysis is key. In sociology, the interest in surveys came early. Following the concern for modernization, Max Weber carried out a survey on agricultural workers in East Prussia in 1891 (Desrosières 1998: 185). Emile Durkheim regularly used statistics to diagnose society (see, e.g., Durkheim [1897] 1979: 196), and Maurice Halbwachs applied statistics in his thesis on workers’ living standards in the early 1900s (Desrosières 1998: 188). Because the founding sociologists saw society—and very often also religious traditions and institutions—as the primary object of analysis, census data and surveys became part of the sociological discipline from its infancy.

Although research on religion through quantitative means did not become more important until the 1950s and 1960s (Voas 2007: 146), quantitative approaches such as surveys are now included in any textbook on method in the sociology of religion (e.g., Riis 2011). At the University of Copenhagen, where I received my training, sociology of religion has been an independent area of study since 1970. Similar to the other Nordic countries, the Danish branch of the sociological study of religion has sprung from early studies of majority Christian practices and beliefs (e.g., Rod 1961; Salomonsen 1966, 1971). The inclusion of minority religions and a closer engagement with international theories of religion and society since the late 1970s (Warburg 1979) brought the Danish branch of sociology of religion in line with sociological environments in Britain, the United States, and Canada. However, the initial interest in majority Christianity and church-state relations still remains a strong characteristic of the Scandinavian sociology of religion.

This article has its origins in my experience of moving back and forth between qualitative fieldwork and quantitative survey work, and how shifting between these two methods creates flexible ways of engaging with religion. In this sense, sociologists working with mixed methods often navigate between different epistemological perspectives that may also be related to their personal genealogies. In my view, studying religion through numbers means that one is allowed to creatively challenge the complexities of qualitative understandings (verstehen) by converting them into generalizable macro entities. This flexibility speaks directly to my personal enthusiasm for methodological diversity. In 2020, I conducted a national survey on religion at the Knowledge Center for the Evangelical Lutheran Church (Folkekirkens Videnscenter) under the Church Ministry in Denmark. Together with three colleagues, my job was to create a national mapping of the Danish population in terms of religiosity and attitudes toward the Evangelical Lutheran Church, majority church in Denmark. After several test runs, the final questionnaire was sent out by Statistics Denmark to a random populational sample in the spring of 2020. The respondents numbered 4,109 with an answer rate of 51 percent, making it one of the largest Danish surveys on religion to date (Poulsen et al. 2021). The ministerial framing meant that the survey work took place outside of the university, in an interesting space between the majority church and the state. In Denmark, the Evangelical Lutheran Church is not separate from the state, and, as a result, quite a number of studies on majority Lutheran Christianity are funded by the Church Ministry or by independent church organs. Although the close ties between the ministry and the majority church could potentially pose a threat to an independent mapping of populational religiosity, the survey in question simultaneously speaks to a long tradition of sound religious registration in Denmark.

Censuses and Surveys on Religion

Censuses and surveys are part of a contested history as they are conflated with national and regional means of governance. For the sake of clarification, censuses and surveys are distinguishable in terms of scale and method. Censuses map entire national populations with the participation of every citizen, whereas surveys are often based on samples of the population under investigation.1 A strict definition of census requires (1) that a national legal authority is behind the population census, (2) that it is mandatory to fill out the questionnaire,2 and (3) that the census should be performed at regular intervals (Thorvaldsen 2018: 2).3 The survey, on the other hand, operates on a smaller scale and can also target particular groups or geographical regions. With more sophisticated technical support in terms of measurement, the sample survey has replaced the census in several countries (House et al. 2004: 16), just as many quantitative studies on religiosity are survey based.

The first censuses appeared in the 1600s and were related to the administration of the French, British, and Danish colonies (Solli and Thorvaldsen 2012: 109). Because of resistance toward registration in the mainland areas of these countries, census implementation began in the outer regions of the colonial kingdoms. According to Gunnar Thorvaldsen (2018), the earliest nominative census (including both women and men) was gathered in Iceland in 1703 by the Danish administration. Due to volcano eruptions in 1701 with serious human consequences, the colonial administration feared that the Icelandic population would shrink and possibly disappear (ibid.: 20), so it conducted a full-count census. This period marked a mercantilist approach to enumeration where the prosperity and power of a kingdom were expressed through the number of subjects of the crown: the more subjects, the more power (ibid.: 35). In the early censuses, then, enumeration became important as a means of showing greatness in competition with other kingdoms. A point to draw from the mercantilist numerical vision is that populational registration does not necessarily entail a sustained interest in pruning the national population. As Michel Foucault (2007: 71) has argued, the institutional will to govern the populational body through registration only began in the eighteenth century, after the mercantilist era. Besides this goal, surveys were also used to take stock of local environments in the face of rapid industrialization and urbanization, portraying workers and their living conditions (Bulmer 2001: 14469). Probability sampling and complex statistical analysis were only possible after the 1940s, and until then census taking was the most reliable form of registration available (ibid.: 14470).

If we turn to questions of religion in censuses, local variation is vast. The first census to ask a question about religion was the 1776 census for St. John and Nova Scotia, Canadian territories under British rule (Thorvaldsen 2014: 205). From the middle of the nineteenth century several countries integrated enumerations on religion as this was recommended by the international statistical conferences (ibid.). However, some countries have never asked their populations about religious identity or affiliation on censuses. Up until today, the United States has not included a question of religion.4 In Britain, the reintroduced question of religious affiliation in the 2001 census resulted in broad public discussion. It had been removed from the census in 1861 (Howard and Hopkins 2005: 70) and made voluntary in the 2001 version. Religious identity and affiliation were recognized as sensitive issues, which is why several censuses in the late 1800s and into the 1900s did not include religious questions. Religious privacy, then, was respected in several countries before the terrible events of World War II. After the war, the emphasis on religious freedom from registration did not diminish, and only in the beginning of the 2000s, with an increased awareness of Muslim citizens in the European countries, did religiosity surface as a politically interesting topic in national censuses such as the British case. Again, censuses and surveys differ, and surveys are more in tune with the immediate interests of political actors. In the 1980s and 1990s, international surveys, such as Religious and Moral Pluralism (RAMP) in Europe, grew out of a Christian concern about shifting values; the same applies to the European Values Study (EVS) (Arts et al. 2003: 3–6). With the growing visibility of especially Islam in the public sphere of many European countries over the last two decades, several local surveys have been directed at European Muslims with the intention of measuring people's attitudes toward human rights and gender equality (Johansen and Spielhaus 2019). Religious identity has become an entry point into understanding contemporary societal tensions. Subsequently, we see a growth in religious questions on surveys—and also more academic critique of the survey method from scholars of religion.

Danish Registration of Religiosity

Unlike many other countries, the registration of majority religious identity and affiliation has been fully integrated into even the early Danish censuses due to the close cooperation between the majority church, the Evangelical Lutheran Church, and the governments of the time. In Denmark, general census counts stopped in 1970 (Hansen 2007: 38) when the system was replaced by the Danish Civil Registration System (CRS, Det Centrale Personregister).5 Today, anyone in need of information about the Danish population on a general level can easily find answers concerning age, gender, country of birth, majority church membership, and ecclesiastical acts (baptism, weddings, funerals) within the Evangelical Lutheran Church. In terms of religious affiliation, only membership in the majority church is centrally registered. All other religious affiliations have not been enumerated since World War II, and questions of religion disappeared from the general census in 1955 (Thorvaldsen 2014: 207). The comprehensive registration taking place in Denmark has a long history of church-state cooperation wherein the state-employed clergy of the majority church are still responsible for the registration of all Danish citizens regardless of individual religious affiliation. During the early colonial period of census taking, the Danish administration convinced the clergy of the Evangelical Lutheran Church to compile annual reports of demographic data in every parish under the pretext of wanting to build larger churches (Solli and Thorvaldsen 2012: 110). Although the demographic interest was also nurtured by the administration's wish for soldiers, the clergy gathered information on individuals from 1701 onward, thereby laying the ground for the contemporary system of registration service kept within the majority church.6 The Danish colonial administration covered the areas of Norway, Iceland, and the Faroe Islands, as well as the German provinces of Schleswig, Holstein, Oldenburg, and Delmenhorst (Johansen 2004: 26). In other words, a regional tradition of census taking developed independently of the mainland European traditions and has resulted in a wide-ranging and historically long collection of individual data on the Danish population.

In 1981, Denmark became the first nation to carry out a census based entirely on the CRS register. With the amount of information currently available, authorities no longer had to ask questions to individuals directly (Bauer 2014: 196). Today, the Danish model of registration links several types of data, resulting in thorough and longitudinal data on health, work ability, migration, mortality rates, and much more (Schmidt et al. 2014: 543). My point here is that the enumeration of the population is extensive and invisible. Danish citizens are already part of a regime of surveillance in the Foucauldian sense, yet religious identity—except for membership in the majority church—is not part of the governing infrastructure. Instead, (minority) religious identity has only recently grown into a topic of political interest. The separate colonial history and census system of the old Danish administration differ from the mainland European narrative of statistical developments following the standardizing efforts of the renowned Belgian statistician Adolphe Quetelet in the 1800s. Remembering the regional history helps us to contextualize the religious survey and points to how the Danish experience of invisible and extensive registration is also echoed in the before mentioned survey on religiosity and relations to the Evangelical Lutheran Church. The next section addresses two critiques of censuses and surveys that contribute to a nuanced discussion of the limits of the quantitative method in measuring religiosity.

Two Critiques: The Suppression of Minorities and the Reduction of Religious Complexity

Censuses and surveys have mostly been implemented as a means of showing power and giving insights into the demography of populations. In broad terms, this could be recognized as the majority (state) perspective on enumeration. Within the framework of enumeration, religious registration holds a special place, leading to a critical reception of censuses and surveys by many scholars of religion. The first aspect of this critique has to do with census taking as involuntary registration. The understanding of registration as an expression of unwanted control has biblical roots. In the Hebrew Bible/Old Testament, censuses were referenced to the conscription of soldiers, and in the New Testament to the Roman Empire's registration of newborn male children (House et al. 2004: 5). Especially the Christmas story's narrative of how the Romans used registration to administer, control, and defeat is a powerful leitmotif. The fact that the first systematic censuses were performed in former colonies also pays tribute to the close relationship between census taking and unwanted state control. From a post-colonial perspective, censuses are understood to perform a process of objectification. In the words of Bernard Cohn (1987: 230), this objectification entailed that the local population in India needed to explain themselves again and again to the British administration. Initially, the censuses in India consisted of information on the history of the Indian districts and families, whereas similar censuses in South Africa also assessed the distribution of race and ethnicity (Christopher 2002: 401). In these colonial settings, census taking became a part of administrative control that rarely benefited the local population.

The close affinity between state administrations and censuses also incorporated a nationalist outlook in the 1800s where membership in the majority church in the European countries strengthened the national project (Thorvaldsen 2014: 205). The Holocaust was made possible by census data on the Jewish communities (Bouma and Hughes 2014: 439), and in the aftermath of this event, religious registration comes with long shadows in the religious field. In short, it is easy to find examples of damaging registration that could only benefit the state, and these examples are inscribed in powerful narratives of colonialism, nationalism, and World War II. The focus on how census and survey data remain a political power tool finds support with several scholars of religion. Abby Day and Lois Lee (2014: 346) remark how populational statistics can act as “weapons of mass distrust,” informing governments through invasive questions on religious identity. The same concern for survey data as an expression of political intent comes through in the work of Talal Asad. In Secular Translations, Asad (2018: 9) illustrates the suspicion that numeration induces calculative reasoning that distorts the social reality: “When translation of words into numbers takes place, there is also something essential that is discarded so that it belongs only to the afterlife of the original, creating new possibilities and new dangers.” Asad explains that words and numbers represent two forms of translation—the remaking of the body and soul (words) and the translation of political information (numbers) with the goal of transforming collective life. The abstraction and universality brought by numbers enable the political power to mask intent and to “regulate and reinvent the world” (ibid.: 133), thus overriding discursive interpretation and embodied religious practice. Asad is right to point out how numbers have a flexible character that political regimes can use to harm groups and individuals. One of the concerns in applying survey data is that the data set becomes so detached from its original intent that new analyses cannot be trusted. Unlike qualitative material, survey data will often be available as a data set to download from research libraries. The free access and seemingly coherent structure of content may lead to interpretations that cannot be substantiated in the original context of the survey.

However, numeration and registration can also foster resistance. Looking back into history, minority groups have used census and survey data to gain recognition. If we return to the 2001 census in Britain when a question of religious identity was included, the possibility of registering as Muslim rather than being limited to an ethnic category was important to Muslims in Britain (Sherif 2011). Religious recognition has been a main driver for many religious minorities in their approach to registration. In 1833, the Unitarians contributed to the establishment of the Manchester Statistical Society because of their wish for member registration (Chryssides and Geaves 2007: 276). In the US, survey research before the 1940s was mainly applied by social reformers such as Florence Kelley Hull and W. E. B. DuBois to ensure rights for minoritized segments of the population (Bulmer 2001: 14471). Gary Bouma and Philip Hughes (2014: 439) sum up how minority groups use census data in Australia as “support, service and recognition.” Census and survey data can thus work both ways—as a technology of governing and as a tool to claim recognition. Keeping a narrow focus on censuses and surveys as mainly a means of consolidating state power risks overlooking the possibilities that large numbers can bring to minorities.

The second critique is that surveys reduce religious complexity. As Day and Lee (2014: 345) have argued, the “tick-box talk” of surveys cannot satisfactorily grasp religious and especially non-religious identities that do not fit the given categories in the questionnaire (see also Day 2013: 73; Voas 2007). The visibility and authority of the response categories often portray only one version of religious identity that the respondents can either accept or reject (and then potentially drop out of the survey altogether). This argument resonates with Asad's earlier statement that in the process of translating the social into numbers, meaning is lost. A theological reading would also view the attempt to reduce the richness of religious reality as unimaginative and poor in spirit. Questionnaires do reduce complexity, but we could also ask if religious complexity is any different from other types of messy reality. One of the most difficult categories to standardize at the international statistical conferences in the 1800s was occupation and the social status ascribed to individual occupations. Every country boasts of specific work traditions that cannot be translated directly into a neighboring context. Categorizing religiosity has similar problems. In addition, response categories about religiosity fit less easily into complex statistical analysis because questions to questions such as “Do you believe in God?” (Yes, No, I don't know, I don't want to answer) cannot be given any meaningful numeration. The answers to “the God question” therefore remain nominal (Spickard 2017: 116). But although censuses and surveys mainly yield simple insights into a ‘tick-box’ version of religiosity, they are able to show us answering patterns on a large scale that could not be achieved by material gathered through a single researcher's interaction with the religious material. Understanding survey data as only repressive or reductionist risks losing a chance to think creatively about religiosity on a numerical scale.

Lastly, the two critiques point to epistemological questions of what counts as desirable and legitimate knowledge. Framing numeration as calculative political reasoning (Asad 2018: 134) reveals the old—and ongoing—conflict between a (seemingly) positivistic idea of the measurable world and the hermeneutical credo of viewing reality as social and thereby reachable only through interpretation. Surveys combine both positions, although this might be mainly apparent in the process of creating a questionnaire.

Searching for a Religious Mainstream

Returning to the survey on populational religiosity that I conducted on behalf of the Evangelical Lutheran Church and the Danish Church Ministry, it is embedded in a long tradition of church registration in a somewhat religiously homogeneous nation. The purpose of the survey was to give an overview of the Danish population in terms of their individual religiosity and relations to the Evangelical Lutheran Church. As mentioned before, this survey (Poulsen et al. 2021) stood out in terms of size and location. It represents an engagement with religion from actors outside of the university with particular interests from a state and a church perspective. The intention to cover religiosity in broad terms matches the Evangelical Lutheran Church's goal of being a church for the population at large. Asking about the relations and use of the majority church speaks to the ministerial interest in the connections between the population and the church as well as the move toward the commodification of church services. The questionnaire had 55 questions divided into three sections: (1) individuals’ religiosity and religious background, or the absence of one or both, (2) attitudes toward the Evangelical Lutheran Church, and (3) religious membership and religious transmission. Background variables such as gender, age, education, and level of urbanization were available to us through the CRS register. We thereby benefited from the extensive registration of Danish citizens already present. As with earlier studies, we returned to Durkheim's notion of religion as “affiliation, practice, and belief” (Brierley 2010: 69) in an attempt to cover several ways of being religious. Questionnaires are always designed as an integrated experience, a sort of puzzle, where every question triggers a mood that can influence the respondents’ later answers. This is also the reason why we chose to ask about religious membership—if any—at the end of the questionnaire. We did not want to remind people of their affiliation before they answered questions about belief and practice. In a sense, the questionnaire was designed to bring forth a slow and unconscious unpacking of religious content. Although belief is known to be a multi-dimensional concept, we followed the “propositional” line (Day 2013: 6) and asked about belief in healing, reincarnation, and clairvoyance to discern the possibility of holistic environments in Denmark. In addition, we asked about belief in God, but not about belief in Hell, Sin, and Paradise. These dogmatic scales were left out as several theological colleagues commented that including them in the survey would express a fundamentalist religious ideal that does not fit the contemporary Danish reality.

There is an important point here that has to do with the survey method's ability to measure ‘unorthodox’ belief and practice. Earlier in this article, I mentioned how my discipline, the Danish branch of sociology of religion, was shaped by efforts to bring international theories of religion and society into conversation with the Nordic religious field. One of the attempts in the survey genre came with the European Values Study (EVS), which generates data especially fit for theories of value change within Western societies.7 Yet the theories of post-materialism aimed at cross-national comparison (e.g., Inglehart and Welzel 2005: 52) and a standardized set of questions based on dogmatic beliefs (Riis and Gundelach 1992: 20) also limit the understanding of religion in a Nordic context (Willander 2020: 51). Before the EVS, local surveys had shown a much higher level of individual religiosity than the outcome of international surveys would suggest. In fact, most survey questions on religiosity date back to the 1960s when the academic fascination with secularization was at its highest (ibid.: 25). When asking the population about beliefs resting on religious dogma such as Paradise, Hell, and Sin, the result will often be that Swedes and Danes are not believers and that they rarely attend religious services. Through these international comparisons, Sweden and Denmark have earned the reputation as being the most secular countries in the world (Zuckerman et al. 2016: 79; see also Thurfjell 2019). However, the secular ideal is challenged by the historical intertwinement of church and state that characterizes the Nordic countries (Christoffersen 2017). Erika Willander (2020: 5) argues that the almost invisible integration of the majority church in public institutions (schools, kindergarten, hospitals, etc.) creates a commonality among Swedes that she terms the “religious mainstream.” The religious mainstream describes religious similarity within groups and over time (ibid.: 17).8 The commonality in religious expression is directly derived from shared experiences of the majority church and its moral codex in a range of situations during everyday life. In Denmark, 73.2 percent of the population are members of the Evangelical Lutheran Church (FUV 2022). Confirmation practice is fully integrated as a part of the public school system, and 72 percent say that the majority church should be present in school for confirmation preparation (Poulsen et al. 2021: 159). As mentioned, the church is not separate from the state, and the clergy perform the task of registering the demographics of the population through the parish register. The church also holds the funerary authority (Kjærsgaard 2017; Nielsen and Kühle 2011). In this sense, the intricate connections between the majority church, the state, and the population exist in a common mainstream that is just as relevant in Denmark as in Willander's Sweden. I emphasize the notion of the religious mainstream because it builds on the local religious context wherein the history of church-state integration is embedded in the material experience of individual citizens living in this type of majority Christian society. Of course, the idea of shared commonality is vulnerable to critiques of misleading representation. By focusing on the experiences uniting the majority of Danes, religious minorities fade from the analysis. Nonetheless, our survey was designed to map the religious mainstream with all the difficulties that this notion entails. One last point about the survey is that participation was voluntary. It has been pointed out that censuses, surveys, and other forms of registration risk targeting minorities by establishing a norm and subsequently also what should be recognized as deviance from that norm (Foucault 2007: 63; Hacking [1990] 2014: 166). In this sense, minorities are categorized as deviant.

Birgitte Schepelern Johansen and Riem Spielhaus (2019) have argued that populational statistics are applied as techniques for producing and managing public emotions, performing new forms of visibility and new forms of deviance that are especially aimed at Muslims in secular publics. Statistics produce the political and secular imaginary. Although I am convinced by this analysis, the placing of statistics as part and parcel of a political agenda makes it difficult to embark on any survey on religiosity—especially a survey framed by a ministry. My attentiveness toward a common mainstream would then be read as a depreciation of religious minorities. In questionnaires, respondents show their dissatisfaction by either not answering the questions or choosing the option “I don't know” or “I don't want to answer.” Respondents also have the possibility of lying. As researchers, we cannot be sure that the respondents are honest in their responses. In other words, the respondents ‘own’ their answers. The voluntary aspect means that the people who participate are often personally interested in the subject, which can skew the sample. But the voluntary aspect also means that I cannot force people to answer. Unlike census data collected through the CRS register, surveys on minority religious identity and affiliation are always voluntary in a Danish context. The political use of these surveys certainly creates new forms of visibility around ‘deviant’ religiosity as religious identity has become increasingly politicized over the last two decades (Furseth et al. 2018). Yet in a Danish context, registration of populational characteristics has taken place since the 1700s and with varying intentions. In this light, the recent survey invested in mapping the Danish religious mainstream (Trolle 2015) is a product of the national history in which mapping does not necessarily entail exposing deviancy.

Applying Stereotypical Notions of Religiosity

In the previous section, I explained how our survey aimed at mapping the religious mainstream without falling into general questions of dogmatic belief. In this section, I acknowledge that in quantitative work one often needs to analyze religious material within a predefined hypothesis that draws upon stereotypical notions of religiosity. Alain Desrosières (1998: 17) writes that descriptions of the world in surveys are valid only if the objects described are consistent: “But this consistency is not given in advance. It has been created. The aim of surveys is to analyze what makes things hold in such a way that they constitute shared representations, which can be affected by actions with a common meaning.” As a statistician, Desrosières was very much aware of the fallacies of creating a shared representation where there is none, but consistency is necessary for analyzing quantitative data. In the following, I will explain how we worked with consistency through religious ideal types in the survey.

One of the greatest misunderstandings (and dissatisfactions) with existing questionnaires is that they do not adequately reflect social reality. But the point of many questions is not to choose phrases that everybody agrees upon. The questions are chosen because they reflect the hypothesis that is being tested, or because the questions are able to show us how different social groups react to the same religious statement. In a Danish context, only 5 percent of the population disagree with statements such as “As long as you behave properly, you can believe whatever you want” (Trolle 2015).9 Although support for the notion of ‘live and let live’ shows us a commonality that might be part of the religious mainstream, the lack of disagreement also means that it is very difficult to do meaningful analysis of the statement. The strength of surveys is often to uncover the social characteristics of various religious profiles. It could also be to find new patterns between religious attitudes. The objective is to look for interesting difference that can tell us about the commonality that is also present. If the above-mentioned statement had received agreement from only 30 percent of the population, it could tell us more about the social groups behind it. If the statement was supported only by working women, what would that tell us? In the words of David Voas (2007: 145): “Quantification often provides the best or only way of testing theories about the causes and consequences of religious affiliation and involvement. It is the natural perspective to use in discussing trends, and it facilitates international comparisons.” The theory-driven hypothesis looks for difference in explaining societal trends.

When formulating questions that will show us interesting difference, it is helpful to think in ideal types. Here I follow Richard Swedberg (2018) in his rethinking of Max Weber's work. Swedberg argues that sociology should engage the ideal type in empirical research through confronting the ideal with reality (ibid.: 188). The ideal type should be viewed as a heuristic device that can show the social segmentation surrounding particular statements. Subsequently, the ideal type is great for finding interesting difference. In applying the ideal type, the researcher has to make a number of artificial assumptions about the typical individual actor: (1) that the typical actor acts in a rational way, (2) that the typical actor has complete information, (3) that the typical actor is totally aware of what he/she is doing, and (4) that the typical actor does not make any mistakes (Weber, cited in ibid.: 187). All of these assumptions are also integrated in building Desrosières's statistical coherence. Clearly, the four assumptions are not accurate traits of any one individual, but they are good to think with. For Swedberg the social action described through the ideal type should be linked to the motivation of the individual actor. He thereby underlines how meaning should guide the implementation of the ideal type. The same can be said for the sociological study of religion. If the results are not meaningful to the religious actors, then there will often be something wrong with the analysis.

In the questionnaire on religiosity and majority church relations, the ideal types were based on theories of secularization and earlier studies of the Danish population. Individual religiosity was framed by representations of ‘the fervent believer’ to assess how people differ from the ideal. An example could be the Christian, pious, and religiously practicing elderly woman who lives in the countryside. She may not exist in reality, but the characteristics attributed to her are consistent with early forms of secularization theory, which took for granted that urbanization, industrialization, and education would rationalize the working force to become less religious (Andersen and Riis 2002). The ideal types attempt to objectivize reality by creating an extreme form of measurement. Returning to the norm/deviance discourse, the questionnaire then actually needs to integrate quite a lot of religious ‘deviance’ in order to catch the religious ‘normality’ that would make up the religious mainstream. I count on my respondents to disagree with my ideal type. In this sense, the creation of a questionnaire is far from its perceived political use. Although the quantitative analysis seems to create a distance between the researcher and the religious reality of the respondents, it is still a dialectic process between the person asking the questions and the person answering. Put differently, the creation of a questionnaire relies on a form of conversation that may not apply when the survey results are portrayed in the media and acted upon by political actors.

With regard to this survey, I discussed the questions with about 70 people in a pilot study before sending the questionnaire to Statistics Denmark. But the moment that I let go of it, I had no way of knowing how the respondents felt about the content and order of questions. If the incoming data seem to confirm existing knowledge, then I can start to analyze the findings. In the process of creating a questionnaire and analyzing the outcome, the translation of social reality into numbers takes place. Numbers are not neutral in the political use of surveys, but in assigning numbers to categories for statistical analysis, numbers can have a neutral character. In questions of belief, “yes” would be 1, “maybe” 2, “no” 3, and “I don't know” 6, but you could easily reverse the numbers and still do a meaningful analysis. In quantitative analysis, numbers help clarify connections between social categories. In surveys, religious identity, affiliation, and practices are often viewed as something in need of explanation—the dependent variable. Therefore, the social context (such as children, spouses, livelihood, education, and political views) surrounding a religious individual takes center stage. Of course, you can choose to analyze religiosity as the main driver of the other variables, but part of the distance provided by the survey method is that religiosity becomes one variable among many. As I do not meet the respondents personally, I am also given space to analyze their religiosity at an aggregated level, which often yields interesting results that I could not have foreseen through a qualitative study. In other words, the survey introduces a distance that can potentially bring new questions of religious transition and translation.

Conclusion

In my work as a sociologist of religion, I often engage with religion through mixed methods, and my aim in this article has been to explain how a national survey on the Danish population's religiosity and relations to the Evangelical Lutheran Church reflects a long history of church-state integration and shifting forms of enumeration and governance. By underlining the regional history of religious registration, I argue that some of the academic critique of surveys as numbers performing the political imaginary is based on the lack of registration of religious minorities. In Denmark, the state has vast knowledge of its citizens in every aspect of life, yet religious affiliation outside the majority church is not part of any registered information. Although censuses are closely integrated with the state administration, reaching back to colonial time, the purpose of this registration has differed over time. My case of the survey conducted at the Knowledge Center for the Evangelical Lutheran Church exemplifies how analyzing survey material rests on the local and regional religious context. Our intention was to map the religious mainstream by asking questions that would tease out the multi-layered interactions of the population and the majority church. Unlike the political use of survey material, the creation of the questionnaire affords a different engagement with religion. Through the use of ideal types and theory-driven hypotheses, the questions are often meant to provoke a reaction in order to find interesting difference in the large number of respondents. In this sense, I study religion as a puzzle, a house with many rooms, where the social characteristics of the informant frame religion as a dependent variable. This creative aspect of engaging religion through designing a questionnaire is often overlooked in the later dissemination and reception of its results.

Acknowledgments

I would like to thank the Symposium editors, Andreas Bandak and Simon Stjernholm, for their encouraging comments while this article was being written. I also extend thanks to the participants in the “Engaging Religion” workshops and to the two anonymous reviewers who provided very helpful feedback.

Notes

1

In censuses, the idea of who constitutes a citizen may vary. In some of the first censuses conducted in Scandinavia in 1664 and again in 1701, only male adults participated (Solli and Thorvaldsen 2012: 110). In Australia, the first censuses did not include the Indigenous population as they were not considered citizens before 1971 (Bouma and Hughes 2014: 436).

2

Not answering the questionnaire can lead to heavy fines (Thorvaldsen 2014: 215).

3

In some instances, the strict definition also requires a form of enumeration as well as publication of the results (Thorvaldsen 2018: 2).

4

According to Gunnar Thorvaldsen (2014: 212), the missing religious registration is due to Jewish resistance in the United States. In 1957, Catholic organizations almost succeeded in including a question of religion, but the request was turned down in favor of freedom of religion (ibid.).

5

The CRS registers all individuals born alive of a mother already in the CRS, individuals who have their birth or baptism noted in the electronic church register, and individuals who legally reside in Denmark for more than three months (Schmidt et al. 2014: 544).

6

The registration was sometimes divided according to geography. In 1769, magistrates carried out the census in the towns while vicars covered the rural areas (Johansen 2004: 24).

7

The EVS ran its first questionnaire in 1981 with repetitions in 1990, 1999, 2008, and 2017.

8

See Day (2013: 47–73) for a problematization of the claim for a religious mainstream.

9

“Bare man opfører sig ordentligt, kan man tro og mene, hvad man vil.” In Danish, the statement represents an everyday phrase that is often used to convey a position of equality of opinion, and that the speaker is not that interested in arguing about differing opinions.

References

  • Andersen, Peter A., and Ole Riis. 2002. “Religionen bliver privat” [Religion goes private]. In Danskernes værdier 1981–1999 [The values of the Danes 1981–1999], ed. Peter Gundelach. Copenhagen: Hans Reitzels Forlag.

    • Search Google Scholar
    • Export Citation
  • Arts, Wil, Jacques Hagenaars, and Loek Halman. 2003. “The Cultural Diversity of European Unity: An Introduction.” In The Cultural Diversity of European Unity: Findings, Explanations and Reflections from the European Values Study, ed. Wil Arts, Jacques Hagenaars, and Loek Halman, 112. Leiden: Brill.

    • Search Google Scholar
    • Export Citation
  • Asad, Talal. 2018. Secular Translations: Nation-State, Modern Self, and Calculative Reason. New York: Columbia University Press.

  • Bauer, Susanne. 2014. “From Administrative Infrastructure to Biomedical Resource: Danish Population Registries, the ‘Scandinavian Laboratory,’ and the ‘Epidemiologist's Dream.’Science in Context 27 (2): 187213.

    • Search Google Scholar
    • Export Citation
  • Beckford, James A. 2003. Social Theory and Religion. Cambridge: Cambridge University Press.

  • Bouma, Gary D., and Philip J. Hughes. 2014. “Using Census Data in the Management of Religious Diversity: An Australian Case Study.” Religion 44 (3): 434452.

    • Search Google Scholar
    • Export Citation
  • Brierley, Peter. 2010. “Measuring Religious Behaviour.” In Social Measurement through Social Surveys: An Applied Approach, ed. Martin Bulmer, Julie Gibbs, and Laura Hyman, 6988. Farnham: Ashgate.

    • Search Google Scholar
    • Export Citation
  • Bulmer, Martin. 2001. “History of Social Survey.” In International Encyclopedia of the Social and Behavioral Sciences, ed. Neil J. Smelser and Paul B. Baltes, 1446914473. Oxford: Elsevier.

    • Search Google Scholar
    • Export Citation
  • Christopher, A. J. 2002. “‘To Define the Indefinable’: Population Classification and the Census in South Africa.” Area 34 (4): 401408.

    • Search Google Scholar
    • Export Citation
  • Chryssides, George D., and Ron Geaves. 2007. The Study of Religion: An Introduction to Key Ideas and Methods. London: Bloomsbury Academic.

    • Search Google Scholar
    • Export Citation
  • Cohn, Bernard S. 1987. An Anthropologist among Historians and Other Essays. Delhi: Oxford University Press.

  • Day, Abby. 2013. Believing in Belonging: Belief and Social Identity in the Modern World. Oxford: Oxford University Press.

  • Day, Abby, and Lois Lee. 2014. “Making Sense of Surveys and Censuses: Issues in Religious Self-Identification.” Religion 44 (3): 345356.

    • Search Google Scholar
    • Export Citation
  • Desrosières, Alain. 1998. The Politics of Large Numbers: A History of Statistical Reasoning. Trans. Camille Naish. Cambridge, MA: Harvard University Press.

    • Search Google Scholar
    • Export Citation
  • Durkheim, Emile. (1897) 1979. Suicide: A Study in Sociology. Trans. John A. Spaulding and George Simpson. New York: Free Press.

  • Foucault, Michel. 2007. Security, Territory, Population: Lectures at the Collège de France, 1977–1978. Trans. Graham Burchell. New York: Palgrave Macmillan.

    • Search Google Scholar
    • Export Citation
  • Furseth, Inger, Lars Ahlin, Kimmo Ketola, Annette Leis-Peters, and Bjarni Randver Sigurvinsson. 2018. “Changing Religious Landscapes in the Nordic Countries.” In Religious Complexity in the Public Sphere: Comparing Nordic Countries, ed. Inger Furseth, 3180. Cham: Springer International Publishing.

    • Search Google Scholar
    • Export Citation
  • FUV (Folkekirkens Uddannelses- og Videnscenter). 2022. “Folkekirkens medlemstal” [Membership in the Evangelical Lutheran Church]. https://www.fkuv.dk/videnscenteret/kirkestatistik/medlemstal (accessed 24 June 2022).

    • Search Google Scholar
    • Export Citation
  • Hacking, Ian. (1990) 2014. The Taming of Chance. Cambridge: Cambridge University Press.

  • Hansen, Hans Oluf. 2007. “Spredning af religion: Målingsproblemer og nogle demografiske faktorer” [The spread of religion: Measurement problems and some demographic factors]. In Tørre tal om troen: Religionsdemografi i det 21. Århundrede [Dry numbers on faith: Demography of religion in the 21st century], ed. Margit Warburg and Brian Jacobsen, 3851. Aarhus: Forlaget Univers.

    • Search Google Scholar
    • Export Citation
  • House, James S., Eleanor Singer, Robert L. Kahn, Howard Schuman, and F. Thomas Juster. 2004. “The Development and Contribution of Survey Research as a Scientific Instrument and Social Institution.” In A Telescope on Society: Survey Research and Social Science at the University of Michigan and Beyond, ed. James S. House, F. Thomas Juster, Robert L. Kahn, Howard Schuman, and Eleanor Singer, 120. Ann Arbor: University of Michigan Press.

    • Search Google Scholar
    • Export Citation
  • Howard, David, and Peter E. Hopkins. 2005. “Editorial: Race, Religion and the Census.” Population, Space and Place 11 (2): 6974.

  • Inglehart, Ronald, and Christian Welzel. 2005. Modernization, Cultural Change, and Democracy: The Human Development Sequence. Cambridge: Cambridge University Press.

    • Search Google Scholar
    • Export Citation
  • Johansen, Birgitte Schepelern, and Riem Spielhaus. 2019. “Quantitative Knowledge Production on Muslims in Europe as a Practice of ‘Secular Suspicion.’” In Secular Bodies, Affects and Emotions: European Configurations, ed. Monique Scheer, Nadia Fadil, and Birgitte Schepelern Johansen, 171184. London: Bloomsbury Academic.

    • Search Google Scholar
    • Export Citation
  • Johansen, Hans Christian. 2004. “Early Danish Census Taking.” The History of the Family 9 (1): 2331.

  • Kjærsgaard, Anne. 2017. Funerary Culture and the Limits of Secularization in Denmark. Münster: LIT Verlag.

  • Nielsen, Marie Vejrup, and Lene Kühle. 2011. “Religion and State in Denmark: Exception among Exceptions?Nordic Journal of Religion and Society 24 (2): 173188.

    • Search Google Scholar
    • Export Citation
  • Poulsen, Jais, Astrid Krabbe Trolle, Maiken Friischman Larsen, and Bjarke Schønwandt Mortensen. 2021. Religiøsitet og forholdet til folkekirken 2020. Bog 1—Kvantitative studier [Religiosity and the relation to the Evangelical Lutheran Church 2020. Book 1—Quantitative studies]. Copenhagen: Folkekirkens Uddannelses- og Videnscenter.

    • Search Google Scholar
    • Export Citation
  • Riis, Ole, and Peter Gundelach. 1992. Danskernes værdier [The values of the Danes]. Copenhagen: Forlaget Sociologi.

  • Riis, Ole Preben. 2011. “Methodology in the Sociology of Religion.” In The Oxford Handbook of the Sociology of Religion, ed. Peter B. Clarke, 229244. New York: Oxford University Press.

    • Search Google Scholar
    • Export Citation
  • Rod, Jakob. 1961. Folkereligion og Kirke [Folk religion and the church]. Copenhagen: Gads Forlag.

  • Salomonsen, Per. 1966. Religionssociologi: En moderne videnskab om menneskets religiøse adfærd [Sociology of religion: A modern science about man's religious behavior]. Copenhagen: Munksgaards Forlag.

    • Search Google Scholar
    • Export Citation
  • Salomonsen, Per. 1971. Religion i dag: Et sociologisk metodestudium. Bind 1 [Religion today: A sociological study of method. Vol. 1]. Copenhagen: G.E.C. Gads Forlag.

    • Search Google Scholar
    • Export Citation
  • Schmidt, Morten, Lars Pedersen, and Henrik Toft Sørensen. 2014. “The Danish Civil Registration System as a Tool in Epidemiology.” European Journal of Epidemiology 29 (8): 541549.

    • Search Google Scholar
    • Export Citation
  • Sherif, Jamil. 2011. “A Census Chronicle: Reflections on the Campaign for a Religion Question in the 2001 Census for England and Wales.” Journal of Beliefs and Values 32 (1): 118.

    • Search Google Scholar
    • Export Citation
  • Solli, Arne, and Gunnar Thorvaldsen. 2012. “Norway: From Colonial to Computerized Censuses.” Revista de Demografía História 30 (1): 107136.

    • Search Google Scholar
    • Export Citation
  • Spickard, James V. 2017. Research Basics: Design to Data Analysis in Six Steps. Thousand Oaks, CA: Sage.

  • Swedberg, Richard. 2018. “How to Use Max Weber's Ideal Type in Sociological Analysis.” Journal of Classical Sociology 18 (3): 181196.

    • Search Google Scholar
    • Export Citation
  • Thorvaldsen, Gunnar. 2014. “Religion in the Census.” Social Science History 38 (1–2): 203220.

  • Thorvaldsen, Gunnar. 2018. Censuses and Census Takers: A Global History. London: Routledge.

  • Thurfjell, David. 2019. Det gudlösa folket: De postkristna svenskarna och religionen [The godless people: The post-Christian Swedes and religion]. Stockholm: Norstedts.

    • Search Google Scholar
    • Export Citation
  • Trolle, Astrid Krabbe. 2015. “Blandt principielle sekularister og konkrete konservatorer: Danskernes holdning til religion og folkekirke i fire YouGov-undersøgelser” [Between principled secularists and concrete conservators: Danes’ attitudes toward religion and the folk church in four YouGov surveys]. Paper published by the Faculty of Theology, University of Copenhagen. https://teol.ku.dk/cfk/dokumenter/Trolle_Fire_YouGovs_CFK_Okt.pdf (accessed 15 June 2021).

    • Search Google Scholar
    • Export Citation
  • Voas, David. 2007. “Surveys of Behaviour, Beliefs and Affiliation: Micro-Quantitative.” In The SAGE Handbook of the Sociology of Religion, ed. James A. Beckford and N. J. Demerath III, 144166. London: Sage.

    • Search Google Scholar
    • Export Citation
  • Warburg, Margit. 1979. Religionssociologi: Grundbog [Sociology of religion: The basics]. Copenhagen: Hans Reitzel.

  • Willander, Erika. 2020. Unity, Division and the Religious Mainstream in Sweden. Cham: Palgrave Macmillan.

  • Zuckerman, Phil, Luke W. Galen, and Frank L. Pasquale. 2016. The Nonreligious: Understanding Secular People and Societies. New York: Oxford University Press.

    • Search Google Scholar
    • Export Citation

Contributor Notes

ASTRID KRABBE TROLLE is Assistant Professor in the Department of Cross-Cultural and Regional Studies at the University of Copenhagen. She is currently part of the research project “The Catholic Ethic and the Spirit of Global Modernities.” Trolle received her PhD from the University of Copenhagen in 2019 with an ethnography on Filipino Catholics in Denmark. E-mail: a.trolle@hum.ku.dk

  • Collapse
  • Expand

Religion and Society

Advances in Research

  • Andersen, Peter A., and Ole Riis. 2002. “Religionen bliver privat” [Religion goes private]. In Danskernes værdier 1981–1999 [The values of the Danes 1981–1999], ed. Peter Gundelach. Copenhagen: Hans Reitzels Forlag.

    • Search Google Scholar
    • Export Citation
  • Arts, Wil, Jacques Hagenaars, and Loek Halman. 2003. “The Cultural Diversity of European Unity: An Introduction.” In The Cultural Diversity of European Unity: Findings, Explanations and Reflections from the European Values Study, ed. Wil Arts, Jacques Hagenaars, and Loek Halman, 112. Leiden: Brill.

    • Search Google Scholar
    • Export Citation
  • Asad, Talal. 2018. Secular Translations: Nation-State, Modern Self, and Calculative Reason. New York: Columbia University Press.

  • Bauer, Susanne. 2014. “From Administrative Infrastructure to Biomedical Resource: Danish Population Registries, the ‘Scandinavian Laboratory,’ and the ‘Epidemiologist's Dream.’Science in Context 27 (2): 187213.

    • Search Google Scholar
    • Export Citation
  • Beckford, James A. 2003. Social Theory and Religion. Cambridge: Cambridge University Press.

  • Bouma, Gary D., and Philip J. Hughes. 2014. “Using Census Data in the Management of Religious Diversity: An Australian Case Study.” Religion 44 (3): 434452.

    • Search Google Scholar
    • Export Citation
  • Brierley, Peter. 2010. “Measuring Religious Behaviour.” In Social Measurement through Social Surveys: An Applied Approach, ed. Martin Bulmer, Julie Gibbs, and Laura Hyman, 6988. Farnham: Ashgate.

    • Search Google Scholar
    • Export Citation
  • Bulmer, Martin. 2001. “History of Social Survey.” In International Encyclopedia of the Social and Behavioral Sciences, ed. Neil J. Smelser and Paul B. Baltes, 1446914473. Oxford: Elsevier.

    • Search Google Scholar
    • Export Citation
  • Christopher, A. J. 2002. “‘To Define the Indefinable’: Population Classification and the Census in South Africa.” Area 34 (4): 401408.

    • Search Google Scholar
    • Export Citation
  • Chryssides, George D., and Ron Geaves. 2007. The Study of Religion: An Introduction to Key Ideas and Methods. London: Bloomsbury Academic.

    • Search Google Scholar
    • Export Citation
  • Cohn, Bernard S. 1987. An Anthropologist among Historians and Other Essays. Delhi: Oxford University Press.

  • Day, Abby. 2013. Believing in Belonging: Belief and Social Identity in the Modern World. Oxford: Oxford University Press.

  • Day, Abby, and Lois Lee. 2014. “Making Sense of Surveys and Censuses: Issues in Religious Self-Identification.” Religion 44 (3): 345356.

    • Search Google Scholar
    • Export Citation
  • Desrosières, Alain. 1998. The Politics of Large Numbers: A History of Statistical Reasoning. Trans. Camille Naish. Cambridge, MA: Harvard University Press.

    • Search Google Scholar
    • Export Citation
  • Durkheim, Emile. (1897) 1979. Suicide: A Study in Sociology. Trans. John A. Spaulding and George Simpson. New York: Free Press.

  • Foucault, Michel. 2007. Security, Territory, Population: Lectures at the Collège de France, 1977–1978. Trans. Graham Burchell. New York: Palgrave Macmillan.

    • Search Google Scholar
    • Export Citation
  • Furseth, Inger, Lars Ahlin, Kimmo Ketola, Annette Leis-Peters, and Bjarni Randver Sigurvinsson. 2018. “Changing Religious Landscapes in the Nordic Countries.” In Religious Complexity in the Public Sphere: Comparing Nordic Countries, ed. Inger Furseth, 3180. Cham: Springer International Publishing.

    • Search Google Scholar
    • Export Citation
  • FUV (Folkekirkens Uddannelses- og Videnscenter). 2022. “Folkekirkens medlemstal” [Membership in the Evangelical Lutheran Church]. https://www.fkuv.dk/videnscenteret/kirkestatistik/medlemstal (accessed 24 June 2022).

    • Search Google Scholar
    • Export Citation
  • Hacking, Ian. (1990) 2014. The Taming of Chance. Cambridge: Cambridge University Press.

  • Hansen, Hans Oluf. 2007. “Spredning af religion: Målingsproblemer og nogle demografiske faktorer” [The spread of religion: Measurement problems and some demographic factors]. In Tørre tal om troen: Religionsdemografi i det 21. Århundrede [Dry numbers on faith: Demography of religion in the 21st century], ed. Margit Warburg and Brian Jacobsen, 3851. Aarhus: Forlaget Univers.

    • Search Google Scholar
    • Export Citation
  • House, James S., Eleanor Singer, Robert L. Kahn, Howard Schuman, and F. Thomas Juster. 2004. “The Development and Contribution of Survey Research as a Scientific Instrument and Social Institution.” In A Telescope on Society: Survey Research and Social Science at the University of Michigan and Beyond, ed. James S. House, F. Thomas Juster, Robert L. Kahn, Howard Schuman, and Eleanor Singer, 120. Ann Arbor: University of Michigan Press.

    • Search Google Scholar
    • Export Citation
  • Howard, David, and Peter E. Hopkins. 2005. “Editorial: Race, Religion and the Census.” Population, Space and Place 11 (2): 6974.

  • Inglehart, Ronald, and Christian Welzel. 2005. Modernization, Cultural Change, and Democracy: The Human Development Sequence. Cambridge: Cambridge University Press.

    • Search Google Scholar
    • Export Citation
  • Johansen, Birgitte Schepelern, and Riem Spielhaus. 2019. “Quantitative Knowledge Production on Muslims in Europe as a Practice of ‘Secular Suspicion.’” In Secular Bodies, Affects and Emotions: European Configurations, ed. Monique Scheer, Nadia Fadil, and Birgitte Schepelern Johansen, 171184. London: Bloomsbury Academic.

    • Search Google Scholar
    • Export Citation
  • Johansen, Hans Christian. 2004. “Early Danish Census Taking.” The History of the Family 9 (1): 2331.

  • Kjærsgaard, Anne. 2017. Funerary Culture and the Limits of Secularization in Denmark. Münster: LIT Verlag.

  • Nielsen, Marie Vejrup, and Lene Kühle. 2011. “Religion and State in Denmark: Exception among Exceptions?Nordic Journal of Religion and Society 24 (2): 173188.

    • Search Google Scholar
    • Export Citation
  • Poulsen, Jais, Astrid Krabbe Trolle, Maiken Friischman Larsen, and Bjarke Schønwandt Mortensen. 2021. Religiøsitet og forholdet til folkekirken 2020. Bog 1—Kvantitative studier [Religiosity and the relation to the Evangelical Lutheran Church 2020. Book 1—Quantitative studies]. Copenhagen: Folkekirkens Uddannelses- og Videnscenter.

    • Search Google Scholar
    • Export Citation
  • Riis, Ole, and Peter Gundelach. 1992. Danskernes værdier [The values of the Danes]. Copenhagen: Forlaget Sociologi.

  • Riis, Ole Preben. 2011. “Methodology in the Sociology of Religion.” In The Oxford Handbook of the Sociology of Religion, ed. Peter B. Clarke, 229244. New York: Oxford University Press.

    • Search Google Scholar
    • Export Citation
  • Rod, Jakob. 1961. Folkereligion og Kirke [Folk religion and the church]. Copenhagen: Gads Forlag.

  • Salomonsen, Per. 1966. Religionssociologi: En moderne videnskab om menneskets religiøse adfærd [Sociology of religion: A modern science about man's religious behavior]. Copenhagen: Munksgaards Forlag.

    • Search Google Scholar
    • Export Citation
  • Salomonsen, Per. 1971. Religion i dag: Et sociologisk metodestudium. Bind 1 [Religion today: A sociological study of method. Vol. 1]. Copenhagen: G.E.C. Gads Forlag.

    • Search Google Scholar
    • Export Citation
  • Schmidt, Morten, Lars Pedersen, and Henrik Toft Sørensen. 2014. “The Danish Civil Registration System as a Tool in Epidemiology.” European Journal of Epidemiology 29 (8): 541549.

    • Search Google Scholar
    • Export Citation
  • Sherif, Jamil. 2011. “A Census Chronicle: Reflections on the Campaign for a Religion Question in the 2001 Census for England and Wales.” Journal of Beliefs and Values 32 (1): 118.

    • Search Google Scholar
    • Export Citation
  • Solli, Arne, and Gunnar Thorvaldsen. 2012. “Norway: From Colonial to Computerized Censuses.” Revista de Demografía História 30 (1): 107136.

    • Search Google Scholar
    • Export Citation
  • Spickard, James V. 2017. Research Basics: Design to Data Analysis in Six Steps. Thousand Oaks, CA: Sage.

  • Swedberg, Richard. 2018. “How to Use Max Weber's Ideal Type in Sociological Analysis.” Journal of Classical Sociology 18 (3): 181196.

    • Search Google Scholar
    • Export Citation
  • Thorvaldsen, Gunnar. 2014. “Religion in the Census.” Social Science History 38 (1–2): 203220.

  • Thorvaldsen, Gunnar. 2018. Censuses and Census Takers: A Global History. London: Routledge.

  • Thurfjell, David. 2019. Det gudlösa folket: De postkristna svenskarna och religionen [The godless people: The post-Christian Swedes and religion]. Stockholm: Norstedts.

    • Search Google Scholar
    • Export Citation
  • Trolle, Astrid Krabbe. 2015. “Blandt principielle sekularister og konkrete konservatorer: Danskernes holdning til religion og folkekirke i fire YouGov-undersøgelser” [Between principled secularists and concrete conservators: Danes’ attitudes toward religion and the folk church in four YouGov surveys]. Paper published by the Faculty of Theology, University of Copenhagen. https://teol.ku.dk/cfk/dokumenter/Trolle_Fire_YouGovs_CFK_Okt.pdf (accessed 15 June 2021).

    • Search Google Scholar
    • Export Citation
  • Voas, David. 2007. “Surveys of Behaviour, Beliefs and Affiliation: Micro-Quantitative.” In The SAGE Handbook of the Sociology of Religion, ed. James A. Beckford and N. J. Demerath III, 144166. London: Sage.

    • Search Google Scholar
    • Export Citation
  • Warburg, Margit. 1979. Religionssociologi: Grundbog [Sociology of religion: The basics]. Copenhagen: Hans Reitzel.

  • Willander, Erika. 2020. Unity, Division and the Religious Mainstream in Sweden. Cham: Palgrave Macmillan.

  • Zuckerman, Phil, Luke W. Galen, and Frank L. Pasquale. 2016. The Nonreligious: Understanding Secular People and Societies. New York: Oxford University Press.

    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1074 573 100
PDF Downloads 484 108 6