1 Introduction

Throughout Europe the status of religious education at state schools is under discussion (van der Kooij et al. 2017; Jackson 2019; Jackson and Everington 2017; Kuusisto and Gearon 2019). In Germany, denominational religious education still is the default option in twelve out of 16 federal states in Germany (Rothgangel and Ziebertz 2013). In 1947, religious education was confirmed as a regular scholastic subject in the German constitution (Meckel 2011). It has to meet both the educational standards of the school system and the doctrine of the relevant denomination. In consequence, it is up to the state to provide the material basis of religious education (such as providing class rooms and text books, hiring the teachers, and defining basic educational goals), while the religious communities establish the curriculum and determine the teachers’ qualifications. Over the decades, the denominational character of this subject changed (Englert 2015; Dressler 2015). At the beginning, religious education was about familiarizing the students with their relevant church. Today the students are dealing with both the relevant religious community’s doctrine and religious plurality in German society. It is the primary goal of denominational religious education that the students understand how religions perceive reality and that they develop their personal attitude towards this perspective.

In recent times, however, denominational religious education at state schools is under pressure. First, in a secularizing country denominational religious education loses societal support (Gärtner 2015, pp. 163–169). The number of religious people in Germany is declining. Many people regard religion as a private matter and it is hardly tolerated as an argument in civil society. According to the German constitution, the state must be neutral towards differing ideologies and worldviews. All these issues raise the question of why religion should be a subject at school. Since it is a subject at school in most federal states, it must be asked whether it should deal with religions in an objective perspective instead of prioritizing a particular religious tradition.

Second, the pedagogical justification of religious education at state schools loses its plausibility (Gruehn 2020). From an educational point of view, RE should help the students to understand the Christian background of recent society, to develop a personal identity, and to feel responsible for civil society. For an education like this religious narratives, symbols, norms and practices offer an important resource. Often this option is neglected because religions are seen as exclusivist accounts to reality and, therefore, are unable to deal fairly with societal plurality. In this respect, moral education is seen to be more adequate to reach the educational goals described previously. If religion is accepted as moral resource, then the respective subject should deal with all religious traditions on even grounds. An integrative approach to religious education is suggested to replace denominational religious education (Frank 2014).

Third, demography causes problems in organizing homogeneous denominational groups of learners in many German regions (Platzbecker 2019, pp. 46–47). The two Christian churches are losing members every day at a significant rate and prognostics indicate that in 2060 only one third of the German population will belong to a Christian church (Gutmann and Peters 2020). Moreover, Catholics and Protestants are not distributed equally over Germany. Some regions show a Catholic majority, while others show a Protestant one, whereas in Eastern parts both denominations represent minorities. Then, Muslims, Jews, Buddhists, etc. are religious minorities, too, with the right to offer religious education according to its doctrines. On an organizational level, this religious heterogeneity makes it tricky to bring together students from the same denomination or religion and to schedule these subjects, not to mention the many teachers needed to cover this heterogeneity of beliefs. Often these problems are solved by practicing integrative RE without permission (Boschki and Schlenker 2005, p. 388; Hütte et al. 2003). In these cases, the entire class attends religious education regardless of the religious affiliation. To cover up this illegal practice these classes are usually labelled “Catholic RE” or “Protestant RE” and the teachers of these classes are usually trained for denominational RE.

Apart from sophisticated variants within the basic options to teach religious education, there are four such basic options of future religious education (RE) at state schools under discussion in Germany:

  1. (1)

    to go on with denominational RE in its current form (Sajak 2004; Verhülsdonk 2005; Englert 2015). This type of RE is requested by the German constitution, has brought about a rich portfolio of practical experience and theoretical reflection, and has proved to be able to adjust to changing societal environments. Accordingly, it does not seem to be clever to change a running system, in other words: to go on with the tried and tested model of religious education.

  2. (2)

    to slightly modify the form of denominational religious education into a cooperative model (Platzbecker 2019; Gennerich and Mokrosch 2016; Lindner et al. 2017). Religious education still will be taught according to the religious communities’ doctrines like it is typical for the denominational model. The group of learners, however, includes students from two different denominations. The curriculum contains both doctrines and elaborates the commonalities and respects the differences at the same time. In dialogue the students should learn what their personal belief is about and to respect the beliefs of others. Up to now, such cooperative RE is realized by Catholics and Protestants. It is regarded as the form of denominational RE that is more adaptive to the societal environment than the original form with denominationally homogeneous groups of learners.

  3. (3)

    to replace denominational RE by integrative RE. Such RE includes all students in the classroom, regardless of their denominational background. It addresses religious as well as non-religious students, because in a modern pluralistic society everybody has to learn how to deal with religious people and differing beliefs. There are two modes of integrative RE under discussion. Some scholars promote an objective information about the various religious traditions (Alberts 2007; Loobuyck and Franken 2011). An identification of the students with one of these traditions is not part of the model. Since this is the nature of all religiousness, however, others propose a form which allows students to identify with a particular religious tradition (Riegel 2018; Bauer 2019). Despite this difference, integrative RE treats all religions and denominations equally, no particular religious tradition is favoured in regard to the others. The students should learn about the basic doctrine of every religion and to respect religious plurality in society.

  4. (4)

    to replace any religious education by moral education (Wetz 2008). This last option holds that religious education at state schools violates the state’s neutrality in terms of differing worldviews. In consequence, any form of RE has to be abolished from the curriculum. From an educational perspective, however, some sort of moral education is needed to help the students to become competent citizens. The introduction of a new subject such as Moral Education could address ethics and norms from a philosophical perspective. Religious stories might be addressed, too, if they foster moral development. The subject itself, however, is indifferent with regard to any particular religious tradition. Moral Education at state schools has to be secular in its basic account on norms and values.

All in all, the situation and status of religious education at German state schools prove to be complex. To overcome the problems with denominational RE, in three federal states—Baden-Württemberg, Lower Saxony and North Rhine-Westphalia—the schools’ principals may apply for cooperative RE (Sajak 2017). In Bremen and Hamburg, integrative RE is the official form of religious education, and with some particularities in Brandenburg, too. Moral education is taught only in Berlin. Given the controversial public discussion, a prediction of the subject’s future seems to be difficult to make though religious education is still anchored in the German Basic Law. However, going into particularities, the school principals’ attitudes towards religious education may be a suitable indicator in which direction the public debate on religious education in state schools might develop. First of all, a principal is affected by the organizational problems caused by this subject. It is their task to make up the school’s timetable and it is truly difficult to integrate a subject with several groups of learners into one schedule. Besides these organizational problems, the principals are regular citizens and take part in public discourse on religious education. For example, some of them may share the constitutional or educational doubts in regard of religious education as previously described. In private matters this person may hold a secular worldview and, therefore, prefer that state schools abolish RE to comply with the state’s neutrality. Or the principal may have an educational ideal that regards denominational RE as being not able to deal with religious plurality appropriately. Therefore, our research questions read as following:

  1. (1)

    Which form of religious education do principals of state schools prefer for teaching religion?

  2. (2)

    How do organizational requirements and educational attitudes predict these preferences?

2 Method

We answer these research questions with data from a bigger project that evaluates the cooperative RE in North-Rhine Westphalia (NRW). At the beginning of this project, the principals of participating schools have been asked in a questionnaire, why they applied for this form of RE and how they regard alternative forms to teach religion at state schools. The population of the survey, therefore, are the principals of those schools within NRW who took the chance to offer an alternative to denominational RE. This might include a biased view of the principals’ preference regarding the form of religious education. However, since NRW is one of the federal states within Germany with a maximum in religious plurality (Hero et al. 2008), these principals form a suitable example of facing the problems caused by denominational RE.

The principals’ preference concerning the form of religious education has been assessed by an instrument including the four basic options described in the introduction. Please note that these forms are discussed in Germany under other labels. Therefore, we offered the principals the following items:

  • denominational RE: “konfessioneller Religionsunterricht im herkömmlichen Format” (denominational religious education in its common format),

  • cooperative RE: “konfessionell-kooperativer Religionsunterricht mit Lehrerwechsel” (confessional-cooperative religious education with change of teachers),

  • integrative RE: “Religionsunterricht im Klassenverband” (religious education in the classroom), and

  • Moral Education: “Kein Religionsunterricht, stattdessen ein allgemein verbindlicher Ethik- bzw. Werteunterricht” (compulsory ethics or value education instead of religious education).

These four items cover only the basic options of how to teach this subject in future. For each of these options various variants can be identified in the relevant academic discussion. In the case of cooperative RE, for example, some suggest that a Protestant teacher and a Catholic one teach a class in turns. Alternatively, however, it is also being considered whether the cooperative RE could be better taught in team teaching. In the case of denominational RE, there are voices that want to stick to conventional teaching. But it is also suggested that this subject should work more closely with subjects such as ethics or philosophy within a so-called “Fächergruppe” (“subject group model”). Some studies that asked teachers about their preferred models of religious education have presented a list of such variants (e.g., Feige and Tzcheetzsch 2005), others have not (Rothgangel et al. 2017). Because we assume that principals are not experts in the field of religious education, we have chosen the less sophisticated option with four clear alternatives. The principals were asked to decide for each item how desirable it is. They could answer on a Likert scale from “1 = not desirable at all” to “5 = desirable at all”.

To answer the second research question some variables on organizational issues have been used. First, we checked whether the school offers primary or secondary education. Since primary schools normally are not as big as secondary schools, the timetable in primary schools might be easier to construct. Second, we asked for the numbers of classes in which cooperative RE is taught. Again, the idea is that size might make the organization of denominational RE more complex. Third, we assessed the composition of the classes in RE. In detail, the principals have been asked whether the groups of learners in RE include students from one class only or whether they had to build new groups with students from various classes. All these variables were single item ones.

Despite organizational variables we further used those that give evidence of the principals’ educational attitude. First, we asked about the relevance of religious education, which this subject has at the school, with a single item. The principals could asses this relevance on a 5-point Likert scale (“1 = no relevance at all”, “5 = very relevant”). Second, an instrument assessed the educational goals religious education at state schools should meet. This instrument comprises 13 items and the principals have been asked to evaluate how appropriate they consider the single items for religious education at state schools. An exploratory factor analysis (principal component, Varimax rotation, Kaiser–Guttmann-Criterion; KMO = .78; Kaiser and Rice 1974) brought about two factors: church-related goals and life-related goals. While the first factor wants religious education to familiarize the students with their denomination, the second factor takes religious education as a subject to offer orientation in life. Third, another instrument assessed how principals think about teachers taking position in terms of religion and worldview. This instrument includes eight items and has been asked on a 5-point Likert scale, too. Here the exploratory factor analysis (principal component, Varimax rotation, Kaiser–Guttmann-Criterion; KMO = .66) resulted in three factors. The first factor represents a denominational perspective favouring that teachers should stand up for their own belief in religious education. The second factor is about a cooperative perspective and claims that teachers should encourage the dialogue between the denominations. The third factor that contains just one item offers a neutral perspective and wants the teachers to not take position in religious education at all. Finally, a fourth instrument assessed the motives why the school did apply for cooperative RE, including eleven motives. The exploratory factor analysis (principal component, Varimax rotation, Kaiser–Guttmann-Criterion; KMO = .75) on these motives ends up in two factors. The first factor comprises organizational motives, the second one educational ones.

In terms of statistics, we applied both frequencies and descriptive statistics to get an overview on the principals’ preference regarding the four forms of RE. The relationship between these references have been analysed by correlation analysis (Pearson). Both statistical routines answer the first research question. The second research question is answered by regression analysis for each of the four forms of RE. The routines have been run with SPSS 26. It is recognized that these methods have to be considered carefully in case of ordinal variable.

The population of this paper are 305 schools all over NRW offering cooperative RE. 171 (56%) are primary schools, 134 (44%) are secondary schools. We invited all of them to participate in the evaluation by addressing the principals via a letter which explained the nature of the evaluation, offered the link to the principals’ questionnaire, and guaranteed anonymity and data security according to the European General Data Protection Regulation (GDPR). 228 schools filled in the questionnaire, which represents a response rate of 75%. Since bigger schools are headed by a board rather than by a single person, there is a chance that some of the questionnaires have been filled in by the person of the board who manages the school’s timetable. Concerning our research question this is not a problem because it is these people who experience the problems with denominational RE first hand. The distribution of primary schools and secondary schools in the sample matches exactly the proportions in the population: 56% of the responding schools are primary ones, 44% are secondary ones.

3 Results

The principals of our sample prefer integrative RE and cooperative RE (see Table 1). The mean values of both forms are clearly in the positive sector of the response scale. Looking at the frequencies it is interesting to note that the absolute preference for integrative RE (48.8%) is much higher than that for cooperative RE (29.4%). Denominational RE (M = 2.27) does not get any noteworthy support. The principals’ preference for Moral Education (M = 2.57) is slightly better but still clearly in the sceptical sector of the response scale.

Table 1 Frequencies (in percent) and descriptive statistics on models of teaching religious education at state schools
Full size table

The correlation between denominational RE and cooperative RE is positive though not strong (r = .14; see Table 2). It is the only positive relation between the forms of RE. Both forms, denominational RE and cooperative RE, show a negative correlation with Moral Education. Out of these relations only the one between denominational RE and Moral education (r = − .24) explains more than 5% of the variance. Finally, there is a slightly negative correlation between denominational RE and integrative RE (r = − .19).

Table 2 Correlations (Pearson) of models of teaching religious education at state schools
Full size table

Regression analysis on the four forms of RE gives evidence on the aspects which predict the principals’ preferences (see Table 3). To check for such indicators we inserted items regarding the organizational aspect (i.e. type of school, size of school) and educational aspects (i.e. educational goals, taking position in RE) into this analysis as described previously. First, denominational RE is predicted by the number of classes (β = − .23), by a denominational perspective on RE’s goals (β = .14), and by applying for cooperative RE for educational motives (β = − .27). In detail, the principals tend to reject denominational RE less if they think that RE has to serve the interests of the religious communities, if educational motives did not count for applying for denominational RE, and if they head a small school. Integrative RE is predicted by the composition of the classes (β = − .41) and by a cooperative perspective with regard to taking position in RE (β = .24). The principals’ support for integrative RE is bigger if they are able to offer RE within the given system of classes and if they prefer that teachers in RE should promote a comparative perspective on the various religions. Finally, Moral Education is predicted by the type of school (β = − .18), the relevance of RE (β = − .22), church related goals for RE (β = − .39) and a neutral perspective on taking position in RE (β = .25). This means that principals of secondary schools reject Moral Education less than those of primary schools. Moreover, the acknowledgement of this form is even greater if the principals do not cherish RE at their school and if they do not agree with church related goals for this subject. Plus they show minor impediments towards Moral Education if they prefer teachers in RE that are rather objective concerning all religions than taking position in this issue. Note that the explained variance in the case of cooperative RE is too low to indicate any relevant prediction.

Table 3 Linear regression on models of teaching religious education at state schools
Full size table

4 Discussion

This paper analyses the principals’ preferences of how to teach religion at state schools within a sample of schools that applied for organizing RE according to the cooperative model in the year 2019/20. In this regard it could have been expected that the participants favour cooperative RE. The principals in our sample answer accordingly. The preference for cooperative RE, however, is slightly topped by an even bigger preference for integrative RE. Since cooperative RE is a transformation of denominational RE, which is the current established model of RE in the federal state of North-Rhine Westphalia, Germany, this result is not insignificant. At least some of the principals in our sample seem to regard it as more attractive to change the given denominational format of religious education rather by including students of all religious and non-religious backgrounds in this group of learners than by including only Catholic and Protestant ones. From an organizational point of view this is obvious because in the cooperative model the principal has to organize additional groups of learners for the non-Christian and the non-religious students. From an educational perspective the benefit is not as obvious. One might argue that integrative RE resembles better the social and cultural environment of the Christian students who in NRW often live next door to peers of various religious and non-religious backgrounds (Hero et al. 2008). Finnish principals, for example, very often think so (Kallioniemi and Matilainen 2011). Another argument would claim that handling religion in a heterogeneous group of learners in terms of worldview may prepare the students better in dealing with religious diversity in society (Clarke and Woodhead 2015). According to our results the latter seems to predict the principals’ preference for integrative RE to some extent, because it is a cooperative perspective on dealing with religions to modify it. The organizational aspect influences this preference, too. But the effect is counter intuitive. One would expect that it is the principals that have problems to organize cooperative RE within one class to particularly prefer integrative RE. In our sample, however, the principals favour integrative RE the more they are able to organize cooperative RE within the given structure of classes. This is an interesting finding that needs further analysis.

Denominational RE is strictly rejected by the principals. Due to the composition of the sample, one could have expected that the participants are not necessarily in favour of this form of religious education. That they object the current mode of RE in NRW this much, however, is a clear statement. The principals in our sample do not see any good aspects on groups of learners that are homogeneous in terms of denomination or religion, respectively. According to our data, this rejection is driven most of all by organizational matters. The more classes the principals had to handle the more they are critical towards denominational RE. The less important educational motives have been for changing to cooperative RE the bigger is the rejection of denominational RE. In Finland, the critical attitude of principals towards denominational RE is motivated to a great deal by the notion that it does not fit into the religiously plural environment of Finish society (Kallioniemi and Matilainen 2011). Also in the culturally Catholic Spanish region of Catalonia, the interviewed principals consider denominational RE as not appropriate for an educational program of state schools (Garreta-Bochaca et al. 2020). A relevant predictor does not come up in our analysis. Because denominational RE is the usual form of this subject in Germany, the question arises of how representative this result is. Given both the secular character of German society and the state’s neutrality in terms of differing worldviews, one could assume that this rejection reflects both issues. In order to estimate the range of this result it might be important to know that there are about 3,600 schools in NRW that offer religious education if vocational schools are left out of the calculation. Our sample represents < 10% of these schools. Therefore, the strong rejection of denominational RE in this study should be handled with care. We simply do not know whether it represents some general critique on denominational RE or rather reflects an attitude of those who decided to establish an alternative form of RE at school. Still, since many people in Germany are critical in regard of denominational RE, this result should not be neglected.

That many principals reject Moral Education, too, is a striking result. In a secular society, Moral Education is the ethical substitute of religious education (Halstead 2010). This is reflected in various findings across Europe. For example, in an Estonian survey, the moral aspect of religious education was the issue to be most convincing to the principals of state schools (Valk 2010, p. 538). In Norway, many principals do not allow any religious presence in the school building at all (Botvar 2017, p. 82). Only in a Welsh sample no differences of the principals’ attitude towards various forms of RE have been found (Davies and Francis 2007). But this study did not offer the option of Moral Education. These results indicate that Moral Education should be preferred by the participants of our study, too. Obviously, this is not the case. Even in the secular society of NRW, letting go of religious education in state schools seems to be no proper option for those principals who applied for an alternative to denominational RE. The critical stance on denominational RE rather leads to some reformation of RE than to abolish it at all. Again, we are not able to explore this finding further by our data. It is worth a study of its own.

Then, the data show that the four forms of religious education presented in the questionnaire are hardly related to each other for the principals surveyed. All correlations are low and, with one exception, explain < 5% of the variance. These low correlations indicate that the participants consider the four forms of religious education to be independent forms that are hardly related to each other. In particular, the low correlation between denominational RE and cooperative RE is of relevance here (r = .14), as the two German churches emphasize that cooperative RE is a special form of denominational RE (e.g. Platzbecker 2019). This does not seem to be the case for principals in our sample.

Finally, these findings can be compared with the findings of studies that asked German teachers of religious education about the preferred form of this subject (e.g., Feige and Tzcheetzsch 2005, pp. 314–322; Pohl-Patalong et al. 2016, p. 328; Rothgangel et al. 2017, p. 82). The results indicate that both groups agree in their rejection of Moral Education. If denominational RE is regarded the teachers are much more in favour of this model than the principals of our sample. This could have been expected since the surveyed teachers have been trained according to the denominational model and the principals have applied for an alternative to this model. For the same reasons it seems to be plausible that integrative RE seems to get a bigger support underneath the principals than underneath the teachers. These comparisons, however, have to be handled with care because each study has its genuine research questions and particular operationalizations of future forms of religious education.

There are some limitations, too. First, due to the fact that our project evaluates cooperative RE in NRW our sample is biased in this regard. Then, our survey took place in NRW which is located in Germany’s West. This is a culturally Christian region. Perhaps principals of schools in the East of Germany would prefer Moral Education instead of some sort of religious education. Further on, we could not assess personal data of the principals due to privacy issues. Some of the respondents felt offended by the mere fact that we asked them about religious education. Such data, however, would have offered insight into the question of how the principals’ religiousness interferes with their preference towards religious education. Finally, the four items of our instrument on forms of religious education cover only the basic options of how to teach this subject in future. These four options do not cover the full spectrum of sophisticated suggestions on future RE as previously described. All these limitations indicate that further empirical investigations are necessary to clarify how principals at state schools think about religious education. This is important knowledge because it is also the principals who directly experience the obstacles of denominational RE.