Secularization, Religious Denominations, and Differences in Regional Characteristics: The State of Research and a Regional Statistical Investigation for Germany

Round 1

Reviewer 1 Report

I think this is an interesting study on a somewhat overlooked topic. One thing I'd encourage the author to do would be to make a more noticeable list of the theoretical questions that power the study. This does not have to be as formally enumerated as the research hypotheses later in the paper, but perhaps using more headings in the first third of the paper will make it easier to show the organization of the lit review and theoretical organization behind the investigation. 

Author Response

Comments in separate file

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper presents a detailed excellent study of relationships between religion and economic behaviour comparing regions of Germany. Yes, grounded in the Weberian tradition, but using contemporary measures and methods of analysis. While the measure of religion is very thin - Prot vs Cath identification and a third category combining those with no and those identifying as Muslim, Buddhist, Orthodox and other, the measure of economic activity are rich - GDP, employment/unemployment rates, female employment rates and feritlity.

Hypotheses are developed and assessed at National and regional levels. No single hypothesis holds uniformly, although interesting relations are discuverd at regional levels, but none that can be attributed to a 'religious factor'.

I would have expected more attention to be paid to the historic effects of the years of soviet rule in East Germany both in the distribution of religious identity and economic behaviour. The influence of larger culture of which religion may or may not be a carrier is not given as much attention as I would like.

The failure to meaninglfully label the variables in the charts makes them unintelligible. 

While overall very clear there are points at which the English expression is unclear. I cite as examples:   ln 329 'meaning of denomination' ???; ln 277;  ln 272; ln 235 - 'Protestant belief in debauchery' ????;  

Author Response

Comments in separate file

Author Response File: Author Response.pdf

Reviewer 3 Report

This is a really interesting and relevant paper on the connection between the distribution of religious communities and the economic and demographic characteristics in German districts. This article will attract the attention of various fields of research, and it would be beneficial if it were published in Religions.

In my view, however, the paper does not currently meet all the necessary quality standards and should be fundamentally revised.

A first point is that the paper's line of thought could be more clearly elaborated to guide the reader through the complex subject. The paper starts quite directly with a historical overview of religious differences in Germany, but it is not really clear where this will lead. I recommend the author to start by explaining the main problem and the main aim of his research. Why is it important to know about the contexts researched? Why is it necessary to be informed from history etc.? The same applies to the (not mentioned) research questions. What is to be researched in the empirical part of the work and how does this relate to previous research? One would expect this at the end of section 1, before the methodological part begins. In explaining the research questions, it would become clear that the author is not so much interested in differences between religious denominations (the old distinction between Protestants and Catholics), but rather in differences between religious and non-religious people (at least that is how I understood the argumentation in the results section). The results section (section 3.2.1) also formulates hypotheses. This is rather untypical. I would recommend that they be placed in the introductory section immediately after the report on the state of research, because that is where the hypotheses come from.

A second comment concerns the method of the study. I have at least two major concerns about this. The first is that the title of the paper speaks of regional development. However, the data used here are cross-sectional data from 2011 (census), not longitudinal data. So, what the author can show is that in 2011 the distribution of religious affiliation correlates with some economic and demographic characteristics, but not that one has influenced the other over time. I would expect this limitation to be further elaborated in the methods section and in the discussion. The second point is that the author him- or herself refers to the importance of context factors (on page 8) to explain economic success e.g. in Catholic regions in Northern Italy, but that these context factors do not play a role in further analyses. It would strengthen the paper's argumentation if correlations between religion and, in particular, economic characteristics were controlled with other variables that may have an influence on economic development (such as the existence of educational institutions, transport infrastructure, etc.). This could statistically be achieved with partial correlations.

A third observation concerns the theoretical background of the study. As I have already mentioned, I have the impression that the author is more interested in differences between religious and non-religious populations than in differences between certain religions. That is why s/he so often refers to secularisation theory. However, I do not think that the paper clearly establishes the link between secularisation and economic growth. Here and there, there are comments on it in the text. It would be more advantageous for the reader, however, if the author could explain his or her way of thinking on this point at a very prominent place in the text, perhaps right at the beginning. This would help readers to understand how the author views the debate on religious influences on economic development.

Finally, there are some inaccuracies in the text which should be corrected.

On page 3, lines 71-75, the author states: “The notion that large numbers of people are leaving the Christian churches is thus only true to a limited extent.” I cannot follow this interpretation. In 2019, for example, 270,000 people each left the Catholic and Protestant churches in Germany. That is quite a lot. Nor would I argue that a decline in baptism is a demographic phenomenon. Rather, this reflects processes of secularisation among (former) church members.

On page 3 line 81 the author uses the word irreligiousness. This is a rather unusual term. I would recommend using the word non-religiousness.

On page 3 the author also uses the terms significance and importance to talk about the churches, but in a very imprecise way. S/he more or less equates this with membership. In my view, the public significance of a religion is not the same as the number of its members, because significance and importance also relate to public perception (see, for example, the role of Islam) and influence (see, for example, the influence of the two Christian churches in the welfare sector). I therefore ask the author to formulate more precisely what s/he means.

The same applies to page 3 lines 92-94, where the author equates religious belief with attending church services. This is a very limited understanding of what religiosity is, and we know from several studies that people believe in God and see themselves as Christians without participating in the rituals of their church (especially when it comes to the weekly services on Sundays). I would therefore like to ask the author to formulate more carefully here.

On page 14, line 456, the author uses the term religious freedom, but this is not appropriate in this context. What the author means can be described as religious plurality or religious diversity.

Both section 3 and section 4 are entitled Results. It would be good to distinguish between them.

On page 18, the description of the results does not match the above illustrations. I have checked this against the first two hypotheses and then stopped reading, but I think that the illustrations are not in the right order and are not correctly labelled. Please check this. Furthermore, I would recommend to document the correlations in tables with numbers rather than with colour codes for two reasons. Firstly, the correlation coefficients make it easier to assess the strength of the correlations. Secondly, due to some colours (especially the very light blue) I could not see if the results are significant or not (the white stars were hard to see for me).

Author Response

Comments in separate file

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The authors have implemented the suggestions of the experts appropriately. I have no further comments.