Religious Beliefs and Socialization: An Empirical Study on the Transformation of Religiosity in Spain from 1998 to 2018

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a high-quality article with an appropriate theoretical and methodological basis, especially with regard to the statistical data from studies carried out in Spain between 1998 and 2018. As the author points out right at the beginning of his article, Spain, a country with a Catholic tradition, has undergone a process of secularisation following the re-democratisation of 1978. Although he starts from this observation, the author provides a more nuanced analysis by taking into account data from questionnaires conducted over the last two decades. In analysing the data, it was demonstrated that the decline in the importance of religiosity and parental practise over time suggests that generational and ideological factors are increasingly influencing religious beliefs in Spain (p. 9 pdf). Overall, the article is quite clear and objective in terms of its hypotheses and arguments. However, as a suggestion to the author, it would be important to avoid repetitive statements. Between page 1 and 4 he has stated his aims and main hypothesis several times in different words. But the objectives and hypothesis are clearly presented at the end of page 4 of the PDF file. We suggest moving the presentation of these objectives and the hypothesis to the first part of the text and avoiding repetition. In this way, you can give more weight to the analysis of your data and your conclusions.

Comments on the Quality of English Language

As already mentioned, it is a well-written and structured text. Only to make it even better, it would be important to avoid the repetition of ideas, because between pages 1 and 4 of the pdf, the author sometimes repeats his main objectives with the article, which makes the reading a bit repetitive. However, this in no way detracts from the quality of the article.

Author Response

Dear Reviewer 1,

We sincerely appreciate your thorough review and insightful comments on our manuscript. Your feedback has been invaluable in helping us improve the clarity and structure of our paper.

In response to your suggestion regarding repetitive statements, we have carefully revised the introduction section. Specifically, we have removed the following paragraph that was previously at the end of the introduction:

"Integrating these theoretical reflections, we aim to move beyond a juxtaposition of theories to offer a coherent interpretation that connects philosophical discussions with our empirical findings. This contextualizes the evolution of religiosity in Spain within a robust theoretical framework, fostering a dialogue between theory and practice that enhances our understanding of the dynamic interplay between the sacred and the secular in modernity."

We agree that this removal helps streamline the introduction and reduces redundancy. The core ideas expressed in this paragraph are now more concisely integrated throughout the introduction, maintaining the paper's coherence while eliminating unnecessary repetition.

Sincerely,

Reviewer 2 Report

Comments and Suggestions for Authors

This paper is primarily a descriptive study that examines changes in religiosity in Spain over the course of two decades. It focuses on changes in belief, religious controversies, and changes in religious socialization over this time period.

The authors demonstrate that there has been a precipitous decline in belief and religious observance, and in concomitant increase in atheism and irreligion. This shift is most noticeable in younger generations.

The society is becoming increasingly polarized on the observance of religion, and this is interpreted as meaning that demographic and generational variables are more important than the socialization process. This comparison of demography vs. socialization is an important contribution of the paper.

The analysis uses logistic regression models. These are the correct models for dichotomous dependent variables. The models demonstrate explanatory power.

The paper appears to use these models correctly, but the paper should include some diagnostics to examine multicollinearity and the independence of errors. Also, make sure that every step of the analysis—including limitations and assumptions—is properly explained. The paper is relatively short, so there is room to provide more detail here.

I would also include more information on how the variables were operationalized. The criteria for categorizing respondents as atheists, agnostics, and so forth can be made more clear.

Finally, the findings can be better situated within the political and cultural context of contemporary Spain. More detail on policy and the implications of the findings will increase the relevance of the paper.

The quality of the writing is good, but somewhat dense and technical. It might be worth going through the paper again and editing for clarity. The findings make an important contribution to the literature.

Comments on the Quality of English Language

The quality of the writing is good, but somewhat dense and technical. It might be worth going through the paper again and editing for clarity. 

Author Response

Dear Reviewer 2,

We sincerely appreciate your thoughtful comments and valuable suggestions for improving our manuscript. Your insights have been instrumental in enhancing the methodological rigor and clarity of our study. We have carefully addressed your concerns and made several significant additions to the paper.

In response to your suggestions, we have expanded our methodology section to provide a more comprehensive explanation of our analytical approach. Specifically, we have added two detailed paragraphs that outline the various statistical methods employed in our analysis. The first paragraph describes the descriptive statistics, trend analysis, and comparative statistics used in the initial three sections of our results. The second paragraph focuses on the logistic regression models used in the final section, including details on multicollinearity tests, variable coding, and model fit assessment.

Furthermore, we have elaborated on the presentation of the logistic regression analysis. We now provide a more detailed description of the independent variables and their coding. We have also included the results of multicollinearity tests, which demonstrate the absence of severe multicollinearity issues across all three time periods. Additionally, we report the Hosmer-Lemeshow test results for each period, confirming acceptable model fit.

To address your concern about the study's limitations, we have added a reflective paragraph at the end of the discussion section. This new content acknowledges the potential for unmeasured factors influencing religiosity and suggests directions for future research.

Lastly, we have updated our reference list to include Tabachnick and Fidell (2019), which supports our methodological approach to multicollinearity testing.

1. In the Methodology section, we have added two paragraphs explaining our analytical approach:

"This study employs a multi-faceted analytical approach to examine the evolution of religiosity in Spain over two decades. For the first three sections of our results (Analysis of Beliefs and Religiosity, Evolution of Religious Controversies, and Religious Socialisation and Its Impact), we utilize descriptive statistics, trend analysis, and comparative statistics. Specifically, we employ frequency distributions and cross-tabulations to examine changes in religious beliefs and self-reported religiosity over time. Chi-square tests are used to assess the relationship between parental religiosity and individual beliefs, providing insights into the impact of religious socialisation. These analyses offer a comprehensive overview of the changing religious landscape in Spain and the evolving dynamics of various religious positions.

For the final section of our results (Evolution of explanations for belief in God: between socio-demographic and religious socialisation variables), we employ logistic regression models. These models are constructed for each of the three time points (1998, 2008, and 2018) to examine how the influence of various socio-demographic and religious socialisation factors on belief in God has changed over time. To ensure the robustness of our findings, we conduct multicollinearity tests using the correlation matrix for each time point (Tabachnick & Fidell, 2019). The dependent variable 'Belief in God' is dichotomously coded, while independent variables include both categorical and continuous measures. The Hosmer-Lemeshow test is used to assess the goodness of fit for each model, and Nagelkerke's R² is calculated to measure the models' explanatory power. This rigorous approach allows us to identify significant predictors of religious belief and how their influence has evolved over the two decades under study."

2. In the presentation of the logistic regression analysis, we have added:

"Independent variables included: sex (dichotomous: 0 = Male, 1 = Female), age (continuous), education level (ordinal), socioeconomic status (ordinal), ideology (scale from 1 to 10, where 1 represents extreme left and 10 extreme right), parental religiosity (dichotomous: 0 = No, 1 = Yes), and religious practices of the mother, father, and the respondent during childhood (ordinal frequency scale). To assess the validity of the model, multicollinearity tests were performed using the correlation matrix. No correlations exceeding 0.8 were observed among the independent variables across the three analyzed periods (1998, 2008, and 2018), indicating the absence of severe multicollinearity issues. The highest correlation observed was 0.557 between education level and socioeconomic status in 2008. The model's goodness of fit was evaluated using the Hosmer-Lemeshow test. Results for the three periods were as follows: 1998 (χ²(8) = 5.165, p = .740), 2008 (χ²(8) = 15.526, p = .050), and 2018 (χ²(8) = 13.502, p = .096). These findings indicate an acceptable model fit to the data across all three periods."

3. At the end of the Discussion section, we have added a paragraph on limitations:

"It is important to acknowledge certain limitations of this study. While the logistic regression model provides a robust approximation of factors influencing belief in God, the complex nature of religiosity implies that other factors not captured in this analysis may exist. Furthermore, the decline in the model's explanatory power over time suggests that new factors relevant to understanding religiosity in contemporary Spain may be emerging. Future research could explore additional variables and employ mixed methods to capture more nuanced aspects of the evolution of religious beliefs."

4. We have added a new reference to support our methodological approach:

Tabachnick, B.G., and Fidell, L.S. 2019. Using Multivariate Statistics, 7th ed. Boston: Pearson.

We believe these additions address your concerns regarding the clarity of our methodology, the robustness of our analysis, and the acknowledgment of study limitations. Thank you again for your valuable feedback, which has significantly improved the quality of our manuscript.

Sincerely,

Reviewer 3 Report

Comments and Suggestions for Authors

The article is interesting and worth publishing. It is written correctly, and its structure is also clear and correct. Although it does not present new or surprising data, it does collate interesting data collected in the course of studying the changes in Spaniards' religiosity. Based on the analysis of these results, the authors indicate that socio-demographic factors (gender, age, education, ideology), rather than traditional religious socialisation, play an increasingly important role in the evolution of Spaniards' religiosity (lines 406-407).

This is a valid conclusion, but in my opinion, it is nevertheless too shallow. Apart from general statements, neither in the Discussion nor in the Conclusions do the authors say specifically how these socio-demographic factors influence the evolution of Spaniards' religiosity. This would be the added value of an article based on a study already conducted by someone else (the Centro de Investigaciones Sociológicas). In my opinion, the article would have benefited a lot if this conclusion had been deepened. 

Comments on the Quality of English Language

The article's language is correct, but it should be more inclusive. I would very much appreciate an adjustment of the language (e.g., in Table 2: "He has never...", "He doesn't believe...", etc., and Table 3: "His mother's religious practice..." etc.).

Author Response

Dear Reviewer 3,

We sincerely appreciate your thoughtful comments and your attention to detail regarding inclusive language in our manuscript. Your feedback has been invaluable in helping us improve the clarity and inclusivity of our work.

In response to your suggestions, we have made several important changes:

1. We have added a paragraph to the conclusions section to contextualize our findings within recent empirical evidence on the evolution of religiosity in Spain. This addition emphasizes the complexity of secularization and the importance of sociodemographic variables:

"Recent empirical evidence on the evolution of religiosity in Spain underscores the complexity of the secularization phenomenon and the crucial importance of sociodemographic variables in this process. Studies by Rosa-González and Cabrera (2023a, 2023b) corroborate our findings on the general decline in religious identification, while adding important nuances. Their research suggests that religious decline is influenced by a complex interplay of factors, including generational replacement, urbanization, educational level, and regional differences. The persistent, albeit possibly decreasing, gender gap in religiosity and the marked variations between urban and rural areas indicate that secularization in Spain is not a uniform phenomenon, but rather deeply intertwined with broader social changes. For instance, Rosa-González and Cabrera (2023a) found substantial differences in religiosity across municipalities, with larger cities showing lower levels of religious identification. These findings emphasize the need for a multidimensional approach to understanding the evolution of religious beliefs in the contemporary Spanish context, one that considers not only individual characteristics but also the socio-spatial dynamics at play."

2. We have added two new references to support this discussion:

Rosa-González, F.M., and Cabrera, L. 2023a. Religious identification (BELIEVERS) by population size of the municipalities in Spain. Heliyon 9: e19083.

Rosa-González, F.M., and Cabrera, L. 2023b. A Sociological Perspective on Religious Identification in Spain: A Multidimensional Analysis Based on Empirical Data (Over 467,187 Individuals). Secularism and Nonreligion 12: 1-14.

3. Regarding your comments on inclusive language, we have made the following changes:
   • In Table 2, we have revised the row descriptions to directly reflect the survey response options: "I don't believe in God now and I never have" "I don't believe in God now, but I used to" "Total 'Does not believe in God'" "I believe in God now, but I didn't used to" "I believe in God now and I always have"
   • In Tables 4 and 5, we have adjusted the column headers to use more gender-neutral language: Table 4 now uses "Maternal religiosity" and "Paternal religiosity" Table 5 now uses "Maternal religious practice" and "Paternal religious practice"
4. We have also conducted a thorough review of the entire manuscript to identify and modify any language that may not be inclusive.

We believe these changes significantly enhance the contextual depth of our discussion, while also improving the inclusivity and clarity of our presentation. Thank you again for your valuable feedback, which has undoubtedly strengthened our manuscript.

Sincerely,