Modeling Local Variations in Intermarriage
Round 1
Reviewer 1 Report
The paper “Modeling spatial local variations in intermarriage” submitted to Mathematics deals with an interesting topic. The main goal of the article is to model the spatial distribution of intermarriage considering the local characteristics of the marriage market regarding size of the population, heterogeneity of migrant origins and the sex ratios in the migrant and native communities. The study referes to Spain, a mediterranean country of recent immigration. In my opinion the paper is absolutely cutting edge. I fount it very interesting, well written and original. My congrats to the Authors. Here follows some general observations and, after that, some minor revisions. Hope this can help. I believe that the paper deservers to be published
- Authors used “country of citizenship” as criterion to identify migrant population. Do they have think about the possiblity (like happens in Italy, see for example Guetto and Andreotti, 2015, Journal of Ethnic and Migration Studies) that for certain foreign communities the marriage with spaniards can be a tools for achieving spanish citizenship? So that from this point of view the intermarriagge market is biased?
- Did the Authors think about the existence of possible heterogeneities in the foreign population? In the future development maybe they can reflect about run models on different groups of foreigners based on the country of citzenship (north africa, east europe etc.). Or the data doens’t allow this kind of disaggregation?
- Again about inner heterogeneity: do Authors think that migrants coming from Spanish language countries (South America for example) have major propensity to being engaged in mixed marriage?
Minor revisions
- 3, 104-106; it seems to me that something is wrong in the sentence: “Typically…, and more recently”. It seems to me that a part is missing from the sentece.
- 4, 169; references 35 and 36 are not linked to the references list
- Maybe the softwares or libraries used in the study (maps, models etc.) can be briefly described.
Author Response
We thank you for your comments. We found them very useful and have revised our paper accordingly. Our response appears below.
# 1. Authors used “country of citizenship” as criterion to identify migrant population. Do they have think about the possiblity (like happens in Italy, see for example Guetto and Andreotti, 2015, Journal of Ethnic and Migration Studies) that for certain foreign communities the marriage with spaniards can be a tools for achieving spanish citizenship? So that from this point of view the intermarriagge market is biased?
This is a very relevant observation for the purpose of this analysis. Indeed, some migrants might prefer marriage over unmarried cohabitation in order to get Spanish Citizenship. Unfortunately, we do not know the reasons why foreign nationals decide to marry natives. This is something that we cannot measure. However, we do not think that this might bias intermarriage downwards because many foreign-born nationals with non-European nationalities might be more likely to marry than cohabit. In addition, they might be more likely to marry in Spain than abroad. The only source of potential bias refers to the fact that some cross-national unions might not marry if the acquisition of citizenship is not an issue. The other potential source of bias could be that some couples decide to marry abroad if the acquisition of citizenship is not an issue again.
We have acknowledged in the text that this analysis is based on observed marriages in Spain. Marriages that have taken place abroad and couples that decide not to marry are not included in the analysis. We have also referred to this motivation for foreign-born nationals to marry to get the Spanish citizenship in line 5 of the 3rd paragraph of section 4.1. as follows:
“We cannot measure the reasons why foreign nationals decide to marry natives either, e.g., to get Spanish citizenship like happens in other countries ([43])”.
And in the references, we have added:
- Guetto, R.; Azzolini, D. An Empirical Study of Status Exchange through Migrant/Native Marriages in Italy. J. Ethn. Migr. Stud. 2015, 41, 2149–2172, doi:10.1080/1369183X.2015.1037725.
# 2. Did the Authors think about the existence of possible heterogeneities in the foreign population? In the future development maybe they can reflect about run models on different groups of foreigners based on the country of citzenship (north africa, east europe etc.). Or the data doens’t allow this kind of disaggregation?
This is undoubtedly a relevant point that we have not addressed although we do check for homogeneity of origins. This is a very valuable idea for future research which we have incorporated into the last section of the manuscript. It is also important to notice that information on the country of citizenship at the local level might be problematic to obtain due to confidentiality reasons.
“In the context of an increasingly heterogeneous foreign population in Spain, it is also relevant to study whether intermarriage propensity differs according to the origin of migrants, though information on the country of citizenship al the local level might me problematic to obtain due to confidentiality reasons”.
# 3. Again about inner heterogeneity: do Authors think that migrants coming from Spanish language countries (South America for example) have major propensity to being engaged in mixed marriage?
Like the previous comment, this is an interesting point. As the reviewer mentions, we observed in previous research that the Spanish nationals were more likely to marry Latin Americans, especially among Spanish men (Serret et al., 2013). We consider that we captured this idea with the comment we have added in the conclusions.
Minor revisions
# 3, 104-106; it seems to me that something is wrong in the sentence: “Typically…, and more recently”. It seems to me that a part is missing from the sentece.
We are very thankful to the reviewer for detecting both minor errors. We have solved this issue by deleting ‘and more recently’.
# 4, 169; references 35 and 36 are not linked to the references list
We have solved this issue by linking the references.
# Maybe the softwares or libraries used in the study (maps, models etc.) can be briefly described.
Thank you for this appreciation. We have added the following texts:
. In the title of Figure 1: “Maps were generated with the software Map Viewer, version 8.7”.
. In the 1st paragraph of both sections 5.1. and 5.2.: “Model estimations and testing were computed with the software STATA, version 15”.
Referencies:
Serret Sanahuja J, Esteve A, López Gay A. Revelaciones de una década de matrimonios entre españoles y extranjeros, 1998-08. Estadística Española. 2013;181(5):0213-229.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The reviewed article deals with a very interesting and in many respects difficult topic of modeling spatial local variations in intermarriage. In the era of widespread migration, such research topics are very important because they tell us about the possibility of integration and acceptance of migrants in a given country.
The strengths of the article include, first of all, the uniqueness of the data reached by the authors of the publication. This allows, using an appropriate methodology, to achieve the results of analyzes largely corresponding to reality. Secondly, the author's method of research should be appreciated, which is unique and, according to the authors, fills a kind of gap in the current state of research on this subject.
The choice of the research period should be considered slightly controversial. As the authors themselves recall, they chose for their analyzes the data from 2005-2007, when Spain is the peak of the first major stage of international migration in this country. One should consider whether the research should not include data from a later period, when the influence of the great international migration in 2005-2007 will begin to be visible (people need time to get to know each other and decide on a marriage). I also suggest that you correct Figure 1 - add some halo around the communes - as they are very faint visual at present. One should also consider the possible interpretation of the obtained results. In most cases, the value of the Poisson correlation index was very low. Therefore, the relationship between the traits studied does little importance.
Despite the above-mentioned discussion elements, the reviewed article should be considered very novel and innovative. This publication increases the current state of knowledge on international marriages. After minor corrections, it is perfectly suitable for publication.
Author Response
We thank you for your comments. We found them very useful and have revised our paper accordingly. Our response appears below.
# 1. The choice of the research period should be considered slightly controversial. As the authors themselves recall, they chose for their analyzes the data from 2005-2007, when Spain is the peak of the first major stage of international migration in this country. One should consider whether the research should not include data from a later period, when the influence of the great international migration in 2005-2007 will begin to be visible (people need time to get to know each other and decide on a marriage).
This is a good point; we thank the reviewer for highlighting it. We agree that choosing a period is always controversial and that ours has the limitation that the reviewer has mentioned. However, we consider that choosing any other later period may raise even greater controversy. The 2008 financial crisis had a remarkable effect on international migration flows. On the one hand, there was a major slowdown in immigration. On the other hand, there was a substantial outflow from Spain, in the form of movements to other European countries or to the countries of origin (Parella and Petroff, 2014; Prieto et al., 2018). For these reasons we selected the period 2005-2007 for the analysis. We consider that the choice of a later period would have added more complexity to the study of intermarriage due to the varied trajectories of migrants.
To clarify the text, we have emphasized that this period corresponds to the end of the first stage of foreign immigration to Spain (Page 2, lines 67-68). We also have extended the explanation of the 2008 financial crisis on international migration, and added a reference (Page 4, lines 158-159).
# 2. I also suggest that you correct Figure 1 - add some halo around the communes - as they are very faint visual at present
We are thankful for this suggestion, and we agree that the maps on Figure 1 could be improved. As suggested, we have added a thin line in a light grey color around municipalities. We have been very careful in adding the outline to the municipalities because wider lines could make the colors of the smaller municipalities not visible. We also have added the outline of provinces and regions for a better display of the map and for a better understanding of the Spanish geography.
# 3. One should also consider the possible interpretation of the obtained results. In most cases, the value of the Poisson correlation index was very low. Therefore, the relationship between the traits studied does little importance.
We also appreciate your indication because we have realized that this point was not sufficiently clear in the first version of the paper. Although there is a definition of the pseudo-R2 values in the footnote of Table 3, we have added an additional sentence in the second paragraph of section 5.2, where with state that a pseudo-R2 is not equivalent to a proper R2:
Second paragraph of section 5.2.:
“Although the R2 values are somewhat lower than the pseudo-R2 values of the ZIP models, they are not straightforwardly comparable: while the pseudo R2 is computed as the correlation coefficient between the real and estimated dependent variables, the R2 is the determination coefficient of an Ordinary Least Squares (OLS) estimation”.
However, the highly significance of the coefficients of the second model, as inferred by the t-tests, demonstrates that the relationship of their corresponding explanatory variables with the dependent one is very high. Hence, we can rely on the results of this second model.
Referencies:
Parella S, Petroff A. Migración de retorno en España: salidas de inmigrantes y programas de retorno en un contexto de crisis. Anuario CIDOB de la Inmigración. 2014:61-88.
Prieto-Rosas V, Recaño J, Quintero-Lesmes DC. Migration responses of immigrants in Spain during the Great Recession. Demographic Research. 2018 Jan 1;38:1885-932.
Author Response File: Author Response.pdf
