Life Satisfaction of Immigrants and Length of Stay in the New Country

Round 1

Reviewer 1 Report

The paper's statistical data is clearly presented and the literature review seems adequately referenced.  This seems like an ok paper I had a few questions:

1. at points the authors invent explanations for phenomena ie less satisfaction over time, without giving any evidence at all for this. If there is some, for example evidence of declining social trust which would support the idea that their expectations change, then this should be cited. Otherwise, other references should be used eg to ethnographic and interview data. i think the authors should avoid speculating.

2. I find the conflation of race and national origin to be rather problematic in a text purportedly about national origin and which does not really deal with race. These things are not a proxy for each other.

3. insufficent discussion of racism. Are we sure its that the migrants had too high expectations and not that receiving countries are showing (perhaps increased levels of) racism.    Can the authors engage with work on structural racism and immigration? 

Author Response

Letter to reviewer 1

Thank you for a useful review. Here are the main points of concern and reply to them.  I hope these explanations are sufficient.

1. The authors invent explanations for phenomena ie less satisfaction over time, without giving any evidence at all for this. If there is some, for example evidence of declining social trust which would support the idea that their expectations change, then this should be cited. Otherwise, other references should be used eg to ethnographic and interview data. i think the authors should avoid speculating.

Reply,

I totally agree with a reviewer, that speculations should be avoided. We added some sentences to the discussion part to address the situation more clearly (colourful sections are added).

Theoretically, one possible explanation for the lack of improvement in SWB over time is the change of the reference group during the stay in the country. Initial optimism about the future will be replaced by pessimism over time if expectations are not met or the situation does not improve as quickly as expected.

These hypothesis need to be tested in next research projects, because our models were not able to capture a change of a reference group, neither optimism – pessimism about the future.

Also, some solid conclusion about the lower SWB is presented in the discussion part.

Namely: For immigrants who arrived 10-20 year ago limited social contacts and satisfaction with democracy in combination with other studied factors explained to some extent differences of SWB compared to stayers, but some of their lower SWB remained also unexplained. Because the difference of SWB with local people remained even after usage of all variables, we can conclude, that 10-20 years after arrival might be the most critical period for immigrants.

PS The paper states frankly, that the list of variables used in the model did not provide good explanation for lower SWB, and the work should continue.

2. I find the conflation of race and national origin to be rather problematic in a text purportedly about national origin and which does not really deal with race. These things are not a proxy for each other.

REPLY

The questions about attitudes of different race and ethnic group are original questions of European Social Survey already for 20 years, and are meant to measure the closeness of immigrant group with prevailing population (if race or ethnic group is different, the immigrants are more distant). I agree, that perhaps the question is not perfect, but this is the approach ESS is using, and it does not have other alternatives.

3. Insufficient discussion of racism. Are we sure its that the migrants had too high expectations and not that receiving countries are showing (perhaps increased levels of) racism. Can the authors engage with work on structural racism and immigration?

REPLY

The attitudes of local people are very important. They are measured with the question about acceptance of people with different racial or ethnic background. Racism is important theme, but we do not provide more precise regional picture in this paper, and the question is ESS is meant to measure acceptance of immigrants.

Thank you once again for questions and we hope, I could provide some explanation.

Sincerely an author

Reviewer 2 Report

In my review of the paper, I would like to highlight several key points. First and foremost, I agree with the authors that the topic of immigrants' subjective well-being is of significant importance, not only for immigrants themselves but also for the host society. The paper is well-written, and the English is impeccable. The manuscript is easy to follow, and the authors have done an excellent job in presenting their case.

There are many commendable aspects of the paper that deserve recognition. However, I have some reservations about the findings and their interpretation. The primary focus of the paper is on the relationship between the length of stay in the host country and subjective well-being. It is concerning that the differences between the groups, as depicted in Table 1, appear to be negligible. This raises questions about the significance and practical relevance of these findings. 

The paper attempts to explain these seemingly minor differences between groups using a complex approach. While it's commendable that the authors have considered various factors, it has led to a somewhat disorganized analysis. The use of cross-sectional data from the ESS is a significant limitation. It makes it challenging to establish causal relationships between variables or determine if SWB is an outcome or a cause of other factors (and alternatively, some of these variables can be understood as moderators and mediators). 

There are questions about certain choices made in the ESS data, such as the omission of two items in the measurement of institutional trust and the decision not to use the item explicitly about happiness. Clarifying the reasons for these choices would enhance the paper's transparency. 

The extremely low levels of institutional trust among immigrants, as indicated by a score of 1.6 on a 11-point scale, are noteworthy. This deserves more attention and discussion in the paper, as it could have significant implications for immigrants' well-being (but again, it is not the differences between these groups that is interesting, but the overall low level of institutional trust among these groups). 

The conclusion that SWB is declining may be an exaggeration, as the means of the groups are quite similar. Instead of framing it as a decline, it might be more accurate to explore why SWB does not increase as expected (it appears to me that the data tells more of a story of stagnation and not decline).

Finally, I would like to say that I appreciate the opportunity to review this paper and wish the authors good luck.   

Author Response

Letter to reviewer 2

Thank you for a useful review. Here are the main points of concern and reply to them.  I hope these explanations are sufficient.

1. The primary focus of the paper is on the relationship between the length of stay in the host country and subjective well-being. It is concerning that the differences between the groups, as depicted in Table 1, appear to be negligible. This raises questions about the significance and practical relevance of these findings.

REPLY

We use the measures of statistical differences to draw conclusion in this paper. However, I agree, on practical level, all decision makers should think about the practical relevance. From our side we tried to be quite careful with wording of conclusions.

For example the table 1 tells that there is statistical difference between life satisfaction of local born people and two immigrant groups, those who had stayed more than 20 years in the country  and immigrant who arrived in 1994-2003. The simple analyses does not show difference with recently arrived immigrants.

2. The paper attempts to explain these seemingly minor differences between groups using a complex approach. While it's commendable that the authors have considered various factors, it has led to a somewhat disorganized analysis. The use of cross-sectional data from the ESS is a significant limitation. It makes it challenging to establish causal relationships between variables or determine if SWB is an outcome or a cause of other factors (and alternatively, some of these variables can be understood as moderators and mediators). 

REPLY

We agree, that cross-sectional data is limitation, and the longitudinal approach would have been the best selection. This is also addressed in the paper. From our side we tried to be quite careful with wording of conclusions.

At the same time, the immigrants in the study are indeed with different stay period in the country, and after considering all other important background variables, the period factor can be analysed. This is also the aim of the paper to see how different factors might shape the period variable.

3. There are questions about certain choices made in the ESS data, such as the omission of two items in the measurement of institutional trust and the decision not to use the item explicitly about happiness. Clarifying the reasons for these choices would enhance the paper's transparency. 

REPLY

Institutional trust was measued with three items - country’s parliament, the legal system, and the police. Two available variables omitted from the index are trust of politicians, political parties, United Nations and European Parliament

United Nations and European Parliament are omitted because they are more external bodies, rather than state institutions.

politicians, political parties are omitted because 1) trust on politicians depends a lot of political situation and political preferences, and it reflects therefore less trust towards state institutions generally, 2) It has very volatile character (not the best for a measurement of stables phenomena) like state institutions trust.

Why life statisfaction is the main dependent variable instead of happiness? Because these two measures are different parts of subjective well-being. Happiness is affective measure including more emotional components and life satisfaction is more cognitive evaluation, based on assessment of all life. Life satisfaction fits therefore better for measurement of long-term success of life.

4. The extremely low levels of institutional trust among immigrants, as indicated by a score of 1.6 on a 11-point scale, are noteworthy. This deserves more attention and discussion in the paper, as it could have significant implications for immigrants' well-being (but again, it is not the differences between these groups that is interesting, but the overall low level of institutional trust among these groups). 

REPLY

Thank you for a notice. It seems that something went lost during language proof reading. The original explanative text is added to the paper now.

Scores of the average index vary between 0 and 10, where 0 means you can’t be too careful and 10 means that most people can be trusted. A dichotomous variable with two values is used later in the analyses with value 1 for no trust (scores 0-4) and with value 2 for trust (scores 5-10).

Namely, although the trust level for some immigrant groups is lover, it is not critical. PS Also, some previous research papers had found, that many immigrant perceive the European institutions more trustworthy, than their home institutions (comparison).

5. The conclusion that SWB is declining may be an exaggeration, as the means of the groups are quite similar. Instead of framing it as a decline, it might be more accurate to explore why SWB does not increase as expected (it appears to me that the data tells more of a story of stagnation and not decline).

REPLY

Thank you for this point. The discussion part addresses these ideas. Theoretically, everything what leads to decline, also prevents rise. There are also some ideas in the discussion part, why, the SWB do not rise, but this paper did not have data to test them empirically.

For example theoretically, we can expect decline – because of changing reference group, but the only problem is, that we are unable to prove it with this data. However, this research might be good grounds to further investigations.

Thank you once again for questions and we hope, I could provide some explanation.

Sincerely an author

Round 2

Reviewer 1 Report

thanks for the revisions

Author Response

Dear reviewer 1,

It seems that you did not have any particular remarks. There are still some changes in the text of a paper. Please see green areas.

Sincerely an author

Reviewer 2 Report

Dear Author! Thank you for your response and for providing some clarifications about the measures. Here are my comments on these issues: 

1. I believe this is the case of 'agreeing to disagree,' and there may not be much that can be done to address this point. Naturally, when using a sample like the ESS, most group comparisons would exhibit some statistically significant differences. However, it's important to distinguish between statistical significance and meaningfulness (either theoretical or practical meaningfulness). Taking everything into consideration, in your paper, you attempt to explain statistically significant differences between the studied groups that may not necessarily be meaningful. Frankly, I have reservations about this approach.

2. Issue 2 shares a similar nature. It's not an easy fix, however, merely mentioning something as a limitation doesn't remove that limitation. SWB is considered an outcome, and the authors argue that it depends, among other factors, on the lengths of stay in the new country of the immigrants. At least one very important limitation comes from this approach: present-day SWB and lengths of stay are conceptually distant, with numerous other factors that can mediate, moderate, or contribute to one's SWB.  Then, the small differences between the groups, make the reader question whether SWB really has a connection to lengths of stay in the new country or not. Consequently, the explanations surrounding the results become susceptible to various influences.

Don't get me wrong, I believe this issue is prevalent in many studies utilizing similar data (I have used ESS data myself). However, I think that the problems with the data combined with the small differences found in the analyses, genuinely prompt readers to question whether a different type of discussion is warranted for these findings. I must admit that I was somewhat disappointed with the authors' responses, as they did not seem to involve a thorough examination of my comments.

3. I'm fully satisfied with these explanations about measurement choices.

4. I accept this response, however, it raises another question: Why did the authors transform the 0-10 Likert-type scales into dichotomous variables? At first glance, it appears that his transformation would lead to a loss of significant nuances. For instance, Participant A choosing 5 and Participant B choosing 10 would now apparently have the same level of institutional trust, but in reality, this seems to be a poor reflection of the truth. Participant B evidently has a significantly higher level of trust in domestic institutions compared to Participant A. Generally, transforming continuous variables into dichotomous ones is not ideal, however, it is particularly challenging to justify the conversion of an 11-point scale into a 2-level categorical variable. What is the rationale behind this decision?

5. I'm fine with this point.

Author Response

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

Dear Author, thank you for including additional clarifications into the manuscript. Ideally, you should include, at least as Supplementary Material, the analysis without data transformation. I still believe that creating dichotomous variables from 11-point Likert scales is not advisable. Other than that, I have no further comments.

Author Response

Dear Reviewer,

Thank you once again for a valuable input and for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions highlighted in the re-submitted main text file. Changes are marked with green color in a new version of a submitted file.

As a result of your last remarks, I decided to change both trust indicators to continuous ones. Although the initial idea behind the transforming them to dichotomous was to study the low trust group more profoundly, I must confess, that nothing very interesting came out of it, and the final paper even did not address this low trust group in the conclusions. Therefore, thank you for your suggestion, the trust indicators became continuous, all analyses including them were run once again. Transformation of variables did not change the main substantial results.

 The relevant methodological sections were modified and describes the formation of indicators as follows.

Author Response File: Author Response.pdf