Empathy and Socio-Emotional Competencies: The Perception of Portuguese Adolescents in Today’s Changing Environment
Round 1
Reviewer 1 Report
In this paper the authors administer a survey instrument to a sample of convenience of adolescents in Portugal. Recruitment is done via social networks including Facebook, Instagram, and LinkedIn. The authors find that the adolescents report relatively high social emotional competencies and empathy. They also identify gender- and relationship-based differences in these.
The authors try to link their findings to recent world events such as the COVID-19 pandemic, but I didn't find this convincing. There's no baseline to tell us what would have been the case prior to these events. I think this would be okay with some minor changes in emphasis, focusing on the fact that, however these constructs may have changed as a result of recent events, these adolescents are still reporting relatively high SEC and empathy (though the fact that this is a sample of convenience makes it hard to know if this is also true for Portuguese adolescents more broadly). But the authors should NOT claim that they have measured changes (or a lack of change) in these constructs, or that they have (or have not) been impacted by events.
I also found it a little hard to see why the results about gender and relationship status were included and given such prominence, since they don't seem related to the key research question. I don't think there's anything wrong with reporting the findings, except that they do sort of distract from the key ideas of the paper.
Introduction
I don't know enough about this field to say much here.
Materials and methods
It would be useful to know how the sample compares to Portugal as a whole. For example, what percent of Portuguese adolescents live in rural, semi-urban, and urban areas.
Results
This all seems fairly straightforward.
Discussion
I don't really see how this approach can tell us how Portuguese adolescents reacted to recent world events, since there's not really a baseline measurement. I don't think the results from this sample of convenience can be compared to previous results. The authors claim that they cannot verify that recent events have reduced children's SEC, but the study design could not possibly have verified any impact.
It's also unclear what the authors mean when they say that these adolescents will exhibit less problematic and aggressive behaviors. Less than whom? And how strong are the relationships the authors are basing their conclusions on?
In interpreting the results from this study, the authors should be careful to restrict their interpretation to the sample. There's not a good way to generalize these results to adolescents in Portugal more broadly. I would like them to be more explicit about this. I know that the limitation briefly describes the issue, but it should be made more prominent earlier on.
I also am unclear how the findings on gender and relationship status are related to the rest of the paper. While not uninteresting, they seem to have been tacked on.
Author Response
Response to Revisor 1
Thank you very much for your comments, which have enabled us to improve our article.
Comments 1: The authors try to link their findings to recent world events such as the COVID-19 pandemic, but I didn't find this convincing. There's no baseline to tell us what would have been the case prior to these events. I think this would be okay with some minor changes in emphasis, focusing on the fact that, however these constructs may have changed as a result of recent events, these adolescents are still reporting relatively high SEC and empathy (though the fact that this is a sample of convenience makes it hard to know if this is also true for Portuguese adolescents more broadly). But the authors should NOT claim that they have measured changes (or a lack of change) in these constructs, or that they have (or have not) been impacted by events.
Response 1: Thank you very much for your comment. It made us reflect and change the title, the line of reasoning we had defined and even add new results. Thank you. We opted for the title: Empathy and socio-emotional competencies: the perception of Portuguese adolescents in today's changing environment. We adapted the objectives: 1) To assess the level of socio-emotional competencies and interpersonal reactivity of Portuguese adolescents in the current changing environment; and 2) analyze differences due to sex, relationship status, residence and educational level on the dimensions of inter-personal reactivity and social and emotional competencies. Also to check whether the values in this study differed from those before the pandemic, we compared them with those in the IRI validation study (2010). We didn't use the SEC_Q values, because although they were available, they were from 2023 and with a sample of university students.
Comments 2: I also found it a little hard to see why the results about gender and relationship status were included and given such prominence, since they don't seem related to the key research question. I don't think there's anything wrong with reporting the findings, except that they do sort of distract from the key ideas of the paper.
Response 2: Thank you once again for your contribution. We did think it was pertinent to analyze, from what we read in the literature, the effect of these variables (gender, relationship status, residence and educational level) on the dimensions of empathy and socio-emotional competencies. As neither residence, nor educational level presented significant main effects, and also no significant interaction between the effects of residence and educational level was observed and no significant differences were found in empathy or in the SEC when residence and educational level were taken as factors we chose not to put these results (residence and educational level). However, it makes sense to present them. We have introduced these results and we thank you for your recommendations because it gives us a more holistic view of this reality.
Comments 3: The title doesn't mention the gender and relationship status results which occupy a large part of the paper, and more importantly, it asks whether world events have "affected" various constructs, which the paper does not actually measure.
Response 3: Thank you for your comment. See previous replies. The title has been changed and the document adapted.
Comments 4: It would be useful to know how the sample compares to Portugal as a whole. For example, what percent of Portuguese adolescents live in rural, semi-urban, and urban areas.
Response 4: It would indeed be very interesting, but in Portugal we don't have these statistics. Although there is official statistical data that allows us to gauge the distribution of the population by geographical area (NUTS II/ III/ IV) and by age group, when it comes to more specific analyses (NUTS IV) the data provided by age group is only aggregated by major functional groups (0-14; 15-64 and 65+ years), which prevents us from obtaining this information specifically for the group under study. Furthermore, the fact that place of residence did not prove to be a variable with explanatory value (no significant main effects) meant that we did not explore this issue further.
Comments 5: This all seems fairly straightforward. Any conclusions about whether world events have impacted Portuguese children are unwarranted since the authors do not have a baseline for comparison.
Response 5: Thank you for your comment. The document has been adapted. We have reformulated our approach. We have now the comparison of the SEC and empathy by the factors we judged relevant (sex, relationship status, residence and educational level) and a comparison of the values obtained in our study with the values that were found when the Portuguese validation of the scales were made.
Comments 6: I don't really see how this approach can tell us how Portuguese adolescents reacted to recent world events, since there's not really a baseline measurement. I don't think the results from this sample of convenience can be compared to previous results. The authors claim that they cannot verify that recent events have reduced children's SEC, but the study design could not possibly have verified any impact.
Response 6: Thank you for your comment. See previous replies. The document has been adapted.
Comments 7: It's also unclear what the authors mean when they say that these adolescents will exhibit less problematic and aggressive behaviors. Less than whom? And how strong are the relationships the authors are basing their conclusions on?
Response 7: Thank you for your comment. However, we're not sure we understand your question. This statement “ In fact, median levels of empathic concern and perspective taking, and reduced levels of personal distress seem to indicate the probability that in the future these adolescents will exhibit less problematic and aggressive behaviors and more prosocial competencies and problem-solving abilities, as evidenced in the literature” is the result of several studies available in the literature: Van der Graaff, J.; Carlo, G.; Crocetti, E.; Koot, H., & Branje, S. Prosocial behavior in adolescence: Gender differences in development and links with empathy.” Journal of Youth Adolescence 2018, 47, 1086–1099, https://doi.org/gdbdrq; Laghi, F.; Lonigro, A.; Pallini, S., & Baiocco, R. Emotion regulation and empathy: Which relation with social conduct?” The Journal of Genetic Psychology 2018, 179(2), 1-9. https://doi.org/j7xs; Lissa, C.; Caracciolo, M.; van Duuren, T., & van Leuveren, B. Difficult empathy: The effect of narrative perspective on readers’ engagement with a first-person narrator. Interdisciplinary E-Journal for Narrative Research 2016, 5(1), 43-63; and Vinayak, S., & Judge, J. Resilience and empathy as predictors of psychological wellbeing among adolescents. International Journal of Health Sciences & Research 2018, 8(4), 192- 200.
Comments 8: In interpreting the results from this study, the authors should be careful to restrict their interpretation to the sample. There's not a good way to generalize these results to adolescents in Portugal more broadly. I would like them to be more explicit about this. I know that the limitation briefly describes the issue, but it should be made more prominent earlier on.
Response 8: Thank you for your comment. The document has been adapted.
Comments 9: I also am unclear how the findings on gender and relationship status are related to the rest of the paper. While not uninteresting, they seem to have been tacked on.
Response 9: Once again, thank you for your comment. We thought we had answered you earlier. The document has been adapted.
We hope we have answered all your questions and promoted improvements. Thank you very much.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
See detailed comments section
The manuscript presents the results of an interesting study on How recent world events affect empathy and socio-emotional competencies of Portuguese adolescents. The theoretical framework is adequate, based on relevant and current literature sources. However, there are some issues that should be analysed:
The title and objectives are based on the idea of analysing how recent world events affect empathy and socio-emotional competences. However, the influence of these world events is not measured. Nor is there any reference to which events are referred to (COVID pandemic, war in Ukraine, economic crisis...), nor is there any comparison between the scores obtained by the current sample and those of a sample prior to the events that were considered important. Another option could have been to ask respondents whether a world event (e.g. the COVID pandemic) they believe has influenced their social competence and coping (and a comparison could have been made between those who believe it has and those who do not). In short, it is not possible to determine a cause-effect relationship, in which the results are a consequence of global events (which, as I mentioned earlier, are also not identified). My advice is to eliminate everything related to recent world events because it is not possible to contract their influence with the available data. Therefore, the first objective cannot be checked, and the second objective can. The second objective can be. Another thing is that the manuscript focuses on the relationship between social-emotional competence and empathy in relation to sex and relationships in adolescents.
Why was the gender and marital status variable chosen? And not others, such as place of residence (which is something that studies have linked to the influence of COVID measures, very different in urban and rural environments. Or level of education. On the other hand, ... has only marital status been considered as having or not having a partner? At these ages it is possible that some have had a partner for months and others for days.
If psychometric properties of the instruments are calculated with the data from the study sample, it would be convenient to use McDonald's Omega statistic, since Cronbach's Alpha has many limitations (it overestimates the value) with ordinal variables (such as Likert scale items) and when there are many variables. On the other hand, it would be useful to provide more information on psychometric properties, such as factorial validity (if available).
Using the mean value of a questionnaire as the mean value in the population is very dubious... the mean value of a scale does not necessarily correspond to what is expected in the majority of subjects in the population. If there is no information on scales, it would be better to put this as a limitation and, in any case, to insist that the information be taken with caution. In the case of the IRI, there is average data provided by the study that adapted the scale. This could be a reference value.
In the section on data analysis, reference should be made to the calculation of the effect size for those variables in which statistically significant differences occur.
As mentioned above, the results provided in section 3.1. are exploratory, since: (1) the population value is not known, (2) the previous values are not known, e.g. prior to the pandemic, (3) an inferential test is not applied that allows us to state whether it is higher or lower. For the IRI, the values from the validation study are known. It would be necessary to consider the date on which they were carried out. In this case, they could serve as comparison values (although the sample is not large either) by applying an inferential test. On the other hand, stating that these values are higher or lower than previous ones do not determine that the cause is due to social events experienced. It could be for many other reasons.
Table 2. Putting only one decimal place in the SD is too low, since, by range of scores, most of them will be below the unit. Two or even three decimal places should be used. With the mean, two decimal places should be used (table 2).
Table 3. Putting only "with" and "without" in the table is too short. You should put it in full or put acronyms with a footnote with their meaning.
As the two independent variables are dichotomous, no further evidence is needed to know between which groups the significant differences occur. However, the effect size of those variables where statistically significant differences occur should be provided.
Discussion. As discussed above, how has the influence of world events been measured, calculated and analysed, and has it discriminated between adolescents who perceive that they have been influenced by recent world events?
Author Response
Response to Revisor 2
Thank you very much for your comments, which have enabled us to improve our article.
Comments 1: The title and objectives are based on the idea of analysing how recent world events affect empathy and socio-emotional competences. However, the influence of these world events is not measured. Nor is there any reference to which events are referred to (COVID pandemic, war in Ukraine, economic crisis...), nor is there any comparison between the scores obtained by the current sample and those of a sample prior to the events that were considered important. Another option could have been to ask respondents whether a world event (e.g. the COVID pandemic) they believe has influenced their social competence and coping (and a comparison could have been made between those who believe it has and those who do not). In short, it is not possible to determine a cause-effect relationship, in which the results are a consequence of global events (which, as I mentioned earlier, are also not identified). My advice is to eliminate everything related to recent world events because it is not possible to contract their influence with the available data. Therefore, the first objective cannot be checked, and the second objective can. The second objective can be. Another thing is that the manuscript focuses on the relationship between social-emotional competence and empathy in relation to sex and relationships in adolescents.
Response 1:
. Thank you very much for your comment. It made us reflect and change the title, the line of reasoning we had defined and even add new results. Thank you. We opted for the title: Empathy and socio-emotional competencies: the perception of Portuguese adolescents in today's changing environment.
. Indeed, the option “could have been to ask respondents whether a world event (e.g. the COVID pandemic) they believe has influenced their social competence and coping (and a comparison could have been made between those who believe it has and those who do not).” It would be very interesting and we will take it into account in a future study.
. The objectives were adapted as follows: 1) To assess the level of socio-emotional competencies and interpersonal reactivity of Portuguese adolescents in the current changing environment; and 2) analyze differences due to sex, relationship status, residence and educational level on the dimensions of inter-personal reactivity and social and emotional competencies.
Comments 2: Why was the gender and marital status variable chosen? And not others, such as place of residence (which is something that studies have linked to the influence of COVID measures, very different in urban and rural environments. Or level of education. On the other hand, ... has only marital status been considered as having or not having a partner? At these ages it is possible that some have had a partner for months and others for days.
Response 2: Thank you very much for your question. We thought it was pertinent to analyze, from what we've read in the literature, the effect of these variables (gender, relationship status, residence and educational level) on the dimensions of empathy and socio-emotional competencies. As neither residence, nor educational level presented significant main effects, and also no significant interaction between the effects of residence and educational level was observed and no significant differences were found in empathy or in the SEC when residence and educational level were taken as factors we chose not to put these results (residence and educational level). However, it makes sense to present them. We have introduced these results.
Comments 3: If psychometric properties of the instruments are calculated with the data from the study sample, it would be convenient to use McDonald's Omega statistic, since Cronbach's Alpha has many limitations (it overestimates the value) with ordinal variables (such as Likert scale items) and when there are many variables. On the other hand, it would be useful to provide more information on psychometric properties, such as factorial validity (if available).
Response 3: We appreciate it. The reliability of the scales of the instruments used were now calculated using Omega statistic, as can be seen in section 2.2
Comments 4: Using the mean value of a questionnaire as the mean value in the population is very dubious... the mean value of a scale does not necessarily correspond to what is expected in the majority of subjects in the population. If there is no information on scales, it would be better to put this as a limitation and, in any case, to insist that the information be taken with caution. In the case of the IRI, there is average data provided by the study that adapted the scale. This could be a reference value.
Response 4: Please, see response 6.
Comments 5: In the section on data analysis, reference should be made to the calculation of the effect size for those variables in which statistically significant differences occur.
Response 5: The reference of effect size calculation was added.
Comments 6: As mentioned above, the results provided in section 3.1. are exploratory, since: (1) the population value is not known, (2) the previous values are not known, e.g. prior to the pandemic, (3) an inferential test is not applied that allows us to state whether it is higher or lower. For the IRI, the values from the validation study are known. It would be necessary to consider the date on which they were carried out. In this case, they could serve as comparison values (although the sample is not large either) by applying an inferential test. On the other hand, stating that these values are higher or lower than previous ones do not determine that the cause is due to social events experienced. It could be for many other reasons.
Response 6: Comparisons were made, against the values obtained in the validation study of the IRI. The same procedure was not applied to the SEC_Q, because the validation study is dated from 2023.
Comments 7: Table 2. Putting only one decimal place in the SD is too low, since, by range of scores, most of them will be below the unit. Two or even three decimal places should be used. With the mean, two decimal places should be used (table 2).
Response 7: The tables 3 and 5, which present descriptive values (means and SD) of both the IRI and SEC are now with two decimal cases.
Comments 8: Table 3. Putting only "with" and "without" in the table is too short. You should put it in full or put acronyms with a footnote with their meaning.
Response 8: In the footnotes we added the complete description of these two terms.
Comments 9: As the two independent variables are dichotomous, no further evidence is needed to know between which groups the significant differences occur. However, the effect size of those variables where statistically significant differences occur should be provided.
Response 9: In tables 4 and 6, which present the results of the MANOVAs performed, a column was added with the effect size for each F statistic, calculated using the partial eta squared coefficient.
Comments 10: Discussion. As discussed above, how has the influence of world events been measured, calculated and analysed, and has it discriminated between adolescents who perceive that they have been influenced by recent world events?
Response 10: Once again, thank you for your comment. We thought we had answered you earlier. The document has been adapted.
We hope we have answered all your questions and promoted improvements. Thank you very much.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
All recommendations have been considered.
All recommendations have been considered
