Generational Climate Engagement in Liquid Modernity: Eco-Anxiety, Environmental Activism and Pro-Environmental Behavior Among Older Adults in Spain

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is an interesting and generally well written manuscript reporting on survey data from adults in Spain, looking at how eco-anxiety, environmental activism and pro-environmental behavior vary across the life course/age groupings. While looking at results across age groupings, the authors focus specifically on how their findings speak to ideas about older adults.

Especially for those who are not familiar with Spain, what is the sampling frame that was used to get a stratified sample based on gender and age (as well as “autonomous community”—which is not clear)? Also, was this a proportional stratified sample which was intended to match the breakdown in gender/age within the population? Or was this disproportional in order to get a certain number of individuals per each gender/age grouping? (The analysis seems to indicate the latter, but this is not clear.)

The authors note a sample size of 3000—how many people were contacted in order to get this completed number? What was the response rate?

Given their role in the analyses, all other measures should be included in the methods section. While gender, age and education seem relatively straightforward, how was social class measured?

Throughout the article, the authors make a big deal of the notion of “self-efficacy” and they even phrase H3 with this notion. For a while, I assumed that this was also measured in the survey, especially in order to actually test H3. However, it is not measured, but a decline in self-efficacy is assumed to occur with age. This may need greater explanation. H3 should be written to reflect what is actually tested.

Author Response

Author's Reply to the Review Report (Reviewer 1)

Comment 1.1: This is an interesting and generally well written manuscript reporting on survey data from adults in Spain, looking at how eco-anxiety, environmental activism and pro-environmental behavior vary across the life course/age groupings. While looking at results across age groupings, the authors focus specifically on how their findings speak to ideas about older adults.

Response 1.1: Thank you very much for your kind words and help in improving our work. We believe that your insightful and constructive comments have contributed to improving the clarity and overall quality of our contribution. Below, we explain how we have addressed all your comments. In our replies, we mention the lines that have been changed in the marked-up manuscript version.

Comment 1.2: Especially for those who are not familiar with Spain, what is the sampling frame that was used to get a stratified sample based on gender and age (as well as “autonomous community”—which is not clear)? Also, was this a proportional stratified sample which was intended to match the breakdown in gender/age within the population? Or was this disproportional in order to get a certain number of individuals per each gender/age grouping? (The analysis seems to indicate the latter, but this is not clear.)

Response 1.2: Thank you for this insightful comment. We have revised and introduced new sentences in section 3.1. Sampling Strategy and Data Collection to clarify this. The new sentences are the following: “A stratified random sample of 3,000 residents aged 18 and older was drawn using census data from the Spanish National Institute of Statistics (www.ine.es) as the sampling frame. The sample was stratified by sex, five-year age group, and 19 regions (17 Autonomous Communities and 2 Autonomous Cities) to ensure coverage across key demographic and geographic subgroups. A disproportional stratified sampling approach was used to ensure sufficient sample sizes in each age/sex/region cell, and population weights were applied in the analyses to reflect the distribution of gender, age, and region in the Spanish population.” L 282-290

Comment 1.3: The authors note a sample size of 3000—how many people were contacted in order to get this completed number? What was the response rate?

Response 1.3: We appreciate this observation. We have included this information in section 3.1. Sampling Strategy and Data Collection as follows: “To achieve the final sample of 3,000 completed interviews, a total of 23,614 telephone numbers were contacted, yielding a response rate of 12.7%. The majority of non-completions were due to non-contacts (61.8% no answer), while refusals represented 2.8% of all calls. These outcomes are consistent with response patterns commonly observed in large-scale telephone surveys.” L 290-295

Comment 1.4: Given their role in the analyses, all other measures should be included in the methods section. While gender, age and education seem relatively straightforward, how was social class measured?

Response 1.4: Social class was measured via self-placement in five categories (upper/upper‑middle; middle; lower‑middle; working; lower/poor). We have revised and included this clarification in Data analysis section: “Finally, hierarchical regression models were applied to evaluate the relative contributions of gender, education, and subjective social class (measured via a five-point self-placement scale), followed by the incremental influence of age—entered as both a linear and a quadratic term—on eco-anxiety, environmental activism, and pro-environmental behavior.” L 334-338. Moreover, we have consistently referred to this measure as “subjective social class” throughout the manuscript for clarity and ease of reading.

Comment 1.5: Throughout the article, the authors make a big deal of the notion of “self-efficacy” and they even phrase H3 with this notion. For a while, I assumed that this was also measured in the survey, especially in order to actually test H3. However, it is not measured, but a decline in self-efficacy is assumed to occur with age. This may need greater explanation. H3 should be written to reflect what is actually tested.

Response 1.5: Thank you very much for your insightful comment. As you indicated, the original version of the study did not make clear how self-efficacy was operationalised. We have now revised the original hypotheses of the study and refined them. Specifically, H3 has been rephrased to reflect what is actually measured in the survey: perceived efficacy (both personal and collective) is positively associated with environmental activism and pro-environmental behavior, and it partially mediates age differences in these outcomes. The revised hypothesis now reads: “H3: Perceived efficacy—personal and collective—is positively associated with environmental activism and pro-environmental behavior, and it partially mediates age differences.” L 229-231. Accordingly, we have complemented the Survey Instruments section of the manuscript. 

 

Reviewer 2 Report

Comments and Suggestions for Authors

Comments to the Authors:

I commend your work on the manuscript “Generational Climate Engagement in Liquid Modernity: Eco-Anxiety, Environmental Activism and Pro-Environmental Behaviour Among Older Adults in Spain.” The study is timely, interdisciplinary, and conceptually sophisticated, offering important contributions to the intersection of climate psychology, ageing studies, and sociocultural theory. Life-course analysis, combined with large-scale survey data and robust psychometric validation, makes this a valuable and original addition to the literature.

The manuscript is fluently written, theoretically grounded, and empirically rigorous. However, some areas could benefit from clarification, refinement, and further development to maximise their scholarly impact. The comments below are meant to support further enhancing this contribution.

General Comments

• Ensure consistent terminology across sections (e.g., “older adults,” “seniors,” “late life,” etc.) and clearly define age groupings where applicable.
• Improve transitions between theoretical sub-sections and between analyses in the results, especially when shifting from descriptive to inferential findings.
• Consider slight reductions in sentence length in more analytically dense sections (particularly the theoretical framework and discussion) to improve readability.
• Ensure all figures and tables are clearly labelled, fully described in the narrative, and formatted to publication standards.

Specific Comments

Introduction:

• The introduction is strong and well-contextualised. It articulates a clear research gap and justifies the life-course focus effectively.
• Consider reducing minor repetition (e.g., “may exhibit comparatively lower levels of climate anxiety…” appears twice in the same sentence).
• While the introduction mentions a "CATI survey of 3,000 residents aged 18 and older," it would be helpful to clarify at the outset that the primary analytic focus is on older adults, even though younger participants are included. This distinction could help sharpen the research objectives and better manage reader expectations.
• The term "older adults" is central to the study, but its operational definition (e.g., 60+, 65+) is not provided in the introduction. Including this detail would improve clarity, particularly for readers from disciplines where age categorisations vary.
• While the focus on Spain is appreciated and relevant, the introduction would benefit from a brief justification for why Spain, in particular, is a meaningful or revealing case for exploring the intersection of eco-anxiety and ageing. For instance, is there something unique about Spain's demographic, environmental, or political context?
• The invocation of liquid modernity is compelling, though the concept could be slightly more concretely linked to environmental behaviours earlier in the introduction.

Theoretical Framework:

• The structure is comprehensive and conceptually rich. However, the relationships between the theories (e.g., VBN model, Activity Theory, Theory of Planned Behaviour) could be more clearly synthesised in a concluding paragraph.
• While each model is introduced appropriately in its respective subsection, their relationships could be more explicitly synthesised. For instance, how might the Activity Theory and VBN model intersect in predicting activism among older adults? Or how does the Theory of Planned Behaviour interact with the eco-anxiety catalyst/inhibitor model? A concluding paragraph tying these frameworks together would enhance coherence and analytical strength.
• The section rightly acknowledges the challenge of distinguishing cohort from life-stage effects. However, further elaboration on how the study design attempts to disentangle these (or if it doesn't, how it controls for this limitation) would clarify the theoretical stakes. This is particularly relevant to H1 and H3, which implicitly rely on temporal assumptions.
• The discussion of obstacles to activism (e.g., the digital divide, mobility limitations) is compelling but might benefit from brief engagement with structural-environmental justice literature. Including a reference or two on how infrastructural inequalities shape opportunities for action would broaden the scope of the argument and strengthen the theoretical grounding.
• The term "pro-environmental behaviour" is rightly described as encompassing a range of actions. Still, it would be useful to more clearly differentiate “low-cost symbolic behaviours” from “high-cost structural ones” in theoretical terms, perhaps referencing Kollmuss & Agyeman (2002) or other foundational models. This would provide a stronger conceptual bridge to the hypotheses in 2.4.
• The emphasis on moral responsibility and legacy is compelling. However, there may be some benefit in acknowledging that not all older adults frame their environmental engagement in this way. Briefly noting ideological or cultural variability would introduce helpful complexity.
• Each subsection is internally coherent. However, the transitions between 2.1–2.2 and 2.2–2.3 could be improved with short framing sentences that explain how the dimensions interrelate (e.g., from emotional distress to action to behaviour).
• The writing is scholarly and fluent, but a few long sentences could be streamlined. For example, in 2.3, the sentence “In everyday practice, these logics intersect…” could be split for easier digestion.

Materials and Methods:

• Sampling design and CATI methodology are appropriate and well-explained. It would strengthen the study by including the survey response rate (e.g., completed interviews vs. total contacts attempted) and how missing data were handled.
• The adaptation and validation of scales are robust and well documented. A 0–10 range for all items is justified, though a brief explanation of how this affects comparability with the original scales would be helpful.
• The psychometric reporting is excellent, particularly regarding Cronbach’s alpha values and item-total correlations.
• Minor clarification needed on whether back-translation or other procedures were used to ensure semantic equivalence in scale translation.
• Ensure scale names are italicised when first introduced (e.g., Eco-Anxiety Scale).

Data Analysis

• Indicate whether any variables were transformed (e.g., normalised or log-transformed) for regression assumptions.
• Consider stating how missing values were treated (e.g., listwise deletion, imputation) to clarify data handling procedures.
• Report the specific assumptions tested for the hierarchical regression (e.g., multicollinearity, homoscedasticity), as this would increase transparency.

Results:

• Results are structured, with excellent interpretation of life-course patterns. The use of five-year brackets provides high granularity.
• Figures and tables (referenced but not shown here) should be checked for alignment with the text. Make sure all statistical values are reported with significance levels and effect sizes.
• Clarify whether statistical tests (e.g., ANOVAs or t-tests) were corrected for multiple comparisons. This would strengthen confidence in reported significance.
• The hierarchical regression models are well-explained. It would be useful to briefly state whether the variance inflation factor (VIF) or other multicollinearity checks were performed, especially since quadratic terms are included.
• The “partial decoupling” insight is powerful. A diagram or conceptual summary (e.g., path model or summary matrix) might help readers visualise how eco-anxiety, activism, and behaviour diverge and overlap across age groups.
• Emphasising the cross-analysis of the three constructs is a major strength of this section—consider making this more prominent in the summary.

Discussion:

• Excellent integration of empirical results with theory. The “partial decoupling” discussion between emotion, activism, and behaviour is conceptually strong.
• The cultural framing, particularly through Bauman’s concept of liquid modernity, is innovative and lends the study a distinctive voice. Consider explicitly linking this framing to potential implications for intervention or policy.
• Consider structuring each paragraph of the discussion around the three hypotheses to clarify alignment and highlight divergences.
• Expand briefly on the practical implications of the findings for climate health policy, community programs, or intergenerational engagement.
• The sentence “Finally, it is needed to investigate…” should be revised for grammar (e.g., “Finally, it is necessary to investigate…” or “Further research should explore…”).
• It may be helpful to suggest potential qualitative complements to this quantitative design (e.g., life-course interviews or diary studies).

Conclusions:

• The conclusion is strong and resonates well with the rest of the manuscript. The characterisation of older adults as “societal anchors” is especially powerful.
• Consider emphasising implications for climate resilience planning, civic inclusion, and public communication for older adults.
• The recommendations for future research are clear and well justified, especially the call for longitudinal and mixed-method designs.
• Minor revisions to sentence structure would enhance impact (e.g., “not only improves the well-being of older people, but also strengthens...” should remove “but also” to match “not only”).

Author Response

Author's Reply to the Review Report (Reviewer 2)

Comment 2.1: Comments to the Authors: I commend your work on the manuscript “Generational Climate Engagement in Liquid Modernity: Eco-Anxiety, Environmental Activism and Pro-Environmental Behaviour Among Older Adults in Spain.” The study is timely, interdisciplinary, and conceptually sophisticated, offering important contributions to the intersection of climate psychology, ageing studies, and sociocultural theory. Life-course analysis, combined with large-scale survey data and robust psychometric validation, makes this a valuable and original addition to the literature.

The manuscript is fluently written, theoretically grounded, and empirically rigorous. However, some areas could benefit from clarification, refinement, and further development to maximise their scholarly impact. The comments below are meant to support further enhancing this contribution.

Response 2.1: Thank you very much for your kind words and help in improving our work. We believe that your insightful and constructive comments have contributed to improving the clarity and overall quality of our contribution. Below, we explain how we have addressed all your comments. In our replies, we mention the page number and lines that have been changed in the marked-up manuscript version.

Comment 2.2: General Comments

• Ensure consistent terminology across sections (e.g., “older adults,” “seniors,” “late life,” etc.) and clearly define age groupings where applicable.
• Improve transitions between theoretical sub-sections and between analyses in the results, especially when shifting from descriptive to inferential findings.
• Consider slight reductions in sentence length in more analytically dense sections (particularly the theoretical framework and discussion) to improve readability.
• Ensure all figures and tables are clearly labelled, fully described in the narrative, and formatted to publication standards.

Response 2.2: Thank you for your constructive general comments. We have carefully reviewed the manuscript to address the issues you raised. As suggested, we have harmonized the terminology throughout the text, consistently using "older adults" to ensure clarity. To provide a more precise demographic focus, we have defined the age threshold (≥65 years) in the Introduction. We have also enhanced the flow of the manuscript by adding smooth transitions between theoretical sub-sections and between the descriptive and inferential results. Additionally, we have verified the alignment of all figures and tables to improve the visual presentation. The specific changes made in response to your detailed comments are addressed individually below.

Comment 2.3: Specific Comments, Introduction: The introduction is strong and well-contextualised. It articulates a clear research gap and justifies the life-course focus effectively.

Response 2.3: Thank you very much for the positive comment.

Comment 2.4: Specific Comments, Introduction:  Consider reducing minor repetition (e.g., “may exhibit comparatively lower levels of climate anxiety…” appears twice in the same sentence).

Response 2.4: Thank you for noticing that. We have removed the repeated sentence.

Comment 2.5: Specific Comments, Introduction: While the introduction mentions a "CATI survey of 3,000 residents aged 18 and older," it would be helpful to clarify at the outset that the primary analytic focus is on older adults, even though younger participants are included. This distinction could help sharpen the research objectives and better manage reader expectations.

Response 2.5: We appreciate your comment. We have revised and reworded the sentences to  clarify this issue. The new sentences are the following: “The present study—based on a CATI survey of 3,000 residents aged 18 and older—aims to address this gap by exploring the evolution of eco-anxiety, environmental activism, and pro-environmental behavior from a life course perspective.  Although the survey includes adults of all ages, the primary analytic focus is on older adults (aged ≥65 years), with particular emphasis on the later life segment.” L 97-102.

Comment 2.6: Specific Comments, Introduction:  The term "older adults" is central to the study, but its operational definition (e.g., 60+, 65+) is not provided in the introduction. Including this detail would improve clarity, particularly for readers from disciplines where age categorisations vary.

Response 2.6: Please refer to response 2.5 above.

Comment 2.7: Specific Comments, Introduction:  While the focus on Spain is appreciated and relevant, the introduction would benefit from a brief justification for why Spain, in particular, is a meaningful or revealing case for exploring the intersection of eco-anxiety and ageing. For instance, is there something unique about Spain's demographic, environmental, or political context?

Response 2.7: Thank you for the comment. We have included in the Introduction a brief justification of the relevance of Spain as a case study: “Spain provides a pertinent context for examining aging and climate engagement, given its rapidly aging population, recurrent heatwaves and droughts, and strong tradition of local civic participation. These characteristics make the country particularly informative for studying how these phenomena manifest across the adult life course, with a focus on older adults.” L 93-97.

Comment 2.8: Specific Comments, Introduction:  The invocation of liquid modernity is compelling, though the concept could be slightly more concretely linked to environmental behaviours earlier in the introduction.

Response 2.8: We thank the reviewer for this insightful comment. In response, we have revised the Introduction to more concretely link the concept of liquid modernity to environmental behaviors earlier in the text. Specifically, we now highlight how the characteristics of liquid modern societies—such as uncertainty, individualization, and fragile social frameworks—can shape perceptions of environmental risk and influence eco-anxiety, environmental activism, and pro-environmental behavior. “The life course is not merely a biological sequence; it also involves social trajectories, shared historical milestones, and culturally embedded repertoires of meaning that shape how individuals at different life stages perceive and respond to environmental threats [4]. As Bauman [5] suggests, we inhabit a condition of liquid modernity, in which frameworks of meaning and social certainties are increasingly fragile. These conditions influence how people perceive environmental risks and engage with sustainability issues, shaping eco-anxiety, environmental activism, and pro-environmental behavior. In this context, generational interpretations of environmental risk take on particular importance, especially as traditional ways of understanding humanity’s relationship with nature and the viability of a shared future become increasingly fragile and contested. (…)” L 57-66

Comment 2.9: Theoretical Framework: The structure is comprehensive and conceptually rich. However, the relationships between the theories (e.g., VBN model, Activity Theory, Theory of Planned Behaviour) could be more clearly synthesised in a concluding paragraph.

While each model is introduced appropriately in its respective subsection, their relationships could be more explicitly synthesised. For instance, how might the Activity Theory and VBN model intersect in predicting activism among older adults? Or how does the Theory of Planned Behaviour interact with the eco-anxiety catalyst/inhibitor model? A concluding paragraph tying these frameworks together would enhance coherence and analytical strength.

Response 2.9: Thank you for your constructive and insightful comment. We have included a concluding paragraph integrating comments 2.9 and 2.11: “Taken together, these frameworks provide a coherent understanding of climate engagement in older adults. The Activity Theory of Aging highlights how participation in meaningful activities reinforces identity and well-being, the VBN model emphasizes moral and value-driven motivations, and the Theory of Planned Behavior focuses on attitudes, norms, and perceived control. From an environmental justice perspective, infrastructural inequalities—such as limited access to digital technologies or transportation—can further constrain opportunities to act [22]. Together, these frameworks illustrate how psychological, moral, social, and structural factors interact to shape eco-anxiety, environmental activism, and pro-environmental behavior across the adult life course.” L 200-209

• Schlosberg, D. Reconceiving Environmental Justice: Global Movements and Political Theories. Environmental Politics, 2004, 13(3), 517–540. https://doi.org/10.1080/0964401042000229025

Comment 2.10: Theoretical Framework: The section rightly acknowledges the challenge of distinguishing cohort from life-stage effects. However, further elaboration on how the study design attempts to disentangle these (or if it doesn't, how it controls for this limitation) would clarify the theoretical stakes. This is particularly relevant to H1 and H3, which implicitly rely on temporal assumptions.

Response 2.10: Thank you for this valuable comment. We agree that clarifying the distinction between cohort and life-stage effects is essential for a nuanced interpretation of our findings. We have therefore expanded the Limitations section to explicitly acknowledge this issue. The new text reads: “The cross-sectional nature of the survey data constrains our ability to establish causality and prevents a complete disentanglement of age, period, and cohort effects. Longitudinal designs would be required to examine these dimensions more systematically. To mitigate this limitation, we employed hierarchical regression models that estimate the incremental contribution of age once socio-demographic variables are controlled. While this strategy does not eliminate potential cohort effects, it reduces confounding and enhances the robustness of our life-course interpretations.” L 586-592

Comment 2.11: Theoretical Framework: The discussion of obstacles to activism (e.g., the digital divide, mobility limitations) is compelling but might benefit from brief engagement with structural-environmental justice literature. Including a reference or two on how infrastructural inequalities shape opportunities for action would broaden the scope of the argument and strengthen the theoretical grounding.

Response 2.11: It is done. Please refer to response 2.9 above.

Comment 2.12: Theoretical Framework: The term "pro-environmental behaviour" is rightly described as encompassing a range of actions. Still, it would be useful to more clearly differentiate “low-cost symbolic behaviours” from “high-cost structural ones” in theoretical terms, perhaps referencing Kollmuss & Agyeman (2002) or other foundational models. This would provide a stronger conceptual bridge to the hypotheses in 2.4.

Response 2.12: We thank the reviewer for this helpful suggestion. In response, we have clarified the distinction between low-cost symbolic behaviors and high-cost structural behaviors within the theoretical framework (section 2.3). Specifically, we now reference Kollmuss & Agyeman (2002) to emphasize that low-cost actions, such as recycling or reducing plastic use, are primarily accessible and habitual, whereas high-cost structural behaviors, like adopting low-meat diets, installing solar panels, or investing in ethical funds, require greater resources, planning, and perceived efficacy.  Please see L 168-199

• Kollmuss, A.; Agyeman, J. Mind the Gap: Why Do People Act Environmentally and What Are the Barriers to Pro Environmental Behavior? Environmental Education Research 2002, 8(3), 239–260. https://doi.org/10.1080/13504620220145401.

Comment 2.13: Theoretical Framework: The emphasis on moral responsibility and legacy is compelling. However, there may be some benefit in acknowledging that not all older adults frame their environmental engagement in this way. Briefly noting ideological or cultural variability would introduce helpful complexity.

Response 2.13: Thank you for the comment. We have revised the text and incorporate a new sentence to explain that as follows “The Theory of Planned Behavior [19] places perceived control at its core; in older adults, this perception may be weakened by fixed incomes, mobility limitations, or technological barriers. Meanwhile, from a generational context perspective, older adult pro-environmental behavior is often interpreted as an expression of legacy and moral responsibility toward descendants. This framework helps explain the persistence of conservationist habits even as physical capacities decline. However, it is important to also recognize that this is not the only factor at play, as ideological and cultural variability, along with personal experiences, can also significantly influence how individuals from this cohort relate to environmental issues, adding a layer of complexity to the phenomenon.” L 177-186

Comment 2.14: Theoretical Framework: Each subsection is internally coherent. However, the transitions between 2.1–2.2 and 2.2–2.3 could be improved with short framing sentences that explain how the dimensions interrelate (e.g., from emotional distress to action to behaviour).

Response 2.14: We appreciate your comment. We have incorporated two new sentences to improve the transitions between sections.

• Section 2.1 – 2.2: “These emotional responses can serve as a powerful impetus for collective action, a topic we explore in the following section on environmental activism.”. L 132-134
• Section 2.2 – 2.3: “Beyond formal activism, these motivations and resources also shape a wide range of everyday pro-environmental behaviors”. L 165-167

Comment 2.15: Theoretical Framework:

The writing is scholarly and fluent, but a few long sentences could be streamlined. For example, in 2.3, the sentence “In everyday practice, these logics intersect…” could be split for easier digestion.

Response 2.15: We appreciate your suggestion. We have revised and rewritten the sentence to make it easier to read. The new sentences are the following: “In everyday practice, these logics intersect. For example, an older adult with moderate eco-anxiety and high self-efficacy may choose to engage in high-cost structural behaviors, such as installing solar panels—an action with a high initial cost but tangible long-term benefits. In contrast, another individual with high anxiety but low perceived control may limit their contribution to low-cost symbolic behaviors”. L 191-196

Comment 2.16: Materials and Methods: Sampling design and CATI methodology are appropriate and well-explained. It would strengthen the study by including the survey response rate (e.g., completed interviews vs. total contacts attempted) and how missing data were handled.

Response 2.16: We appreciate this observation. We have detailed the response rate in section 3.1. Sampling Strategy and Data Collection as follows: “To achieve the final sample of 3,000 completed interviews, a total of 23,614 telephone numbers were contacted, yielding a response rate of 12.7%. The majority of non-completions were due to non-contacts (61.8% no answer), while refusals represented 2.8% of all calls. These outcomes are consistent with response patterns commonly observed in large-scale telephone surveys.” L 249-254. In addition, we have explained how missing data were handled in the Data Analysis section “Cases with missing data on the variables included in the models were excluded, resulting in a total of  92 cases removed from the analysis (n = 2,908 valid cases).” L 346-347.

Comment 2.17: Materials and Methods: The adaptation and validation of scales are robust and well documented. A 0–10 range for all items is justified, though a brief explanation of how this affects comparability with the original scales would be helpful.

Response 2.17: Thank you for your comment. We have included a brief explanation in the revised version of the manuscript as follows: “The uniform 0–10 range was chosen for consistency and ease of use, and while this may affect direct comparisons with original scales, the instruments' psychometric properties remained robust, ensuring their validity for this study. ” L 295-297

Comment 2.18: Materials and Methods: The psychometric reporting is excellent, particularly regarding Cronbach’s alpha values and item-total correlations.

Response 2.18: Thank you very much for the positive comment.

Comment 2.19: Materials and Methods: Minor clarification needed on whether back-translation or other procedures were used to ensure semantic equivalence in scale translation.

Response 2.19: Thank you for your valuable feedback. We have added a clarification to the Methods section to address this point. The new text specifies that we used a rigorous translation process that included back-translation by our research team and validation by bilingual Spanish-English speakers to ensure the scales' semantic equivalence. This procedure guarantees that the Spanish versions of the scales accurately retained the meaning of the originals, thereby strengthening the methodological rigor and validity of our study. “The scales were administered following a strict protocol that included obtaining informed consent from participants and ensuring their anonymity. Each instrument was presented in a translated and culturally adapted version, thereby confirming its cultural validity. Linguistic adaptations followed established translation guidelines, including a rigorous back-translation process conducted by our research team and subsequently validated by bilingual Spanish-English speakers. All methodological decisions aimed to guarantee the reliability and reproducibility of the collected data.” L 265-270

Comment 2.20: Materials and Methods: Ensure scale names are italicised when first introduced (e.g., Eco-Anxiety Scale).

Response 2.20: Thank you for the suggestion. We have revised the text to ensure that the names of the three scales used in the study are in italics the first time they are mentioned.

Comment 2.21: Data Analysis: Indicate whether any variables were transformed (e.g., normalised or log-transformed) for regression assumptions.

Response 2.21: Thank you for your constructive feedback. We have expanded the Data Analysis section to clarify and other details associated with comment 2.23. The new sentences are the following: “We verified standard assumptions, including homoscedasticity and normality of residuals, and residual plots suggested no serious departures from these conditions. Therefore, no transformations of the variables were necessary to meet these assumptions. In addition, we conducted diagnostic checks for multicollinearity, with all Variance Inflation Factor (VIF) values falling below the conventional threshold of 5, indicating that the predictor variables were not excessively correlated. ”. L 340-346

Comment 2.22: Data Analysis: Consider stating how missing values were treated (e.g., listwise deletion, imputation) to clarify data handling procedures.

Response 2.22: Thank you for the suggestion. We have clarified this in the Data Analysis section: “Cases with missing data on the variables included in the models were excluded, resulting in a total of 92 cases removed from the analysis (n = 2,908 valid cases)”. L 346-347

Comment 2.23: Data Analysis:  Report the specific assumptions tested for the hierarchical regression (e.g., multicollinearity, homoscedasticity), as this would increase transparency.

Response 2.23: Thank you for this helpful recommendation. As detailed in comment 2.21, we have expanded the Data Analysis section to specify the assumptions tested for the hierarchical regression models. The new text reads: “We verified standard assumptions, including homoscedasticity and normality of residuals, and residual plots suggested no serious departures from these conditions. Therefore, no transformations of the variables were necessary to meet these assumptions. In addition, we conducted diagnostic checks for multicollinearity, with all Variance Inflation Factor (VIF) values falling below the conventional threshold of 5, indicating that the predictor variables were not excessively correlated. ” L 340-346

Comment 2.24: Results: Results are structured, with excellent interpretation of life-course patterns. The use of five-year brackets provides high granularity.

Response 2.24: Thank you very much for the positive comment.

Comment 2.25: Results:  Figures and tables (referenced but not shown here) should be checked for alignment with the text. Make sure all statistical values are reported with significance levels and effect sizes.

Response 2.25: Thank you for your valuable feedback. We have thoroughly revised the manuscript to ensure all figures and tables are properly aligned with the text. We have also verified that all statistical values in the manuscript and tables are reported with their corresponding significance levels (p-values) and effect sizes (η2). This information is  presented in Table 3, where the ANOVA results are summarized, and within the text of the Results section, ensuring full transparency and clarity.

Comment 2.26: Clarify whether statistical tests (e.g., ANOVAs or t-tests) were corrected for multiple comparisons. This would strengthen confidence in reported significance.

Response 2.26: Thank you for this valuable comment. We agree that clarifying the treatment of multiple comparisons is important to strengthen confidence in our results. We have therefore expanded the Data Analysis section with the following sentence: “To control for family-wise error, we applied the Holm–Bonferroni correction to all p-values derived from ANOVAs and t-tests.” L 334-336

Comment 2.27: Results: The hierarchical regression models are well-explained. It would be useful to briefly state whether the variance inflation factor (VIF) or other multicollinearity checks were performed, especially since quadratic terms are included.

Response 2.27: Please refer to response 2.23 above.

Comment 2.28: Results:  The “partial decoupling” insight is powerful. A diagram or conceptual summary (e.g., path model or summary matrix) might help readers visualise how eco-anxiety, activism, and behaviour diverge and overlap across age groups.

Response 2.28: We appreciate this insightful suggestion. To enhance the clarity of our “partial decoupling” finding, we added a conceptual summary Table (Table 2) that synthesizes how eco-anxiety, environmental activism, and pro-environmental behavior evolve and interact across the life course. This complements Figure 1 by offering a more integrative view of points of convergence and divergence among the three constructs.  L 390

Comment 2.29: Results:  Emphasising the cross-analysis of the three constructs is a major strength of this section—consider making this more prominent in the summary.

Response 2.29: Thank you for your valuable comment. We agree that emphasizing the cross-analysis of the three constructs is a major strength of our findings and should be more prominent. To address this, we have revised the Results section to explicitly highlight this key insight in the summary. The new text states: "The simultaneous comparison of the three constructs reveals patterns of disconnection between emotion and action in older adults. There is a significant age-related partial decoupling among eco-anxiety, environmental activism, and pro-environmental behavior—a key insight that merits closer inspection. (…)" L 418-421

Comment 2.30: Discussion: Excellent integration of empirical results with theory. The “partial decoupling” discussion between emotion, activism, and behaviour is conceptually strong.

Response 2.30: Thank you very much for the positive comment.

Comment 2.31: Discussion: The cultural framing, particularly through Bauman’s concept of liquid modernity, is innovative and lends the study a distinctive voice. Consider explicitly linking this framing to potential implications for intervention or policy.

Response 2.31: Thank you for your valuable feedback. We have revised the Discussion section to address this. We have added a new sentence that directly connects Bauman's concept of liquid modernity to the practical implications of our findings as follows: “The findings call for policies that counter the atomization of liquid modernity [5] by supporting community-based and intergenerational initiatives, which could provide older adults with the stable social context needed to translate concern into sustained action”. L 549-552

Comment 2.32: Discussion: Consider structuring each paragraph of the discussion around the three hypotheses to clarify alignment and highlight divergences.

Response 2.32: Thank you for your valuable feedback. We agree that structuring the discussion around the hypotheses would significantly improve the clarity and impact of the manuscript. To address your suggestion, we have reorganized the Discussion section: 1) Each subsection now has a heading that directly references the corresponding hypothesis (e.g., "4.1. H1: Eco-Anxiety and the Life Course"): 2) the first sentence of each paragraph now explicitly states whether our findings confirm or diverge from the hypothesis; and, 3) we also have slightly reordered some sentences and ideas within each subsection to create a more cohesive flow. Please see the Discussion section (L 499-607).

Comment 2.33: Discussion:  Expand briefly on the practical implications of the findings for climate health policy, community programs, or intergenerational engagement.

Response 2.33: We thank the reviewer for this valuable suggestion. Following the comment, we have expanded the discussion to briefly address the practical implications of our findings for climate health policy, community programs, and intergenerational engagement. Specifically, we have added the following paragraph: “From a climate health policy perspective, these findings suggest the importance of integrating mental health screening, community resilience, and climate literacy into national and local adaptation plans. By recognizing eco-anxiety as both a health and social issue, policies can support preventive interventions and resource allocation that strengthen older adults’ agency. For community programs, the evidence underscores the value of intergenerational platforms where seniors contribute lived ecological knowledge, while also benefiting from peer support and collective efficacy. Finally, in terms of intergenerational engagement, structured mentorship schemes and climate storytelling initiatives can function as low-cost, high-impact strategies that not only foster pro-environmental behavior but also enhance well-being, social cohesion, and trust between generations.” L 575-582

Comment 2.34: Discussion:  The sentence “Finally, it is needed to investigate…” should be revised for grammar (e.g., “Finally, it is necessary to investigate…” or “Further research should explore…”).

Response 2.34: Thank you for the comment here. We have revised and rewritten the sentence as follows: “Finally, further research should explore how socioeconomic inequalities influence the age–climate relationship—particularly in contexts of energy poverty and unequal exposure to extreme events—and to compare Spain’s situation with that of other middle- and low-income countries, in order to develop a truly inclusive perspective on the ecological transition”. L 602-606

Comment 2.35: Discussion: It may be helpful to suggest potential qualitative complements to this quantitative design (e.g., life-course interviews or diary studies).

Response 2.35: Thank you for your valuable comment. We fully agree that suggesting qualitative complements is an important way to strengthen our discussion on future research. As you suggested, we have revised the text to explicitly include this recommendation, stating that "Future research may benefit from combining this quantitative approach with qualitative methods, such as in-depth interviews or diary studies, to gain a deeper understanding of personal motivations and lived experiences." L 597-600.

Comment 2.36: Conclusions: The conclusion is strong and resonates well with the rest of the manuscript. The characterisation of older adults as “societal anchors” is especially powerful.

Response 2.36: Thank you very much for your support.

Comment 2.37: Conclusions: Consider emphasising implications for climate resilience planning, civic inclusion, and public communication for older adults.

Response 2.37: Thank you for your constructive feedback. We agree that emphasizing the practical implications for climate resilience planning, civic inclusion, and public communication is crucial to strengthening our conclusions. To address this, we have revised the conclusions and included this sentence: “This stabilizing role holds significant implications for climate resilience planning, civic inclusion, and public communication, as supporting older adults’ contributions can strengthen community-wide responses to climate disruption.” L 622-626

Comment 2.38: Conclusions: The recommendations for future research are clear and well justified, especially the call for longitudinal and mixed-method designs.

Response 2.38: Thank you very much for the positive comment.

Comment 2.39: Conclusions: Minor revisions to sentence structure would enhance impact (e.g., “not only improves the well-being of older people, but also strengthens...” should remove “but also” to match “not only”).

Response 2.39: We appreciate your comment. We have revised and rewritten the sentence to facilitate readability while also addressing comment 2.37. The new sentence is: “This stabilizing role holds significant implications for climate resilience planning, civic inclusion, and public communication, as supporting older adults’ contributions can strengthen community-wide responses to climate disruption.” L 622-626

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have responded nicely to my prior comments, especially those involving information about sampling/sample size and the measurement of social class.

I still find H3 problematic. H1 and H2 both state life course expectations for two of the key variables, eco-anxiety and environmental activism. It would make sense to have H3 present a parallel statement about changes in pro-environmental behavior across the life course. The inclusion of perceived efficacy in the hypothesis is still confusing—this was not measured directly but they note that it was embedded in the measure of pro-environmental behavior (line 272). So, the hypothesis essentially says that the measure of pro-environmental behavior (which includes the efficacy idea) is related to itself (as well as environmental activism).

In addition to fixing the wording of H3, it may be useful for the authors to include an H4 which states their expectations regarding how the three concepts/measures will interrelate (after all, they DO discuss this in the results). This is currently somewhat included in H3 but, as noted above, H3 is confusing as written.

Also, without more information about the measure of pro-environmental behavior, readers cannot evaluate how well the measure gets at both behaviors as well as efficacy issues as the authors state. Relatedly, while the Hogg scale is used exactly as published so people can readily find it, the authors note that the measures of environmental activism and pro-environmental behavior “draws inspiration from” (line 261) or are “based on” (line 268) the work of others. It would be very helpful for readers to have ready access to the actual measures that were used (providing an appendix would be very helpful so that readers don’t have to reach out to the authors for this information as noted in line 244).

One minor note: The authors may wish to clarify what they intend by stating that some activities are “symbolic” (e.g., line 157). Do they mean that such activities have no effect or perhaps are more limited in effect than others? Or do they mean they are meant to be openly visible to others? Or…?

Author Response

Author's Reply to the Review Report (Reviewer 1- Round 2)

Comment 1.1: The authors have responded nicely to my prior comments, especially those involving information about sampling/sample size and the measurement of social class.

Response 1.1: Thank you very much for your kind words and help in improving our work. We believe that your insightful and constructive comments have contributed to improving the clarity and overall quality of our contribution. Below, we explain how we have addressed all your comments. In our replies, we mention the lines that have been changed in the marked-up manuscript version.

Comment 1.2: I still find H3 problematic. H1 and H2 both state life course expectations for two of the key variables, eco-anxiety and environmental activism. It would make sense to have H3 present a parallel statement about changes in pro-environmental behavior across the life course. The inclusion of perceived efficacy in the hypothesis is still confusing—this was not measured directly but they note that it was embedded in the measure of pro-environmental behavior (line 272). So, the hypothesis essentially says that the measure of pro-environmental behavior (which includes the efficacy idea) is related to itself (as well as environmental activism).

Response 1.2: Thank you for your constructive feedback. We have revised the manuscript to align H3 with H1 and H2. The reframed hypothesis now reads: "Pro-environmental behavior follows a curvilinear trajectory across the life course, increasing through midlife and declining in older age, possibly due to the increasing influence of practical barriers (e.g., health, income, mobility, and digital barriers)." L 384-387.

Comment 1.3: In addition to fixing the wording of H3, it may be useful for the authors to include an H4 which states their expectations regarding how the three concepts/measures will interrelate (after all, they DO discuss this in the results). This is currently somewhat included in H3 but, as noted above, H3 is confusing as written.

Response 1.3: Thank you for the suggestion to make our expectations about the interrelationships explicit. We now add H4 to anticipate the associations among the three constructs, in line with the analyses reported in the Results:
“Additionally, we formulate a fourth hypothesis to articulate the expected associations between these constructs. H4: The three key constructs—eco-anxiety, environmental activism, and pro-environmental behavior—are positively associated with one another: H4a. Eco-anxiety is positively associated with environmental activism; H4b. Eco-anxiety is positively associated with pro-environmental behavior, though this association is weaker than for activism (partial decoupling); H4c. Environmental activism is positively associated with pro-environmental behavior.” L 388. Accordingly, we have made adjustments throughout the manuscript to ensure alignment with these hypotheses, specifically:

• Clarified the corresponding analysis in the Methods section (L351–354).
• Added a new subsection in the Results section, including Table 4, to present the findings related to this hypothesis clearly and systematically (L417–439).
• Included a new subsection in the Discussion section to address these results (L601–667).

Comment 1.4:  Also, without more information about the measure of pro-environmental behavior, readers cannot evaluate how well the measure gets at both behaviors as well as efficacy issues as the authors state. Relatedly, while the Hogg scale is used exactly as published so people can readily find it, the authors note that the measures of environmental activism and pro-environmental behavior “draws inspiration from” (line 261) or are “based on” (line 268) the work of others. It would be very helpful for readers to have ready access to the actual measures that were used (providing an appendix would be very helpful so that readers don’t have to reach out to the authors for this information as noted in line 244).

Response 1.4: Thank you for your suggestion. We agree that readers should be able to evaluate the content of our measures. We have therefore expanded the method section to clarify the measures of environmental activism and pro-environmental behavior by providing all items in Appendix A.

Comment 1.5: One minor note: The authors may wish to clarify what they intend by stating that some activities are “symbolic” (e.g., line 157). Do they mean that such activities have no effect or perhaps are more limited in effect than others? Or do they mean they are meant to be openly visible to others? Or…?

Response 1.5: We thank the reviewer for this helpful comment. In the manuscript, “symbolic” behaviors refer to low-cost, straightforward actions, such as recycling, reducing plastic use, or adjusting the thermostat. While their direct environmental impact is smaller than that of higher-cost structural behaviors, they can generate indirect effects through norm signaling and diffusion (e.g., public commitments, visible pledges) and may also serve social or identity functions. This distinction has been clarified in the revised manuscript: “Pro-environmental behavior ranges from low-cost, straightforward symbolic behaviors —such as recycling, reducing plastic use, or adjusting the thermostat—to high-cost structural behaviors including adopting low-meat diets, installing solar panels, or investing in ethical funds [17]. Symbolic behaviors generally have a smaller direct environmental impact than structural behaviors, but can produce indirect effects through norm signaling and diffusion, and may serve social or identity functions.” L 171-177

Round 3

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have addressed all of my comments nicely.

One minor issue that I noticed--there are two "3.2" subsections. Subsections in section 3 will need slight renumbering.

Author Response

Author's Reply to the Review Report (Reviewer 1- Round 3)

Comment 1.1: The authors have addressed all of my comments nicely.

Response 1.1: We sincerely thank the reviewer for the kind words and continuous support in improving our work throughout the review process. We greatly value the insightful and constructive comments, which have significantly contributed to enhancing the clarity and overall quality of our manuscript. Below, we address the remaining minor issue.

Comment 1.2: One minor issue that I noticed--there are two "3.2" subsections. Subsections in section 3 will need slight renumbering.

Response 1.2: We thank the reviewer for carefully checking our revised manuscript. We appreciate the observation regarding the duplicated subsection numbering. The numbering in the sections has been corrected to ensure consistency throughout the manuscript.