Review Reports - Responding to Precarity: Young People’s Ambiguity Aversion, Resilience, and Coping Strategies

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I am attaching the evaluation report. Success to the authors!

Comments for author File: Comments.pdf

Author Response

Comments 1: General assessment

Response 1: With regard to your general assessments, we thank you for your comments.

Comments 2:

Response 2: In relation to points 1-7, we thank you for your comments. Regarding your concern in point 6 to provide evidence of the tools’ validation in the three countries of our sample, we have added new citations in the texts of our Measures section, where we indicated which scales have been previously validated in the Norwegian, Indonesian, and Bangladeshi samples. We acknowledge, however, that not all the scales of our study have been validated in the local contexts, so we have noted this limitation in lines 693-697 of our revised manuscript. Additionally, in relation to your question about the acceptability of using the English version of the scales in our data collection in Bangladesh, scholars have reported Bangladeshis’ use of and proficiency in the English language in public settings (e.g., educational systems; Banu & Sussex, 2001). Considering the English language is widely used in their country (Ara, 2020), we proceeded with our data collection in Bangladesh using the original English versions of the study measures. Again, we recognize the limitation of this approach when the local validation study of the measure is not available.

References:

Ara, R. (2020). A foreign language or the second language: The future of English in Bangladesh. International Journal of Language Education, 4(2), 81-95. https://eric.ed.gov/?id=EJ1249887

Banu, R., & Sussex, R. (2001). English in Bangladesh after independence: Dynamics of policy and practice. In B. Moore (Ed.), Who's centric now? The Present state of post-colonial Englishes (pp. 122-147). Oxford University Press.

Comments 3:

Response 3: We agree with your comment and acknowledge the limitations of using versions of the said scales without previous validation studies in the said countries. As we wrote in Response 2 above, we have provided additional information in our revised manuscript’s Measures section regarding which scales have existing validation studies for each country. We also reported the results of our confirmatory factor analyses in each country and tests for measurement invariance across countries in our manuscript’s supplementary material. However, we still acknowledge in our study limitations (found in lines 693-697 of the updated manuscript) the need for local validation studies to better confirm the psychometric properties of our study scales in each country context.

Comments 4:

Response 4: In our manuscript revision, we have prepared a supplementary material, where we report more information on the reliability coefficients and factor structures of the translated Indonesian measures.

Comments 5:

Response 5: We have considered your comments and suggestions in which we retained our reports of the model fit indices attained for each measure in our study using our overall three-country sample. In our supplementary material, we reported the results of the measurement invariance testing that we carried out in a previous study, where we used the same measures in a different sample that shares similar characteristics as the present study (i.e., young people from Norway, Indonesia, and Bangladesh). Through these analyses, we were able to establish measurement invariance across the three countries.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This study is meaningful in that, based on COR theory, it empirically examines the roles of resilience and two different coping strategies in the context of career precarity. It also adopts an international comparative approach, utilizing samples from three countries, which, although with some limitations, provides an exploratory view of cross-national and cultural differences and offers useful groundwork for future research. Still, the paper could be further strengthened if the following points are addressed.

First, the manuscript describes Ambiguity Aversion as a resource in COR theory, but this does not seem appropriate. Ambiguity Aversion can be better understood not as a resource, but as a factor influenced by both the socio-economic environment and individuals’ character toward ambiguity. Rather than explaining it as a resource, I suggest viewing it as a vulnerability to be more consistent with the COR framework. So Ambiguity Aversion can be understood as a resource deficiency and may increase the likelihood of entering a resource-loss spiral.

This issue will be related to the discussion section. Currently, the manuscript focuses only on avoidance coping without addressing how the two coping strategies differ. Since the study findings show that Ambiguity Aversion influences the two coping strategies in different ways, interpreting these results within the framework of the resource-loss spiral could enrich the discussion.

There are also methodological issues to consider. The authors used PROCESS Model 6 twice, once for each coping strategy. However, if Model 81 had been used, which allows for serial and parallel mediation in a single analysis, both strategies could have been examined together, leading to a more rigorous test. The absence of control variables also requires explanation.

If the authors retain the current approach (running Model 6 twice), it would still be important to include as covariates not only the coping subscale excluded from each model but also demographic variables such as country, gender, and age. Doing so would help rule out potential confounds and provide a clearer and more rigorous set of results.

Author Response

Comments 1:

Response 1: Thank you for your comment.

Comments 2:

Response 2: Thank you for the valuable suggestion. We explicitly framed our contribution within the dual-process model (Xu 2021, 2023), clarifying that this theory defines ambiguity aversion in terms of an individual’s dislike or negative orientation to ambiguity that may prompt individuals to adopt avoidance coping strategies. Additionally, by integrating the principles of the resources loss/gain spiral of COR theory (Hobfoll 1989), we explained how ambiguity aversion could affect career anxiety, through resilience and coping strategies. At the same time, we stated that the adoption of approach coping strategies may be influenced by different "ambiguity containing orientations", as hope and calling. Similarly, we suggested that hope and calling may activate the resource gain spiral associated with taking the initiative in the face of career ambiguity.

Comments 3:

Response 3: We have considered your suggestion and included relevant control variables to each tested relationship in Model 6. Specifically, we included as covariates the alternative coping subscale and demographic factors such as gender, age, and national unemployment rate to minimize potential confounding effects and strengthen the rigor of our analysis.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The subject of precarity and its impact on young people’s career development is significant for the nowadays’ reality. The theoretical background is solid, including relevant references for the analysed subject. I particularly liked the fact that the authors developed research hypotheses based on the literature review, which were subsequently tested in the empirical part.

Regarding the methodological part, there are a few observations. First of all, in the abstract, but also in part 1.6, it is mentioned “serial mediation analyses” and, respectively, “serial mediation model”. Since this method may not be familiar to all readers, a brief description could help. Secondly, the sample size (N = 156) is small for cross-cultural comparisons, fact that is also mentioned the authors in the „limitations” part. Therefore, the authors might be more careful when generalizing their results for the entire populations.

Comments on the Quality of English Language

The paper needs proof-reading.

Author Response

Comments 1:

Response 1: Thank you for your comment.

Comments 2:

Response 2: We have added an explanation about serial mediation analysis in lines 453-455 of the revised manuscript.

Comments 3:

Secondly, the sample size (N = 156) is small for cross-cultural comparisons, fact that is also mentioned the authors in the „limitations” part. Therefore, the authors might be more careful when generalizing their results for the entire populations.

Response 3: We agree with your concern; in fact, we adopted an exploratory and descriptive approach to our study’s cross-country comparisons. In our revision of the paper, we rerun our analyses in which we added some control variables (i.e., age, gender, and national unemployment rate) in our tested models to partially control the potential effects of our unbalanced sample. This approach allowed us also to provide alternative explanations when we commented on the differences between countries. In addition, we have revised the texts of our Discussion section to carefully consider the extent of our study findings’ generalizability given our sample limitations.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Dear authors,

I found the article well framed from a theoretical point of view. The theoretical bases that support it are presented and the constructs studied are well articulated.

From a methodological point of view, it presents a somewhat small number of participants for the analyses that are carried out (which are quite sensitive to sample size), which proves to be a limitation of the study. Nevertheless, the fact that it allows for cultural comparison is a positive aspect of the study.

The results are presented succinctly, allowing for a fairly clear reading in light of the hypotheses formulated. They are adequately discussed, referring to the aspects mentioned throughout the introductory part of the study.

The results obtained by comparing the means are also carefully discussed, considering different dimensions that may be associated with the significant differences found.

The limitations presented make sense, and the authors managed to find realistic and feasible suggestions for future research of the same nature.

Finally, the implications are also carefully explored to account for the contribution of the study.

Please check the following lines for corrections:

Line 602: the year of the reference "Takahashi" is missing.

Line 769: it seems to be the only reference where the date (i.e. 2017) shows up between parenthesis.

Author Response

Comments 1:

I found the article well framed from a theoretical point of view. The theoretical bases that support it are presented and the constructs studied are well articulated.

Response 1: Thank you for your comment.

Comments 2:

Response 2: Thank you for your remark. We agree that further research is needed to determine whether and how cultural factors can influence the ambiguity management process.

Comments 3:

The results obtained by comparing the means are also carefully discussed, considering different dimensions that may be associated with the significant differences found

Response 3: Thank you for your comment.

Comments 4:

The limitations presented make sense, and the authors managed to find realistic and feasible suggestions for future research of the same nature.

Finally, the implications are also carefully explored to account for the contribution of the study.

Response 4: Thank you for your comment.

Comments 5:

Please check the following lines for corrections:

Line 602: the year of the reference "Takahashi" is missing.

Line 769: it seems to be the only reference where the date (i.e. 2017) shows up between parenthesis.

Response 5: We have revised these lines to include the reference year and reformat the reference date in the mentioned citations.

Author Response File: Author Response.pdf

Reviewer 5 Report

Comments and Suggestions for Authors

A lot of the literature review seems to be quite segmented. The authors layout the concepts of each construct and describes the meaning of it in the career development context, but some of the descriptions are not necessarily related to their hypotheses or the study model, which makes the readers confusing and the literature review less focused. The authors need a more integrated theory that can explain their suggested model.

The overall sample size appears insufficient to meaningfully represent three distinct cultural contexts while arguing for shared patterns across them. The country-level subsamples are even smaller and highly unbalanced, which further undermines the credibility of claims about similarities across contexts. Although from a purely statistical standpoint the sample size may be adequate for a path model with only four observed variables, the deeper issue is a mismatch between the theoretical ambition and the empirical foundation. In other words, the cross-cultural framing and claims of generalizability are not well supported by the limited and uneven data.

Related to the above concern about small and uneven subsamples, the study does not test for measurement invariance across the three cultural groups. Without at least establishing configural and metric invariance, it is unclear whether the scales are even measuring the same constructs across countries. This issue is particularly salient given that (a) one country used a translated version of the instruments, while others completed the measures in English, and (b) the authors interpret cross-country differences and similarities as evidence of shared processes.

Although the authors found that only avoidance coping significantly mediated the relationship with ambiguity aversion and career anxiety, when looking at the correlation between avoidance and approach coping, it was about r=.43. This means that both copings are used together, probably in the context of career development. This brings up two important questions: (a) can the two be seen as differentiated constructs? (b) while they seem to be used together, why only one of them showed significant findings? (They do seem to have similar effects; it is just that avoidance coping was strong enough to reach statistical significance.) A further examination regarding these findings needs to be discussed.

The authors said that the data were analyzed with path analysis, but I think that is an overstatement. A pat analysis is a special type of structural equation modeling, where there are only observed variables in the model. However, process macro is mainly running a series of regressions. Thus, I see them as slightly different analytical techniques.

In Figure 1, it would be helpful for the readers if expected (positive and negative) associations were marked on the arrows. It makes it easier to see what the expected hypotheses are.

Author Response

Comments 1:

Response 1: Thank you for your feedback. In our revised manuscript, we have reduced the segmentation of our literature to ensure that the logical flow of our argumentation is more well-connected. Specifically, we strengthened the justification of the hypotheses by integrating the description of the dual-process model with COR theory. Firstly, we presented the dual-process model more broadly than in the previous version of the paper. Accordingly, we described the ambiguity management process in parallel with the resources loss/gain spirals to clarify the link between the variables. Section 1.2 introduces "The role of resilience in managing ambiguity", while section 1.3 describes how ambiguity aversion can lead to an increase in career anxiety through the use of approach and avoidance coping strategies.

Comments 2:

Response 2: We agree with you that sample size and its unbalanced composition can be a relevant limitation of the study. Accordingly, we adopted an exploratory and descriptive approach to the cross-country comparison. The main objective of the study is to contribute to the quite recent literature on career ambiguity management by exploring the role of ambiguity aversion in increasing career anxiety through resilience and coping strategies. At the same time, we also found it important to consider and explore any potential differences between countries. Acknowledging your concern, we had rerun our analyses by adding control variables (i.e., age, gender, and national unemployment rate) in our tested models to partially control the potential effects of our unbalanced sample.

Comments 3:

Response 3: We agree with your comment, so we have reported the results of our measurement invariance testing in the supplementary material of our revised manuscript. In summary of the results we attained, we were able to establish measurement invariance across the three countries (i.e., Norway, Indonesia, Bangladesh) included in our research, indicating the similar conceptualizations that our multicountry sample have of our study constructs.

Comments 4:

Response 4: We commented further on our revised manuscript, specifically in lines 575-589, regarding your point about the significant correlation between avoidance and approach coping in which we cite previous literature that found similar results. In general, coping has been theoretically and empirically conceptualized as a multidimensional construct with higher order factors (Duhachek and Oakley 2007; Skinner et al. 2003), which is how we measured the coping strategies in our study. In consideration of the nature of the stressor which an individual responds to, coping responses have been considered by past researchers as a complex phenomenon (Carver er al. 2000). In particular, responding to stressful situations as multifaceted as the ambiguity surrounding modern careers can entail the use of more than one coping strategy to achieve multiple individual goals and regulate the multitude of emotions induced by the stressor. In lieu of the mixed findings we attained regarding the mediating roles of avoidance and approach coping strategies, we cited in our revised Discussion section prior literature that highlighted different associations between factors related to a stressor and specific coping strategies (i.e., avoidance, approach). Additionally, we offer an perspective that suggests two (coping) mechanisms at play when managing ambiguity in the career decision-making process (Xu 2023) that is found in lines 590-602 of the updated manuscript:

“These mixed findings on the mediating roles of coping strategies can also be interpreted by considering two mechanisms that can occur when coping with ambiguity in the career decision-making process (Xu 2023). The first mechanism is based on ambiguity aversion, leading to the adoption of avoidance coping strategies, which have a negative impact on career outcomes (e.g. career anxiety). The second mechanism could be activated by hope or calling, which is postulated to help individuals navigate ambiguity and achieve a career direction (Xu 2023), that can increase the use of approach coping strategies and reduce career anxiety. This perspective aligns with arguments of the resources gain spiral (Hobfoll 2001): those who have more resources are more like-ly to invest them to gain new ones and, consequently, are more effective in preventing negative emotional outcomes of stress, such as career anxiety. Future research should consider the impact of hope and calling in addition to ambiguity aversion on the ambiguity management process."

Comments 5:

Response 5: Thank you for your feedback. We have removed the text previously found on section 3.2 Hypothesis Testing that commented on the use of path analysis.

Comments 6:

In Figure 1, it would be helpful for the readers if expected (positive and negative) associations were marked on the arrows. It makes it easier to see what the expected hypotheses are.

Response 6: We appreciate your suggestions, and we have added the positive and negative signs in Figure 1 of the revised manuscript to better clarify the hypotheses of our research model.

Author Response File: Author Response.pdf

Reviewer 6 Report

Comments and Suggestions for Authors

Dear authors,

I appreciate the authors’ effort to tackle a timely and meaningful (hard) question, on how ambiguity aversion and resilience relate to career anxiety and coping strategies among young people across three national contexts. The theoretical anchoring in Conservation of resources theory and the decision to compare multiple countries are promising choices that, in principle, can yield useful insights about psychological resources and labor market precarity. At the same time, the manuscript in its current form requires substantial revision before it can be considered ready for publication.

My first concern is with the literature and framing. The introduction and theoretical sections contain useful material, but they often run long and sometimes drift into general descriptions rather than building a tight argument that leads directly to the hypotheses. The paper will gain clarity and persuasive force if the literature review is trimmed to the essentials and if each hypothesis is justified with a crisp, context specific rationale. Right now several hypotheses read as statements already established in the literature; the manuscript should make explicit why these relationships might look different in the three national settings under study and what specific theoretical mechanism the study tests or refines.

There are important design and inferential limitations that need to be confronted more rigorously. The total sample (N = 156) and the country subsamples (n = 64, 62, 30) are small and unbalanced. This matters for statistical power, especially for mediation models, cross group comparisons, and any multi parameter model. The authors acknowledge sample size limits but do not present sensitivity or power analyses to indicate whether the study could reasonably detect the effects it tests. I recommend including at least a post hoc power analysis and, if possible, re running key tests with estimators designed for small samples (for example, robust ML with adjusted standard errors, Bayesian estimation, or bootstrapped SEM). These steps will help distinguish true null results from low power artifacts.

Measurement and comparability across countries are another major issue. The manuscript reports translations and use of English versions, but it is not clear whether participants in all settings completed conceptually equivalent instruments. Before interpreting cross national differences, it is essential to test measurement invariance (configural, metric, and ideally scalar ) across the three groups. Without these tests, comparisons of means or structural paths are prone to bias. The CFA results reported are fragmentary; for example, an RMSEA of 0.09 for the anxiety scale should be discussed as borderline and may indicate model misspecification. Please present full psychometric evidence by group: item loadings, reliability coefficients (alpha and omega), fit indices, and any item level decisions (drops, correlated errors). Also report any checks for common method bias given the single wave self report design.

The analytical strategy is reasonable in intent but too narrowly executed in practice. The manuscript relies on serial mediation logic but does not explore plausible alternative models. For example, resilience might be better conceptualized as a moderator rather than a mediator, or coping strategies might function as a proximal outcome influenced by both anxiety and resources in more complex ways. I suggest testing alternative specifications and reporting effect sizes not just p values so readers can judge substantive significance. If the authors retain the mediation framework, provide clear reporting of indirect effects with bootstrap confidence intervals and clarify which effects remain robust under alternative specifications.

Contextual interpretation of cross country contrasts is currently speculative. The authors attribute differences to socioeconomic context but do not bring any macro level indicators or contextual data into the analysis to support these claims. If context matters to the argument, include national level variables such as youth unemployment, GDP per capita, or labor market protection indices, or at least situate qualitative reasoning about country differences in documented indicators. That will make cross national inferences more credible and less impressionistic.

Transparency and reproducibility should be improved. The methods section should specify recruitment channels, inclusion and exclusion criteria, how missing data were handled, the exact phrasing of translated items, and any data quality filters used. Providing an appendix with the questionnaire and a table of CFA loadings would substantially increase the manuscript’s utility. Where data cannot be fully shared for privacy reasons, the authors should at minimum supply analysis scripts and a synthetic or anonymized dataset so that key results can be reproduced.

Finally, the writing and presentation need polishing. Shorter, clearer sentences and more focused paragraphs would help readers follow the argument. The results section tends to enumerate tests without interpreting their practical meaning; the discussion should prioritize which findings are robust and practically important, which are tentative, and what follow up research is required. Policy and practice recommendations must be proportionate to the evidence: avoid broad prescriptions based on marginal or underpowered effects.

In sum, this manuscript addresses a worthwhile question and contains promising material, but it needs a stronger, more rigorous treatment of measurement, statistical power, alternative model specifications, and contextual interpretation. My editorial recommendation is to invite a major revision. If the authors address the points above, test invariance, report fuller psychometrics, justify and where possible augment the analytic approach to account for small subsamples, and tighten the argument and writing, the aper could make a meaningful contribution. I am happy to help further by drafting specific wording edits, preparing a concise revision checklist for the authors, or reviewing a revised version.

Author Response

Comments 1:

Response 1: We thank you for your feedback. Based on the comments you provided, we have worked to improve our manuscript and address each of the points you made below.

Comments 2:

Response 2: Following your suggestion, we partially restructured and reformulated the introduction and literature review sections in order to clarify and strengthen the arguments of our hypotheses. The introduction was shortened to focus on summarizing the theoretical framework and the study's contribution. We have removed the overly descriptive sections of the literature review, focusing instead on discussing the relationship between the variables within the theoretical framework of the dual-process framework (Xu 2023) and COR theory (Hobfoll 1989). Specifically, we presented the ambiguity management process and the loss/gain spiral of resources as the mechanisms on which our hypotheses were based. We believe that the literature on career ambiguity management requires more evidence, and that additional confirmation of how this process works would be beneficial, particularly with regard to young individuals who are often affected by precarious working conditions. Of course, we agree with the reviewer that it would be extremely interesting to know whether these processes work differently in different national contexts. This is the main reason why we introduced the comparison between countries. Unfortunately, however, the small and unbalanced sample suggested us to adopt an exploratory and descriptive approach to the cross-country comparison.

Comments 3:

Response 3: We followed your suggestion, and we conducted a post hoc power analysis using G*Power (Faul et al. 2009). As we had reported in lines 326-328 of the revised manuscript, our analysis revealed that our study has 95% power to detect a medium effect size at α = .05 (Cohen 1988).

Comments 4:

Response 4: In our revised manuscript, we prepared a supplementary material, where we report the results of our tests for measurement invariance from a previous study that employed the same measures in a similar sample of young people from Norway, Indonesia, and Bangladesh. We used the recommended criteria (Putnick and Bornstein 2016) to evaluate measurement invariance across the three countries by estimating a constrained model (i.e., metric) against an unconstrained model (i.e., configural; Darwish et al. 2024). In relation to the results of our confirmatory factor analyses (CFA), we based our evaluation of the model fit in accordance with the threshold values of three indices (i.e., CFI, SRMR, RMSEA; Hu and Bentler 1999). In lines 430-432 of the manuscript, we acknowledged the RMSEA value of the career anxiety scale that was higher than the cited threshold value, and we based the acceptability of the model in lieu of the other model indices that showed acceptable fit. In our supplementary material, we also reported the CFA results for the measures of each country group. To control for common method bias, we followed the recommendations of Kock et al. (2021) in relation to our survey design and statistical control in which we ran Harman’s single-factor test. We described this procedure in more detail in lines 349-357 of the updated manuscript.

Comments 5:

Response 5: We followed your suggestion and explored two plausible alternative models: (1) we considered the moderating role of resilience to facilitate the use of avoidance and approach coping strategies in response to career ambiguity aversion; (2) we also considered avoidance and approach coping strategies as proximal outcomes to career ambiguity aversion through the mediating roles of resilience and career anxiety in serial. The results of these alternative models are reported in the Supplementary Material. In summary of the results, we found the effect size of our study’s reported research model to be stronger in comparison to the other alternative models presented, supporting our original theoretical considerations and hypotheses formulation.

Comments 6:

Response 6: Thank you for your useful suggestion. We added national unemployment indexes in our correlation analyses and hypotheses testing by considering it as a control variable, which we described in line 434-437 of the revised manuscript. Considering the significant correlations between this index and our study variables, we used this as a basis for our interpretation of the exploratory results on country differences found in the Discussion section on p. 16 of the manuscript

Comments 7:

Response 7: We revised the texts of our Methods section in relation to participants and procedure (i.e., lines 340-345 of the updated manuscript) in which we provided additional information regarding the recruitment channels and participant criteria applied in our study. We have also prepared a supplementary material, where we added a table of our translated scale items in Indonesian. In the Data Availability statement found in our manuscript, we indicated that an anonymized version of our dataset will be available when requested from the corresponding author.

Comments 8:

Response 8: Thank you for your overall feedback. In our revision of the manuscript, we have polished our writing to ensure the texts (i.e., sentences and paragraphs) are clearer with its arguments. In our Results section, we have also added concluding sentences after each test of hypothesis and exploratory research question that interpret the meaning of the test results. And in our Discussion section, we have reworked the presentation of our findings in which we better emphasized our most important findings, in addition to highlighting which research findings were more tentative/exploratory that require further research considerations. With regard to the practical recommendations of our research (found in lines 757-778 of our revised manuscript), we have reformulated our texts to highlight the implications of our significant research findings related to the relationships between career ambiguity aversion, resilience, avoidance coping, and career anxiety. Specifically, we elaborated on education and youth policies that should aim to prevent the negative psychological impact of work precarity on individuals. Specifically, we distinguished between primary prevention actions, which we encouraged as part of career education programs at school, and secondary prevention initiatives, which we recommended to help remove young people at risk of entering a spiral of resource loss from a state of precariousness. Lastly, as we had explained in our Responses 2-7, we have incorporated your feedback to improve the methodological aspects and results interpretation of our study.

Author Response File: Author Response.pdf

Round 2

Reviewer 5 Report

Comments and Suggestions for Authors

I think the authors addressed the comments from the previous round of reviews well.

Reviewer 6 Report

Comments and Suggestions for Authors

Dear authors,

I consider you made the needed changes and corrections, so your work now would be appropriate for its publicaction.