Pilot Study on Institutional Trust, Security, and Democratic Support in Ecuador During the 2024 Crisis

Comments 1: I think this is an interesting and timely topic.  There is a great deal of interest among academics and officials about the ongoing security crisis in Ecuador, and how this will shape political developments in the country.  As the authors note, Ecuador had long been a peaceful oasis in a region in which most countries struggled to contain high rates of violent crime.  The rapid deterioration of security in Ecuador, particularly at the hands of organized criminal groups, merits close examination.  While the authors raise several good points in this paper, unfortunately I do not think it is ready for publication at the current time.  I hope that the comments below help the authors as they continue to revise this manuscript, and I think it will eventually make a strong contribution to the literature.

Response 1: Thank you for your thoughtful and encouraging assessment. We greatly appreciate your recognition of the relevance of the topic, and we are committed to addressing each of your comments carefully. Your feedback has been extremely helpful in improving the clarity, structure, and theoretical contribution of the manuscript.

Comments 2: First, the Abstract could use some reworking.  I think the phrase “political-ideological” institutions is a bit clunky, and in the abstract, the authors don’t have space to explain what it means exactly.  I would select a different term to describe these institutions throughout the manuscript, but particularly in the Abstract.  The first sentence of the Abstract is also difficult to follow, and since this is the first sentence introducing the reader to the study, it really should aim to grab the attention of the audience.  I would start the Abstract with the research question:  “How does the ongoing security crisis in Ecuador shape public support for democracy and approval of the incumbent government?”  This way, the reader understands the goal of the paper immediately.  Including the research question up front would also help follow the summary of the main results – in its current form, it is a bit hard to follow the detailed summary of results without knowing exactly what the research question is.  On a positive note, the authors provide a great recommendation in the last sentence of the Abstract.

Response 2: Thank you for your helpful comments regarding the abstract. We have fully implemented your recommendation. The revised abstract now begins with a version of the core research question: “This pilot study seeks to answer the question: How does the ongoing security crisis in Ecuador shape public support for democracy and approval of the incumbent government?” This change helps to immediately orient the reader and clarify the purpose of the study.

In addition, we have removed the term “political-ideological institutions” from the abstract and throughout the manuscript. It has been replaced with the clearer and more conventional term “political institutions,” which distinguishes them from “security institutions.” We agree that this improves the clarity and readability of the abstract, particularly in such a condensed space. We appreciate your suggestion and have ensured that the summary of findings is easier to follow considering this adjustment.

Comments 3:  The Introduction tries to cover too much ground.  The first paragraph is fine, but in the second, the authors have a digression where they discuss the differences between the ideologically-motivated insurgency in the 1970s and 1980s and organized crime today.  This level of detail should go in the body of the paper.  Finally, the last sentence in that second paragraph is a fragment. 

Response 3: Thank you for your observation. In response, we have removed this paragraph from the Introduction and relocated the relevant content to the theoretical background section (Section 1.2), where it now helps contextualize the changing nature of non-state violence and its implications for democratic stability in Ecuador.

Comments 4: The literature review begins by focusing on studies of sociopolitical conflicts in Ecuador.  I would broaden this discussion to include literature on the region of Latin America as a whole, as other countries share similar security challenges.  What makes Ecuador unique is that this violence erupted and escalated quite quickly, and as the authors note, this was a huge change in the peaceful status quo.  As the literature on public reactions to crime indicates, it is likely that average citizens will react quite differently in this context compared to ongoing violence in places like Honduras.  

Response 4: Thank you for this valuable observation. We have addressed this point by integrating a new paragraph into section 1.1 that incorporates comparative literature on public reactions to crime across Latin America. We now contrast Ecuador’s case—where violence escalated rapidly in a previously stable context—with countries like Honduras or El Salvador, where chronic insecurity has shaped long-term citizen expectations. This comparative perspective strengthens the analytical framework of the paper and helps contextualize the reactions of Ecuadorian citizens in a broader regional landscape. We also discuss how sudden transformations in public security may trigger different dynamics in institutional trust and democratic legitimacy than those observed in societies with long-term exposure to organized violence.

Comments 5: In the literature review, it would be helpful if the authors discussed their arguments in greater depth.  The authors refer to a “double dilemma,” and I found this unclear at first.  It seems that the authors argue that the current insecurity crisis can weaken democratic governance in two ways:  (1) by eroding the legitimacy of the democratic government, which could open the door to authoritarian measures; (2) relying upon militarized (and perhaps extralegal) measures to fight crime, which can in turn weaken democracy.  I personally don’t know if I would call this a double dilemma, but if the authors wish to retain this term, I would spell out more clearly exactly the various ways in which they expect insecurity to weaken democratic governance.

Response 5: Thank you for your thoughtful observation. We agree that the term “double dilemma” may have introduced unnecessary ambiguity. In response, we have removed this expression and revised the paragraph to present our argument more clearly. The new version outlines two specific risks posed by the security crisis to democratic governance: (1) the potential erosion of democratic legitimacy due to ineffective government response, and (2) the risk that expanded reliance on military and police forces may lead to the normalization of exceptional, authoritarian measures.

Comments 6: Later on the same page, the authors argue that the relationship between support for the armed forces and support for the incumbent government and democracy is not uniform or unidirectional.  They note that strengthening the armed forces can lead to reduced government approval or weakened support for democracy.  I don’t see how this contradicts their earlier claim about a double dilemma.  To clarify some of this ambiguity, I think the authors need to walk the reader through their arguments more carefully.  For example, what exactly is the “complex interplay between security and political stability”?  What are the multiple ways in which security affects public support for democracy and the incumbent government?  I think developing the theoretical expectations in greater depth would strengthen the paper.

Response 6: Thank you for this observation. We have clarified the “complex interplay between security and political stability” by expanding the theoretical explanation following the contextual discussion in Section 1.1. In particular, we now outline two distinct logics that link trust in security institutions to political attitudes: one that boosts support for the incumbent government through perceived effectiveness, and another that risks eroding democratic legitimacy if militarization becomes excessive or permanent.

Comments 7: The authors also argue that “trust in the Armed Forces tends to correlate with a positive impact on the perception of the government.”  Does this relationship vary by the ideological orientation of the government?  For example, would one expect that this relationship is stronger for governments that identify more with the right, and/or are more overtly nationalistic?  Since the authors cite the case of the United States here, I wondered if this relationship between trust in the Armed Forces and trust in the incumbent government would vary by presidential administration (i.e., be stronger in the case of a Trump-like administration but less so under a Biden or Obama style administration).  On a smaller note, in this paragraph I am not sure why the authors use the phrase “Armed Forces (Armed Forces).” 

Response 7: Thank you for this insightful observation. Indeed, we agree that the relationship between trust in the Armed Forces and support for the incumbent government could vary depending on the ideological orientation of the government, especially in regimes with strong presidential leadership. The reference to the United States was included as an illustrative case where this pattern has been studied extensively, particularly in recent administrations where ideological identity and nationalistic discourse were more prominent. However, we acknowledge that exploring this dimension in the Ecuadorian context requires a more detailed comparative framework and more specific data, which go beyond the scope of this pilot study. We plan to examine this question further in future stages of the research. But, we have written some lines extra in order to clarify the argument.

 Additionally, the repeated expression “Armed Forces (Armed Forces)” was a typographical oversight and has been corrected in the revised manuscript.

Comments 8: The authors need to explain more clearly what are “political-ideological institutions.”  As noted above, I would jettison this terminology and replace it with something more succinct and clearer.  I do get the authors argument, however.  They are trying to distinguish between institutions that are overtly political (e.g., political parties and the legislature) and those that are supposed to be more apolitical and neutral (e.g., the courts).  Stolle et al have analyzed this, and their work could be used to discuss this distinction in greater detail.  Indeed, rather than just look at the impact of trust in overtly political institutions and security institutions, the authors could also include trust in the courts as another point of comparison.

Response 8: Thank you for this valuable observation. We agree that the expression “political-ideological institutions” may be unnecessarily complex and potentially ambiguous. In response, we have simplified the terminology throughout the manuscript to refer to them simply as “political institutions,” in contrast with “security institutions.” This revised language enhances clarity and aligns with more established conceptual distinctions in the literature.

Our initial usage of the term aimed to differentiate overtly political institutions (such as political parties and legislatures) from institutions perceived as less partisan or more neutral, like courts. While our pilot data did not include courts, we acknowledge the relevance of incorporating trust in judicial institutions in future studies. As suggested, we will cite and integrate theoretical contributions by Dietlind Stolle and colleagues, who examine how trust in different types of institutions can shape political behavior and civic engagement. This body of work will help us better substantiate our institutional distinctions and refine the theoretical framing.

In the revised manuscript, we have clarified the institutional categories, provided a more grounded conceptual explanation, and added a note in the concluding section to state that future analyses will include trust in courts as an additional institutional variable, particularly relevant to theories of institutional legitimacy and democratic accountability.

It was included in: Approach to Literature and Theory.

Comments 9: On page 3, the authors start to report different levels of trust that Ecuadorians have in different institutions.  This discussion would be easier to follow with a visual aid – perhaps one graph that shows average levels of trust in various institutions using the LAPOP data (i.e., the b series). 

Response 9: We appreciate the suggestion. While we agree that a visual representation of institutional trust levels can enhance clarity, we opted to keep this section in narrative form to reduce the number of figures and preserve the textual flow.

Comments 10: On page 3 paragraph 3, the authors highlight the need to examine the role of citizens in consolidating/weakening democracy, as was recently done in Peru.  I am not sure why one isolated study from Peru is mentioned here, where the literature on the role of citizen and public attitudes in democratic consolidation/erosion is very extensive. Bermeo wrote perhaps the most influential book on citizens in democratic breakdown, Ordinary Citizens in Extraordinary Times.  I think the authors need to provide a far more thorough overview of the literature on this point.   In this paragraph, the authors state that they will focus on citizens’ ideological orientations on democratic sustainability, and I think a far more rigorous justification for this focus is needed here.  Why ideology, and not satisfaction with the economy, or perceptions of public security, for example? 

Response 10: Thank you for your observation regarding the need to broaden the literature on citizens’ roles in democratic consolidation and to justify more rigorously our focus on ideological orientation. We have revised the manuscript accordingly.

First, we now incorporate foundational literature that emphasizes the decisive role of citizen attitudes and participation in both the erosion and consolidation of democracy. As Nancy Bermeo highlights in Ordinary People in Extraordinary Times, democratic breakdowns are not merely elite-driven; ordinary citizens’ support—or withdrawal of support—plays a critical role in shaping regime trajectories. Similarly, McCoy, Anderson and Philip underscore that in the Latin American context, citizen mobilization, frustration, and ideological disaffection have often triggered or resisted democratic backsliding, it shows more cases than Perú. These studies strengthen the conceptual foundation for analyzing public opinion as a key driver of democratic outcomes, beyond institutional or elite-level factors. (It was included in Literature and Theory).

Second, regarding the focus on citizens’ ideological self-placement, we clarify in the revised manuscript that this variable was initially included to test the hypothesis—common in the literature—that ideological polarization may correlate with declining support for democracy or openness to authoritarian alternatives. However, our findings indicate that university students in Ecuador show very limited ideological polarization; most respondents locate themselves near the center of the ideological spectrum (positions 5 and 6 on a 10-point scale). As a result, while we retain the analysis of ideological orientation, we now clearly state that it plays a marginal role in explaining democratic support or institutional trust in this pilot sample. We also acknowledge that other factors—such as economic satisfaction and perceptions of security—merit inclusion in future survey waves, and we plan to expand our model accordingly in the broader research project. (It was included in 3.1. Descriptives)

Comments 11: On page 3, lines 116-125, the authors introduce several important ideas, but need to discuss them in greater depth and more concretely.  Sentences like, “the president’s weight is particularly significant” are just too vague.  Perhaps here the authors could draw upon some of Linz’s work on presidentialism in Latin America. 

Response 11: We appreciate the reviewer’s suggestion to deepen the discussion on presidentialism. We have revised the paragraph to include a key theoretical contribution by Juan Linz, emphasizing how the institutional design of presidential systems—particularly fixed terms and limited flexibility—can increase the risk of democratic instability in contexts of executive-legislative conflict.

Comments 12: The authors’ survey data is unique and very interesting, but I’m not sure they fully leverage it.  Originally when I read of the authors’ monthly surveys, I thought they would examine how levels of support/approval varied over time, in light of key events occurring in Ecuador.  I thought the authors would provide a qualitative assessment of key security indicators (e.g., homicide rates, highly publicized attacks perpetrated by organized criminal groups, emergency decrees, political discourse, etc.), and determine how this security context shaped public opinion, and the impact of trust in security organizations.  The methodology employed by the authors does not really provide this temporal overview, we don’t get a full picture of what has transpired in Ecuador over the course of the survey administration.  The authors do a good job of increasing the degree to which their sample reflects different provinces and socio-economic groups, although I would like to see more of a discussion about the potential impact of the gender imbalance in their sample. 

Response 12: We appreciate the reviewer’s thoughtful feedback and recognition of the uniqueness of our longitudinal survey design. While we agree that an in-depth temporal analysis would provide important insights—particularly by linking public opinion shifts to major security-related events such as homicide spikes, states of emergency, or changes in official discourse—our primary objective in this pilot was to test general associations between institutional trust and support for government and democracy. Nonetheless, we recognize the potential of our monthly panel data to explore the dynamic evolution of perceptions, and we are currently planning follow-up studies that incorporate time-series components and contextual indicators. We have clarified this in the revised version of the Temporary Unit of Observation section. Additionally, while our sampling strategy improved coverage across different territories and socioeconomic backgrounds, we also acknowledge the gender imbalance and have noted its implications in the “Final Considerations and Limits” section as a limitation to be addressed in future phases of our research.

Comments 13: The hypotheses could be refined.  Ideally, there is one hypothesis for each independent variable – i.e., as trust in the military increases, trust in the incumbent government will also increase.  I was not clear on the logic for the second hypothesis – why does trust in the Armed Forces and Police reduce the Approval of Democracy?  This is particularly unusual for the police, as other scholars (e.g., Cruz) have found that trust in the police is positively linked to system support and support for democracy.

Response 13: We appreciate the reviewer’s suggestion regarding the refinement of hypotheses. In the revised manuscript, we have restructured the hypotheses to ensure that each one corresponds clearly to a specific independent variable and to clarify their theoretical underpinnings. In particular, we now distinguish the different roles that trust in the Armed Forces, Police, and political institutions may play in shaping government approval and support for democracy. These changes appear in the section 1.2.

Comments 14: I appreciate the authors’ efforts to address the problem of multicollinearity, but I am not sure that they have done so fully.  The indicators that they analyze typically tend to be highly correlated together.  For example, when LAPOP includes measures of the armed forces, police, legislature, political parties, and incumbent government, typically the Chronbach’s alpha is over .8.  This is indicative of very high multicollinearity, as attitudes towards all of these institutions is highly inter-related.  Indeed, if one does a factor analysis, there is usually one common dimension, indicating that public attitudes towards these institutions are based upon some core commonality.  I encourage the authors to think more about the implications that these high levels of correlation have on their analysis, not just empirically but theoretically.

Response 14: We appreciate the reviewer’s insightful comment on the potential multicollinearity among trust indicators, a common feature in analyses of institutional trust in Latin America. We recognize that measures of trust in institutions such as the Armed Forces, police, legislature, and political parties often correlate highly, reflecting what some scholars have referred to as a diffuse institutional confidence syndrome—a core evaluative dimension rooted in broader legitimacy perceptions. However, recent theoretical developments in comparative political behavior stress the growing divergence in how citizens perceive security versus political institutions, especially under conditions of state fragility, populist disruption, or acute insecurity. In this context, high correlations may coexist with meaningful distinctions in the political consequences of trust. For example, while trust in political parties may reflect partisan identification or dissatisfaction with representational politics, trust in the Armed Forces may be more linked to performance-based legitimacy in security provision or nostalgic views of order, particularly in crisis contexts. Therefore, despite shared variance, we believe it remains analytically valuable to disaggregate these components and assess their differential effects, especially given our empirical finding that trust in security institutions (especially the Armed Forces) is significantly associated with government approval but not with democratic support.

We agree that a confirmatory factor analysis or structural equation model could further clarify the latent structure of institutional trust in future work. However, our current pilot study aims to test basic associations and remain cautious in its interpretive claims, particularly given the sample size and exploration scope. We included some lines in 3.2. section.

Comments 15: The authors do a good job of using graphs to illustrate trust in various institutions, but I think they could probe their findings in greater depth.  The graphs indicate that there are some interesting trends that can be hidden by focusing on average levels of support.  For example, in Figure 1, particularly for the Armed Forces, there are more respondents located on the extremes of the range (i.e., no support and a lot of support). This indicates that there is some polarization in terms of attitudes towards security forces, and it would be interesting if the authors examined this polarization in greater depth, particularly in terms of how people on the two extremes support the incumbent government and democracy.  There is also a bit of polarization evident in Figure 3 too.  I would also encourage the authors to edit the axes of these graphs as they are hard to read.

Response 15: We thank the reviewer for highlighting the polarization apparent in our data, particularly the bimodal distribution of trust in the Armed Forces (Figure 1) and the similar, albeit less pronounced, pattern in Figure 3. Indeed, a subset of respondents report either very low or very high support for the security forces, indicating a polarized view. We acknowledge that this is an interesting and important phenomenon. However, our study’s primary focus is on overall trust levels and their general trends, so we have not analyzed this polarization in depth. We do mention the observed polarization briefly in the manuscript, but not extensively, as a detailed investigation of its causes and implications (such as how those at the two extremes feel about the incumbent government or democracy) lies beyond our current scope. We agree that exploring these extreme groups’ attitudes would be valuable, and we plan to pursue this analysis in future research.

Regarding the graphs, we appreciate the suggestion to improve the readability of the axes. Clarity in our figures is important. While it is true that some visual elements (such as font size or label orientation) could be refined, we believe the current format effectively conveys the findings and maintains consistency across the visualization of results. We have therefore chosen to retain the existing style of the plots. Nonetheless, we will keep the reviewer’s feedback in mind and will continue to improve the clarity of our figures in this and future work as needed.

Inclusion of Polarization Discussion in Manuscript: In the revised manuscript, we have added a short discussion acknowledging the observed polarization in trust levels. This addition appears in the Discussion section, where we note that the distribution of trust in certain institutions (e.g., the Armed Forces) is polarized. We highlight that an analysis of how the most extreme responses relate to support for the incumbent government and democratic values would be insightful, but is outside the scope of the present study. By placing this note in the 3.1. Descriptives, we ensure that readers are aware of the polarization pattern and understand that while we recognize its significance, a thorough examination is deferred to future research.

Comments 16: On page 9, lines 355-358, the authors have some interesting findings.  They note that trust in the Armed Forces is linked to trust in the Incumbent Government, but not Support for Democracy. This is an interesting finding, although please see my comment earlier about whether this relationship depends upon the character of the incumbent government.  In contrast, Trust in Police, the institution responsible for internal security in a democratic regime, increases both support for incumbent government and democracy.  The contrast between these two security institutions is interesting and merits more discussion. The authors lump findings about security institutions together, and say that trust in them is “more explanatory of government approval than support for democracy.” I don’t think this is the case.  When you separate the impact of these two security institutions, Trust in Police operates differently than Trust in the Military, and this distinction is theoretically interesting.  In democracies, police are the institution that should have a monopoly on internal security, as they are a civilian force trained to work with domestic populations.  The training of the military is very different, given that they are supposed to be an armed force focused on external threats in democratic regimes (although often times this distinction does not hold in Latin American countries, with implications for democratic stability). 

Response 16: Absolutely. We fully agree with your observation and have added specific lines in the discussion to highlight this distinction. Trust in the Police—given their civilian role and constitutional mandate over internal security—is positively associated with both government approval and support for democracy. In contrast, the Armed Forces operate under a different logic and training, which may explain why their influence is limited to government approval. We appreciate your insight, which helped us sharpen this theoretical distinction (3.2 section). Specific discussion was included in 4.1 section.

Comments 17: In section 4.1, I am not sure the empirical findings support the authors’ argument.  Trust in the Military did not reduce support for democracy, the two were just not statistically related.  If they are not significantly related, I am not sure how Trust in the Armed Forces could “incentivize the search for non-institutional solutions . . . which could generate instability in the democratic regime.”  On page 10 lines 383-389, I don’t necessarily disagree with the authors, I am just not completely convinced that their findings support these arguments. 

Response 17: Thank you for your thoughtful feedback on the interpretation of our findings related to trust in the Armed Forces and support for democracy. You are right to point out that the relationship between these two variables is not statistically significant. In response, we have revised Section 4.1 to clarify that our data do not support the claim that trust in the military directly weakens democratic support.

Instead, we now interpret this lack of association as a form of normative decoupling: while trust in the Armed Forces significantly increases government approval, it does not appear to influence citizens’ support for democracy. We have reformulated our argument more cautiously to reflect this asymmetry. Rather than suggesting a direct democratic threat, we now point to the theoretical possibility—grounded in the literature on militarization and presidential regimes—that such institutional trust could, under certain political conditions, be instrumentalized in ways that bypass institutional checks. This possibility is now presented as a theoretical avenue for future research, not a conclusion supported by our data.

We appreciate your observation, which helped us align our interpretation more precisely with the empirical evidence.

Comments 18: In section 4.2, the authors have an interesting discussion about Trust in Political Parties and the personalization of political leadership – this is a great addition.  The discussion on ideology and stability/security also is very good (lines 426-433).

Response 18: Thank you very much for your kind words. We truly appreciate your recognition of the discussion on political parties, personalization of leadership, and the role of ideology in relation to stability and security.

Comments 19: On page 8, line 306, the authors include trust in the Church as a variable.  This appears out of no where, and then kind of abandoned.

Response 19: You are absolutely right. The reference to trust in the Church reflected an earlier version of our analysis that intended to explore this institution further. However, due to space limitations and the need to maintain a coherent analytical focus, we decided to set aside this line of inquiry for a future paper. Accordingly, we have removed references to trust in the Church from the current version of the manuscript to avoid confusion and maintain thematic consistency.

Comments 20: Finally, while well-written overall, the manuscript would benefit from another round of thorough editing and revision.

Response 20: Thank you for your recommendation. Following your suggestion, we have conducted a thorough revision of the manuscript to improve clarity, coherence, and academic style.

Comments 21: I recognize that I have been critical of this manuscript, but I hope that this criticism is constructive.  I think the authors have something important to say, and that with some revisions, they will be able to say it persuasively and make an important contribution to the literature. 

Response 21: We sincerely appreciate your detailed and constructive feedback. We have carefully addressed each of your comments and implemented substantial revisions based on your suggestions. Your insights helped us clarify our theoretical framework, improve our methodological transparency, and refine both the structure and language of the manuscript. We are confident that the paper has significantly improved as a result and that it now makes a more robust contribution to the literature on democratic support and institutional trust during security crises.

4. Response to Comments on the Quality of English Language

Point 1: All comments.

Response 1: Thank you for your comment. We have engaged the services of a copy editor to review and improve the English throughout the manuscript. We trust that the revised version meets the journal’s language standards.

5. Additional clarifications

NA