\documentclass{article}
\usepackage{geometry}
\geometry{a4paper, margin=1in}
\usepackage{xcolor}
\usepackage{hyperref}
\hypersetup{
    colorlinks=true,
    linkcolor=blue,
    filecolor=magenta,
    urlcolor=cyan,
}
\urlstyle{same}

\title{Response to Reviewers}
\author{}
\date{}

\begin{document}

\maketitle

\section*{Cover Letter}

\begin{flushleft}
\textbf{To the Editor,}

We thank you and the reviewers for the opportunity to revise and resubmit our manuscript, \textbf{"Empirical Communication and Trust Construction in ACLED Data of the 2023-2024 Palestine–Israel Conflict"} (Manuscript ID: [PLACEHOLDER]). We are grateful for the reviewers' thoughtful, detailed, and constructive feedback, which has been invaluable in strengthening our work.

In this revision, we have undertaken substantial revisions to address the core methodological and conceptual concerns raised. The key improvements include:
\begin{itemize}
    \item \textbf{Methodological Rigor:} We have added inferential statistical analysis (OLS regression models) to move beyond descriptive statistics, explicitly addressed the limitations and potential biases of the ACLED dataset, and enhanced the transparency of our qualitative analysis by reporting inter-coder reliability and providing a detailed description of our coding process.
    \item \textbf{Conceptual and Analytical Refinement:} We have reframed the study's focus to critically analyze the ACLED dataset as a \textit{communicative artifact} whose features are designed to build credibility, rather than presupposing its neutrality. In direct response to a major critique, we have removed the speculative demographic estimates (formerly Table 5) and refocused the qualitative analysis on the contextual function of keywords, not just their frequency.
    \item \textbf{Reproducibility and Transparency:} We have committed to sharing our processed dataset and analysis scripts in a public repository upon acceptance and have significantly expanded the methodological details in the manuscript.
\end{itemize}

We believe these revisions have directly addressed the reviewers' primary concerns, resulting in a more rigorous, transparent, and conceptually sound manuscript. Our detailed point-by-point responses follow.

\vspace{2em}
Sincerely,\\
The Authors
\end{flushleft}

\section*{Response to Reviewers}

\noindent \textbf{Reviewer 1}

\textit{Comment 1: "Methodological Rigor: Incorporate inferential statistics (e.g., regression models) to test relationships between variables (e.g., event types and fatalities)."}  
\textbf{Response:} We thank the reviewer for this crucial suggestion. To move beyond descriptive and bivariate statistics, we have conducted multiple Ordinary Least Squares (OLS) regression analyses. A primary model regresses fatality count on event type dummies, region dummies, and month fixed effects. The results confirm the strong, positive association between air/artillery strikes and higher fatality counts ($\beta = 3.92$, $p < 0.001$) even after controlling for other factors. We have added a new subsection, \textbf{"Regression Analysis Summary"} (Section 5.5, page 10), to present these findings. The full regression table is provided in Appendix A. We also note checks for model assumptions (linearity, homoscedasticity, multicollinearity) in the Methods section (Section 4.2, page 7, lines 210-225).

\textit{Comment 2: "Address ACLED data limitations explicitly: discuss sourcing biases, verification processes, and how missing data may affect findings."}  
\textbf{Response:} We have significantly expanded our discussion of ACLED's limitations. In the Methods section, we now explicitly acknowledge: "We explicitly acknowledge the limitations of the ACLED sourcing methodology, including potential geographic and reporting biases due to access restrictions, reliance on a finite set of media and NGO sources, and the challenges of real-time verification in active conflict zones. These limitations are factored into the interpretation of all quantitative results." (Section 4.2, page 7, lines 200-205). This caution is reiterated in the Discussion (Section 6, page 13, lines 315-320).

\textit{Comment 3: "For qualitative analysis, report inter-coder reliability metrics and provide a codebook in supplementary materials."}  
\textbf{Response:} We have enhanced the transparency of our qualitative analysis. We now report: "To ensure analytical rigor and reliability, two researchers independently coded a random subset of 500 narrative entries. Inter-coder reliability was assessed using Cohen's Kappa, which yielded a score of 0.78, indicating substantial agreement." (Section 4.4, page 8, lines 245-248). Furthermore, we state: "The full codebook and detailed analytical memos are available in the supplementary materials to ensure auditability." (Section 4.6, page 9, lines 275-277).

\textit{Comment 4: "Theoretical Contribution: Clarify how the study advances epistemic trust or moral witnessing theory beyond applying existing frameworks."}  
\textbf{Response:} We have refined our theoretical contribution throughout the manuscript. The core advance is framed as a methodological and empirical one: applying a concurrent mixed-methods design to deconstruct the trust-building features \textit{within} a dataset itself. We state in the Introduction: "This study distinguishes itself by critically examining the ACLED dataset as a constructed representation of conflict, analyzing the interplay between its quantitative metrics and qualitative narratives to understand the mechanisms of trust construction..." (Section 1, page 2, lines 45-50). The Discussion elaborates that a key finding is "that trust is constructed through a synergy of internal consistency... narrative framing... and the sheer persistence of documentation..." (Section 6, page 12, lines 290-295).

\textit{Comment 5: "Reproducibility: Share processed datasets and analysis scripts in a repository. Detail qualitative coding procedures."}  
\textbf{Response:} We commit to full reproducibility. In the Methods, we now state: "All data and analysis code are archived in a public repository to ensure full reproducibility (repository link anonymized for review)." (Section 4.2, page 7, lines 195-197). We have also significantly detailed the qualitative coding procedures, describing the stages of open coding, axial coding, and thematic development, and emphasizing the move beyond keyword counting to interpretive analysis (Section 4.4, page 8, lines 235-260).

\textit{Comment 6: "Minor Flaws: Reduce jargon, include visualizations, fix formatting."}  
\textbf{Response:} We have edited the text to improve clarity and reduce unnecessary jargon where possible. While we have not added new figures due to space and focus considerations, we have added references to supplementary scatter plots (Appendix B) to support correlation analyses. All formatting inconsistencies (title, acronym definition) and the erroneous date ("July 2025") have been corrected.

\vspace{1em}
\noindent \textbf{Reviewer 2}

\textit{Comment 1: "Fatal Flaw in Causal Logic / Circular Reasoning: The study claims to investigate 'how trust and credibility are constructed' but uses the ACLED data as its primary source. This is circular reasoning."}  
\textbf{Response:} We thank the reviewer for this fundamental critique, which prompted a significant reframing of our study's aim. We agree that using a dataset to externally validate its own credibility would be circular. Therefore, we have reconceptualized our analysis. We now explicitly position the ACLED dataset as the \textit{object of analysis}—a communicative text whose structural and narrative features are designed to signal credibility to its users. We state in the Introduction: "This study distinguishes itself by critically examining the ACLED dataset as a constructed representation of conflict... rather than presupposing the dataset's inherent neutrality or completeness." and "These questions are investigated with an awareness of the methodological circularity... this study positions the ACLED data as the object of analysis—a communicative artifact whose structure and content reveal practices of trust construction—rather than as an unproblematic ground truth." (Section 1, page 2, lines 45-60). This framing is reiterated in the Method (Section 4.1, page 6, lines 175-180).

\textit{Comment 2: "Remove the Demographic Estimation: Excise Table 5 and all conclusions drawn from it entirely. It is irredeemable without a fully specified and validated model."}  
\textbf{Response:} We agree with the reviewer that the demographic estimation presented in the original Table 5 was insufficiently justified. In accordance with this recommendation, we have removed the table and all specific percentage estimates (e.g., "42\%") from the manuscript. The qualitative analysis now focuses on the prevalent \textit{mention} of vulnerable groups as a narrative feature. We state: "The qualitative analysis instead notes the prevalent \textit{mention} of specific vulnerable groups (e.g., children) as a narrative feature, without attributing precise statistical proportions to the actual casualty population." (Section 4.4, page 8, lines 255-258).

\textit{Comment 3: "Conduct a Genuine Qualitative Analysis: ...move beyond keyword counts. Perform a rigorous, interpretive thematic analysis..."}  
\textbf{Response:} We have substantially revised our description and execution of the qualitative analysis to address this concern. We now explicitly distinguish between simple frequency counts and interpretive analysis: "Beyond simple keyword counting, the analysis interpreted the context and function of terms like 'children' or 'hospital'—examining whether they served to specify location, denote victims, invoke legal protections, or elicit moral response. This interpretive layer is crucial for understanding narrative construction." (Section 4.4, page 8, lines 250-254). This interpretive focus is carried into the Results and Discussion, where we discuss the communicative function of these keywords (Section 5.6, page 11, lines 280-285; Section 6, page 12, lines 300-305).

\textit{Comment 4: "Statistical Overreach... The 'estimation' of civilian age distribution... is methodologically indefensible."}  
\textbf{Response:} As noted in our response to Comment 2, we have removed the contested demographic estimation. We acknowledge the reviewer's point about the need for a fully specified model, which is beyond the scope and data available for this study.

\textit{Comment 5: "Add Context and Baselines: Provide pre-conflict baselines and comparisons to other conflicts..."}  
\textbf{Response:} While providing comprehensive pre-conflict baselines or cross-conflict comparisons is a valuable avenue for future research, it extends beyond the primary focus of this paper, which is the analysis of trust construction within a specific dataset during a defined period. However, we have added contextual statements where possible, such as noting that the fatality rate per event in Gaza was approximately ten times higher than in Israel (Section 5.2, page 9, lines 225-227).

\textit{Comment 6: "Tone Down the Rhetoric: Replace advocacy-oriented language... with neutral, scientific terminology."}  
\textbf{Response:} We have carefully reviewed the manuscript to ensure the language is analytical and scholarly. We have replaced phrasing that could be perceived as advocacy with more precise, evidence-based descriptions. Our theoretical lens (epistemic justice) is now presented more as an analytical framework than a predetermined conclusion. We also explicitly acknowledge our positionality as a potential boundary condition: "We recognize that our theoretical orientation towards epistemic justice creates a particular interpretive lens; we have sought to counter confirmation bias by applying systematic, transparent coding procedures..." (Section 4.6, page 9, lines 285-288).

\textit{Comment 7: "Lack of Comparative Data or External Validation..."}  
\textbf{Response:} We agree that comparative analysis with other datasets would be a powerful method for directly assessing credibility. We now frame this as a critical limitation and a primary direction for future work. In the Discussion, we state: "A primary limitation... is the inherent circularity... future research should triangulate findings across multiple independent data sources (e.g., UN reports, satellite imagery, health ministry data) to empirically assess credibility gaps and convergences." (Section 6, page 13, lines 320-325). We also suggest future studies that "measure trust perceptions or compare multiple datasets to assess credibility construction from an external vantage point." (Section 7, page 14, lines 350-355).

\section*{Closing Note}

We again express our sincere gratitude to both reviewers for their rigorous and insightful critiques. Their comments have challenged us to substantially improve the methodological rigor, conceptual clarity, and transparency of our work. We believe the revised manuscript is stronger, more nuanced, and makes a clearer contribution to the literature on conflict data, communication, and trust. We hope it now meets the journal's standards for publication.

\end{document}