Earticle

초록

영어

Understanding what makes online reviews helpful remains a central concern in information systems research and platform design. While prior studies have examined sentiment and structure, less is known about how distinct linguistic tones―analytical, authentic, and mixed emotional―signal reviewer credibility in text-only environments. Drawing on Source Credibility Theory (SCT) and Cognitive Fit Theory (CFT), we propose that these tones act as stylistic heuristics for perceived expertise and trustworthiness, particularly when identity cues are absent and tone aligns with the reader’s evaluative goals. Using LIWC to analyze 54,759 Amazon reviews across search and experience goods, we find that all three tones enhance perceived helpfulness, but their effects vary by product type. Analytical tone is more persuasive for search goods, where structured reasoning signals expertise and supports attribute-based comparisons. Authentic tone is more effective for experience goods, where sincerity supports trust and affective simulation. We also introduce mixed emotional tone―the co-expression of positive and negative affect―as a novel signal of deliberation, which improves helpfulness ratings, especially for search goods. These findings clarify prior inconsistencies in sentiment research and advance a theory-driven framework for how tone–product fit influences review helpfulness in digital platforms.

목차

ABSTRACT
Ⅰ. Introduction
Ⅱ. Literature Review and Theoretical Background
2.1. Tone and Linguistic Style in OnlineReviews
2.1. Tone and Linguistic Style in Online Reviews
2.2. Source Credibility Theory and Linguistic Tone
2.3. Cognitive Fit Theory and Tone–Product Alignment
Ⅲ. Hypothesis Development
3.1. Analytical Tone and Perceived Expertise
3.2. Authentic Tone and Perceived Trustworthiness
3.3. Mixed Emotional Tone and Balanced Credibility Signals
3.4. Product Type as a Moderating Context
Ⅳ. Research Design and Methodology
4.1. Data Collection and Sampling Strategy
4.2. Linguistic Inquiry and Word Count(LIWC)
4.3. Measurement of Key Variables
4.4. Model Specification and Estimation Approach
Ⅴ. Results
5.1. Descriptive Statistics and Correlations
5.2. Effects of Linguistic Tone and Product Type Interactions on Review Helpfulness
5.3. Robustness Checks and Alternative Specifications
Ⅵ. Discussion
6.1. Summary of Key Findings
6.2. Theoretical Contributions
6.3. Practical Implications for Platform Design and Review Writing
6.4. Limitations and Future Research
Ⅶ. Conclusion

저자

• Yongsuk Kim [ Associate Professor, SKK Business School, Sungkyunkwan University, Korea ] Corresponding Author
• Xilan Jiang [ M.S. Graduate, SKK Business School, Sungkyunkwan University, korea ]

참고문헌