This paper, published in Frontiers in Business, Economics and Management, explores the psychological mechanics behind online hotel bookings. By applying the Stimulus-Organism-Response (SOR) framework, the researchers demonstrate how specific features of online reviews—such as timeliness and reviewer prestige—act as "stimuli" that reduce a consumer's "perceived risk," ultimately driving the "response" of a completed booking.
TL;DR
Online reviews are more than just star ratings; they are critical risk-mitigation tools for experiential products like hotel stays. This study finds that review quality, volume, timeliness, reviewer prestige, and positive sentiment all significantly boost purchase intention by lowering the consumer's perceived risk.
What Problem Does This Paper Address?
Booking a hotel is an act of faith. Unlike a physical product you can inspect in a store, a hotel stay is an "experiential product"—you don't truly know the quality until you’ve already paid and checked in. This creates significant information asymmetry between the hotel and the traveler.
The authors tackle the challenge of "perceived risk": the uncertainty and potential for regret a consumer feels in a virtual shopping environment. While we know reviews matter, this paper seeks to deconstruct which specific features of a review most effectively bridge the trust gap and how they influence the internal psychological state of the buyer.
Methods
The researchers employed a quantitative approach based on the Stimulus-Organism-Response (SOR) model:
- Stimulus: Five characteristics of online reviews (Content Quality, Reviewer Prestige, Timeliness, Number of Reviews, and Positive Emotional Tendency).
- Organism: The consumer’s internal state, specifically their Perceived Risk.
- Response: The final Purchase Intention.
Data was collected via an online survey platform (Questionnaire Star). The authors used SPSS to perform:
- Reliability and Validity Analysis: Using Cronbach’s alpha and KMO/Bartlett’s tests to ensure the survey data was robust.
- Exploratory Factor Analysis (EFA): To confirm that the survey questions accurately mapped to the intended variables.
- Regression Analysis: To test the direct and mediating relationships between review features, risk, and purchase intent.
Datasets
The study relied on a primary dataset generated from 360 valid survey responses (from an initial pool of 433).
- Demographics: The sample was primarily aged 18-35, representing the "digital native" demographic that travels frequently and relies heavily on e-commerce.
- Criteria: Respondents were screened to ensure they had actual hotel booking experience and habitually browsed reviews.
Key Insights and Ideas
- The "Organism" Matters: The most important takeaway is that reviews don't just "cause" a purchase; they change the consumer's internal psychological state. Specifically, they act as a "risk buffer."
- Reviewer Prestige is a Trust Proxy: The "rank" or "credit level" of a reviewer significantly impacts how much a review reduces risk. Users perceive high-prestige reviewers as more objective and authentic.
- Timeliness as a Quality Signal: In the hotel industry, quality can fluctuate rapidly (e.g., a change in management or a recent renovation). Therefore, the "freshness" of a review is a critical indicator of current conditions, and old reviews are often discounted as "risky."
- Quantity as Social Proof: A high volume of reviews serves as a market performance indicator. Even if some are mixed, a large number of reviews suggests the hotel is a "popular" choice, which inherently feels safer to a consumer.
Key Results
The study confirmed all proposed hypotheses with statistical significance ($p < 0.01$ for most variables):
- Risk Reduction: All five review features (Quality, Prestige, Timeliness, Quantity, and Sentiment) had a significant negative impact on perceived risk. Reviewer Prestige (-0.197) and Positive Sentiment (-0.164) were particularly strong drivers in the initial model.
- Purchase Intent: Perceived risk was shown to have a significant negative effect on purchase intention (-0.237).
- Partial Mediation: Perceived risk was found to be a partial mediator. This means that while reviews directly influence purchase intent, a significant portion of that influence happens specifically because the reviews make the consumer feel less "at risk."
- Model Fit: The final regression model (Model 3) explained 28.5% of the variance ($R^2$) in purchase intention.
Limitations and Future Work
The authors acknowledge several areas for further exploration:
- Hotel Segmentation: The study treats the "hotel industry" as a monolith. Future research could investigate if these review features carry different weights for luxury resorts versus budget hostels.
- Content Format: The current study focuses on text-based features. With the rise of social commerce, the impact of visual content (images and short-form video reviews) remains a fertile ground for study.
- Geographic Specificity: The survey was conducted within the context of the Chinese tourism market; cross-cultural studies could determine if "Reviewer Prestige" or "Review Quantity" is valued differently in other global markets.
Summary updated