| In the age of the Internet,online reviews,as a new form of word-of-mouth,have become a vital source for consumers to acquire product information,replacing and complementing traditional offline word-of-mouth communication regarding product quality.Online reviews are regarded as a cost-effective sales aid that significantly influences consumer purchasing decisions and product sales.Prior research on online reviews has primarily centered on the traditional e-commerce landscape,with substantial attention given to online retail lowinvolvement products and relatively little research conducted on offline high-involvement products.For high-involvement products,potential consumers have a strong incentive to gather comprehensive product information and engage in thorough evaluation.However,for both low and high-involvement products,the influence of online reviews on consumer purchasing decisions and product sales outcomes varies.The findings for low-involvement products cannot be extrapolated to high-involvement products.As a typical high-involvement product,automobiles possess intricate product features and a substantial price tag,necessitating consumers to invest a lot of time in information retrieval and thoughtful deliberation before committing to a purchase.Simultaneously,the automotive industry constitutes a cornerstone of the national economy.Precise forecasting of automobile sales,effective promotion of automobile sales,comprehension of consumer buying behavior,and the development of sound marketing strategies are paramount in preserving the competitiveness of automobile manufacturers and sellers in a fiercely competitive market.Therefore,this study focuses on the Chinese automobile sales market and endeavors to examine the utility of online reviews in the realm of automotive marketing.This study explores the correlation and causation between online reviews and car sales,along with the spillover effect of online reviews in automobile sales,and reveals in-depth the intrinsic influence mechanism of online reviews on consumers’ decision to purchase automobiles,aiming to offer insights and guidance for related studies.The primary research work and findings are as follows:First,this study explores the correlation between online reviews and car sales.Utilizing data mining techniques,online reviews from the Autohome word-of-mouth channel are extracted.Sentiment analysis algorithms are used to quantify textual information and derive the sentiment score,i.e.,review valence.Sentiment scores are adjusted using prospect theory and the entropy method.An autoregressive distributed lag model is constructed to analyze in-depth the long-run equilibrium relationship and short-run dynamic effects between the valence and volume of reviews and car sales across various sales volumes.Additionally,the predictive value of online reviews for car sales across different volumes is tested.The findings indicate that in a long-run equilibrium state,online review valence is significantly negatively related to car sales,while review volume exhibits a significant positive correlation with car sales.In the short run,review valence is significantly and positively related to car sales,whereas a significant positive correlation is observed between review volume and car sales for higher sales,with an uncertain association for lower sales.This reveals the difference in the relationship between review volume and car consumption groups with varying sales levels.Moreover,combining online review data significantly improves the predictive accuracy of car sales,with a more pronounced improvement observed for high-sales cars compared to low-sales cars.Second,this study explores the causal relationship between online reviews and car sales.A two-way fixed-effects model is constructed to examine the impact of online review valence and volume on car sales,and to explore the nonlinear relationship between online review valence and car sales,as well as the effects of the interaction between review valence and review volume on car sales.The findings reveal an inverted U-shaped relationship between online review valence and car sales.Specifically,when online review valence is low,enhancing review valence leads to increased car sales.However,as review valence further increases,its impact on car sales diminishes,and excessively high review valence impedes car sales.Furthermore,online review volume exerts a significant positive influence on car sales,with an increase in review volume translating into increased car sales.Additionally,there is a significant negative interaction effect between review valence and review volume,with review valence negatively moderating the relationship between review volume and car sales.Third,this study examines the spillover effect of online reviews of competitive cars on car sales.A two-way fixed-effects model is constructed to analyze the impact of online review valence and volume for competing cars on the sales of the focal car.A heterogeneity analysis is conducted on car price level to explore the effects of review valence and review volume of competing cars in different price levels on the sales of the focal car.The findings reveal that neither the average valence nor the average volume of online reviews for competing cars has a significant impact on the sales of the focal car when considering various influencing factors.However,online reviews of competing cars at different price levels exert varying influences on the sales of the focal car.For high-end cars,the average review volume for competing cars exhibits a significant negative impact on focal car sales,while the average review valence does not show a significant effect.For low-end cars,both the average review valence and average review volume for competing cars do not have a significant impact on focal car sales.Moreover,among high-end cars,the average review valence of competing cars negatively moderates the relationship between the review valence of the focal car and its sales.In sum,online reviews exhibit spillover effects only on high-end car sales.Specifically,the higher the number of reviews for competing car models directly dampens focal car sales,while the higher the review valence for competing cars diminishes the impact of the focal car’s review valence on its sales.Fourth,this study explores the inherent influence mechanism of online reviews on car sales from a consumer psychology perspective.The research model is based on the S-O-R theoretical model.The model takes consumers’ perceived usefulness and perceived information risk as mediating variables and need for cognition as moderating variables to explore the impact paths of the valence and volume of online reviews on consumers’ decisions to purchase cars.A random sample questionnaire is used to collect data from users who have purchased automobiles.The structural equation model is established and the empirical research is carried out.The results show that the valence and volume of online reviews have a positive direct effect on consumers’ decision to purchase cars.Perceived usefulness and perceived information risk mediate the relationship between online reviews and consumers’ decisions to purchase automobiles.Increasing review valence and review volume will increase consumers’ perceived usefulness of the information,which in turn will facilitate their purchasing decisions.Boosting review volume will reduce consumers’ perceived information risk,which in turn will stimulate purchase decisions.And an increase in review valence increases consumers’ perceived information risk,which in turn can inhibit their purchasing decisions.This result explains the conclusion of the study in Chapter 4,that there is an inverted U-shaped relationship between the valence of online reviews and car sales.Furthermore,consumers’ need for cognition moderated the relationship between review valence,review volume,and consumers’ perceived usefulness,perceived information risk.Need for cognition positively moderates the relationship between review valence and consumers’ perceived usefulness;whereas need for cognition negatively moderates the relationship between volume of reviews and consumers’ perceived usefulness.Need for cognition positively moderates the relationship between review valence and consumers’ perceived information risk.This study explores the relationship between online reviews and automobile sales and the inherent influence mechanism using the research methods of data mining and empirical analysis.The findings reveal that for high-involvement products the relationship between online reviews and car sales is nonlinear,unlike low-involvement products for which online reviews have a direct contribution to their sales.It enriches and improves the theories related to online reviews,offline product sales,and consumer purchase decisions,and provides theoretical references and methodological references for related research in this field.Meanwhile,it also provides useful guidance and suggestions for enterprises to accurately predict product sales,promote product sales,enhance product competitive strength,and formulate online marketing strategies. |
