Based on the ‘wisdom of the crowd’ effect (Surowiecki, 2005), consumers make use of reviews to make accurate product evaluations. However, due to the large amount of information and conflicting opinions in reviews, it becomes difficult for them to identify and consider the attributes that are relevant to their consumer situation.

Imagine you are browsing a webstore, looking for a new camera to take on your backpacking trip. For this situation, you prefer a camera that is lightweight, easy to use, shock-resistant and cheap. You don’t have a lot of experience with camera’s, so you decide to look at the reviews of other consumers that bought Camera X. As you browse through several reviews, you start to notice that a lot of reviews mention things like FPS, image stabilization, Wi-Fi connection and GPS tracking. However, the reviews are in conflict about the quality of the image stabilization and many mention the lack of a Wi-Fi connection. After reading most of the reviews, you decide that you want to look for a camera that has better image stabilization and a Wi-Fi connection, attributes which you originally didn’t pick as relevant for your situation …

The scenario above, is what Liu & Karahanna (2017) describe as the ‘swaying’ effect. After reading reviews, people might over-weigh irrelevant attributes and under-weigh relevant attributes. They suggest that attribute preferences are more heavily influenced by characteristics of the online reviews rather than by the relevance of the attributes to the consumers decision context.

Theory development & methodology

Liu & Karahanna (2017) developed their theory from the constructive preference perspective theory (Bettman, Luce, & Payne, 1998; Payne, Bettman, Coupey, & Johnson, 1992). This theory suggests that preferences are shaped by the interaction between the properties of the information environment of the choice problem and the properties of the human information-processing system. Liu & Karahanna (2017) propose that three characteristics of online reviews affect the assessment of attribute preference and theorize that these characteristics together may ‘sway’ attribute preferences.

1. the amount of information about attribute level performance,
2. the degree of information conflict about attribute level performance
3. the overall numeric rating and the attribute-level performance information

They conducted three studies, in which they provided the participants with a consumer scenario, asked them to weigh different attributes in terms of relevance and made them evaluate a digital camera based on reviews.

In study 1 they manipulated the three hypothesized factors and examined their effects on the attribute preferences. In study 2, they reproduced this study but added a monetary incentive to induce high motivation to process review information. The third study was a free simulation experiment to provide more realism and to allow for higher generalizability, in which verbal protocol analysis was used to capture and measure the factors.

Main findings

When the participants were asked to weigh the attributed based on the provided scenario, they placed more weight on the relevant attributed than the irrelevant attributes (in the scenario above, the attributes cost, ease-of-use and weight are relevant attributes, whereas image stabilization is not). But when they had to evaluate the camera based on reviews (that contained an uneven amount of information across different attributes, varying degrees of information conflict, and a numeric overall rating), the relevance of the attributes did not have a significant impact on attribute preferences.

The amount of attribute information in the reviews had the greatest impact on attribute preferences. Study 2 showed that the degree of attribute information conflict only affects attribute preferences when people have high motivation to process information. Study 3 showed consistent results. The studies provided evidence that attribute preferences that result from reading the reviews are primarily driven by the review characteristics and not by attribute relevance, thus supporting the hypothesized ‘swaying’ effect of online product reviews.

Practical implication.

What implications can be derived from these results? To support informed consumer decision making, it should be investigated how reviews should be organized and presented and how making sense of information conflicts can become less cognitively demanding. The effectiveness of some practical suggestions, such as providing a short description of the reviewer’s background (newegg.com), displaying the amount of positive and negative comments on an attribute (Q. (Ben) Liu, Karahanna, & Watson, 2011) and allowing people to see the overall rating from reviewers who have similar decision context, need to be investigated. Implementation of these suggestions allows consumer to filter reviews from people in a similar consumer scenario, makes making sense of conflicts become less demanding and causes the numeric overall rating to make more sense.

Strengths, weaknesses, suggested improvements

By conducting multiple studies with consistent results, the article provides strong evidence for generalizability & robust hypotheses, which enhances the external validity of the results. Nevertheless, there are some limitations. The study only examines a single product category (camera) and a single scenario. Additionally, the samples only consisted of students with a similar expertise of cameras. It would be interesting to examine whether the effects differ based on the consumer’s level of expertise with the product category (camera) or the product category itself. Additionally, to increase the generalizability of this study, it would be interesting to see if these results also apply on a sample that is more representative of the population (not only students).

I would love to hear your opinions on this. Do you recognize yourself in the ‘swaying’ effect? Are reviews influencing your preferences? 

References

Bettman, J. R., Luce, M. F., & Payne, J. W. (1998). Constructive Consumer Choice Processes. Journal of Consumer Research, 25(3), 187–217. https://doi.org/10.1086/209535

Liu, Q. (Ben), Karahanna, E., & Watson, R. T. (2011). Unveiling user-generated content: Designing websites to best present customer reviews. Business Horizons, 54(3), 231–240. https://doi.org/10.1016/j.bushor.2011.01.004

Liu, Q. Ben, & Karahanna, E. (2017). The dark side of reviews: The swaying effects of online product reviews on attribute preference construction. MIS Quarterly, 41(2), 427–448. https://doi.org/10.25300/misq/2017/41.2.05

Payne, J. W., Bettman, J. R., Coupey, E., & Johnson, E. J. (1992). A constructive process view of decision making: Multiple strategies in judgment and choice. Acta Psychologica, 80(1–3), 107–141. https://doi.org/10.1016/0001-6918(92)90043-D

Surowiecki, J. (2005). The Wisdom of Crowds. American Journal of Physics, 75(908), 336. https://doi.org/10.1038/climate.2009.73