Deception in the context of online product reviews: The effects of online deception cues and Awareness of online deception on attitude towards the product, online retailer trustworthiness and
product purchase intention
Author: Bogdan Stroe
Student number: 10858024 Master’s Thesis
Graduate School of Communication
Master Programme Communication Science
Thesis supervisor: Sandra Zwier
Abstract
The issue of anonymity in the online marketplace is raising many questions regarding the vulnerability of online communities and the likelihood of people to be deceived online. This paper particularly addresses the issue of online deception on the Amazon marketplace. Since Amazon does not have a non-verified purchase disclosure for its reviews, this study tested whether disclosing non-verified purchases on online product reviews can impact people’s attitudes towards the product, online retailer trust and product purchase intentions, when compared to disclosing verified purchases. Further on, this study tested whether this effect becomes stronger in case people are made aware of reading a potentially deceptive review in advance to reading it. Lastly, it was tested whether people’s involvement with the product moderates these effects. The study failed to confirm the hypotheses and the implications of this research highlight that identifying and combating deception through online reviews is a complex process that will need to take into account all variables of the online platform, people’s
preliminary knowledge about online deception and their familiarity with the product, brand and online retail platform itself.
Introduction
Worldwide retail ecommerce continues to be on the rise nowadays. Only the year 2016 has seen a 6% increase in online sales globally compared to the year 2015 and this growth is expected to continue in the following years (emarketer.com, 2016). The fast growth of the online marketplace is urging brands to closely monitor developments in order to be able to keep identifying and addressing their consumers. The rapid shift from traditional to the online media has also challenged brands and organizations to ensure that their online communities remain stable and reliable. That is, in order to maintain relationships between brands and consumers, online
retailers need to make sure their shopping network environment is a safe and trustworthy place to be. However, online privacy has always been a major concern for internet users and the internet itself remains a relatively incognito global network nowadays. The freedom of choice and selection that the online shopper owns has become endangered by the freedom of identity that people have online. In regard to this concern, there is more and more research being done on online deceptive acts and the consequences they can have. The present research will look into one of the many ways in which online deception occurs due to the vast usability of the internet.
Previous research on people’s online shopping behavior has paid special attention to the online retailer Amazon, recognized nowadays as the biggest online retailer worldwide and used by over 688.690 unique brands (360pi.com, 2016). The product diversity that is offered through the Amazon marketplace is continuously increasing and brands are seeking new opportunities by paying more and more unique attention to Amazon. For instance, the famous car brand Hyundai has recently dedicated an online marketing campaign only for premium users of the
considering Amazon as a singular key intermediary between them and their communities. Since brand communities on Amazon can be primarily identified through product reviews (Racherla, Mandviwalla & Connolly, 2012), my research will particularly address the issue of online deceptive acts through Amazon’s product review system.
Research to date indicates the impact that online reviews can have on people’s brand attitudes and purchase intentions (Doh & Hwang, 2009; Walther et al., 2012). In addition to that, research is also starting to investigate degrees of trust towards online retailers (Ruimei et al, 2012). The vulnerable nature of the internet raises many questions regarding issues of deceit and trust that can affect the way people behave and develop their attitudes and perceptions in the online environment. Distance and anonymity facilitate a loss of inhibition of internet users, allowing deceivers to be morally involved to a very small extent (Squicciarini & Griffin, 2012). To picture the size of this issue, it has been estimated that one third of all online product reviews are not authentic, i.e. placed by people who have not actually purchased the product of service (Streitfeld, 2012). Because of that, this research aims to investigate the issue of online deception in the context of online product reviews by looking at the effects of online deception cues and awareness of online deception on people’s attitudes toward the product, online retailer
trustworthiness and purchase intentions.
Another concept that grounds this research is represented by involvement with the
reviewed product. Racherla et al. (2012) demonstrated that the more people are involved with the product, the more they will pay attention to the content of a product review, whereas the less involved people are, the more inclined they will be to pay attention to other characteristics rather than the content of an online review per se. This raises the question if deception in online product
reviews will still have an impact on people’s attitudes towards the product, online retailer trust and purchase intentions in case they are less involved with the reviewed product.
All these considerations lead to the following research question: To what extent can deceiving people and creating awareness of deception in the context of online product reviews
influence their attitudes towards the product, online retailer trustworthiness and purchase
intention? Does people’s involvement with the product moderate this effect?
Theoretical background
Electronic word of mouth and online customer reviews
Muntinga, Moorman and Smit (2011) define electronic word of mouth (eWOM) as a form of online communication between consumers that is related to brands. The content that is being created as a result of this interaction is recognized under the term ‘user-generated-content’ (UGC), and it remains available for people to consume information about brands on the internet. A very important form of UGC are online customer reviews (Kroenke, 2016), defined as the online sharing of personal experiences with the products and services of brands and organizations (Chen & Xie, 2008). The valence of online reviews can affect attitudes (Walther et al., 2012) and product sales (Moe & Trusov, 2011). Moe and Trusov (2011) draw attention to the fact that online customer reviews have become a requisite of the internet and especially of online retailing. Research on online reviews helps brands understand consumers’ feedback to their products and allows companies to better define their objectives and goals when developing new products (Ku, Wei & Hsiao, 2012).
Online deception
eWOM is becoming more and more powerful because the amount of online communication between consumers related to brands is vastly increasing, but also because of its anonymous nature (Doh & Hwang, 2009). The anonymity characteristic, for instance, has led researchers to question issues of trust and deceit that can occur through eWOM (Drouin et al., 2016). The issue of online fake identities and online deception is heavily discussed by researchers nowadays (Munzel, 2016; Wu & Zhou, 2015; Tsikerdekis & Zeadally, 2014) but has been taken in consideration from an early stage of researching the internet world (Caspi & Gorski, 2006; Grazioli & Jarvenpaa, 2000).
What definition grounds the concept of online deception? The nature of deception is to hide the truth from the message recipient (Cambridge Dictionary, 2016). Buller and Burgoon (1996) define deception as a misleading act against others, while recipients are not consciously aware that such an act might occur to them. The deceiver intends to influence others by giving them a confusing or inaccurate belief about a specific object of communication. When looking at the online environment, deception can occur in various ways on different internet channels. For example, it can occur through fraudulent websites that imitate platforms of online retailers (Grazioli & Jarvenpaa, 2000). Also, Rathod and Pattewar (2015) highlight that hacking and malicious attacks can occur through email spam, a form of unrequested and fake communication that can cause the spreading of viruses, money loss or stealing confidential data. Another way in which online deception can manifest is through social media. Tsikerdekis and Zeadally (2014) identified two different types of online deception that can occur in social media. The most common strategy is known as content deception, which refers to manipulating information in
order to persuade its recipients. Another strategy is known as sender deception, referring to manipulating the identity information of the sender. The focus of the present study will be on content deception in the online retail environment, being contextualized to the example of online customer reviews, as they represent the social poll of online retailing (Racherla et al., 2012). Further, this research will focus on content deception in online customer reviews on Amazon because the latter is known to be one of the most recognized online retail platforms worldwide nowadays. Amazon is being practically used as a reference by many researchers who study online reviews and people’s online shopping behavior (David & Pinch, 2006; Chevalier & Mayzlin, 2006; Kokkodis, 2012; Walther et al., 2012).
Online deception awareness
In the present study, the concept of online deception awareness relates to informing a person that he or she may be reading a potentially deceptive online product review. Tsikerdekis and Zeadally (2014) suggest that awareness of potential deception is an essential factor that can lead to
identifying a misleading act. In support for this statement comes the Persuasion Knowledge Model, a theoretical framework developed by Friestad and Wright (1994), emphasizing the importance of disclosure cues in shaping people’s knowledge of persuasion. Training people and raising awareness can indeed contribute to protecting users of online media against delusive actions that have the purpose of persuading people’s perceptions in order to induce a false belief about products or services (Bambauer-Sachse & Mangold, 2013).
The anonymous nature of the internet can lead to a certain amount of reservation which can protect consumers from the act of deception. George, Giordano and Tilley (2016) indicate that people will tend to believe information they find online only if they perceive it as being
communicated in a sincere way. Otherwise, they may apply strategies to resist persuasion such as derogating the source of the message (Jacks & Cameron, 2003). When a person believes that an act of deception might occur to them on the platform of an online retailer, their satisfaction with the retailer will be negatively influenced (Román, 2010). Another tendency that may arise from the side of the reader would be to become skeptical towards the review (Sher & Lee, 2009). Fransen, Smit and Verlelgh (2015) did a thorough overview of resistance to persuasion strategies that people can use. The authors explain how effects of persuasive messages can be avoided if people employ avoidance strategies, for example. In such cases, people may either mechanically, physically or cognitively avoid a message. Another strategy is related to contesting the content of a message, such that if a message receiver has certain beliefs, then arguments against those beliefs may be disregarded/refuted.
In order to lessen the likelihood of deception, Amazon uses a “Verified purchase” icon next to reviews that have been verified to belong to the persons who truly purchased the reviewed products (Amazon.com, 2016). In case a review belongs to a person whose purchase hasn’t been verified, Amazon does not give a non-verified purchase disclosure.
Product involvement
The Elaboration Likelihood Model (ELM) of persuasion (Petty & Cacioppo, 1986) is a
theoretical framework that explains how people process persuasive information. The ELM is a dual processing model, indicating that persuasion can take place through a central or a peripheral route. When people are highly involved with processing information, persuasion will take place centrally via deep processing and will be based on the quality of the information. In case people are not highly involved with processing information, on the other hand, persuasion will take place
peripherally via more superficial processing (people will be rather persuaded by e.g. colors, images, sounds). Being involved with processing information means that an individual must be able and motivated sufficiently enough to retain messages. The process describing a person’s involvement with analyzing information is also called elaboration. A high level of elaboration indicates a high level of message processing, meaning that a person will carefully evaluate the content of a message by attentively analyzing the quality of its arguments. Oppositely, a low level of elaboration suggests that the message receiver may not be interested in that particular
information, making less cognitive effort to process information and being more likely to be persuaded by other factors rather than argument quality.
Racherla et al. (2012) did a study in which they looked into the key factors that determine people to trust online product reviews. In line with the ELM, their study indicates that people who are more involved with reading an online review will be more influenced by its content. Conversely, participants who were in the low involvement mode were more influenced by peripheral cues, such as information about the reviewer. The ELM is frequently used in marketing communication research because brands often rely on it when developing new strategies of influencing people’s attitudes towards them and their products (Lin, Lee & Horng, 2011). Depending on whether involvement is high or low, people may behave differently when searching or evaluating products in the online environment (Micu & Pentina, 2014).
Li and Hitt (2010) discuss the importance of price effects in people’s evaluations of online product reviews. The authors highlight the idea that the perceived quality of a product can be particularly influenced by the price of the product. Depending on the configuration of the review system, the extent of this bias can be higher or lower. A review system that doesn’t
separate quality reviews from price reviews will most likely face a diffusion due to the price effects, especially if the person reading the review is not familiar with the product. Linking this idea to ELM, my study wants to test if product price can be a key determinant of people’s involvement with a product. Moreover, such a finding may also highlight the need of separating review systems depending on price versus quality. People exposed to a cheap product may be less influenced by the content of the review, whereas people exposed to an expensive product may be more influenced by the content of the review. Wu, Huang and Fu (2011) also associated the role of involvement with that of product price, arguing that such a link reflects the consumer’s evaluation of the product when reading product information.
Brand attitudes
Congruity theory (Osgood & Tannebaum, 1955) highlights the function of opinion triggers by underlining that an individual’s attitude toward an object of communication can be influenced by the strength of the individual’s relationship with the source of information. That source can either express arguments pro or against that object, and depending on this valence one’s attitude can be influenced.
Dellarocas, Zhang and Awad (2007) discuss the importance of online product reviews in impacting the attitudes of consumers towards brands and their products and services. Positive reviews are expected to influence the reader to adopt a positive perception of a brand and/or its reviewed product/service, whereas negative reviews are expected to lead to an opposite effect, influencing the reader to adopt a more negative perception of a brand and/or its reviewed product/service. Research following the above-mentioned study confirmed this rationale (Doh and Hwang, 2009; Walther et al., 2012).
In case the reader is made aware of being exposed to a potentially deceptive review, he/she will derogate the source (Jacks & Cameron, 2003) or will become skeptical towards the review (Sher & Lee, 2009). These potential effects suggest that if the reader is more aware of deception, then attitudes will develop in a direction that is incongruent with the content of the review. In the opposite case, if the reader is less aware of deception, he/she will be more susceptible to being deceived. This means that the likelihood of acting upon the intent of the review will be higher and people will be more likely to develop attitudes that are congruent with the content of the review. The following hypotheses arise:
H1a: When online deception cues are absent in an online review, attitudes toward the product will be more congruent with the message intent than when online deception cues are present.
H1b: When readers are made aware of the potential presence of online deception cues in online product reviews, the effect of the presence of online deception cues on attitudes toward the product will be stronger than when readers are not made aware of the potential presence of online deception cues.
H1c: The effect of the presence of online deception cues in online product reviews on attitudes toward the product will be stronger for product reviews of high involvement products (high price) than for product reviews of low involvement products (low price). H1d: When readers are made aware of the potential presence of online deception cues, the effect of the presence of online deception cues on attitudes toward the product will be stronger than when readers are not made aware of the potential presence of online
deception cues, but this effect will be stronger for product reviews of high involvement products (high price) than product reviews of low involvement products (low price). Online retailer trustworthiness
Due to the large amount of online retailers, each of them faces the vital challenge of building a strong relationship with their clients and customers. Most importantly, depending on the relationship of the retailer with the customer, a brand can decide to use an online retailer’s platform in order to sell its products and services. Ruimei et al. (2012) investigated the notion of E-satisfaction, on which the trust that a consumer holds for an online retailer can also be based. Their study indicates that the more secure customers feel about an online retailer, the higher their trust will be towards it.
However, in case the practice of fake reviews will continue to rise, a lack of trust may expand towards the online retailer as a whole, imperiling an important source of information that consumers relate to and harming people’s perceptions and behaviors towards brands and online retailers (Li, Chou & Yu, 2013)
The present study will particularly investigate whether deception cues in online product reviews will have a detrimental effect on online retailer trustworthiness. Based on the findings of Racherla et al. (2012), it is also expected that the effects of deception cues and deception
awareness on people’s trust towards the online retailer will be stronger for people who are more involved with reading the online review when compared to people who are less involved. The following hypotheses are formulated:
H2a: When online deception cues are absent in an online review, online retailer trustworthiness will be more congruent with the message intent than when online deception cues are present.
H2b: When readers are made aware of the potential presence of online deception cues in online product reviews, the effect of the presence of online deception cues on people’s trust towards the online retailer will be stronger than when readers are not made aware of the potential presence of online deception cues.
H2c: The effect of the presence of online deception cues in online product reviews on people’s trust towards the online retailer will be stronger for product reviews of high involvement products (high price) than for product reviews of low involvement products (low price).
H2d: When readers are made aware of the potential presence of online deception cues, the effect of the presence of online deception cues on people’s trust towards the online
retailer will be stronger than when readers are not made aware of the potential presence of online deception cues, but this effect will be stronger for product reviews of high involvement products (high price) than product reviews of low involvement products (low price).
Purchase intention
Past research highlights the importance of online customer reviews in determining consumers’ purchase intentions (Chevalier & Mayzlin, 2006; Dellarocas et al., 2007). Ketelaar et al. (2015) indicate that people’s purchasing intentions can depend on the valence of online reviews,
meaning that more positive reviews lead to higher purchasing intentions, whereas more negative reviews lead to lower intentions to purchase a product.
Park, Lee and Han (2007) further indicate that the quality of online reviews can positively affect product purchase intention. Their statement is supported by the Elaboration Likelihood Model (Petty & Cacioppo, 1986), in that that the more a reader is cognitively involved with reading an online review, the more he/she will base their purchase intention on the perceived argument quality. O’Keefe andMcEachern (1998) suggest that a high price will motivate consumers to process more information before purchasing a product, leading to the assumption that a review of a more expensive product will be read with more attention than a review of a cheap product.
Wu et al. (2011) found that perceived risk has a negative influence on product purchase intention. In order to link this finding to the context of online deception, my study will rely on the idea that by increasing awareness of a potentially deceptive online review, people will perceive the risk of being deceived, and this will have a negative impact on their purchase intention. The following hypotheses arise:
H3a: When online deception cues are absent in an online review, people’s purchase intention will be more congruent with the message intent than when online deception cues are present.
H3b: When readers are made aware of the potential presence of online deception cues in online product reviews, the effect of the presence of online deception cues on purchase intention will be stronger than when readers are not made aware of the potential presence of online deception cues.
H3c: The effect of the presence of online deception cues in online product reviews on people’s purchase intention will be stronger for product reviews of high involvement products (high price) than for product reviews of low involvement products (low price). H3d: When readers are made aware of the potential presence of online deception cues, the effect of the presence of online deception cues on purchase intention will be stronger than when readers are not made aware of the potential presence of online deception cues, but this effect will be stronger for product reviews of high involvement products (high price) than for product reviews of low involvement products (low price).
All hypotheses have the same structure for all three dependent variables, as shown in the conceptual model in Image 1.
Methods section
Selection of research units (respondents, organizations, media)
For this research, a convenience sample of participants was used, mainly consisting of English-speaking students and young professionals that own social media accounts. The reason for this is that students and young professionals are familiar with the online environment of retailer
platforms such as Amazon, as well as being at ease with taking an online survey in a time- and cost-effective manner. The participants were recruited through social media (i.e. Facebook, Skype) and were directed to the online survey platform that belongs to Qualtrics Survey Software Tools. All participants were informed at the start of the survey that they were taking part in a study investigating people’s ideas of products sold through Amazon, conducted for the Graduate School of Communication, part of The University of Amsterdam. All respondents also actively agreed with an informed consent before taking part in the study.
Characteristics of research units
This research was conducted with a sample of 125 participants, out of which 84 (32 male and 52 female) were eligible for the analysis. The reason for excluding participants was because 27 of them did not complete the online survey and the rest of 14 had a variance coefficient which was less than 1 on more than two dependent variables.
Respondents were aged 24 on average (Mage = 24.44, SDage = 2.96), 42 (50%) of them had a bachelor’s level degree and 31 (36.9%) of them were master graduates. The remaining 11 (13.1%) participants had finished their upper secondary education as their highest level of education.
Few respondents were regular Amazon users (n = 5, 6%); most of them (n = 39, 46.4%) indicated that they hadn’t purchased any items through Amazon in the past 6 months and another 30 (35.7%) reported to have done this once or twice. There were more participants (n = 35, 41.7%) who had zero awareness of the Philips Wake-up Light product, the first product that was used as one of the stimuli in this study, when compared to the 22 (26.1%) participants who indicated a high level of awareness for this product. In case of the Philips Light Bulbs, the other product used as a stimulus in the study, 24 (28.6%) participants reported to have not been familiar with Philips Light Bulbs at all, and 22 (26.2%) of them indicated a high level of awareness for these products.
When asked about their preliminary awareness about online deceptive reviews (M = 3.64, SD = 1.07), 11 (13.1%) respondents indicated low awareness levels, 29 (34.5%) of them reported
to be moderately aware and most of them (n = 44, 52.4%) indicated to have high awareness levels.
Research design
In order to answer the research question and hypotheses, an online experimental study was conducted through Qualtrics Survey Software Tools. A 2x2x2 factorial design was used, having the first factor represented by the Presence of online deception cues, treated as a between-subjects variable with 2 levels: absent vs. present. The second factor concerned inducing Awareness of online deception and this factor was also treated as a between-subjects variable
with 2 levels: less aware and more aware. The third factor finally concerned Product involvement and was treated as a within-subjects factor, being operationalized through exposing participants
to online reviews for a more expensive product, respectively to online reviews for a cheaper product.
Experimental stimulus
In all conditions, participants were shown two online reviews for a Philips product on a mock-up Amazon page. The reviews for both products were positively written, meaning that a successful deceptive act may lead to more positive attitudes, online retailer trust and purchase intentions. The online reviews on the mockup Amazon page as shown to participants in the study can be seen in Appendix 1.
Presence of online deception cues
An article on wikihow.com (2016) indicates that one main indicator that may help determining whether an online review is fake is the disclosure of verified purchases, as is customarily done on Amazon. The two conditions concerning the Presence of online deception cues factor (absent vs. present) were created as follows:
The condition showing absent online deception cues was displaying a review containing the standardized Amazon verified purchase disclaimer next to it (Image 2).
Image 2. Verified purchase disclosure
The condition showing present online deception cues was created by displaying reviews containing a non-verified purchase disclaimer next to it (Image 3).
Image 3. Non-verified purchase disclosure
Note that, in reality, there exists a disclosure on Amazon when a review belongs to a person with a verified purchase similar to the one used in this study. However, there is no existing disclosure on Amazon reflecting that a review belongs to a person whose purchase has not been verified by Amazon, and this disclosure was thus especially created for the present study. This research tested the use of a non-verified purchase disclosure on Amazon reviews in order to highlight the idea that non-verified reviews are potentially deceptive and that people could be wrongly influenced by them.
Awareness of online deception
The second experimental factor was represented by induced Awareness of online deception. This factor was treated as a between-subjects variable with 2 levels: less aware and more aware. The two conditions underlying the induced online deception awareness factor were created as follows:
The condition in which participants were made more aware about online deception was created by drawing the reader’s attention towards the purchase disclosure icon in advance of reading the review. This was done by the message “Mind that the review comes from a user whose purchase has been verified” that would appear preliminary to showing an absent online deception cues condition. Respectively, the message “Mind that the review comes from a user whose purchase hasn’t been verified” would appear preliminary to showing a present online deception cues condition. The condition in which participants were made less aware about online
deception was created by not preliminarily pointing the reader towards the purchase disclosure icon.
Product involvement
The third and final experimental factor, Product involvement, was treated as a within-subjects factor, being operationalized through showing reviews for a higher involvement (expensive) product, as well as a lower involvement (cheap) product, belonging to the same category of product and brand category, and by also mentioning their price. All participants saw an Amazon review twice. First they saw a review for an expensive product, followed by a review for a cheap product. The product which was used for the high involvement condition was a Wake-up Light developed by Philips, having the price of 133.56$ (Image 4).
Image 4. High involvement (expensive) product
The product used for the low involvement condition was a Philips Light Bulb, having the price of 4.93$ (Image 5).
Image 5. Low involvement (cheap) product
The online reviews as shown to participants in the study can be seen in Appendix 1. As can be seen there, the reviews for both products were positively written, meaning that a
successful deceptive act may lead to more positive attitudes toward the product, online retailer trust and purchase intentions.
Dependent measures
Attitude toward the product. The first measured variable was represented by attitudes
toward the product. 3 items were used to measure this variable, having a measurement scale from 1 (strongly disagree) to 5 (strongly agree). The 3 items were created by the author of this research paper and consisted of the following statements: “I think the Wake-up Light/ Light Bulb works well”; “I like this product”; “I think this product offers a great wake-up/light experience”. All variables reflecting these items were used to calculate the mean attitude toward the product variables. Attitude toward the Wake-up Light (high involvement product) proved to be a reliable
scale (Cronbach’s α = .75, M = 3.65, SD = .72), as well as attitude toward the Light Bulb (low involvement product) which had Cronbach’s α = .74 (M = 3.46, SD = .70).
Online retailer trustworthiness. The second measured variable was online retailer
trustworthiness. 5 items were used to measure this variable, having a measurement scale from 1 (distrust it greatly) to 5 (trust it greatly). The 5 items were created by the author of this research paper and were represented by the following statements: “If I had to think of ordering a Wake-up Light/ Light bulb from Amazon, I would…”; “If I had to read product reviews about the Wake-up Light/ Light bulbs on Amazon, I would…”; “If I had to read product information about the
Wake-up Light/ Light bulbs on Amazon, I would…”; “Thinking of Amazon as a safe/unsafe online retail platform for the Wake-up Light/ Light bulbs, I…”; “If I had to compare Amazon to an offline retailer in purchasing the Wake-up Light/ Light bulbs, I would…”. All 5 items were used to calculate the mean online retailer trust variable. Online retailer trust that was measured after showing the high involvement product indicated a reliable scale (Cronbach’s α = .73, M = 3.71, SD = .58) as well as online retailer trust measured after showing the low involvement product (Cronbach’s α = .82, M = 3.47, SD = .72)
Purchase intention. The third dependent variable that was measured was product purchase
intention. 3 items were used to measure this variable, having a measurement scale from 1 (strongly disagree) to 5 (strongly agree). The 3 items were created by the author of this research paper and were as following: “I would consider purchasing the Wake-up Light/ light bulb in the future.”; “I am willing to invest in a product that brings me a better wake-up experience/ quality light.”; “I would buy this product as a gift for someone/ I would buy this product because I am concerned with my light experience.”. All variables assigned to the 3 items that were just
described have been used to compute the mean purchase intention variables. Purchase intention for the Wake-up Light indicated a reliable scale (Cronbach’s α = .75, M = 3.39, SD = .97), as well as purchase intention for the Light Bulb (Cronbach’s α = .74, M = 3.25, SD = .89).
Biographical items
After showing the products for both involvement levels and measuring the dependent variables, respondents’ general information was collected. Measured were: gender, age, education level, Amazon product purchasing frequency, frequency of reading online reviews, previous awareness of Philips Wake-up Light products, previous awareness of the Philips Light Bulb products, and preliminary online deception awareness. Please check Appendix 2 for a full overview of how these items were measured.
Manipulation checks
Lastly, two manipulation check questions were asked. The first question measured whether the respondents had noticed the verified purchase or non-purchase purchase icon:
“Which of the two icons below did you see in the Amazon examples that you have read?
Verified purchase
Non-verified purchase
Not sure/ Don’t know”
The second question measured whether people were more involved with the Wake-up Light product or with the Light Bulb product:
“Would you think longer before buying a Wake-up Light than a light bulb, or would you think longer before buying a light bulb than a Wake-up Light?
Think longer before buying a Wake-up Light
Think longer before buying a light bulb”
The end of the survey debriefed respondents about the true purpose of the research, informing them that the information presented in the survey was made for the specific purpose of the research and that it did not represent a real case of online reviews on Amazon.
Results
Analysis of manipulation checks
In order to test whether the manipulation was successful, an Independent Samples t-test was conducted, having Presence of online deception cues as independent variable (0 - absent, 1 - present) and the manipulation check item for perceived disclosure icon as dependent variable. The test was not statistically significant, t(82) = -.98, p = .328. The difference between the means of absent deception cues (M = 2.14, SD = 1) and present deception cues (M = 2.33, SD = .75) was not statistically significant. This means that participants who were shown the non-verified
purchase disclosure did not have a significantly higher chance of being deceived when compared to the group of respondents who saw the verified purchase disclosure.
In order to test whether the Wake-up Light was indeed perceived by the participants of this research as a high-involvement product in comparison to the Light Bulb, a frequency analysis was conducted to show how many people would think longer before buying a wake-up light, and how many people would think longer before buying a light bulb (manipulation check item 2).
The analysis showed that out of 84 respondents, 77 (91.7%) would think longer before buying a Wake-up Light and only 7 (8.3%) would think longer before buying a light bulb. A Paired
Samples t-test was conducted, with one variable representing people who reported to think longer before a Wake-up Light, and the other variable representing people who reported to think longer before buying a light bulb. The result was statistically significant, such that the amount of respondents who think longer before buying a Wake-up Light (M = .91, SD = .27) was
significantly higher than the amount of people who think longer before buying a light bulb (M = .08, SD = .27), t(82) = -13.73, p < .00, 95% CI [-.95, -.71]. This finding allows this study to treat the Wake-up Light as a high-involvement product and the Light Bulb as a low-involvement product.
Main analyses
The effect of online deception and awareness of online deception on attitudes toward the product.
A two-way Analysis of Variance, with Presence of online deception cues (0 - absent, 1 - present) and Awareness of online deception (0 - less aware, 1 - more aware) as independent variables and attitude towards the product as dependent variable, indicated that neither Presence of online deception cues, F(1, 80) = 1.96, p = .165, nor Awareness of online deception F(1, 80) = 1.21, p = .273, nor the interaction between both variables, F(1, 80) = 0.32, p = .858, had a significant effect on attitude towards the product. People who were in the absent deception condition indicated higher attitude levels (M = 3.63, SD = .45) when compared to the respondents who were in the present deception condition (M = 3.48, SD = .59). In other words, the direction of the effect was correctly predicted, however the effect was not statistically significant and the results cannot be generalized to the entire population. People who were in the less awareness of deception
condition reported lower attitude levels (M = 3.50, SD = .53) than respondents who were in the more awareness of deception condition (M = 3.61, SD = .53), however the effect was also not statistically significant. H1a and H1b are not supported.
In order to compare the effects of the Presence of online deception cues between the group of respondents who saw the high-involvement product (Wake-up Light) and the group who saw the low involvement product (Light Bulb) on their attitude towards the product they were exposed to, two Independent Samples t-tests were conducted. The first test used Presence of online deception cues (absent vs. present) as an independent variable and attitude towards the Wake-up Light as dependent variable. The test was not statistically significant, t(82) = -.10, p = .921, meaning that there was no significant difference between the effects of being exposed to a verified purchase icon/ absent deception condition (M = 3.64, SD = .69) and the effects of being exposed to a non-verified purchase icon/ present deception condition (M = 3.65, SD = .75) on people’s attitudes towards the high-involvement product. The second test used Presence of online deception cues (absent vs. present) as an independent variable and attitude towards the Light Bulb as dependent variable. The result was statistically significant, t(82) = 2.10, p < .05. The difference between being exposed to a verified purchase icon/ absent deception condition (M = 3.62, SD = .55) and being exposed to a non-verified purchase icon/ present deception condition (M = 3.30, SD = .80) on people’s attitudes towards the Light Bulb was statistically significant, such that people who saw the non-verified purchase icon had significantly lower attitudes toward the Light Bulb than the people who saw the verified purchase icon.
The differences between the means of the effects of online deception cues on attitude towards the high-involvement product, Mdif = -.01, 95% CI [-.33, .30], and attitude towards the
low-involvement product, Mdif = .31, 95% CI [.01, .61], had overlapping Confidence Intervals, meaning that involvement with the product (high price vs low price) did not bring a significant difference between the effects of present online deception and absent online deception on product attitudes. Thus, it cannot be stated that the one effect was greater than the other depending on people’s involvement with the product they were exposed to. Thereby, H1c is not supported.
Further analyses tested the effects of Awareness of online deception of the group of respondents who saw the high-involvement product (Wake-up Light) and the group who saw the low involvement product (Light Bulb) on their attitude towards the product they were exposed to. An Independent Samples t-test, with Awareness of online deception (less aware vs. more aware) as independent variable and attitude towards the high-involvement product as dependent variable, indicated that participants who were less aware of reading potentially deceptive reviews (M = 3.54, SD = .78) had lower attitude levels than the respondents who were more aware of reading potentially deceptive reviews (M = 3.75, SD = .65). However, this result was not statistically significant, t(82) = -1.32, p = .190, thus it could not be determined whether increasing people’s awareness of reading potentially deceptive reviews would significantly affect their attitude towards the high-involvement product. A second Independent Samples t-test, with Awareness of online deception (less aware vs. more aware) as independent variable and attitude towards the lowinvolvement product as dependent variable, was also not statistically significant, t(82) = -1.12, p = .903. There was no statistically significant difference between the effects of being less aware of reading potentially deceptive reviews (M = 3.45, SD = .66) and being more aware of reading potentially deceptive reviews (M = 3.47, SD = .74) on people’s attitude towards the low-involvement product.
The mean differences between the effects of Awareness of online deception on attitude towards the high-involvement product, Mdif = -.20, 95% CI [-.52, .10], and attitude towards the low-involvement product, Mdif = -.01, 95% CI [-.32, .28], had overlapping Confidence Intervals. This finding indicates that involvement with the product (high price vs low price) did not bring a significant difference between the effects of making people less, respectively more aware of reading potentially deceptive reviews on their attitudes towards the product. There are no grounds to state whether one effect is greater than the other depending on people’s involvement with the product they were exposed to. Thereby, H1d is not supported.
The effect of online deception and awareness of online deception on online retailer
trustworthiness.
A two-way Analysis of Variance, with Presence of online deception cues (0 - absent, 1- present) and Awareness of online deception (0 - less aware, 1 - more aware) as independent variables, and online retailer trust as dependent variable, reflected no statistically significant effect on trust towards the online retailer for neither Presence of online deception cues, F(1, 80) = .961, p = .330, Awareness of online deception, F(1, 80) = .242, p = .624, or the interaction between these two variables F(1, 80) = .000, p = .984. Respondents who saw the verified purchase icon (absent deception condition) reported higher levels of online retailer trust (M = 3.65, SD = .51) than participants who saw the non-verified purchase icon (present deception condition), M = 3.53, SD = .67. Even though the predicted direction of the effect could be noticed, the difference between the two conditions was minimal and not statistically significant. Respondents who were less aware of reading potentially deceptive reviews reported lower levels of online retailer trust (M = 3.56, SD = .52) than participants who were more aware of reading potentially deceptive reviews
(M = 3.62, SD = .65), however this effect was not statistically significant. These results indicate that it cannot be concluded that online deception or online deception awareness affect people’s trust towards the online retailer. H2a and H2b are not supported.
The following analysis measured the effects of Presence of online deception cues on people’s trust towards the online retailer after they were exposed to the review of the high-involvement product (Wake-up Light). An Independent Samples t-test was conducted, with Presence of online deception cues (absent vs. present) as independent variable and online retailer trust as dependent variable. The results were not statistically significant, t(82) = .110, p = .912. This finding indicates that after respondents were exposed to the review of the high-involvement product, there was no statistically significant difference between the effect of seeing the verified purchase disclosure icon (absent deception), M = 3.71, SD = .56 and the effect of seeing the non-verified purchase disclosure icon (present deception) on their trust towards the online retailer, M = 3.70, SD = .62. Further on, a statistical analysis measured the effects of Presence of online deception cues on people’s trust towards the online retailer after they were exposed to the review of the low-involvement product (Light Bulb). An Independent Samples t-test was conducted, having Presence of online deception cues (absent vs. present) as independent variable and online retailer trust as dependent variable. Even though the direction of the predicted effect was correct, the results were was not statistically significant, t(82) = 1.47, p = .144. After people were
exposed to the low-involvement product, there was no significant difference between the effect of seeing the verified purchase disclosure icon (absent deception), M = 3.59, SD = .56 and the effect of seeing the non-verified purchase disclosure icon (present deception), M = 3.36, SD = .85, on their reported trust towards the online retailer.
The mean difference of the effect of Presence of online deception cues on trust towards the online retailer after respondents saw the high-involvement product, Mdif = .01, 95% CI [-.24, .27], had overlapping Confidence Intervals with the mean difference of the effect of Presence of online deception cues on trust towards the online retailer after respondents saw the
low-involvement product Mdif = .23, 95% CI [-.08, .54]. This result indicates that people’s involvement with the product did not significantly influence these effects, and therefore the effects cannot be compared depending on how involved people were with the product they were exposed to. H2c is not supported.
Further on, an Independent Samples t-test was run in order to measure what were the effects of Awareness of online deception on people’s trust towards the online retailer after being exposed to the review of the high-involvement product (Wake-up Light). The independent variable was represented by Awareness of online deception (less vs. more) and the dependent variable was online retailer trust. The result was not statistically significant, t(82) = -.47, p = .639. After people saw the review for the high-involvement product, there was no statistically
significant difference between online retailer trust reported by the group who was less aware of being exposed to a potentially deceptive review (M = 3.68, SD = .57) and online retailer trust reported by the group who was more aware of being exposed to a potentially deceptive review (M = 3.74, SD = .60). A second Independent Samples t-test was conducted in order to measure what were the effects of Awareness of online deception on people’s trust towards the online retailer after they saw the review of the low-involvement product (Light Bulb). Awareness of online deception was treated as independent variable (less aware vs. more aware) and online retailer trust was the dependent variable. The result was not statistically significant, t(82) = -.28, p = .780.
After respondents saw the low-involvement product, there was no statistically significant difference between online retailer trust reported by the group who was less aware of being exposed to a potentially deceptive review (M = 3.45, SD = .63) and online retailer trust reported by the group who was more aware of being exposed to a potentially deceptive review (M = 3.50, SD = .81).
The mean difference of the effect of Awareness of online deception on trust towards the online retailer after respondents saw the high-involvement product, Mdif = -.06% CI [-.31, .19], had overlapping Confidence Intervals with the mean difference of the effect of Presence of online deception cues on trust towards the online retailer after respondents saw the low-involvement product Mdif = -.04, 95% CI [-.36, .27]. People’s involvement with the product did not have a significant influence on the effects of making people less, respectively more aware of reading potentially deceptive reviews on their trust towards the online retailer. It cannot be determined which effect is stronger depending on how involved people were with the product they were exposed to. H2d is not supported.
The effect of online deception and awareness of online deception on purchase intention.
A two-way Analysis of Variance was conducted in order to determine the effects of the Presence of online deception cues and Awareness of online deception on people’s purchasing intention. Presence of online deception cues (0 - absent, 1 - present) and Awareness of online deception (0 - less aware, 1 - more aware) were the independent variables of the analysis, whereas purchase intention was the dependent variable. The results show that neither Presence of online deception, F(1, 80) = .871, p = .354, Awareness of online deception, F(1, 80) = 1.30, p = .257, or the
effect on people’s purchase intention. Respondents who were in the absent deception condition reported higher purchase intention levels (M = 3.41, SD = .76) than respondents who were in the present deception condition (M = 3.24, SD = .72). The direction of the effect was correctly predicted, however the results were not significant. Respondents who were less aware of being exposed to a potentially deceptive review reported higher levels of purchase intention (M = 3.43, SD = .72) than participants who were made more aware of being exposed to a potentially
deceptive review (M = 3.23, SD = .76). Similarly, even though the direction of the effect was correctly predicted, the results were not statistically significant. These results suggest that there are not enough grounds to state that Presence of online deception cues or Awareness of online deception have an effect on people’s purchase intention. H3a and H3b are not supported.
The following analysis tested the effects of Presence of online deception cues on purchase intention for the high-involvement product (Wake-up Light). An Independent Samples t-test was run, with Presence of online deception cues as the independent variable (absent vs. present) and Wake-up Light purchase intention as the dependent variable. The results were not statistically significant, t(82) = .22, p = .824. Respondents who saw the verified purchase icon (absent deception condition) had slightly higher purchasing intentions for the Wake-up Light (M = 3.42, SD = .93) than respondents who saw the non-verified purchase icon (present deception
condition), however this difference was not statistically significant. A second Independent Samples t-test was conducted, in order to test the effects of the Presence of online deception cues on purchase intention for the low-involvement product (Light Bulb). Presence of online deception cues was treated as independent variable (absent vs. present) and Light Bulb purchase intention as dependent variable. The results were not statistically significant, t(82) = 1.51, p = .134. People
who were in the absent deception condition had a higher purchase intention for the Light Bulb (M = 3.40, SD = .92) than people who were in the present deception condition (M = 3.11, SD = .84), however this difference was not statistically significant.
The mean differences between the effects of the Presence of online deception cues on Wake-up Light purchase intention, Mdif = -.04, 95% CI [-.37, .47], and Light Bulb purchase intention, Mdif = .29, 95% CI [-.09, .67], had Confidence Intervals that overlapped. Involvement with the product (high price vs. low price) did not significantly affect the difference between these effects, therefore it cannot be determined whether one effect is greater than the other based on people’s involvement with the product to which participants were exposed to. H3c is not supported.
Further on, statistical analyses were conducted in order to test the effects of Awareness of online deception of the group who saw the high-involvement product (Wake-up Light) and the group who saw the low involvement product (Light Bulb) on their reported purchase intention for the product to which they were exposed to. Firstly, an Independent Samples t-test was conducted, having Awareness of online deception (less aware vs. more aware) as independent variable and purchase intention for the high-involvement product as dependent variable. The results were not statistically significant, t(82) = -.92, p = .356. There was no statistically significant difference between the effects of being less aware of reading potentially deceptive reviews (M = 3.50, SD = .93) and the effects of being more aware of reading potentially deceptive reviews (M = 3.30, SD = 1) on people’s purchase intention for the Wake-up Light. A second Independent Samples t-test was conducted, with Awareness of online deception (less aware vs. more aware) as independent variable and purchase intention for the low-involvement product as dependent variable. The test
result was not statistically significant, t(82) = 1.06, p = .291. The difference between the effects of being less aware of reading potentially deceptive reviews (M = 3.36, SD = .92) and the effects of being more aware of reading potentially deceptive reviews (M = 3.15, SD = .86) on people’s purchase intention for the Light Bulb was not statistically significant.
The mean differences between the effects of Awareness of online deception on Wake-up Light purchase intention, Mdif = .19, 95% CI [-.22, .61], and Light Bulb purchase intention, Mdif = .20, 95% CI [-.18, .59], had overlapping Confidence Intervals. This result suggests that
Involvement with the product (high price vs. low price) did not significantly affect the difference between the effect of making people less aware of reading potentially deceptive reviews and the effect of making people more aware of reading potentially deceptive reviews on people’s purchase intention for the product they were exposed to. The effects cannot be compared, thus H3d is not supported.
Discussion
The present study aimed to investigate a concept that has become more and more prominent for social research nowadays: online deception. Within the broad spectrum of this concept, the main focus of this research was from an online retail perspective and it particularly addressed its social dimension through online product reviews. Contrary to the expectations, this study did not manage to confirm the theoretical assumption that using online deceptive cues may influence review readers to build attitudes, online retailer trust and purchase intentions which are less congruent with the intent of the review. The assumption that online deception awareness may strengthen this effect was also not confirmed by this study.
Several plausible reasons arise in explaining the unsuccessful results of this study. Firstly, as Munzel (2016) highlights, it is important recognize the importance of persuasion knowledge activation. Most participants in this study (n = 44) indicated to be familiar with the topic of online deceptive reviews and this predisposed condition may explain the non-findings of this study. Even though most participants indicated to have rarely used Amazon in the past 6 months, this does not mean that they are not familiar with the potentially deceptive purpose that may exist behind an online product review. In support of this reasoning, it may be that participants used resistance to persuasion strategies (Fransen et al., 2015) in order to avoid being persuaded by an online review. Another aspect that may have lessened the chances of finding a significant result could be linked to the review valence that was used in this experiment. Kusumasondjaja, Shanka and Marchegiani (2012) have found that even though a positive review leads to a greater initial message trust when compared to a negative review, a negative review is overall considered to be more credible and therefore more persuasive. Future research on this topic might focus more on reviews that have a negative valence when examining the potentially deceptive effects of online product reviews.
In order to measure the isolated effects of deception and deception awareness in the context of online product reviews, this study has avoided the use of other factors that are
normally present on Amazon. These factors include number of reviews for the product, reviews previously placed by the reviewer and averaged star ratings. George et al. (2016) stress the importance of these factors by arguing that they can be used as main determinants of purchasing decisions. This aspect suggests that there may be other factors that are indeed more important for consumers when reading online reviews rather than seeing a verified purchase disclosure. People
may pay attention to more than a single review when they want to find out about others’
experiences with products, and their attitudes and perceptions may only be formed once they are able to compare and weight a certain amount of reviews. Also, the verified purchase disclosure icon does not have a big size on the Amazon website, and this study confirms that people may not pay much attention to the sign. Indeed, nearly half of participants did not remember what
purchase disclosure (verified purchase or non-verified purchase) they were exposed to. However, the purpose of this study was to bring a fair approach to the purchase disclosure element that is present on Amazon reviews, in such a way that non-verified purchases should be equally
disclosed, as verified purchases are. From the user’s point of view, this approach would be more ethical and this observed imbalance is what led to conducting this research.
The sample composition may represent another factor which does not come in support for the external validity of this study. Doing research on a convenience sample of bachelor and master graduates does not truly reflect the profile variety of online retail sites customers. A sample with a more varied background information (e.g. age, education levels, and online deceptive review awareness) might have reported different results which could have been more likely to be generalized to the population.
Product involvement was associated with product price in this study, assuming that people will be more involved with the product and the review if the product is more expensive and respectively will be less involved with the product and the review if the product is cheap. This assumption was indeed correct because most participants reported that they would think longer before buying a Wake-up Light (n=77) rather than thinking longer before buying a light bulb (n=7).
Both products that were used in this research belong to the Philips brand, a brand with a strong global identity. Doing research on the example of Philips products may also have been an impediment for this research, since people may already have a formed opinion about this brand. Even though some participants were not familiar with the products that were shown in this study, their familiarity with the Philips brand as a whole was not assessed and this could have also diminished the probability of finding effects of online deception on people’s attitude towards the Philips brand.
All these considerations combined justify the inability of the present study to provide a definitive answer to the research question, and more research is needed.
Conclusion
Investigating the social dimension of online retailers such as Amazon will remain of high relevance for social and online marketing research. Through such platforms online communities are continuously being born, and that is where they can grow and diversify very fast nowadays. The topic of online deception will also remain of high relevance for this context and its
importance will undoubtedly increase with the growth of people’s online presence. Other researchers may readdress the research question of this study by particularly paying attention to the limitations of this research. Online deception can be analyzed from the many angles
highlighted by Tsikerdekis and Zeadally (2014) and depending of the type of deception that is being questioned, research patterns could emerge in order to bring a better understanding of the ramifications of anonymity in the online environment.
Companies and marketers might particularly pay attention to the issue of online deception in order to create transparency between them and their communities. People can freely share their thoughts and experiences with products and services with others, and at the same time they should learn and be aware of online deceptive purposes that exist in the context of online product reviews. Combating online fake reviews will contribute to improving performances of both companies and online retailers and thereby create the basis for a safer online marketplace.
Reference list
360pi.com. (2016). How many products does Amazon actually carry? And in what categories? Retrieved October 1st, 2016 from http://360pi.com/press_release/many-products-amazon-actually-carry-categories/
Amazon.com. (2016). Philips 452978 60-Watt Equivalent SlimStyle A19 LED Light Bulb Soft White, Dimmable. Retrieved October 8th 2016 from https://www.amazon.com/Philips-60-
Watt-Equivalent-SlimStyle- Dimmable/dp/B00I134ORI/ref=sr_1_5?ie=UTF8&qid=1476431414&sr=8-5&keywords=philips++led+light
Amazon.com. (2016). Philips HF3520 Wake-up Light With Colored Sunrise Simulation, White. Retrieved October 1st 2016, from https://www.amazon.com/Philips-HF3520-Wake-Up- Colored-Simulation/dp/B0093162RM/ref=sr_1_1_a_it?ie=UTF8&qid=1476431255&sr=8-1&keywords=philips%2Bwake%2Bup%2Blight&th=1
Bambauer-Sachse, S., & Mangold, S. (2013). Do consumers still believe what is said in online product reviews? A persuasion knowledge approach. Journal of Retailing and Consumer Services,20(4), 373-381.
Buller, D.B., & Burgoon, J.K. (1996). Interpersonal deception theory. Communication Theory, 6:3, 203–242.
Cambridge Dictionary. (2016). Retrieved September 10th, 2016 from http://dictionary.cambridge.org/dictionary/english/deception
Caspi, A., & Gorsky, P. (2006). Online Deception: Prevalence, Motivation, and Emotion. CyberPsychology & Behavior,9(1), 54-59.
Chen, Y., & Xie, J. (2008). Online consumer review: Word-of-mouth as a new element of marketing communication mix. Management Science, 54(3), 477-491.
Chevalier, J. A., & Mayzlin, D. (2006). The effect of word-of-mouth on sales: Online book reviews. Journal of Marketing Research,43(3), 345-354.
David, S., & Pinch, T. (2006). Six degrees of reputation: the use and abuse of online review and recommendation systems. First Monday,11(3).
Dellarocas, C., Zhang, X. & Awad, N. F. (2007). Exploring the value of online product reviews in forecasting sales: The case of motion pictures. Journal of Interactive Marketing,21(4), 23-45.
Doh, M.S. & Hwang, J-S. (2009). How consumers Evaluate eWOM (electronic word-of-mouth) messages. CyberPhychology & Behavior, 12:2.
Drouin, M., Miller, D., Wehle, S. M., & Hernandez, E. (2016). Why do people lie online? “Because everyone lies on the internet”. Computers in Human Behavior,64, 134-142.
Emarketer.com. (2016). Worldwide retail ecommerce sales will reach $1915 trillion this year. Retrieved October 5th, 2016 from https://www.emarketer.com/Article/Worldwide-Retail-Ecommerce-Sales-Will-Reach-1915-Trillion-This-Year/1014369
Fortune.com. (2016). Amazon will now deliver a car to your doorstep. Retrieved October 1st, 2016 from http://fortune.com/2016/08/19/amazon-prime-now-hyundai/
Fransen, M. L., Smit, E. G., & Verlegh, P. W. (2015). Strategies and motives for resistance to persuasion: an integrative framework. Frontiers in Psychology,6.
Friestad, M., & Wright, P. (1994). The persuasion knowledge model: How people cope with persuasion attempts. Journal of Consumer Research, 21(1), 1.
George, J. F., Giordano, G., & Tilley, P. A. (2016). Website credibility and deceiver credibility: Expanding prominence-interpretation theory. Computers in Human Behavior,54, 83-93.
Grazioli, S., & Jarvenpaa, S. (2000). Perils of Internet fraud: an empirical investigation of deception and trust with experienced Internet consumers. IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans,30(4), 395-410.
Jacks, J. Z., & Cameron, K. A. (2003). Strategies for resisting persuasion. Basic and Applied Social Psychology,25(2), 145-161.
Ketelaar, P. E., Willemsen, L. M., Sleven, L., & Kerkhof, P. (2015). The good, the bad, and the expert: How consumer expertise affects review valence effects on purchase intentions in online product reviews. Journal of Computer-Mediated Communication, 20(6), 649-666.
Kroenke, D. M. (2016). Experiencing MIS, global edition (7th ed.). Harlow: Pearson Education Limited.
Kokkodis, M. (2012). Learning from positive and unlabelled Amazon
reviews. Proceedings of the 21st international conference companion on World Wide Web - WWW '12 Companion. doi:10.1145/2187980.2188119
Ku, Y., Wei, C., & Hsiao, H. (2012). To whom should I listen? Finding reputable reviewers in opinion-sharing communities. Decision Support Systems, 53(3), 534-542.
Kusumasondjaja, S., Shanka, T., & Marchegiani, C. (2012). Credibility of online reviews and initial trust: The roles of reviewer's identity and review valence. Journal of Vacation
Marketing,18(3), 185-195.
Li, J., Chou, T., & Yu, X. (2013). The effect of trust signals on online retailer's sales: An example of Taobao. 2013 10th International Conference on Service Systems and Service
Management. doi:10.1109/icsssm.2013.6602645
Li, X., & Hitt, L. M. (2010). Price effects in online product reviews: An analytical model and empirical analysis. SSRN Electronic Journal, 34(4), 809-831
Lin, C.L., Lee, S.H., & Horng, D.J. (2011). The effects of online reviews on purchasing intention: The moderating role of need for cognition. Social Behaviour and Personality,
39(1), 71-82.
Micu, A. C., & Pentina, I. (2012). Integrating advertising and news about the brand in the online environment: Are all products the same? Journal of Marketing Communications,20(3), 159-175.
Moe, W. W., & Trusov, M. (2011). The value of social dynamics in online product ratings forums. Journal of Marketing Research,48(3), 444-456.
Muntinga, D. G., Moorman, M., & Smit, E. G. (2011). Introducting COBRAs: Exploring motivations for brand-related social media use. International Journal of Advertising, 30(1), 13.
Munzel, A. (2016). Assisting consumers in detecting fake reviews: The role of identity information disclosure and consensus. Journal of Retailing and Consumer Services,32, 96-108.
O’Keefe, R. M. & McEachern, T. (1998). Web-based customer decisionsupport system. Communications of the ACM, 41(3), 71–78.
Osgood, C. E., & Tannenbaum, P. H. (1955). The principle of congruity in the prediction of attitude change. Psychological Review, 62, 42–55.
Park, D., Lee, J., & Han, I. (2007). The Effect of On-Line Consumer Reviews on
Consumer Purchasing Intention: The Moderating Role of Involvement. International Journal of Electronic Commerce,11(4), 125-148.
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. Advances in Experimental Social Psychology, 19, 123-205.
Rarcherla, P., Mandviwalla, M., & Connolly, D.J. (2012). Factors affecting consumers’ trust in online prduct reviews. Journal of Consumer Behavior, 11, 94-104.
Rathod, S. B., & Pattewar, T. M. (2015). Content based spam detection in email using Bayesian classifier. 2015 International Conference on Communications and Signal Processing (ICCSP). doi: 10.1109/iccsp.2015.7322709
Román, S. (2010). Relational consequences of perceived deception in online shopping: The moderating roles of type of product, consumer’s attitude toward the internet and consumer’s demographics. Journal of Business Ethics,95(3), 373-391.
