The process behind the influence of social
proof on purchase intention in an online world
The impact of simplification on this process
The process behind the influence of social
proof on purchase intention in an online world
The impact of simplification on this process
Name: Anouk van der Zwaag
Department: Marketing
Qualification: Master Thesis
Date: January 8
th, 2020
Address: Westerhaven 6a, 9718AV,
Groningen
Phone number: +31630750086
E-mail address: anoukvdzwaag@hotmail.com
Student number: S2757338
First supervisor: dr. M. Keizer
Abstract
Online consumer reviews play an important role in influencing consumers’ purchase intention. Online firms acknowledge this and have adopted their marketing strategy
accordingly to increase their sales. This study investigates the mediation role of consumers’ product perception to examine the process behind this influence. In addition, the influence of simplification of seller-created product information on product perception was explored. The study produces four findings: (1) the presence of online reviews has a positive effect on consumers’ purchase intention, (2) product perception does not mediate this relationship, (3) purchasing intention increases as consumers’ product perception increases, and (4)
simplification of product information does not affect consumers’ product perception. The results and implications of this study are discussed.
Keywords: Social proof; Online consumer reviews; Consumer-created information;
Table of contents
1. Introduction ... 4
2. Literature review and hypotheses ... 6
2.1. Social proof ... 6
2.2 Consumer’s perception of a product ... 7
2.3. Readability level ... 10
2.4. Effects of gender and Internet shopping frequency on online purchasing intention ... 12
3. Research methodology ... 14
3.1. Participants and design ... 14
3.2. Experimental product ... 14
3.3. Procedure ... 15
3.4. Independent variables ... 16
3.4.1. Social proof ... 16
3.4.2. Consumers’ product perception ... 17
3.4.3. Readability ... 18
3.5. Dependent variable ... 19
3.5.1. Consumers’ intention to purchase online ... 19
3.6. Control variables ... 20
3.6.1. Gender ... 20
3.6.2. Internet shopping frequency ... 20
3.7. Data analysis ... 20 3.7.1. Subjects ... 20 3.7.2. Cronbach’s alphas ... 21 3.7.3. Analysis plan ... 21 4. Results ... 22 4.1. Manipulation checks ... 22 4.2. Descriptive statistics ... 23 4.3. Dependent variables ... 23 4.3.1. Purchase intention ... 23 4.3.2. Product perception ... 25 4.4. Covariates ... 26 5. Discussion ... 27
5.1. Summary of the results ... 27
5.2. Implications ... 27
5.3. Practical implications ... 29
6. Limitations and future research directions ... 30
References ... 33
Appendices ... 38
Appendix 1: No social proof, regular readability ... 38
Appendix 2: No social proof, simple readability ... 43
Appendix 3: Social proof, regular readability ... 48
1. Introduction
The word-of-mouth (WOM) phenomenon is widely accepted in traditional marketing
research. WOM plays an increasingly important role in influencing consumers’ attitudes and purchasing intentions. In general, WOM refers to the distribution of information (e.g.,
recommendations and opinions) through communications between people (Chen, Wang, & Xie, 2011). The rise of the Internet has led to a transformation of traditional WOM into electronic WOM (e-WOM) communications (Lee, Park, & Han, 2008). Nowadays, WOM is prevalent in the online shopping environment as well.
The online market place allows consumers to share and obtain information about products they consume through consumer reviews, recommendations or star ratings. Where once consumers relied on WOM from family and friends, today they search for -and trust- the online opinions of other consumers. This consumer-created information is useful when consumers make purchase decisions, because it provides them with indirect experiences of products (Park, Lee, & Han, 2007). Consumers rely on the shared information of other consumers to get certainty in unclear or ambiguous situations without direct product experience (Guadagno, Muscanell, Rice, & Roberts, 2013). Previous studies have
consistently found that consumers, in these situations, are sensitive for the opinions of others (Chen & Xie, 2008; Dellarocas, 2003). This is known as the social proof principle, which states that people follow the lead of similar others (Cialdini, 2001). In general, social proof techniques are effective in influencing consumers’ purchasing decisions, as well as making consumers’ attitudes towards products more favourable, and are especially effective in situations characterised by uncertainty and ambivalence (Amblee & Bui, 2011; Fennis & Stroebe, 2016; Guadagno et al., 2013). Hence, online sellers feel encouraged to use these ‘online’ social proof techniques to drive their consumers’ sales. Many online sellers, including one of the largest online retailers Amazon, encourage e-WOM by giving consumers the opportunity to share online product evaluations on the products they offer at their own website. These online reviews serve as an independent product-information resource and have successful features, because they have been shown to play a significant role in influencing consumers’ purchase intention (Talib & Saat, 2017; Chen & Xie, 2008; Erkan & Evans, 2018; Park et al., 2007).
opportunity to share their product evaluations and the reliance on this shared information is extremely high. The information available in the reviews could help consumers to develop perceptions about products in unfamiliar and ambiguous situations. In other words, the presence of online reviews could influence consumers’ product perception (Duan, Gu, & Whinston, 2008). Meanwhile, the related literature also suggests that there exists a relationship between product perception and purchase intention. More specifically, a consumer perception of a product is considered to influence the consumer’s purchase intention (Grunert, 2005). This suggests that the relationship between social proof and consumers’ purchase intention can be direct as well as indirect; it can be mediated by consumers’ perception of a product as well. Therefore, in this paper the focus is put on the mediating role of consumer perception of a certain product in the relationship between social proof and consumers’ purchase intention. This is particularly the case for major purchases, these purchases ask for a more conscious purchasing process, a process in which peoples’ involvement is higher than the involvement in everyday purchases.
Another important product-information resource influencing consumers’ product perception, particularly in unfamiliar and ambiguous situations, is seller-created information. This is objective, product orientated information and presented in a standard form (Bickart & Schindler, 2001; Park et al., 2007). There is an increase in awareness among marketers on the importance of the readability of their product-related information in getting consumers engaged in their products (Colmer, 2017). Clearly, readability as a message feature matters and is playing an important role in driving consumer behaviour (Pancer, Chandler, Poole, & Noseworthy, 2019). Therefore, the impact of readability levels in relationship with consumers’ product perception is of importance for marketers. When seller-created information, e.g., a product description, is difficult to read and/or to understand, it can alter consumers’ product perception, because it can make the product itself, its use and its function unclear or
ambiguous. In these cases, simplifying the seller-created information will greatly increase the likelihood that these pieces of information are better understood (Leroy, Endicott, Kauchak, Mouradi, & Just, 2013). This raises however the question how this improved understanding of text affects the effectiveness of the seller-created information. More specifically, how does simplification of this seller-created information affect consumers’ product perception? There is a significant gap in current knowledge in this specific field and this is remarkable given (a) the large amount of online purchase information available to consumers and (b) the large amount of online purchases by consumers.
purchase intention in an e-commerce setting. In addition, it will show how this process is affected by different readability levels of seller-created information. To investigate this, the focus has been put on the following research questions:
(1) Does social proof act as a promotor of consumer purchase intention?
(2) Does consumer product perception mediate the relationship between social proof and consumer purchase intention?
(3) How are these effects changed by different readability levels?
2. Literature review and hypotheses
As outlined above, the aim of this research paper is to investigate the process behind the relationship between social proof and purchase intention. More specifically, the mediating role of consumers’ product perception in this relationship will be explored. A close
examination of this process explores the influence of different readability levels of seller-created information on this process. In this section of the article relevant literature will be introduced and based on this, the study proposes six hypotheses.
2.1. Social proof
Social proof is one of the six fundamental principles of persuasion. The social proof principle states that people follow the lead of similar others (Cialdini, 2001). People heavily rely on the people around them to determine how they should feel, think and act, especially when they view those others as similar to themselves. An action is seen as more appropriate when other people are doing it: when many people are doing something, they must know something we do not know and it should be the right thing to do. Social proof is most powerful when people are unsure about themselves or when a situation is unfamiliar, ambiguous or unclear. In these situations, social proof provides an easy and comfortable shortcut for determining how they should behave (Cialdini, 1993). Since marketers have recognised the influence of this principle in advertising, social proof is frequently used in advertising messages (Fennis & Stroebe, 2016).
During the last decade, an increasing number of online firms are taking advantage of this and have adopted their marketing strategy accordingly. They provide, next to seller-created information, consumer-created information about products on their website. They offer users of purchased products an option to post their personal product evaluations on the sellers’ websites (Chen & Xie, 2008). This consumer-created information describes -in a subjective or objective way- product features in terms of usage situations and measures the product performance in terms of strengths and weaknesses from a user’s perspective (Bickart & Schindler, 2001; Park et al., 2007). This consumer-created information consists mostly of customer-ratings (often between 1 and 5), recommendations and (the number of) consumer reviews (Amblee & Bui, 2011; Dellarocas, 2003).
A study by BrightLocal reported that on average 86% of online buyers indicated that they read consumer reviews of products in which they were interested. Notably, this included 95% of people between 18 and 34 years old and 91% of them trusted these reviews as much as personal recommendations (Murphy, 2018). Moreover, consumers do not only read and trust those reviews and recommendations, they also rely on this consumer-created information and it is influencing their purchasing decisions (Chen et al., 2011; Chen & Xie, 2008; Zhu & Zhang, 2010). According to Park et al. (2007) an increase in the number of listed ‘positive’ product reviews lead to an increase in consumers’ purchasing intention. This effect can be explained by the fact that the number of online consumer reviews of a product is often related to the product’s popularity, because it is reasonable to believe that the number of consumer reviews is representing the number of customers who have purchased the product (Chatterjee, 2001; Chen & Xie, 2008).
Thus, other things being equal, positive online customer reviews have a positive effect on consumers’ intention to purchase online. Therefore, the following hypothesis is proposed: H1. Positive online consumer reviews, as a consequence of the social proof principle, will positively affect consumers’ purchase intention.
2.2 Consumer’s perception of a product
In this way, consumer product perception is the process by which a consumer selects, organises and interprets stimuli or information to establish an image about a firms’
offerings/products. A consumer’s perception about a product can be positive or negative and is based on their expectations and experiences (Emilien, Weitkunat, & Lüdicke, 2017). Consumers can base their product perception on several factors as they are perceiving different aspects of products (Mityko & Teiu, 2012). According to Zeithaml (1988) consumers perceive a product based on perceptions about its price, quality and value. In the following paragraph, relevant literature is used to describe and define each concept.
Perceived product quality can be defined as the consumer’s judgment of the product’s overall superiority or excellence (Zeithaml, 1988). A product’s perceived price can be defined as the consumer’s subjective perception of the product’s objective price (Jacoby & Olson, 1976). The trade-off between a product’s perceived quality and its perceived price is defined as the product’s perceived value. More specifically, perceived value is the ratio between the
product’s perceived benefits relative to its perceived risks (Dodds, Monroe, & Grewal, 1991). Along this line, Emilien et al. (2017) expanded the product perception view by adding
additional aspects in perception, like perception of product risks and benefits. Together the perception of these different elements of a product resulted in an overall product perception, which is used for product evaluation, interpretation and any purchase and usage decisions. In the context of e-commerce, consumers can still make perceptions about a product’s price, quality and value, however, they cannot physically touch or try the product. This makes online shopping more ambiguous than offline shopping, especially when consumers are unfamiliar with a product, for example, when they buy a product for the first time. They have less available information online to base their product perception on. As a result, consumers request more information about products when shopping online compared to when they are exposed to the same product in an offline shopping environment (Schlosser, 2003).
consumer reviews could influence consumers’ perception about a product. Based on the discussion above, the following hypothesis can be proposed:
H2a. Positive online consumer reviews, as a consequence of the social proof principle, will positively affect consumers’ product perception.
Meanwhile, the causal relationship between consumer product perception and consumer purchase intention has also been recognised by researchers (Grunert, 2005). Prior studies have consistently found that the perceptions of product quality, price and value influence consumer purchase intentions.
Tsiotsou (2006) studied the effects of perceived product quality, values, involvement and overall satisfaction on purchase intentions and found that perceived quality had a positive direct effect on purchase intentions. That is, a higher perceived product quality resulted in a higher purchase intention for consumers. Bao, Bao and Sheng (2011) conducted an
empirical study to examine the effects of intangible extrinsic cues on consumer quality perception and purchase intention of private brands, and reported a similar finding; quality perception of a product has a direct effect on purchase intention. Chang and Wildt (1994) examined the relationship between price, nonprice product information and purchase intention, together with the intervening variables of perceived price, perceived quality, and perceived value. They demonstrated that a product’s perceived value is negatively influenced by its perceived price and positively influenced by its perceived quality. In other words, it is a trade-off between perceived price and perceived quality. In addition, they showed that purchase intention is positively influenced by perceived value. Moreover, perceived quality and perceived price have, besides their indirect effect on purchase intention through
perceived value, a direct positive effect on purchase intention. Cheng-Ping (2017) observed the reactions of 380 individuals to examine the effects of perceived quality, perceived price, perceived value and brand image on purchase intentions towards tourism, sightseeing and sports products. This investigator found that, although brand image had no significant effect on purchase intention, perceived- price, quality and value all had a positive influence on purchase intention.
H2b. Consumers’ product perception is positively related to consumers’ intention to purchase online.
Based on the above line of reasoning, consumer product perception is expected to act as a mediator between social proof and consumers’ intention to purchase online. Therefore, the following hypothesis can be formulated:
H2. The effect of social proof on consumers’ purchase intention is mediated by the perception of the product.
2.3. Readability level
Product perception can also be altered by the readability of seller-created product
information. When this information is difficult to read and/or to understand it can make the product unclear or ambiguous. According to Pancer et al. (2019) an important feature of a message is its readability. Klare (1963, p. 1) broadly defines readability as “the ease of understanding or comprehension due to the style of writing”. A more specific definition of readability is formulated by Dale and Chall (1949, p. 5): “The sum of total (including all the interactions) of all those elements within a given piece of printed material that affect the success a group of readers have with it. The success is the extent to which they understand it, read it at an optimal speed, and find it interesting”.
The readability of a message can be measured among two variables: syntactic complexity and word complexity (DuBay, 2004; Pancer et al., 2019). The average length of a sentence is often referred to as syntactic complexity (Dale & Chall, 1948), whereas word complexity is defined as the number of characters in one word or the number of syllables per word (Flesch, 1948; Mc Laughlin, 1969; Stoel-Gammon, 2010). However, there are no strict rules about the maximum word- or sentence length or the maximum number of syllables, because
sometimes a short word is more difficult than a longer one. To cite some examples, “extol” and “nadir” are rated as easier by most readability metrics than “communication” and
These readability guidelines are based on the fluency effect, which is referred to as the ease with which people process information (Lee & Labroo, 2004). According to Alter and
Oppenheimer (2009) the fluency effect influences peoples’ judgments: especially when information is easily processed people tend to enjoy it more. In an empirical study, these investigators studied different manipulations of processing fluency (e.g., linguistic fluency manipulation and visual ease manipulation) to examine this effect. They found that every fluency manipulation exerts the same influence on peoples’ judgments, independently of how the fluency is generated. In addition, Lerory et al. (2013) examined the effect of simplification on both perceived and actual text difficulty by using an online study. In their study, lexical simplification -replacing difficult terms with appropriate, easier alternatives of equivalent meaning- was used to made the text simpler (Leroy et al., 2013; Paetzold & Specia, 2017). They found that, after text simplification, the text was actually perceived as simpler. In
addition, the simplification led to better understanding of the text, which was measured by the difference in the number of correct answers about the content of the text between the original and simplified version.
Therefore, it is expected that a consumer who is viewing simplified seller-created information will most likely have a more positive perception of the product, because the information is easier processed and better understood. Therefore, the following hypothesis is predicted: H3. Simplified seller-created information will have a positive effect on consumers’ product perception.
Based on the literature described above, it is proposed that both social proof and simplification of the product information provided by sellers are expected to increase a
consumer’s product perception. For this reason, the relationship between these two concepts is of particular interest. Since the presence of social proof will deduct unfamiliarity or
ambiguousness in certain situations, social proof will enhance a consumer’s product
perception. Similarly, simplified seller-created information could increase consumers' product perception, because easily processed information is enjoyed more. Moreover, simplified information also decreases ambiguousness feelings regarding the information, because the information is already understood and perceived better. In comparison with regular product information, it is more likely that simplified information diminishes ambiguousness when reading this information. This makes it likely that people reading simplified product
Based on the above, it is proposed that simplification of seller-created information will moderate the impact of social proof. More specifically, simplification will diminish the
effectiveness of social proof on consumers’ product perception. In other words, it is expected that people will be less sensitive to social proof when confronted with simplified
seller-created information. Therefore, the last hypothesis is:
H4. The positive effect of social proof on product perception is less strong for simplified seller-created information than for regular seller-created information.
2.4. Effects of gender and Internet shopping frequency on online purchasing intention
At the early start of the Internet, men where almost the only users. However, during the last decade the Internet has evolved to a common, everyday tool and the gender gap of Internet users has narrowed: the number of female users is now equal to the number of male users (Hernández, Jiménez, & Martín, 2011; Weiser, 2000). Moreover, gender differences in consumers’ willingness to purchase online do no longer exist (Bae & Lee, 2011).
Although the gender gap for Internet users and buyers faded out, there are still significant differences between males and females regarding their use of and response to online consumer reviews in making online purchase decisions (Bae & Lee, 2011). Females’ purchase intention is influenced more by online consumer reviews than males’ purchase intention (Garbarino & Strahilevitz, 2004; Yoon & Occeña, 2015). This can be explained by the fact that females are more affected by perceived risk in e-commerce (Lin, Featherman, Brooks, & Hajli, 2018). In line with this, Kim, Mattila and Baloglu (2011) found that women read reviews for the purpose of risk reduction in addition to convenience and quality reasons. As a result, they are more likely to rely on other consumers’ recommendations, because they think that other consumers have more knowledge and useful information than themselves. Therefore, gender should be considered as a significant factor that can influence consumers’ intention to shop online and it is useful to empirically test the role of this factor in
e-commerce.
Another important variable which could have a direct effect on consumers’ intention to purchase online, is Internet shopping frequency. Consumer experience in the online
Piew (2010) previous online purchase experience has an effect on consumers’ online purchase intention. Similarly, Hong and Cha (2013) found that Internet shopping frequency significantly increases consumers’ online purchase intention. Therefore, Internet shopping frequency should also be considered as a significant factor that can influence consumers’ online purchase intention and it is useful to empirically test the role of this factor in e-commerce as well.
In summary, it is hypothesised that social proof will increase consumers’ intention to purchase online. Furthermore, consumers’ product perception is expected to mediate the relationship between social proof and consumers’ intention to purchase online. However, the intention to purchase online will not be the same for everybody as it depends on gender and Internet shopping frequency. Finally, simplification of seller-created product information will have a direct effect on consumers’ product perception and plays a moderated role in the relationship between social proof and consumers’ product perception. In figure 1 the end result, a first-stage moderated mediation model, is shown.
3. Research methodology
An experimental method is used for this study. This method was designed to examine how consumers react to readability levels in combination with social proof and how this
combination influences their online purchase intention. Social proof and readability were both manipulated into two levels: social proof and no social proof; regular seller-created
information and simplified seller-created information. Through a pre-test the readability levels and the total number of reviews was determined. In the main experiment, potential
confounds were controlled.
3.1. Participants and design
A 2 (social proof: yes and no) x 2 (readability: regular and simple) between-subjects factorial design was used to test the hypotheses. The dependent variable was the intention of
consumers to purchase online. The participants were 166 randomly selected adults (47.6% female, 52.4% male; Mage = 27.2 years, SD = 1.28) who participated voluntarily in the experimental field study. They were randomly assigned to one of the four conditions and no rewards were given for taking part. Most participants already had online shopping experience (99.4%), with a mean of at least one online purchase every three months. Moreover, most participants (92.8%) gave their English proficiency a rating higher than the basic level A1-A2 (see table 1 for frequencies).
Table 1.
English proficiency levels.
Levels Frequency Percent
A1 – A2 (basic level) B1 – B2 (independent level) C1 – C2 (advanced level) 12 59 95 7.2% 35.6% 57.2% 3.2. Experimental product
3.3. Procedure
An online survey questionnaire was created which was spread via LinkedIn. Participants were asked to fill in a survey, which was said to be part of a graduation research examining consumer differences in online shopping behaviour. The latter was misleading information aimed at covering the studies actual purpose. Data was collected during 2.5 weeks from the 20th of November till the 9th of December.
At the start of the survey questionnaire, the participants were informed that they should carefully read the instructions provided in the survey and complete the survey independently. Furthermore, the instructions informed the participants that they could stop at any point during the survey and that their participation would remain confidential. The participants were first asked to fill in a short questionnaire containing questions on their gender and age, Internet shopping frequency and English proficiency. The latter was included to measure how participants rate their English reading level.
After filling in these questions, the participants received instructions in the second part of the survey. They were informed that they should imagine needing to buy a new electronic device. More specifically, the participants received the following message, “Imagine that you
need or want to buy a new digital camera in order to take beautiful images on your next vacation. There are multiple different cameras which you can choose, each with different features, qualities and prices. One of them is the following product”. Each participant entered
a virtual shopping website. The website contained a picture of the given item and seller-created information (either regular or simple), which included a brief product description (e.g., its price) and its main functions (Exhibit 1B, Appendix 1). The brand name of the product was hidden to avoid any effects of brand and brand category. Depending on the condition in which the participant was, the seller-created information is accompanied by consumer-created information. This information included an average star rating (from 1 to 5 stars) for the product and a number of available reviews (Exhibit 3B, Appendix 3).
Moreover, participants were randomly assigned based on the exposure to consumer-created information or not. Participants in the social proof condition were shown consumer reviews and a product star rating next to the seller-created information. Whereas participants in the control group -in which there was no consumer-created information presented- only saw the seller-created information.
After looking at the product and the presented information, participants were asked to fill in a questionnaire on their perception of this product. This questionnaire contained questions about the product’s perceived price, quality and value (Exhibit 1C-1F, Appendix 1), followed by a short questionnaire on their purchase intentions of this product (Exhibit 1G, Appendix 1). At the end of the survey, two manipulation checks were used. Participants were asked to fill in four questions on their perception of the readability of the seller-created information, followed by one or two questions, depending on their condition, on their perception of the consumer-created information (Exhibit 1H, 1I & 1J, Appendix 1).
Finally, participants were thanked for their participation and they could leave their email-address for the real purpose and results of the study (Exhibit 1K, Appendix 1).
3.4. Independent variables 3.4.1. Social proof
Participants in the social proof condition were exposed to social proof by exposing them to consumer-created information next to seller-created information (i.e., a picture of the given item and a brief product description). Participants in the ‘no social proof’ control condition were only presented the seller-created information.
To vary the extent of social proof both online consumer reviews and product star ratings were used. A product star rating is an overall evaluation of a product’s quality (Kim, Maslowska, & Malthouse, 2018). A combination of these two is used, because the
quantitative aspects of reviews often consist of (1) an average star rating and (2) a number of reviews (i.e., a written explanation for this rating) (Kim et al., 2018; Schlosser, 2011). An average star rating of four out of five stars was used for the product for three reasons: (1) the star rating needs to correspond adequately with the framed positive consumer reviews, therefore a star rating higher than three (the middle star) out of five stars would be
under the name of the product and on the right-side of the product’s picture, as shown in Exhibit 3B, Appendix 3.
Real reviews from online shopping websites were used to create the reviews used in the survey. According to Chevalier and Mayzlin (2006), the length of reviews can affect the information quantity and quality. Therefore, the length of the reviews was controlled. While shopping on the Internet, consumers generally read about four to eight reviews of three or four lines each (Lee et al., 2008; Park et al., 2007). In line with this, four consumer reviews each of three lines were used for this study. Importantly, participants were able to see how many reviews were available for the product, yet, they could only read four.
The reviews were written in Arial, had a font size of 11-point type, included a title, the publication date and the reviewer’s name (mostly a nickname) and were located under the seller-created information, as shown in Exhibit 3B, Appendix 3. The following figure shows an example of a review, which were shown to the participants.
Figure 2. Example of a review.
Before the actual experiment, six subjects (who did not participate in the actual experiment) were asked to evaluate the usefulness of the reviews. Eight different reviews were shown to the subjects. They were asked if they would be more likely to buy a product if these reviews were presented. Moreover, they were asked to rate the reviews based on their quality. Based on the results, the four reviews which were perceived as most useful were, after some small revisions, used for the actual experiment.
3.4.2. Consumers’ product perception
Perceived product quality. A semantic differential scale adopted from Grewal, Krishnan,
Bakker and Borin (1998) was used to measure the perceived quality of the product. This scale consisted of four questions (Exhibit 1E, Appendix 1): e.g., “To what extent do you rate
the expected quality of this product?”. Participants had to complete the statement on a
five-point rating scale – ranging from for example “low quality” to “high quality”.
Perceived price of the product. Perceived price is used to indicate individual differences in
price perception of the product (high vs. low). Differences in price perception might lead to variation in purchase intention (Cheng-Ping, 2017). Consumers’ price perception of the product was measured on a 5-point Likert scale adapted from a scale published by Dodds et al. (1991) and consisted of three items (Exhibit 1D, Appendix 1): e.g., “The product is a great
deal” and “The price is less than what I expect it to be”. The scale has five response
categories, ranging from “strongly agree”, indicated by a 5, to “strongly disagree”, indicated by a 1: it requires the participants to point out a degree of agreement of disagreement with each statement.
Perceived product value. The perceived value of the product is measured by a 5-point Likert
scale consisting of four statements: e.g., “I think this product will satisfy my needs and
wants”. The scale is developed based on prior research from Zeithaml (1988) and Cronin,
Brady and Hult (2000). Participants had to indicate how well each statement characterised them on a 5-point rating scale – ranging from “strongly agree” indicated by a 5, to “strongly disagree”, indicated by a 1 (Exhibit 1F, Appendix1).
3.4.3. Readability
The study employed a replication of a product information page of a digital camera of a real Internet shopping website. The page showed a picture of the digital camera and seller-created information, which included a brief product description and its main functions. The latter was manipulated to vary the readability level of the seller-created information.
A readable text should have a Flesch-Kincaid Grade Level and a Gunning Fox Index lower than 10 and the Flesch Reading Ease should be higher than 60 points.
Participants in the regular readability condition were confronted with the original seller-created information (Exhibit 1B, Appendix 1). This seller-seller-created information was rated as a D; had a reach of 61%; a Flesch-Kincaid Grade Level of 13.2; a Gunning Fog Index of 15.2; and a Flesch Reading Ease of 43.5. Based on the aforementioned scales, the original seller-created information is relatively hard to read.
In the simple readability condition, participants saw the adjusted seller-created information. The original seller-created information was made simpler by making the following
adjustments. To make the text simpler: difficult words were replaced by common words; long sentences were made shorter; the font type was changed into Arial; and more white spaces and bold headers were used. Before the actual experiment, a pre-test (six subjects) was conducted to evaluate whether the simplified version was perceived as intended. The subjects of the pre-test did not participate in the actual experiment, but were asked to evaluate the seller-created information according its readability. The simplified version was shown to the subjects. They were asked to rate the readability of the text on a 5-point rating scale ranging from “very easy to read”, indicated by a 5, to “very hard to read”, indicated by a 1. All the subjects indicated either a 4 or 3. In addition to this, they were asked to give
feedback in order to make the simplified text simpler. Based on the results, the simplified version of the seller-created information was revised: more white spaces were added and some words were replaced by synonyms.
This revised seller-created information was used for the actual experiment (Exhibit 2B, Appendix 2). The simplified text was rated as an A instead of a D; the reach was 100%; the Flesch-Kincaid Grade Level was 6.9; the Gunning Fog index was 9.1; and the Flesch Reading Ease was 67.3. Therefore, the simplified text was relatively easy to read.
3.5. Dependent variable
3.5.1. Consumers’ intention to purchase online
Consumers’ online purchase intention was measured on a scale adapted from a scale published by Erkan and Evans (2016). The scale consisted of four statements (Exhibit 1G, Appendix 1): e.g., “It is very likely that I will buy the product” and “I will definitely try the
product”. A 7-point Likert scale – ranging from “strongly agree”, indicated by a 7, to “strongly
participant. Participants who rated the statements as “strongly agree” or “agree” are
associated with a high purchase intention, whereas participants who rated the statements as “strongly disagree” or “disagree” would have a relatively low intention to purchase.
3.6. Control variables
Gender and Internet shopping frequency were measured in order to determine if these factors are differences between participants in these two variables, as this could possibly influence peoples’ intention to purchase online (Exhibit 1A, Appendix 1).
3.6.1. Gender
Differences in gender might lead to variation in peoples’ intention to purchase online (Bae & Lee, 2011; Brown, Pope, & Voges, 2003). Moreover, females’ purchase intention is
influenced more by online consumer reviews than males’ purchase intention (Garbarino & Strahilevitz, 2004; Yoon & Occeña, 2015). The short question “What is your gender” was used to examine these differences. The question has three response categories: “male”, “female” and “other”.
3.6.2. Internet shopping frequency
Internet shopping frequency was measured in order to determine whether individual differences in this variable existed and if this possibly affects the consumer’s intention to purchase online (Exhibit 1A, Appendix 1). According to Hong and Cha (2013), a positive relationship between Internet shopping frequency and consumers’ online purchase intention exists. Internet shopping frequency was measured based on a single question: “How many
times during the last year did you shop online?”. The question has six response categories,
ranging from “never” to “at least once a week” (Hong & Cha, 2013).
3.7. Data analysis 3.7.1. Subjects
3.7.2. Cronbach’s alphas
The construct purchase intention was measured with four questions (Exhibit 1G, Appendix 1). A correlation analysis was conducted to measure if these questions form a reliable scale. The correlations were significant at the 0.01 level. The questions were combined as one, average, factor (Cronbach’s alpha: 0.912), which was used in subsequent analysis (M = 3.74, SD = 1.52).
Product perception was measured based on three constructs (price, quality and value) and these constructs were each measured with multiple statements (Exhibit 1D, 1E & 1F,
Appendix 1). Price perception consisted of three questions. A correlation test and Cronbach’s alpha were used to assess the reliability of this construct. The multiple questions were all related at a significance level of 0.01 and the construct had an acceptable level of internal consistency, as determined by the associated Cronbach’s alpha of 0.728. This procedure was repeated for the quality and value constructs. Quality perception consisted of four questions. The correlations were significant at the 0.01 level and the construct had an acceptable level of internal consistency, as determined by the associated Cronbach’s alpha of 0.882. Value perception consisted also of four questions. The multiple questions were related at either a significance level of 0.05 or 0.01, after revision of one of the statements. The Cronbach’s alpha (0.705) had an acceptable level of internal consistency. An average value for each construct was made (𝑀"#$%& = 3.20, 𝑆𝐷"#$%& = 0.68; 𝑀)*+,$-. = 3.34, 𝑆𝐷)*+,$-. = 0.71; 𝑀/+,*& = 3.30, 𝑆𝐷/+,*& = 0.61).
A Cronbach’s alpha test (0.712) was used to assess the reliability of product perception based on these three constructs. The average value of these three product perception items was used in subsequent analysis (M = 3.28, SD = 0.53).
3.7.3. Analysis plan
To test the hypotheses of the moderated mediation model, the Process Macro was
Both a linear regressions analysis and a Sobel test were performed to compare the outcomes with the Process Macro, to see if the outcomes were in line when a different method was used.
4. Results
4.1. Manipulation checks
A readability manipulation check was employed to measure the level of perceived text difficulty during the experiment, which consisted of four statements (Exhibit 1I, Appendix 1). A correlation test was employed to assess if these statements correlated with each other. All the statements correlated with each other, after reversion of two of those statements. Thereafter, a Cronbach’s alpha test was used to assess the reliability of this readability manipulation check. The manipulation check had an acceptable level of internal consistency, as determined by the associated Cronbach’s alpha of 0.875. Hence, the four statements were all measuring the same construct, i.e., the level of perceived text difficulty.
A one-way between-subjects ANOVA was performed to compare the effect of the different readability levels on the readability manipulation check in the ‘regular’ and ‘simplified’ conditions. There was a significant effect of the readability level on the readability
manipulation check between individuals assigned to ‘regular’ and ‘simplified’ conditions: F (1, 164) = 10.07, p < .01. In line with expectations, participants in the ‘simplified’ condition reported a significant lower level of perceived text difficulty during the experiments (M = 3.45,
SD = 1.32), in comparison with participants in the ‘regular’ condition (M = 4.13, SD = 1.44).
This suggests that the readability level was manipulated successfully.
The subject’s responses on the manipulation check relevant to the presence of reviews was also examined. In order to analyse whether or not participants in the social proof and ‘no social proof’ condition differ with respect to their perception of consumer reviews, a cross table with Chi-square was performed. The Chi-square test was significant, Chi-square (1) = 102.07, p < .001. Participants in the social proof condition indicated a higher perception of reviews (86.9%) than those in the ‘no social proof’ condition (8.5%). So, most participants correctly indicated the presence/absence of the reviews.
changes when these participants were excluded. Therefore, the participants with incorrect answers were not excluded in the subsequent analyses.
4.2. Descriptive statistics
Table 2 shows the descriptive statistics for the variables relevant in this study. Table 2.
Descriptive statistics.
N Mean SD Minimum Maximum
Predictor variables
Purchase intention 166 3.74 1.52 1 6.5
Product perception 166 3.28 0.53 1.33 4.33
Control variable
Internet shopping frequency 166 4.51 0.95 1 6
4.3. Hypothesis testing
4.3.1. Direct effects on purchase intention
According to the Process Macro, the overall model for consumers’ purchase intention was significant (𝑅1 = 0.46, F (2, 163) = 70.82, p < .001). The result showed a direct effect of social proof on purchase intention (B = 0.48, t (163) = 2.76, p < .01).
A linear regression analysis was also conducted to measure the influence of social proof on consumer purchase intention. The regression analysis was significant (𝑅1 = 0.06, F (1, 164) = 10.44, p < .001). The result showed a main effect of social proof (B = 0.74, t = 3.23, p < .001).
Both results showed that the subjects who viewed the product description accompanied by consumer reviews, indicated that they had a higher intention to purchase the product than those who viewed the product description without consumer reviews (see table 3 for means and standard deviations). This table indicates that the presence of social proof, the positive online consumer reviews, represents an increase in consumer purchase intention. This suggests that hypothesis 1 should be accepted.
Table 3.
Means and standard deviations of consumer purchase intention as a function of social proof.
Means (standard deviations) 4.10 (1.53),
n = 82
3.36 (1.42),
n = 84
Moreover, the significant Process Macro model for purchase intention showed a direct effect of product perception on purchase intention (B = 1.84, t (163) = 11.11, p < .001). Again, a linear regression was also conducted to measure the influence of consumer product perception on consumer purchase intention. The regression analysis was significant (𝑅1 = 0.44, F (1, 164) = 128.87, p < .001). The result showed a main effect of product perception (B = 1.90, t = 11.35, p < .001). This result was the same as the outcome from the Process Macro, suggesting that the influence of product perception on purchase intention is so strong that variation in social proof does not change the effect on purchase intention.
The result suggested that subjects with a higher product perception had a higher intention to purchase than those with a lower product perception (see table 4 for means and standard deviations). This table indicates that higher product perception increases consumer purchase intention. So, it can be concluded that hypothesis 2b should also be accepted.
Table 4.
Means and standard deviations of consumer purchase intention as a function of product perception.
Product perception: High+ Low9
Means (standard deviations) 4.64 (1.20) 2.79 (1.21)
a Higher than the median of 3.3333 (median split); b Lower than the median of 3.3333 (median split).
4.3.2. Mediating effect of product perception
To explore the mediation role of product perception in the relationship between social proof and purchase intention, the direct effect of social proof on product perception was taken into account. According to the Process Macro, the overall model for consumers’ product
perception was not significant (𝑅1 = 0.03, F (3, 162) = 1.85, p = .14). The result showed no direct effect of social proof on product perception (B = 0.31, t (162) = 1.21, p = .23).
A different way to describe the mediation analysis is by use of the Sobel test. In line with the previous findings, no significant evidence for (partial or full) mediation was found (z = 1.69, p = .09). Consequently, hypothesis 2 and 2a are not supported: positive online consumer reviews did not positively affect consumers’ product perception and this perception does not mediate the effect of the reviews on consumers’ purchase intention.
4.3.3. Direct effects on product perception
According to the Process Macro, the overall model for consumers’ product perception failed to reach significance (𝑅1 = 0.03, F (3, 162) = 1.85, p = .14). As indicated above, the result showed no direct effect of social proof on product perception (B = 0.31, t (162) = 1.21, p = .23). Moreover, the effect of simplification also failed to reach significance (B = 0.29, t (162) = 1.13, p = .26). This was of particular interest, because it suggested that the presence of consumer reviews and a simplified product description are both unable to boost consumers’ product perception. Moreover, the analysis failed to produce any significant interaction results (i.e., interaction between social proof and readability level): B = -0.12, t (162) = -0.71,
p = .48.
This was supported by the finding that the conditional indirect effect of X on Y was not significant at different values of the moderator, as zero was present in the confidence intervals (regular readability: B = 0.36, 95% confidence interval: -0.12 – 0.81; simple
readability: B = 0.15, 95% confidence interval: -0.21 – 0.52). Thus, hypotheses 3 and 4 were rejected.
Table 5 presents the means and standard deviations of product perception for each condition. This table indicates that both the presence of social proof and a simplified
readability level increase product perception, however, these effects are too small to produce any significant effect.
Table 5.
Means and standard deviations of consumer product perception as a function of social proof and readability level.
Social proof: Social proof No social proof
All the above-mentioned analyses were conducted again for only the subjects who rated their English proficiency as either B1-B2 or C1-C2, this in order to determine if the subject’s English proficiency had any influence on the analyses. The subjects who rated their English proficiency as A1-A2 were excluded because, possibly, these subjects were unable to understand the seller- or consumer-created information. In this way, their answers are meaningless. However, similar to the previous findings, the results revealed a direct effect of social proof on purchase intention (B = 0.49, t (151) = 2.67, p < .01) and a direct effect of product perception on purchase intention (B = 1.83, t (151) = 10.77, p < .001). Moreover, no main effect of social proof (B = 0.32, t (150) = 1.15, p = .25) on product perception, no main effect of readability level (B = 0.29, t (150) = 1.03, p = .30) on product perception, and no interaction effect between social proof and readability level (B = -0.11, t (150) = -0.63, p = .53) were found.
4.4. Covariates
The Process Macro was conducted to determine if the control variable gender had any influence on our analysis. Similar to the previous findings, the results revealed that the impact of social proof on purchase intention was significant (B = 0.49, t (162) = 2.77, p < .01) after controlling for gender. The effect of product perception on purchase intention was significant: B = 1.83, t (162) = 11.06, p < .001, whereas the direct effect of social proof on product perception failed to reach significance (B = 0.31, t (161) = 1.20, p = .23). This suggests that the covariate did not differentially affect our main effects. In addition, this can be supported by the fact that gender failed to reach significance in both outcome models: B = -0.07, t (162) = -0.41, p = .68 and B = -0.03, t (161) = -0.31, p = 0.75, respectively the
purchase intention model and the product perception model.
This analysis was repeated for the control variable Internet shopping frequency that also failed to reach significance in both outcome models (B = 0.00, t (162) = -0.04, p = .97) and (B = 0.00, t (161) = -0.11, p = .91), respectively the purchase intention model and the product perception model. Again, the results were in line with the previous findings by showing a direct effect of social proof on purchase intention (B = 0.48, t (162) = 2.75, p < .01). The direct of social proof on product perception failed to reach significance (B = 0.31, t (161) = 1.19, p = .24), however, there is a significant effect of product perception on purchase
5. Discussion
As outlined previously, the aim of this study was to examine the process behind the influence of social proof on consumers’ purchase intention in an e-commerce setting. It was assumed that consumer product perception would mediate the positive effect of social proof on purchase intention. More specifically, it was expected that simplification of seller-created product information would positively influence product perception and negatively affect the influence of social proof on consumer product perception. Several hypotheses were proposed and an experiment was conducted to test these hypotheses. Six major findings appear from this study.
5.1. Summary of the results
In line with the first hypothesis, individuals exposed to positive consumer reviews, in addition to the product information, showed a significantly higher intention to purchase compared to individuals allocated to the ‘no social proof’ condition. In line with this, a higher product perception lead to a significantly higher purchase intention, compared to purchase intention when a lower product perception was experienced. However, there was no convincing evidence that product perception could mediate the influence of social proof on consumers’ purchase intention. The analysis demonstrated that the presence of online consumer reviews was not able to significantly influence product perception. This suggests that product
perception does not explain the process behind the influence of social proof on purchase intention. Against expectations, there was also no convincing evidence that simplification of seller-created information was qualified to significantly influence product perception. That is, individuals exposed to the simplified product information had no significantly higher product perception compared to individuals allocated to the ‘regular product information’ condition. Moreover, this study reveals that this simplification was also not able to significantly change the influence of social proof on product perception.
5.2. Implications
In line with previous research showing the effectiveness of online consumer reviews on purchase intention (Chen et al., 2011; Park et al., 2007; Zhu & Zhang, 2010), this study demonstrated that the presence of these reviews -as a consequence of the social proof principle- is effective in raising purchase intention. So, showing consumer-created
Cialdini (2001), people follow the lead of similar others in determining how to behave. As this result is conducted in an e-commerce setting, the finding corroborates with several lines of research suggesting that the social proof principle is not limited to the offline world (Fennis & Stroebe, 2016; Guadagno et al., 2013). So, the social proof principle is also an effective influence principle in an online context. Together this suggests that the manipulation, as performed in this study, was successful in affecting peoples’ decisions.
Concerning the influence of consumer product perception on consumers’ purchase intention, the finding is in line with prior research (Grunert, 2005). This study reveals that perceived product perception had a positive effect on purchase intention. Earlier studies found that this positive influence is due to a product’s perceived quality, perceived price and perceived value, which together shape a consumer’s product perception (Bao et al., 2011; Chang & Wildt, 1994; Cheng-Ping, 2017; Tsiotsou, 2006).
Although a lot of research is done on what makes a text readable and how text simplification affects text understanding, there is still little empirical evidence on the exact effects of text simplification on consumer behaviour. An important consequence of text simplification is the decrease in perceived text difficulty, resulting in an increase in text understanding (Leroy et al., 2013). Moreover, simplification could influence peoples’ judgements, because easily processed information is enjoyed more (Alter & Oppenheimer, 2009). However, in the present study it was not found that text simplification increases a consumer’s product perception. A possible explanation for this is that respondents are used to these kinds of online seller-created information, i.e., product descriptions. Another explanation could be that people, when buying new electronic equipment, are scanning the text on features they want to have in their devices, without reading the whole text. A lot of eye-tracking research has been done on reading on the Web. According to the studies, many people scan web pages, looking for individual words that capture their attention, instead of reading word by word (Cooke, 2005; Nielsen, 1997). In both cases, the difference between the original and simplified seller-created product description remains unnoticed.
5.3. Practical implications
The current findings have some important practical contributions. The major result
emphasizes the importance for online sellers to show consumer reviews on their web shops. Since online reviews are successful in influencing consumers’ purchase intentions, sellers can use them strategically as an independent product-information resource. The online reviews can function as informants: providing user-oriented product information, e.g., product performance (Bickart & Schindler, 2001; Park et al., 2007). They could also function as a recommender: showing the total number of reviews and a star rating can relate to the product’s popularity (Chatterjee, 2001; Chen & Xie, 2008). Thus, online sellers should give consumers the opportunity to share their product evaluations. Nowadays, almost every online seller gave consumers this possibility. If an online seller wants to encourage consumers to share their feedback they can, for example, offer incentives (e.g., 10%
discount on their next purchase) or send a general e-mail to consumers to asks them for their feedback.
new product. Thus, online marketers could enhance their consumers’ purchase intentions by increasing their products perceptions as well as presenting consumer reviews.
The present study suggests that both the presence of online consumer reviews and
simplification of product information are ineffective approaches to raise product perception. Marketers should find and adopt more effective alternative tactics to increase the perception of their products. For example, firms can use a celebrity as an endorsement for their product. This is already a frequently used approach in marketing and could possibly affect consumers’ product perception. In addition, firms can show prices of competitors that sell the same or a related product on their websites, which could influence the perception of their own product. Taken all data together, the present research found no convincing evidence for both the mediating effect of product perception and the positive effect of product information simplification. However, based on the findings, both the presence of online consumer reviews and product perception will increase purchase intention and are recommended for marketers in practice.
6. Limitations and future research directions
Regarding the present study, a few limitations could be distinguished which may provide opportunities for future research. First, due to the short timeline, the results are based on a smaller than desirable group of respondents. Also, it should be noted that many participants in this study were non-native English speakers, which makes the research results less realistic compared to a study using native speakers. Although more than half of the participants rated their English proficiency as C1-C2, it is likely that some of these
participants overestimated their proficiency or did not exactly know which level to indicate. It can also be assumed that the text perception of the regular seller-created information is different for native English speakers compared to non-native speakers. Probably, native speakers perceive the regular seller-created information as less difficult. Therefore, the effect of simplification would be different between native and non-native speakers. Hence, future studies should include more, native speaking people to test the effects.
consumers’ purchase intention might change when there are also negative reviews. The latter may contain negative product features or product performance which were not mentioned in the seller-created information. This possibly changes the reader’s product perception or purchase intention. In the same way, if the quantity of total available online consumer reviews is not fixed, this might have an influence on consumers’ product
perception. The number of consumer reviews is often related to the product’s popularity: it represents the number of consumers who purchased the product. When this number is changing it might change consumers’ perception about the product. Thus, it is interesting to analyse in future research if the effect of social proof on product perception will change if the reviews are no longer exclusively positive and a fixed amount.
Thirdly, the experimental product used in the present study was an electronic device. Although there were grounded reasons for the use of this product type in this study, it is, as mentioned above, possible that the chosen product influences peoples’ perception outside of the influence of the online consumer reviews on this perception. For example, due to
(substantial) familiarity with this product or product type. Therefore, it is interesting in future research to examine if the same effect will be found when another, relatively new product is used for the analysis.
Fourthly, the present study assessed the effects of simplification of seller-created
information. Although it was found that this simplification was not effective in raising product perception, it might be that there is an interaction between online consumer reviews and simplification of product information on product perception if the readability of the reviews is also taken into account. An important characteristic of a review is its readability: higher quality results in a more effective review (Mackiewicz & Yeats, 2014). Since reviews and product information interact, it is interesting in future research to examine whether different results are found when the readability of the reviews is also taken into account or
manipulated. The level of perceived text difficulty of the reviews should be measured with the same construct as the perceived text difficulty of the text used in the current study.
effect of online consumer reviews on product perception. The influence of these variables needs to be analysed in future research.
All in all, the present study investigates the mediating role of consumer product perception to explain the influence of online consumer reviews on consumers’ purchase intention. In addition, it explores how simplification of seller-created product information affects consumer product perception. The data suggests that product perception did not mediate the effect of online consumer reviews on purchase intention. In addition, simplification of product
References
Alter, A. L., & Oppenheimer, D. M. (2009). Uniting the tribes of fluency to form a metacognitive nation. Personality and Social Psychology Review, 13(3), 219-235. Amblee, N., & Bui, T. (2011). Harnessing the influence of social proof in online shopping: The
effect of electronic word of mouth on sales of digital microproducts. International Journal
of Electronic Commerce, 16(2), 91-114.
Bae, S., & Lee, T. (2011). Gender differences in consumers’ perception of online consumer reviews. Electronic Commerce Research, 11(2), 201-214.
Bao, Y., Bao, Y., & Sheng, S. (2011). Motivating purchase of private brands: Effects of store image, product signatureness, and quality variation. Journal of Business Research,
64(2), 220-226.
Bart, Y., Shankar, V., Sultan, F., & Urban, G. L. (2005). Are the drivers and role of online trust the same for all web sites and consumers? A large-scale exploratory empirical study. Journal of Marketing, 69(4), 133-152.
Bickart, B., & Schindler, R. M. (2001). Internet forums as influential sources of consumer information. Journal of Interactive Marketing, 15(3), 31-40.
Brown, M., Pope, N., & Voges, K. (2003). Buying or browsing?: An exploration of shopping orientations and online purchase intention. European Journal of Marketing, 37(11/12), 1666-1684.
Chang, T. Z., & Wildt, A. R. (1994). Price, product information, and purchase intention: An empirical study. Journal of the Academy of Marketing Science: Official Publication of the
Academy of Marketing Science, 22(1), 16-27.
Chatterjee, P. (2001). Online Reviews: Do Consumers Use Them? Advances in Consumer
Research, 28, 129-133.
Chen, Y., Wang, Q. I., & Xie, J. (2011). Online social interactions: A natural experiment on word of mouth versus observational learning. Journal of Marketing Research, 48(2), 238-254.
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.
Cheng-Ping, L. D. (2017). Effects of Brand Image, Perceived Price, Perceived Quality, and Perceived Value on the Purchase Intention towards Sports and Tourism Products of the 2016 Taichung International Travel Fair. The Journal of International Management
Studies, 12(2), 97-107.
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.
Cialdini, R. B. (1993). Influence: Science and Practice (3rd ed.). London: HarperCollins
Publishers.
Colmer, R. (2017, February 22). The importance of readability of user generated content. Retrieved from https://readable.com.
Cooke, L. (2005). Eye tracking: How it works and how it relates to usability. Technical
Communication, 52(4), 456-463.
Cronin, J. J., Brady, M. K., & Hult, G. T. M. (2000). Assessing the effects of quality, value, and customer satisfaction on consumer behavioral intentions in service environments.
Journal of Retailing, 76(2), 193-218.
Dale, E., & Chall, J. S. (1948). A formula for predicting readability: Instructions. Educational
Research Bulletin, 27(2), 37-54.
Dale, E., & Chall, J. S. (1949). The concept of readability. Elementary English, 26(1), 19-26. Dellarocas, C. (2003). The digitization of word of mouth: Promise and challenges of online
feedback mechanisms. Management Science, 49(10), 1407-1424.
Dodds, W. B., Monroe, K. B., & Grewal, D. (1991). Effects of Price, Brand, and Store
Information on Buyers’ Product Evaluations. Journal of Marketing Research, 28(3), 307-319.
Duan, W., Gu, B., & Whinston, A. B. (2008). Do online reviews matter? - An empirical investigation of panel data. Decision Support Systems, 45(4), 1007-1016. DuBay, W. H. (2004). The Principles of Readability. Online Submission.
Emilien, G., Weitkunat, R., & Lüdicke, F. (Eds). (2017). Consumer perception of product risks
and benefits. Cham: Springer International Publishing AG.
Erkan, I., & Evans, C. (2016). The influence of eWOM in social media on consumers’ purchase intentions: An extended approach to information adoption. Computers in
Human Behavior, 61, 47-55.
Erkan, I., & Evans, C. (2018). Social media or shopping websites? The influence of eWOM on consumers’ online purchase intentions. Journal of Marketing Communications, 24(6), 617-632.
Fennis, B. M., & Stroebe, W. (2016). The psychology of advertising (2nd ed.). London:
Routledge.
Flesch, R. (1948). A new readability yardstick. Journal of Applied Psychology, 32(3), 221-233.
Garbarino, E., & Strahilevitz, M. (2004). Gender differences in the perceived risk of buying online and the effects of receiving a site recommendation. Journal of Business
Research, 57(7), 768-775.
Grewal, D., Krishnan, R., Baker, J., & Borin, N. (1998). The effect of store name, brand name and price discounts on consumers’ evaluations and purchase intentions. Journal of
Retailing, 74(3), 331-352.
Grunert, K. G. (2005). Food quality and safety: Consumer perception and demand. European
