Master thesis – MSc Business Administration, Strategy and Innovation
The effect of verbal and visual content on website satisfaction:
The mediating roles of uncertainty and website usability
Aydin Ilhan
Student number: S1785168 Email address: doruk@live.nl Supervisor: Dr. Florian Pallas Second supervisor: Dr. Pedro de Faria
Rijksuniversiteit Groningen
Word count: 11283
21
thof August, 2014
2
Abstract
This study examined the effect of visual and verbal content on website satisfaction, mediated by uncertainty and website usability. Four independent experimental groups faced four different manipulations (mock-up websites). The results showed that verbal content has an indirect positive effect on website satisfaction, mediated by uncertainty.
Furthermore, no significant effect was found of visual content and the interaction between visual and verbal content on uncertainty and website usability. Verbal content does not affect usability, nor does visual content have an influence on uncertainty. This indicates that a high level of verbal content can be provided together with a high level of visual content, which lowers uncertainty and thereby enhances website satisfaction.
Keywords: E-commerce, verbal and visual content, website satisfaction, uncertainty, website
usability, website aesthetics
3
Table of contents
1. Introduction ... 5
2. Literature Review ... 7
2.1 Search Engine Optimization ... 7
2.2 Website usability ... 7
2.2.1 Effect of visual content on website usability. ... 8
2.2.2 Effect of verbal content on website usability. ... 8
2.2.3 Effect of the interaction between verbal and visual content on website usability. .... 8
2.3 Uncertainty ... 9
2.3.1 Effect of verbal content on uncertainty. ... 9
2.4 Website satisfaction ... 10
2.4.1 Indirect effect of visual content on website satisfaction, mediated by website usability. ... 10
2.4.2 Indirect effect of verbal content on website satisfaction, mediated by website usability and uncertainty. ... 11
3. Methodology ... 12
3.1 Pre-test ... 12
3.2 Sample ... 12
3.3 Design and procedure ... 13
3.3.1 Mock-up websites. ... 13
3.3.2 Procedure. ... 13
3.4 Measurements ... 13
3.4.1 Website usability and website satisfaction. ... 13
3.4.2 Uncertainty. ... 14
3.4.3 Age. ... 14
3.4.4 Sex. ... 14
3.4.5 Internet use. ... 14
4. Results ... 15
4.1 Manipulation checks ... 15
4.2 Analyses ... 16
4.3 Mediation ... 17
4.4 Control variables ... 18
5. Discussion and Conclusion ... 19
5.1 Direct effect of verbal and visual content ... 19
5.1.1 The effect of verbal content on uncertainty and website usability. ... 19
4
5.1.2 The effect of visual content on website usability. ... 20
5.2 The indirect effect of verbal and visual content ... 20
5.2.1 The indirect effect of verbal content on website satisfaction... 20
5.2.2 The indirect effect of visual content on website satisfaction. ... 21
6. Implications and Future Research ... 22
6.1 Theoretical implications ... 22
6.2 Managerial implications ... 22
6.3 Limitations and future research ... 23
7. References ... 24
8. Appendix ... 32
8.1 Appendix A: Mock- up Websites ... 32
8.1.1 Mock-up website high visual and high verbal ... 32
8.1.2 Mock-up website low visual and high verbal ... 34
8.1.3 Mock-up website high visual and low verbal ... 36
8.1.4 Mock-up website low visual and low verbal ... 38
8.2 Appendix B: Literature Review Scales ... 40
8.3 Appendix C: Scales Questionnaire Variables ... 41
5
1. Introduction
The use of information and communication technologies to facilitate service delivery is increasing in popularity for companies (MacDonald & Smith, 2004). An important innovative service channel is web-initiated ordering, of which the implementation can lead to improvements in overall consumer satisfaction (Birgelen et al., 2005). However, the successful use of these e-service channels is dependent on the level of online consumer satisfaction, hereafter called website satisfaction (Schaupp & Belanger, 2005). Website satisfaction is an important driver of repurchase intentions (Fang et al., 2011). If customers are satisfied with the website, they will keep using the website; if customers get frustrated and dissatisfied with the online system, they will be unlikely to come back for a visit (Xiao &
Dasgupta, 2005). A main predictor of website satisfaction is the usability of a website (Fang et al., 2011). A lack of usability on websites can therefore lead to business failures (Becker &
Mottay, 2001).
A main contribution of website usability is that it promotes the interaction between the website and visitor (Corritore et al., 2003), which keeps the potential consumers on the website and encourages them to purchase a product online (Belanche et al., 2012). However, before an online manager can satisfy its visitors with an impressive usability, visitors need to be attracted to a website, for which search engine optimization (SEO) is an effective strategy to use (Frydenberg & Miko, 2011).
Search engine optimization (SEO), which is the process of improving the position of websites on search engines by making the website content more relevant for consumers (Berman & Katona, 2013), is an important online strategy for managers since it leads to an increase in overall site traffic and improves the interaction with customers (Kritzinger &
Weideman, 2013). SEO is considered also a key factor in website success, as it can make unknown brands appear ahead well-known ones (Dou et al., 2010). Due to the technological innovations within the search industry (Green, 2003), it is more and more challenging for companies to implement SEO and adapt their strategies to the online environment.
While SEO attracts online visitors and usability keeps visitors on websites, investing in both SEO and website usability could be the key online channel strategy for achieving success. However, SEO focuses on integrating a substantial (and relevant) amount of verbal content on a website (Malaga, 2007) with a limited amount of visual content, since search engines cannot read images (Wang et al., 2012), whereas website usability is positively influenced by visual content with a limited amount of verbal content (Griffith et al., 2001;
Nielsen & Loranger, 2006) – showing the possible contradicting relationship between SEO and website usability. Should companies focus on optimizing online verbal content (which is one of the important features of SEO) in order to attract visitors, or on visual content to maximize website usability and website satisfaction?
The finding of a balance between visual and verbal information is important as managers are increasingly overloaded with information in online environments (Lurie, 2004).
Online firms are therefore competing now by offering attractive, high quality content to users
in order to meet consumer needs (Blanco et al., 2010) and gain a competitive advantage on
the market (Lynch & Ariely, 2000). Although the presentation of visual and verbal content is
considered to play an important role in explaining information processes and decision making
6
on websites (Chau et al., 2000; Hong et al., 2004; Schlosser, 2003), there is limited research of the effect of visual and verbal content on website satisfaction. Kim & Lennon (2008) and Ballantine (2005) showed the dominating role of verbal content (rather than visual content) in influencing website satisfaction. Based on their findings, Kim & Lennon (2008) concluded that detailed verbal descriptions of products on websites are important to influence positive attitudes of consumers towards a website, assuming that it is because of a lowered uncertainty. However, verbal content can also negatively affect the usability of a website (Flavian et al., 2008) and since a lower website usability can decrease website satisfaction (Casalo et al., 2008; Belanche et al., 2012), verbal content can be an indirect threat to website satisfaction. This indicates that verbal content can both positively and negatively affect website satisfaction, by respectively uncertainty and website usability.
Yet, there is no research that has investigated the mediation effects of website usability and uncertainty in the relationship between verbal content and website satisfaction. On the other hand is visual content in the form of graphics and images considered to enhance website usability (Flavian et al., 2008; Tan & Wei, 2006; George, 2005), which in turn enhances website satisfaction (Casalo et al., 2008; Belanche et al., 2012). However, there is also no research that has tested the mediating effect of website usability in the relationship between visual content and website satisfaction.
In order to better understand the indirect relationship between visual and verbal content and website satisfaction, this study will look whether website usability and uncertainty mediate the relationship between website content (in verbal and visual form) and website satisfaction. More precisely, this study will look whether there is a significant effect of visual and verbal content on website usability and uncertainty and whether website usability and uncertainty have a significant effect on website satisfaction. From a managerial perspective - as research concluded that both website usability and SEO are crucial for website success (Palmer, 2002; Zviran et al., 2006; Frydenberg & Miko, 2011; Malaga, 2007) - it is important to know how both can be managed at the same time (and thus please both human audience and search engine crawlers). Only focusing on SEO elements (optimizing verbal content) would mean that the online consumer experience will be ignored and thus visitors cannot be kept satisfied and retained on the website. On the other hand, not taking SEO into account would mean that an opportunity will be missed to attract online visitor traffic. Based on the research problem, this study will try to answer the following research questions:
1) What is the direct effect of visual and verbal content on uncertainty and website usability?
2) What is the indirect effect of visual and verbal content on website satisfaction, through the mediation effect of uncertainty and website usability?
This paper is structured as follows. In the following section, prior literature regarding website
satisfaction, uncertainty and website usability will be reviewed. Based on the literature,
hypotheses will be formulated about the effects of visual and verbal content on the dependant
variables. Next, the methodology will be explained, followed by the presentation of the
experimental results. Finally, the results, implications and limitations of the study will be
discussed.
7
2. Literature Review
2.1 Search Engine Optimization
Online search engines are among the most popular tools that consumers use to discover information on the Internet. As a result, search engine marketing is becoming a dominant form of online advertising (Berman & Katona, 2013). The process of promoting a website so that it becomes easily indexed by search engines is known as search engine optimization, also called SEO (Malaga, 2007). As a higher placement within search results will lead to more traffic to a website, SEO is an important tool for many companies (Frydenberg & Miko, 2011). More than 73% of the search engine users never look beyond the first page of results (Frydenberg & Miko, 2011). Because of this importance of high search engine rankings and the profits involved, search engine optimizers look for tools that will help them to obtain high rankings (Malaga, 2010). Companies that try to improve its position on search engines can do this by making the website content more relevant for consumers (Berman & Katona, 2013).
Since site content can be in verbal or visual form and is most often used in a mixed form (Kim
& Lennon, 2008), it is important to know what the effect is of different forms of verbal and visual content on website usability and uncertainty.
2.2 Website usability
Usability is a key factor in a website’s success (Palmer, 2002; Zviran et al., 2006). Authors (Palmer, 2002; Zviran et al., 2006; Casalo et al., 2008; Flavian et al., 2006; Venkantesh &
Agarwal, 2006) have identified different usability elements in the past, trying to determine what on a website is part of usability and what is not. Nielsen (1994) has formulated ten different usability elements: visibility of a system status; match system with real world; user control and freedom; consistency and standards; error prevention; recognition; efficiency of use; aesthetic design; help user recover errors and documentation. More specifically, website usability elements can be distinguished in the ease of understanding the structure of the website; simplicity of the use of the website in initial stages; speed with which the users can find what they are looking for; the perceived ease of site navigation in terms of time required;
action necessary in order to obtain desired results and the ability of users to control what they are doing (Casalo et al., 2008). In short, usability is associated to the easy-of-use of a website (Flavian et al., 2006).
Website usability is considered as one of the most important factors for determining
website quality (Yang et al., 2005; Yang & Fang, 2004). It contributes to make information
more transparent and promotes the interaction between the website and the visitor which
simplifies the transaction process and helps users to find what they are looking for (Corritore
et al., 2003). Because website usability makes websites easier to use, it has a direct influence
on the consumer, including attitude towards a website (Lui & Arnett, 2000), consumer
satisfaction (Casalo et al., 2008; Belanche et al., 2012) and online purchase behaviour
(Venkantesh & Agarwal, 2006). Less research has been done on the factors that affect website
usability (usability as a dependant variable). Since website content can be either visual or
verbal, it is expected that both have an effect on website usability.
8
2.2.1 Effect of visual content on website usability. Visual information provides images or pictures of products, which can be displayed in different sizes, angels and perspectives (Blanco et al., 2010). Tan & Wei (2006) state that a good design in the form of graphics and images should make the website experience more enjoyable and pleasant, thereby enhancing website usability. Furthermore, visual content enhances perceptions of information quality, measured as the users’ impression of accuracy and completeness of website information (Kim et al., 2008). It is expected that with a positive impression of accuracy and completeness of website content, the perceived ease of website navigation will be high. Finally, visual aspects of a website can display information more briefly (Lurie &
Mason, 2007), leading to an increase in the decision-making efficiency of consumers (Khakimdjanova & Park, 2005). A more efficient decision-making process indicates that it is easier to obtain the desired results, which is one of the indicators of website usability (Casalo et al., 2008). Thus, it is expected that visual content will have a positive effect on website usability.
H1: Online visual content has a positive effect on website usability.
2.2.2 Effect of verbal content on website usability. Verbal content describes content with words and offers specific and detailed information about product characteristics (Blanco et al., 2010). It is argued that verbal content (words) is an abstract representation of distal things (Amit et al., 2012). Verbal information can be presented in two ways, either in a paragraph or schematic. The experiment of Flavian et al. (2009) showed that when the information was presented in a schematic way (product characteristics listed in a table or a chart with less amount of information), users perceived a higher degree of usability than when the information was given in a paragraph. Additionally, also the perceived quality of the information was higher when information was schematically presented. An explanation is that online visitors scan information online (Griffith et al., 2001) instead of reading it completely.
Visitors stay in average 45-60 seconds on a webpage (Nielsen & Loranger, 2006), making it hard for visitors to read plenty of information during the limited time period, leading to a damaged perceived usability. Therefore, it is expected that verbal content will have a negative effect on website usability.
H2: Online verbal content has a negative effect on website usability.
2.2.3 Effect of the interaction between verbal and visual content on website usability.
Since product information is most often presented as a combination of both visual and verbal
form (Kim & Lennon, 2008) it is important to examine the interaction effect between visual
and verbal content. Blanco et al. (2010) showed that the combination of visual and verbal
content positively affects the recall of product information and the perceived ease (through a
positive usability) with which the user recalls product information. The same study proved
that visual content does not have a significant effect without the presence of textual content,
which shows that an interaction is needed to positively influence information re-call. Thus,
the third hypothesis of this study is:
9
H3: The interaction between online verbal and visual content has a positive effect on website usability.
2.3 Uncertainty
Uncertainty is mainly related to the product’s true attributes and future performance (Dimoka et al., 2012). Product uncertainty consists of two components according to Dimoka et al.
(2012): description uncertainty and performance uncertainty. Online sellers may be unable to perfectly describe the product through their online shops. They can be also unaware of the hidden defects that can affect the product’s performance, referring to the difficulty for consumers to predict how the product will perform in the future. Product performance uncertainty plays an important role in the e-commerce world (Bhatnagar & Ghosh, 2004), since online consumers are unable to examine the product while making choices (Forsythe et al., 2006) and consumers look for both immediate benefits and long-term implications of the purchase (Sweeney et al., 1999). Prior studies showed that uncertainty is negatively related to customer satisfaction (Chaudhuri, 1998; Homburg et al. 2006 and Smith & Bolton, 2002).
Since product descriptions and performance can be provided in textual form (Spiller & Lohse, 1998), uncertainty is expected to be affected by verbal content.
2.3.1 Effect of verbal content on uncertainty. When consumers feel more knowledgeable about the product for which they are shopping, their affective and cognitive attitudes may be positively affected (Kim & Lennon, 2008). Research evidence supports that increased verbal information leads to more knowledgeable consumers who are then able to make more informed decisions (Cook & Coupey; 1998, Glazer, 1991). Smith (1991) found that verbal information in ads makes explicit, specific claims about product attributes or performance, which facilitate inferences about unknown information about a product. In addition, websites can offer more information at the point of choice, thereby helping customers to make better choices. Some websites have multiple layers of web pages with detailed information. If the (verbal) information is multi-layered and rich, customers will tend to value the service encounter more than when the information is very superficial (Glazer, 1991), thereby lowering uncertainty. Thus, it is expected that verbal content will have a negative effect on uncertainty.
H4: Online verbal content has a negative effect on uncertainty.
10
2.4 Website satisfaction
Satisfaction has been the primary focus in the marketing literature (Oliver, 1980; Gustafsson et al., 2005). It is defined by Oliver (1981) as “the summary psychological state resulting when the emotion surrounding disconfirmed expectations is coupled with the customer’s prior feelings about the consumption experience.” In other words, satisfaction is an evaluation of the relationship history between the company and the consumer (Casalo et al., 2008). For the online environment, additional concepts of e-satisfaction have been proposed, which is also named as “website satisfaction”, conceptualized as the emotional reaction to the experience provided by a website (Park & Kim, 2003). Hence, in this study website satisfaction will be the evaluation of visitors’ website experience.
Within the e-commerce environment, website satisfaction determines the customer channel preference (Devaraj et al., 2002), positively influences the loyalty of consumers toward online shopping (Chiu et al., 2009) and leads to a re-usage of a website (Belanche et al., 2012). Prior research did not only focus on website satisfaction as an independent variable that affects other variables, but due to unsatisfied visitors also the question of what affects customers’ evaluation of their (online) buying experience (website satisfaction as a dependant variable) has been the focus in the literature. Website satisfaction is positively influenced by a wide range of factors, including shopping convenience and privacy (Schaupp & Belanger, 2005); interactivity and product information (Ballantine, 2005); image attributes (Yun &
Good, 2006); vendor reliability and safe purchasing (Kim & Kim, 2006). However, it is also possible that the online environment could decrease satisfaction due to the perceived lack of privacy, human contact and poor design of interface (Meuter et al., 2000).
2.4.1 Indirect effect of visual content on website satisfaction, mediated by website usability. Visual content is considered to positively affect website satisfaction (Lurie &
Mason, 2007; Lui & Stout, 1987). Prior research showed that website usability has a positive effect on website satisfaction (Belanche et al., 2012; Casalo et al., 2008) because website usability makes it easier for consumers to use a website, which leads to positive attitudes towards a website (Liu & Arnett, 2000). Because of the positive relationship between visual content and website usability, it is expected that visual content has an indirect positive effect on website satisfaction, through website usability.
H5: Online visual content has an indirect positive effect on website satisfaction through the
mediation effect of website usability.
11
2.4.2 Indirect effect of verbal content on website satisfaction, mediated by website usability and uncertainty. Research showed that verbal content can positively influence website satisfaction (Kim & Lennon, 2008; Ballantine, 2005). Kim & Lennon (2008) proved that verbal superiority occurred, by showing that a higher amount of verbal content had a positive effect on the attitudes of visitors on websites and purchase intentions, assuming that it is due to a lowered uncertainty. Another study that also referred to the important role of verbal content is that of Ballantine (2005), showing that the amount of verbal information provided by the website increases online consumer satisfaction. These positive attitudes on a website can be explained by affect (Homburg et al., 2006; Oliver, 1981).
Homburg et al. (2006) concluded that the experienced affect during the consumption of a product or service can have a significant effect on satisfaction. When affect takes the form of uncertainty, positive feelings during consumption can decrease (Chaudhuri, 1998).
This is because uncertainty causes consumer anxiety during the purchase process (Luo et al., 2012). Therefore uncertainty can cause negative affective reactions (Bhatnager et al., 2000), which in turn decreases satisfaction (Homburg et al., 2006). Since verbal information can lower uncertainty (Glazer, 1991; Smith, 1991; Kim & Lennon, 2008) and because of the negative relationship between uncertainty and website satisfaction, it is expected that verbal content indirectly increases website satisfaction, through the mediating effect of uncertainty.
However, because of the proposed negative relationship between verbal content and website usability, website satisfaction can also be negatively affected by verbal content through the mediation effect of website usability. Thus, the following two hypotheses are formulated:
H6: Online verbal content has an indirect positive effect on website satisfaction through the mediation effect of uncertainty.
H7: Online verbal content has an indirect negative effect on website satisfaction through the mediation effect of website usability.
Figure 1 shows a conceptual model of these proposed relationships.
Figure 1: Conceptual model
12
3. Methodology
A 2 (visual content: high/low degree) x 2 (verbal content: high/low degree) factorial “between subjects” experiment was conducted to examine the effect of online verbal and visual content on uncertainty, website usability and the indirect effect on website satisfaction. Within this experiment, there were four independent (experimental) groups that faced four different manipulations (mock-up websites). The four experimental groups are shown in table 1:
Table 1: Experimental groups
3.1 Pre-test
A pre-test of the perceived amount of visual and verbal content was performed, in order to be sure that the four experimental conditions were perceived as intended. An online survey was conducted among 30 people that rated the perceived amount of verbal and visual content for each experimental website on a scale between 1 (low amount) and 7 (high amount). The order of the websites was randomized among the participants of the pre-test, so that each participant rated the experimental conditions in a different order. ANOVA analyses showed that the amount of visual content had a significant positive effect on the perceived amount of visual content (F = 113.69; P < 0.001) with a mean difference of 2.41 (M= 2.23 for low visual, whereas M = 4.61 for high visual). For verbal content, the significant effect was higher (F = 332.25; P < 0.001) with a mean difference of 3.68 (M = 2.15 for low verbal and M = 5.83 for high verbal). The interaction effect of visual and verbal content on the perceived amount of both visual content (F = 0.65; P = 0.420) and verbal content (F = 1.53; P = 0.219) were not significant. Thus, the experimental manipulations were received as intended.
3.2 Sample
The sample consisted of four groups with each 30 participants, making it a total of 120 participants. The participants were mainly students from the Fontys University of Applied Sciences in Eindhoven, where the experiment was conducted. In order to enhance the validity, all participants were randomly assigned to one of the four experimental conditions, so that there were equally distributed groups.
Verbal content
Visual content Low amount High Amount
Low amount Low visual and low verbal Low visual and high verbal High amount High visual and low verbal High visual and high verbal
13
3.3 Design and procedure
Four different mock-up websites were created that simulated a virtual store selling mobile phones. On the website popular mobile phones were presented to the participants, because electronic goods are the most purchased goods online (EIAA, 2008) and mobile phones are popular among students. The participants were physically present at a special computer room.
When each participant logged in, he or she was randomly assigned to one of the four mock-up websites (the experimental conditions).
3.3.1 Mock-up websites. The websites were created according to the four experimental groups, meaning that each website included a low or high amount of verbal and visual content. For verbal content, the content presentation of Blanco et al. (2010) was used, displaying it in either a paragraph form or a schematic form. The schematic form represented a low amount of verbal content, whereas the paragraph form represented a high amount of verbal content. Since Kim & Lennon (2008) examined the effect of visual content by using large and small images, this study used either more and bigger images (higher amount of visual content) or less and smaller images (lower amount of visual content). Each mock-up consisted of a homepage and a product page. The four mock-up websites that were used in the experiment are visible in Appendix A.
3.3.2 Procedure. Before the experiment began, each participant read a short introduction where the hypothetical situation was explained. The participants were asked to act like if they were real consumers and their goal was to make a top three list of mobile phones (they were warned that there could be questions about it at the end of the questionnaire). The reason why they were asked to make a top three mobile phone list was because it could enhance their focus on the website, rather than just navigating through the website without any goal, also called a task-oriented activity with an utilitarian orientation (Kaltcheva & Weitz, 2006). After the participants were done with their task to make a top three list of mobile phones, they switched to the online survey where they answered questions regarding uncertainty, website usability and website satisfaction, using a Likert scale from 1 (totally not agree) to 7 (totally agree).
3.4 Measurements
3.4.1 Website usability and website satisfaction. Website usability and website satisfaction were both measured by using the instruments of Casalo et al. (2008). Casalo et al.
(2008) developed the scale, based on the review of the most relevant literature on relationship
marketing and e-marketing (see Appendix B).
14
3.4.2 Uncertainty. Uncertainty was measured using the instruments of Dimoka et al.
(2012), regarding product uncertainty. All constructs were measured by a multi-item scale, as shown in Appendix C. Control variables included in the survey were age, sex and Internet use (see table 2 for descriptive statistics).
3.4.3 Age. According to Thomas & Streib (2003), demographic characteristics, including age, play a critical role in technology adoption behaviour. It could be that people that are older or younger can use the Internet differently, due to a different cognitive age (Eastman & Iyer, 2005). Age was measured using a distinction in three groups: below 20 years, between 20 and 25 years and above 25 years.
3.4.4 Sex. McGiven et al. (1997) argued that there are sex differences in visual recognition memory, stating that women process information better than men, due to a more efficient visual recognition capability. Therefore, sex was also taken as a control variable, to test whether sex will influence the experimental results or not, since there is a chance that women are more attracted to the experimental conditions where visual content is dominating.
3.4.5 Internet use. Belanger & Carter (2006) found that the frequency of Internet use impacts the use of online services of the government. People that are using the Internet more frequently, could evaluate a website with different forms of verbal and visual content differently than people that visit websites less frequently. A possible reason is that for users with less Internet experience, the importance of visual and verbal content on a website can be lower. Internet use was measured by hours of Internet use per week and was divided into four groups: less than 10 hours a week; between 10 hours and 20 hours a week; between 20 and 30 hours a week and more than 30 hours a week.
Frequency %
Age
<20 36 30.00
20 – 25 70 58.33
> 25 14 11.67
Sex
Male 81 67.50
Female 39 32.50
Internet use (hours per week)
< 10 29 24.16
10 - 20 24 20.00
20 - 30 25 20.84
> 30 42 35.00
Table 2: Overview of descriptive statistics
15
4. Results
In order to test the construct validity and the reliability of the used constructs, different analyses were conducted. First of all, the validity was tested using a factor analysis. A factor analysis with principal components was conducted to determine whether each item of the construct had an eigenvalue greater than 1. Also, the factor loadings needed to be higher than 0.5 with a significant total explained variance (Hair et al., 1998). The results of the factor analysis met the requirements, for each of the dependant variables there was one factor with items that had a factor loading higher than 0.5: website satisfaction (explained variance = 87.12%; KMO = 0.73; Barlett test χ
2(3) = 291.83; P < 0.001), uncertainty (explained variance
= 56.08%; KMO = 0.80; Barlett test χ
2(15) = 318.79; P < 0.001) and website usability (explained variance = 65.97%; KMO = 0.87; Barlett test χ
2(15) = 413.18; P < 0.001).
Second, the reliability was tested using the Cronbach alpha indicator (Cronbach, 1970), which required a minimum value of 0.7 (Nunnally, 1978). The results were overall satisfactory: website satisfaction (α = 0.93), uncertainty (α = 0.84) and website usability (α = 0.89). The reliability and validity results are shown in table 3.
Cronbach’s Alpha All factor loadings >
0.5
Explained variance
Website Satisfaction 0.93 Yes 87.12%
Uncertainty 0.84 Yes 56.08%
Website Usability 0.89 Yes 65.97%
Table 3: Reliability and Validity
4.1 Manipulation checks
To check whether the amount of visual and verbal content was perceived as intended, the participants were asked to rate the amount of visual and verbal content. ANOVA analyses were performed with visual content, verbal content and the interaction between visual and verbal content as the independent variables; the perceived amount of visual and verbal content were the dependant variables. The results showed that the amount of verbal content had a significant effect on the perceived amount of verbal content (F = 73.37; P < 0.001).
Participants that were exposed to more verbal content perceived the amount of verbal content larger (M = 5.32; SD = 1.53) than participants that were exposed to less verbal content (M = 3.00, SD = 1.40). Nor visual content (F = 0.00; P = 0.951) or the interaction (F = 0.19; P = 0.667) had a significant effect on the perceived amount of verbal content.
The same was true for the perceived visual content: the amount of visual content had a
significant influence on the perceived amount of visual content (F = 14.59; P < 0.001) without
any significant effect of verbal content (F = 0.13; P = 0.794) or the interaction between visual
and verbal content (F = 0.15; P = 0.696). Participants that were exposed to more visual
content perceived the amount of visual content larger (M = 3.58; SD = 1.27) than those
exposed to less visual content (M = 2.61; SD = 1.50). Thus, the manipulations were
successful.
16
4.2 Analyses
The hypotheses were tested using a multivariate analysis of variance (MANOVA) with uncertainty and website usability as dependant variables. Results are visible in table 4. This analysis showed that verbal content had a significant effect on the set of dependant variables (Wilks’s λ = 0.91; F = 5.40; P = 0.006; power = 0.84). Visual information had no significant effect (Wilks’s λ = 0.95; F = 2.94; P = 0.057; power = 0.56). Also no significant effect was found for the interaction of visual and verbal information (Wilks’s λ = 0.95; F = 2.88; P = 0.060; power = 0.43).
Table 4: Results multivariate analysis of variance
In order to test the individual effect of verbal content on each of the dependant variables, an univariate analysis of variance (ANOVA) was performed. The ANOVA (results are visible in table 5) showed that verbal information had a significant effect on the uncertainty of the participants (F = 6.81; P = 0,010; power = 0.74). The uncertainty was lower when the participants were exposed to more verbal information (M = 3.77; SD = 0.16) than when they were exposed to less verbal information (M= 4.34; SD = 0.16), in support of H4. There was no main effect of verbal content on website usability (F = 2.30; P = 0.132; power = 0.33), thereby rejecting H2. Due to the non-significant effects of visual content and the interaction on the dependant variables (as the MANOVA showed), the hypotheses H1 and H3 were also not supported. The means of the dependant variables within each experimental condition are shown in table 6.
Table 5: Results univariate analysis of variance
1 Although the ANOVA here is significant, according to Mayers (2013) it is only relevant to look at the individual ANOVA effects, if the MANOVA results are also significant. Since the MANOVA showed no significant effect of visual content, there was no meaning to look at the significant ANOVA effect of visual content.
Effect F Sig. Observed Power
Visual content 2.94 .057 .56
Verbal content 5.40 .006 .84
Interaction 2.88 .060 .43
Source Dependent Variable DF Mean Square F Sig. Power
Visualcontent Website usability 1 6.23 4.65 .0371 .56
Uncertainty 1 3.56 2.47 .191 .34
Verbalcontent Website usability 1 3.22 2.30 .132 .33
Uncertainty 1 9.82 6.82 .010 .74
Visualcontent * Verbalcontent
Website usability 1 .27 .19 .663 .07
Uncertainty 1 8.36 4.34 .018 .67
17 Dependent Variable Visualcontent Verbalcontent Mean Std. Error
Website usability Lowvisual Lowverbal 5.11 .22
Highverbal 4.88 .22
Highvisual Lowverbal 4.75 .22
Highverbal 4.33 .22
Uncertainty Lowvisual Lowverbal 4.43 .22
Highverbal 3.33 .22
Highvisual Lowverbal 4.25 .22
Highverbal 4.21 .22
Table 6: Means of dependant variables in the four experimental conditions
4.3 Mediation
In order to find an indirect effect of verbal content on website satisfaction, mediation analyses were performed, using the bootstrapping method with 5000 bootstrap samples on a 95%
confidence interval (Preacher & Hayes, 2004) with verbal content as the independent variable;
uncertainty and website usability as the mediators and website satisfaction as the dependant variable. According to Zhao et al. (2010), mediation will occur when there is a significant a- path (effect of the independent variable on the mediator) and a significant b-path (effect of the mediator on the dependant variable), thus concluding that there is no need for a significant c- path (direct effect of independent variable on dependent variable). The multiple regression analyses showed that the a-path – the effect of verbal content on uncertainty - was significant (B = -0.57; t = -2.56; P = 0.011). The b-path – the effect of uncertainty on website satisfaction – was also significant (B= -0.19; t = -2.15; P = 0.034). As for the c-path: the effect of verbal content on website satisfaction was not significant (B = 0.27; t = 0.97; P = 0.336). To check for an external effect on the dependant variable, visual content was used as a covariate in the mediation model: no significant partial effect was found of visual content on website satisfaction (B = 0.12; t = 0.58; P = 0.560).
Second, the confidence interval excluded 0 (0.003 – 0.300), which is a necessary
requirement for the establishment of a mediation (Zhao et al., 2010). These results offered
support for H6, due to the presence of an indirect-only mediation (Zhao et al., 2010). Thus,
the results confirmed the mediating role of uncertainty in the relationship between verbal
content and website satisfaction. There was no support found for H7, due to lack of
significance of the a-path in the relationship between verbal content and website satisfaction
with website usability as the mediator (B = -0.33; t = -1.52; P = 0.131). However the b-path
(relationship between website usability and website satisfaction) was significant (B = 0.79; t =
8.67; P = < 0.001), indicating that a higher level of website usability increases website
satisfaction. The results are illustrated in figure 2.
18
4.4 Control variables
The used control variables were age, sex and Internet use. An overview with frequencies and percentages is visible in table 2. These variables were used as covariates in the MANOVA analysis, to check for an external effect on website satisfaction. There was no significant effect of age (F= 2.326; P = 0.102), gender (F = 0.983; P = 0.377) or Internet use (F = 0.334;
P = 0.717) on the dependant variables.
Figure 2: Results experiment
19
5. Discussion and Conclusion
The objective of this study was to investigate whether visual and verbal content had both an effect on website satisfaction, with the mediation roles of uncertainty and website usability.
This study first examined the direct effect of visual and verbal content on website usability and uncertainty. Secondly, the mediating roles of website usability and uncertainty were tested regarding the relationship between visual and verbal content and website satisfaction.
An experiment was conducted with four experimental conditions to find out the effect of visual and verbal content on the mediators uncertainty and website usability and on website satisfaction. The two research questions (regarding the direct and the indirect effect of verbal and visual content) will be discussed in more detail in the following subsections.
5.1 Direct effect of verbal and visual content
5.1.1 The effect of verbal content on uncertainty and website usability. The results of the experiment showed that verbal content has a significant negative effect on uncertainty.
This means that online visitors that are exposed to more verbal content will perceive less uncertainty than visitors that are exposed to less verbal content. This finding is in support with the research of Kim & Lennon (2008) who suggested that when there is verbal superiority (a high amount of verbal information), online consumers feel more knowledgeable about the product they want to buy and due to the lowered uncertainty, their cognitive attitudes are positively affected.
Additionally, another possible explanation is that verbal information makes explicit and specific claims about a product, which facilitates inferences about unknown information (Smith, 1991), thereby reducing uncertainty. As for the effect of verbal content on website usability: no significant effect was found. This finding contradicts with the study of Flavian et al. (2008) that showed that a higher degree of website usability was perceived when information was presented in a schematic form with less verbal content. The reason for this non-significant effect could be that website usability consists of a lot of elements and is influenced by different factors, like the speed of the website, the error prevention, the aesthetic design and the ability of users to control what they are doing (Nielsen, 1994; Casalo et al., 2008). This means that website usability is mainly associated with how easy it is to use the website. While the assumption was that verbal content can negatively affect the easiness of navigating through the website due to an overload of textual information, other factors like the menu, breadcrumbs and the time that is required to navigate through the website can be dominating regarding the effect on website usability. Therefore, it may be that visitors only perceive it as an overload of information, when it exceeds a higher (critical) amount of verbal content.
Another explanation is that verbal content may not affect website usability, but rather
the overall perceived aesthetics, due to the fact that a high amount of verbal content is not
quite an attracting design aspect, which influences website aesthetics (Sonderegger & Sauer,
2010). The usability of a website is distinct from website aesthetics. It is not about the degree
to which a person believes that the website is aesthetically pleasing to the eye (Van der
20
Heijden, 2003) but about how easy it is to use a website (Flavian et al., 2006). Thus while verbal content can negatively affect the aesthetics of a website, it could still be combined with a relatively high level of website usability.
5.1.2 The effect of visual content on website usability. In contrary with the propositions, no significant effect was found of visual content on website usability. A possible explanation is that a high degree of task engagement during the experiment could have shifted the focus on verbal information, while ignoring the visual aspects of the website (Flavian et al., 2010). The experiment asked the participants to act as task oriented as possible (looking for the top three mobile phones). This could have made verbal content predominant, as the specifications, general information and review of mobile phones (that can considerably determine a mobile phone as a favourite one) are mainly provided by verbal content, rather than visual content. In contrary, in the real world there is the search for entertainment (Sanjosé-Cabezudo et al., 2009), where visual aspects of a website can be in the centre of attention.
Additionally, like explained before, while visual aspects can enhance the aesthetics of a website (Han & Mills, 2006), it may not affect the usability, as website usability is related to the use of a website. However, prior research did conclude a positive relationship between perceived usability and product aesthetics (Tractinsky et al., 2000; Sonderegger & Sauer, 2010; Schenkman & Jonsson, 2000). Thus, it may be that website aesthetics has a mediation effect in the relationship between visual content and website usability.
5.2 The indirect effect of verbal and visual content
5.2.1 The indirect effect of verbal content on website satisfaction
.In support with hypothesis H6, a significant indirect effect of verbal content on website satisfaction was found, with uncertainty as a mediator. This indicates that a higher amount of verbal content increases website satisfaction indirectly through the mediating role of uncertainty. This finding is again in line with Kim & Lennon (2008) and Ballantine (2005) that showed a positive relationship between verbal content and website satisfaction. They assumed that a decrease of uncertainty was the reason why this effect occurred.
This study empirically proved that uncertainty is a mediator in the relationship between verbal content and consumer satisfaction. When there is a substantial amount of verbal content present on a website, it will negatively affect uncertainty, as explained in the previous research question. The significant b – path (the negative relationship between uncertainty and website satisfaction) can be explained with prior findings (Chaudhuri, 1998;
Bhatnager et al., 2000; Homburg et al., 2006; Luo et al., 2012) that related uncertainty to
consumer anxiety, leading to negative affective reactions. This negative affect that occurs
could decrease website satisfaction. As for the mediating role of usability: the a-path was
already found non-significant in hypothesis 2, for which a potential explanation was made in
the previous research question. However, a positive significant relationship between website
usability and website satisfaction was found (the b – path), which supports the findings of
prior research (Liu & Arnett, 2000; Belanche et al., 2012; Casalo et al., 2008) that also
concluded a positive effect of website usability on website satisfaction. Website usability
21
makes it easier for online consumers to use a website, because the website made it possible for them to achieve their goal (navigate through the website and buy a particular product for instance), leading to an increase in website satisfaction.
5.2.2 The indirect effect of visual content on website satisfaction. Lastly, the study found no significant indirect effect of visual content on website satisfaction, contradicting prior findings (Lurie & Mason, 2007; Lui & Stout, 1987) that showed a positive effect of visual content on website satisfaction. Due to a lack of a significant effect of visual content on website usability, no mediation effect was found. However, like stated before, a double
mediation is proposed to be present, with visual content as an independent variable, website
satisfaction as a dependant variable and website aesthetics and website usability as the two
mediators that positively affect each other.
22
6. Implications and Future Research
6.1 Theoretical implications
This research makes several contributions to the literature. First, it showed that verbal content on a website negatively influences uncertainty of visitors, which was not examined in previous research. Furthermore, this study demonstrated that uncertainty mediates the relationship between verbal content and website satisfaction. These results provide insight into the important role of uncertainty in explaining the indirect relationship between verbal content and website satisfaction: providing online verbal content will not only decrease uncertainty, but also increase website satisfaction due to the mediating role of uncertainty.
The experimental findings confirmed the verbal superiority where Kim & Lennon (2008) and Ballantine (2005) talked about. However, it was not clear how exactly verbal content was influencing website satisfaction. This study is the first that empirically proved that verbal content indirectly enhances website satisfaction through the mediation effect of uncertainty.
6.2 Managerial implications
This study showed that visitors are not having any trouble with a substantial amount of verbal content regarding the usability of the website and they profit from it by a decrease in perceived uncertainty, which in turn enhances their satisfaction as an online consumer. These experimental results entail different managerial implications.
First of all, online managers should provide verbal content on their websites in order to lower down the perceived uncertainty, thereby indirectly affecting website satisfaction. The presence of website satisfaction means that online managers successfully make use of their e- service channels, which can lead to improvements in overall consumer satisfaction and the further implementation of novel service channels (Birgelen et al., 2005). Furthermore, nowadays online markets still face a barrier in physical experience of products that cannot be easily described via the Internet, leading to information asymmetry and an increase in the uncertainty of online consumers (Dimoka et al., 2012). Thus providing verbal content will decrease these barriers for consumers to gain knowledge about products and firms that sell these products.
Second, a high amount of verbal content will also contribute to SEO, since SEO is based on providing a substantial amount of verbal optimized content (Malaga, 2007), which will attract visitor traffic, thereby improving interaction with customers (Kritzinger &
Weideman, 2013; Pan et al., 2012). Thus providing verbal content will also lead to more visitors with a higher interaction (Kritzinger & Weideman, 2013). More visitors means a potential increase in clients, which is especially important now the growing competition in the e-commerce market makes it difficult for online companies to increase the client base (Casalo et al., 2008). Investing in SEO also means that companies will adapt to the changing (innovating) online environment, which is beneficial from innovation management perspective.
Third, as verbal content does not affect website usability and visual content does not
affect uncertainty - managers should not give priority to visual content in order to increase
usability, rather they should use visual content combined with verbal content in order to affect
23
website satisfaction. This is possible in two different ways: by a lower uncertainty – influenced by verbal content - and by an increase in aesthetic appeal, influenced by visual content (Wang et al., 2010). This also means that focusing on SEO by providing optimized verbal content will attract search engines to get visitors, which at the same time will not affect the visitor experience on a website, since verbal content does not influence website usability.
Thus pleasing both search engines and humans is possible, indicating that managers can simultaneously adapt to the IT environment without a sacrifice in meeting online consumer needs.
6.3 Limitations and future research
This study also contains several limitations. First of all, Dutch students with similar educational backgrounds were used as participants in the experiment, which can limit the generalization of this study to other countries and to people with different educations. A wider sample of different cultures and educational backgrounds is a suggestion for future research.
Second, this study examined the effect of visual and verbal content on experimental websites where mobile phones were displayed. The effects of different presentation modes can vary across product types. For the products used in the experimental websites, online consumers can use more utilitarian motives such as ease of use and functionality, rather than hedonic motives (Flavian et al., 2010). Therefore, it would be better to examine the effect of different product presentations where both the hedonic and utilitarian benefits are mixed. This can be done with an online shop with different product categories where more visual product presentation modes are integrated, containing for instance 3D visualization tools that can improve the hedonic website experience.
Third, the participants were asked to adopt a task-oriented method (make a list of the top three mobile phones) during their visit on the website. This may have focused the attention on verbal content, while ignoring the visual content on the website (Flavian et al., 2010). Future research should also take the hedonic-oriented method into account, which means that enjoy and entertainment should be also part of their goal during the visitor’s stay on the website. This can examine the effect of visual content better, since aesthetics elements can then have a greater influence (Schlosser, 2003).
Lastly, as the study did not find any significant effect of visual content or the interaction of visual and verbal content on uncertainty and website usability, it is assumed that instead of website usability, rather website aesthetics is influenced by visual content (Sonderegger & Sauer, 2009). Since website aesthetics is positively affecting website satisfaction (Wang et al., 2010), future research should examine the mediation effect of website aesthetics in the relationship between visual content and website satisfaction.
Additionally, website aesthetics could also mediate the relationship between visual content
and website usability, because of the positive relationship between website aesthetics and
perceived usability (Sonderegger & Sauer, 2009). Thus visual content could have an indirect
effect (rather than a direct effect) on website usability. Future research could examine also
this potential indirect effect of visual content on website usability, with website aesthetics as a
mediator.
24
7. References
Amit, E., Wakslak, C., & Trope., Y. (2012). The use of visual and verbal means of communication across psychological distance. Personality and Social Psychology Bulletin, 39(1), 43-56.
Ballantine, P.W. (2005). Effects of interactivity and product information on consumer satisfaction in an online retail setting. International Journal of Retail & Distribution Management, 33(6), 461-71.
Becker, S.A., & Mottay, F.E. (2001). A global perspective on web site usability. IEEE Software, 18, 54-61.
Belanche, D., Casalo, L., & Guinaliu, M. (2012). Website usability, consumer satisfaction and the intention to use a website: the moderating effect of uncertainty. Journal of retailing and consumer services, 19(1), 124-132.
Belanger, F., & Carter, L. (2006). The digital divide: an empirical investigation. Proceedings of the Hawaii International Conference on System Sciences, 1-7.
Berman, R., & Katona, Z. (2013). The role of search engine optimization in search marketing.
Marketing Science, 32(4), 644-651.
Bhatnagar, A., Misra, S., & Rao, H. R. (2000). On risk, convenience, and Internet shopping behaviour. Communications of the ACM, 43(11), 98-105.
Bhatnagar, A., & Ghosh, S. (2004). A latent class segmentation analysis of E-Shoppers.
Journal of Business Research, 57(7), 758-767.
Bhattacherjee, A. (2001). Understanding information system continuance: an expectation- confirmation model. MIS Quarterly, 25(3), 351–370.
Birgelen, M., Ghijsen, P., & Janjaap, S. (2005). The added value of web innovation for customer satisfaction: Experiences with a barbeque catering services. Managing Service Quality, 15(6), 539-554.
Blanco, C.F., Sarasa, R.G., & Sanclemente, C.O. (2010). Effects of visual and textual
information in online product presentations: looking for the best combination in website
design. European Journal of Information Systems, 19, 668-686.
25
Casalo, V.L., Flavian, C., & Guinaliu, M. (2008). The role of satisfaction and website usability in developing customer loyalty and positive word-of-mouth in the e-banking services. International Journal of Bank Marketing, 26(6), 399-417.
Chau, P.Y.K., Au, G., & Tam, K.Y. (2000). Impact of information presentation modes on online shopping: an empirical evaluation of a broadband interactive shopping service. Journal of Organizational Computing and Electronic Commerce, 10(1), 1-22.
Chaudhuri, A. (1998). Product class effects on uncertainty: the role of emotion. International Journal of Research in Marketing, 15, 157-68.
Chen, Z., Dubinsky, & A.J. (2003). A conceptual model of perceived customer value in e- commerce. Psychology & Marketing, 20(4), 323-347.
Chiu, C.M., Lin, H.Y., Sun, S.Y., & Hsu, M.H. (2009). Understanding customers’ loyalty intentions toward online shopping: an integration of the technology acceptance model and fairness theory. Behaviour and Information Technology, 28(4), 347-60.
Cook, D., & Coupey, E. (1998). Consumer behavior and unresolved regulatory issues in electronic marketing. Journal of Business Research, 41(3), 231-8.
Corritore, C.L., Kracher, B., & Wiedenbeck, S. (2003). On-line trust: concepts, evolving themes, a model. International Journal of Human-Computer Studies, 58(6), 737-758.
Cronbach, L.J. (1970). Essentials of psychological testing. Harper and Row, New York.
Cyr, D., Head, M., Larios, H., & Pan, B. (2009). Exploring human images in website design:
a multi-method approach. MIS Quarterly, 33(3), 539-566.
Devaraj, S., Fan, M., & Kohli, R. (2002). Antecedents of B2C channel satisfaction and preference: validating e-commerce metrics. Information Systems Research, 13(3), 316-33.
DeLone, W.H. & McLean, E.R. (2003). The DeLone and McLean model of information systems success: a ten-year update. Journal of Management Information Systems, 19, 9–30.
Dimoka, A., Hong, Y., & Pavlou., P. (2012). On product uncertainty in online markets: theory and evidence. MIS Quarterly, 35, 1-32.
Dou, W., Lim, K.H., Su, C., Zhou, N., & Cui, N. (2010). Brand positioning strategy using search engine marketing. MIS Quarterly, 34(2), 261-279.
Dowling, G.r., & Staelin, R. (1994). A model of uncertainty and intended risk-handling activity. Journal of Consumer Research, 21, 119-134.
Edwards, S. M., Lee, J. K., & Ferle, C. L. (2009). Does place matter when shopping online?
perceptions of similarity and familiarity as indicators of psychological distance. Journal of
Interactive Advertising, 10(1), 35-50.
26
Eastman, J.K. & Iyer, R. (2005). The impact of cognitive age on Internet use of the elderly: an introduction to the public policy implications. International Journal of Consumer Studies, 29 (2), 125-136.
EIAAA. (2008). EEIA online shoppers. Viewed 10 April 2014. Report available at:
http://advertising.microsoft.com/sverige/wwdocs/user/sv-
se/newsandevents/events/EIAA%20Mediascope%20Europe%20and%20Online%20Shoppers
%20-%20Sweden%20Focus28.01.08.pps
Eroglu, S.A., Machleit, K.A., & Davis, L.M. (2003). Empirical testing of a model online store atmospherics and shopper responses. Psychology & Marketing, 20(2), 139-150.
Evans, M.P. (2007). Analyzing Google rankings through search engine optimization data.
Internet Research, 17(1), 21-37.
Fang, Y.H, Yat-Sen, C.C., & Wang, E.T.G. (2011). Understanding customers' satisfaction and repurchase intentions: An integration of IS success model, trust, and justice. Internet Research, 43(1), 479–503.
Flavian, C., Guinaliu, M., & Gurrea, R. (2006). The role played by perceived usability, satisfaction and consumer trust on website loyalty. International journal of information systems applications, 43(1), 1-14.
Flavian, C., Gurrea, R., & Orus, C., (2008). The effect of product presentation mode on the perceived content and continent quality of web sites. Online Information Review, 33(6), 1103- 1128.
Forsythe, S., Liu, C. L., Shannon, D., & Gardner, L. C. (2006). Development of a scale to measure the perceived benefits and risks of online shopping. Journal of Interactive Marketing, 20(2), 55-75.
Frydenberg, M., & Miko, J. (2011). Taking it to the top: a lesson in search engine optimization. Information Systems Education Journal, 9(1), 24-40.
George, C.A. (2005). Usability testing and design of a library website: an interactive approach. OCLC System & Services, 21(3), 167−180.
Gefen, D., Straub, D. (2000). The relative importance of perceived ease-of-use in is adoption:
a study of e-commerce adoption. Journal of AIS, 1(8), 1-30.
Geyskens, I., Steenkamp, J., & Kumar, N. (1999). A meta-analysis of satisfaction in marketing channel relationships. Journal of Marketing Research, 36(5), 223–238.
Glazer, R. (1991). Marketing in an information-intensive environment: strategic implications
of knowledge as an asset. Journal of Marketing, 55, 1-19.
27
