Consumer behaviour in the travel industry
“What motivates university students to book their travel itinerary online?”

25 November 2014 Management Summary
Due to the rise of information technology and especially the Internet, the tourism industry has altered tremendously. Nowadays, signifcantly more consumers, especially young, self-determing, brand loyal and price sensible consumers, prefer to book their travels online instead of visiting a travel agency. However, it is currently unknown which factors are of influence on the intention of booking a trip online.
Therefore, the following main problems of issue has been investigated: “Which factors influence the intention of students to purchase travels via online channels? And what is the mediating effect of perceived value of an online service on the relationship between price and students’ online purchase intentions?”
The literature study examined several contemporary scientific theories with respect to students’ online purchase intention, and found that preceived value could potentially serve as an mediator on the relationship between price and consumers’ online purchase intentions. Furthermore, trust, convenience, web-design, reference groups, and promotion, are found to be also of influence on the consumers, online purchase intentions. Based on these relations found in the literature study, the conceptual model (i.e. descriptive representation) and hypothesis were developed.
Based on the conceptual model a research methodology was designed. The research methodology was of quantitative nature and conducted by means of administering questionnaires. Convenience sampling was used to collect all the data. The sample size consisted of 120 studens studying at Tilburg University.
The findings indicated that trust is the only independent variable that has a significantly positive influence on the intention of students to purchase travels via online channels. Additionally, perceived value of an online service appeared to have no meadiating effect on the relationship between price and students online purchase intentions. For this reason, next to deduction, induction was used in order to find if one of the independent variables could be used as a mediator.
Within the discussion section, the researchers mainly determined whether trust could alternatively serve as a mediator on the relationships between all six independent variables and the dependent variable. The additional mediation analysis indicated that trust only has a mediating effect on both “convenience” and “reference groups”, and implies an indirect-only mediation since a x b is significant while c remains insignificant.
Table of Contents
Management Summary i
Chapter 1: Introduction 1
1.1 Induction 1
1.2 Goal of Research 2
1.3 Problem Statement 2
1.4 Research Questions 2
1.5 Relevance 2
1.5.1 Academic Relevance 2
1.5.2 Practical Relevance 3
1.6 Demarcation 3
1.6.1 Demarcation Based on Population 3
1.6.2 Demarcation Based on Variables 3
1.7 Definitions 4
1.7.1 Price 4
1.7.2 Perceived Value 4
1.7.3 Trust 4
1.7.4 Convenience 4
1.7.5 Web-design 4
1.7.6 Reference Groups 4
1.7.7 Promotion 4
1.7.8 Online Purchase Intention 5
1.8 Planning and Structure of Report 5
1.8.1 Structure 5
Chapter 2: Exploration Literature study 6
2.1 Price and Perceived Value 6
2.2 Trust 7
2.3 Convenience 7
2.4 Web-design 8
2.5 Reference Groups 9
2.6 Promotion 10
2.7 Conceptual Model 10
Chapter 3: Research Design 10
3.1 Designs and Procedure 11
3.2 Operationalization 11
3.3 Sample Strategy and Size 13
3.4 Respondent Characteristics 14
3.5 Reliability Analysis 14
Chapter 4: Data analysis 15
4.1 Descriptive Statistics 15
4.1.1 Descriptive Statistics on Item and Variable Level 15
4.1.2 Testing for Non-Normality 16
4.1.3 Correlation Matrix (Independent Variables on Dependent Variable) 16
4.2 Regression Analysis 17
4.2.1 Mediation Analysis 17
4.2.2 Statistical Validity 18
4.2.3 Face Validity 18
4.3.4 Hypothesis Testing 18
4.3.5 Durbin-Watson 20
4.3.6 Multicollinearity: Variance Inflation Factor 20
4.4 Conclusions 21
Chapter 5: Policy advice 22
5.1 Discussion 23
5.2 Limitations 23
5.3 Recommendations 24
5.3.1 Practical Recommendations 25
5.3.2 Recommendations for Future Research 25
Chapter 6: Bibliography 27
Chapter 7: Appendices 35
Appendix 1 35
Appendix 3 36
Appendix 4 66
Appendix 5 90

Chapter 1: Introduction
The first chapter covers the induction that describes the motives and background of this research. Furthermore, the goal of the research, problem statement, research questions, relevance, demarcation, definitions, planning and structure of the report will be addressed.
1.1 Induction
The tourism industry has been greatly impacted by the rise of information technology and especially the Internet (Ho & Lee, 2007). ING proclaims that the preference of consumers to rather explore online options has a substantial effect (ING, 2013). The significant change in proportion of online bookings underlines ING’s claim, as in 2003 only 25 percent accounted for online bookings, whereas in 2012 this number had skyrocketed to 76 percent (ING, 2013). Especially for young, self-determining, brand disloyal and price sensible consumers, the web is a preferred option to book travels instead of visiting a travel agency (Weber, 2013). However, it is currently unknown which factors influence the purchase intentions with regards to booking trips.
When it concerns booking trips online, Kim, Woo, & Han (2006) state that consumers place price transparency and low prices on a high pedestal. Empirical research has concluded that there is a positive relationship between price and perceived value. It is implied that high quality is associated with higher prices (Leavitt, 1954; Tull, Boring, & Gonsior; Gabor & Granger, 1966; McConnell, 1989; Olson, 1977). Moreover, Quelch and Klein (1996) note that: “Trust is a critical factor in stimulating purchases over the Internet, especially at the early stage of commercial development”. In the online market, researchers demonstrated through empirical studies that trust affects purchase intentions (Sirkka L. Jarvenpaa, 2000; Schlosser, White , & Lloyd , 2005; Venkatesh Shankar, 2009; Yoon, 2002). Jeong & Lambert (2001), refer to a study conducted by White and Manning (1998) who identified website attributes that contribute to a higher likelihood of making online purchases; amongst these attributes, a website ought to be useful, clear, attractive and informative. Additionally, consumers are inclined to engage in online shopping due to the efficient (de Ruyter, Dellaert, & TP y Monsuwe, 2004) and timely-manner (de Ruyter, Dellaert, & TP y Monsuwe, 2004; Szymanski & Hise, 2000) in which this can be done with minimum irritation (de Ruyter, Dellaert, & TP y Monsuwe, 2004). Next to this, both on and offline Word-of-mouth (WOM) are of vital importance for the buying decision (Hennig-Thurau, Gwinner, Walsh, & Gremler, 2004). Positive WOM is one of the factors that allegedly positively influences purchase intentions (Maxhamm III, 2001). Research concerning web-based hedonic and utilitarian shopping objectives found that providing consumers with utilitarian benefits instead of hedonic value may boost online purchases (Bridges & Florsheim, 2008), thus suggesting the importance of providing utilitarian benefits such as discounts (monetary incentives), enhanced shopping convenience (nonmonetary incentives), etc.
1.2 Goal of Research
This paper will identify the main influential factors of students to book their travel itinerary online. The outcome will provide insights for travel agencies on how to improve their services targeted towards students.
1.3 Problem Statement
The percentage that accounts for online travel bookings increased from 25 percent in 2003 to 76 percent in 2012 (ING, 2013). Thus undeniably inferring a trend in online bookings especially amongst young consumers (Weber, 2013). Therefore, it is of importance to travel agencies to consider the motives for students to book their trips online, which leads to the following problem statement: “what motivates university students to book their travel itinerary online?”
1.4 Research Questions What factors influence the intention of students to purchase travel via online channels? What is the mediating effect of perceived value of an online service on the relationship between price and students’ online travel purchase intentions?”

1.5 Relevance
1.5.1 Academic Relevance
With respect to the already examined literature, this paper attempts to determine the main factors that influence students’ online travel purchase intentions. Additionally, this paper aims to apply the general model of perceived value and purchase intention to gain a deeper understanding of the perceived value of price and its mediating effect on online travel purchase intention. The model stresses that perceived value is influenced by the investment required to obtain the product and its likely beneficial return (Monroe, 1979). Within this paper, the differentiation point is next to conducting research on pertinent factors, also on taking into account how and/or whether perceived (price) value affects students’ online travel purchase intentions. Moreover, students are taken as a focal point because their behaviour tends to differ significantly in comparison with the entire population.
1.5.2 Practical Relevance
Travel agencies are able to pinpoint how to improve its services and in which areas targeted at students, as the information in this paper aims to present the pertinent factors considered by students when purchasing online travel.
Depicting these main influential factors could e.g. lead to website improvement. Maintaining a user-friendly website (Kim, Woo, & Han, 2006), as well as the overall image (Yoon, 2002) is of importance, as oftentimes websites are used to get acquainted with products and services (van der Heijden, Verhagen, & Creemers, 2003).
1.6 Demarcation
1.6.1 Demarcation Based on Population
For this research, the sample population is limited to university students only. More precisely, the focus is on Dutch students from Tilburg University between the ages of 18 to 25. This group is distinctive and well versed in the field of technology (Prensky, 2001). Students nowadays represent the first generation that grew up with today’s technology. They have been surrounded by digital equipment their entire lives which has been deeply integrated into their lives (Prensky, 2001). Therefore, it is of interest to determine how these technologies, such as the Internet, influence students’ travel purchase intentions.
Practical considerations to focus on Dutch students studying at Tilburg University owes to the time constraint inhibiting the researchers to broaden the research scope beyond the Netherlands. Another reason is that the researchers are based in the Netherlands and are students at Tilburg University.
In this research, travel agencies are categorised as follows: All physical travel agencies (brick shops) and All online travel websites.
1.6.2 Demarcation Based on Variables
For this research the variables are subdivided into six different independent variables and a mediator; (X1) trust, (X2) convenience, (X3) price, (X4) website-design, (X5) reference groups, (X6) promotion and perceived value (M1). These variables are selected based on preliminary research that claims that forces in one’s environment influence buying behaviour (Kotler, Armstrong, Wong, & Saunders, 2008). Additionally, the independent variables will be tested to determine its influence on the dependent variable, which is; (Y1) students’ online purchase intention.
1.7 Definitions
1.7.1 Price
When the law of one price is violated, price dispersion occurs (Baylis & Perloff, 2002). Price dispersion is referred to as the variation of prices among different retailers for homogenous products or services (Stigler, 1961).
1.7.2 Perceived Value
Monroe and Dodds (1985) refer to perceived value as: “a trade off between perceived quality and sacrifice”.
1.7.3 Trust
Lee and Turban (2001) defined consumer trust in Internet shopping as “the willingness of a consumer to be vulnerable to the actions of an Internet merchant in an Internet transaction, based on the expectation that the Internet merchant will behave in certain agreeable ways, irrespective of the ability of the consumers to monitor or control that Internet merchant”.
1.7.4 Convenience
Insinuates that consumers engage in online shopping due to the efficient (de Ruyter, Dellaert, & TP y Monsuwe, 2004) and timely-manner (de Ruyter, Dellaert, & TP y Monsuwe, 2004; Szymanski & Hise, 2000) in which this can be done with minimum irritation (de Ruyter, Dellaert, & TP y Monsuwe, 2004).
1.7.5 Web-design
Web-design is referred to as the elements of a website in terms of features, functionality, information content, and graphic style (Martín & Camarero, 2008; Song & Zahedi, 2005).
1.7.6 Reference Groups
Reference groups are groups that have a direct or indirect influence on persons’ attitudes or behaviour (Kotler, Armstrong, Wong, & Saunders, 2008).
1.7.7 Promotion
Several researchers define sales promotions as "short-term incentives to encourage purchase or sale of a product or service" (Kotler, Armstrong, Wong, & John, 2008) or, similarly, "as temporary and tangible monetary and nonmonetary incentives intended to have a direct impact on consumer behaviour" (Chandon, Wansink, & Laurent, 2000).
1.7.8 Online Purchase Intention
Online purchase intention can be defined as a determent of the strength of a consumer it’s intention to carry out a specified purchasing behaviour via the Internet (Salisbury, Pearson, Pearson, & Miller , 2001).
1.8 Planning and Structure of Report
1.8.1 Structure
This research report is divided into six chapters. The first chapter mainly describes the integral parts of the research proposal; the purpose of the research, research objectives, problem statement, research questions, academic and practical relevance, scope of the study, key definitions, structure, and planning of the research are all discussed in this chapter. Chapter two consists of a thorough literature review concerning the problem of interest, including a conceptual model that reflects the scientific theories and related hypotheses. Chapter three builds on the previous chapters and provides details on the research design (procedures, material, sample and respondents etc.), data collection and analyses (descriptive statistics, testing hypotheses etc.). Chapter four reports on the main results. The fifth chapter presents the final discussion, conclusions, recommendations, and the limitations of this study. In Appendix 1, figure 1.1 the planning can be found. Chapter 2: Exploration Literature study
This chapter examines literature that is believed to be relevant with respect to students’ online purchase intention. Additionally, in this chapter the conceptual model, which is based on the variables identified during the literature review, is introduced with accompanying hypotheses.
2.1 Price and Perceived Value
It is believed that e-commerce markets will eventually become competitive or create price dispersion due to variations in products or services (Brunjolsson & Bakos, 1999). Therefore, consumer motivation to search for price information is high (Urbany, 1986). Bakos (1991) argues that higher market transparency in online markets would lead to higher price competition. Moreover, economist Hall Varian (1980) discussed that price dispersion could be an intentional advertising technique to inspire consumers to discover alternatives of a product or service.
Scitovszky (1944) insinuates that consumers may use price not solely as a cost indicator, but also as an index of product quality. There are a number of empirical studies on this proposed ‘price-perceived value correlation’ (Monroe & Chapman, 1987). Monroe (1979) suggests that a consumer’s value perspective represents a psychological trade-off between the benefits or quality they perceive and the perceived sacrifices in the transaction. Empirical research has concluded that there is a positive link between price and perceived value, because higher quality is associated with higher price (Leavitt, 1954; Tull, Boring, & Gonsior; Gabor & Granger, 1966; McConnell, 1989; Olson, 1977). Dodds and Monroe (1985) suggest that the relationship model of price and perceived value of quality are an important factor in consumers’ purchase intention and state that buyers will purchase a product with high-perceived value. In addition, Wood & Scheer (1996) suggest and continue to use the price and perceived value model as a mediator. Furthermore, a shopper pondering a purchase is likely to evaluate that value=price/quality (Yook, Oh, Song, Kim, & Kim, 2014). Moreover, when it concerns booking trips online, consumers place price transparency and low prices on a high pedestal (Kim, Kim, & Han, 2007).
H1: Higher prices of online travel imply a higher perceived value, which implies a greater intention for students to purchase travel online.
H2: Low prices of online travel imply a higher intention for students to purchase travel online.
2.2 Trust
Quelch and Klein note the following: “Trust is a critical factor in stimulating purchases over the Internet, especially at the early stage of commercial development” (Klein, 1996). Ang and Lee (2000) agree with this statement as they proclaim that: ‘if the web site does not lead the consumer to believe that the merchant is trustworthy, no purchase decision will result.’ In the online market, researchers made evident through empirical studies that trust affects purchase intentions (Sirkka L. Jarvenpaa, 2000; Schlosser, White , & Lloyd , 2005; Venkatesh Shankar, 2009; Yoon, 2002).
Trust is generated once the consumer believes that the supplier has both the motivation and the ability to dependable provide products and services according to the quality expected by the consumer (Schuurman, Mayer, & Davis, 2007). In e-commerce, it is more difficult to obtain trust from consumers compared to a conventional merchant. In a relationship, for example, between the customer and supplier, trust is the critical factor in which the trustor does not have direct control over the actions of the trustee (R.J. Mayer, 1995, pp. 709-734). Donney and Cannon (1997) state that consumers build trust when the seller invests in devoted assets for the relationship, and when there are regular interactions between the consumer and representatives of the supplier’s organization. In Internet commerce, direct contact between the consumer and sales representatives hardly occurs. Principally, consumers depend on an interpersonal electronic platform to act on their behalf (M.J. Culnan, 1998). According to Keen (1997), the lack of consumer trust is the most significant long-term obstacle for achieving the potential of Internet marketing to consumers.
H3: The higher the level of trust, the greater students’ buying intention to purchase travel online.
2.3 Convenience
Consumers are inclined to engage in online shopping due to the efficient (de Ruyter, Dellaert, & TP y Monsuwe, 2004) and timely-manner (de Ruyter, Dellaert, & TP y Monsuwe, 2004; Szymanski & Hise, 2000) in which this can be done with minimum irritation (de Ruyter, Dellaert, & TP y Monsuwe, 2004).
Due the availability and accessibility of an overflow of various types of information, one can easily get acquainted with products (Placeholder1; van der Heijden, Verhagen, & Creemers, 2003) and participate in price comparisons for the lowest fares (Placeholder1). Wicks and Schuett (1991) discuss the importance of information about tourism products in sales media. The writers elaborate that the obtained information is relied on heavily and mention that consumers of tourism products engage in comparisons to determine who will receive the business, thus also implying that the provided information assists in the decision-making.
Consequently, it is pertinent that online travel agencies equip their websites with appropriate content to predict purchase intentions and ensure efficiency (Jeong & Lambert, 2001). Hence, websites are subject to expectations such as the ability to bundle arrangements and to provide additional and informative content relevant for (a) specific audience(s) (Kaynama & Black, 2000). For this reason, the ease of use of a website must be considered and can be defined as the perception that the technology is free of effort (de Ruyter, Dellaert, & TP y Monsuwe, 2004). Subsequently, the ease of use has a positive influence on the evaluation of the information whether to use this for making purchase decisions (Jeong & Lambert, 2001). Research by Jeong, Oh, & Gregoire (2003) made evident that information satisfaction plays a crucial role on purchase intention – an integral factor that contributes to information satisfaction is website quality (Jeong, Oh, & Gregoire, 2003; Udo, Bagchi, & Kirs, 2010).
H4: The convenience associated with online shopping leads to greater purchase intentions by students to book trips online.
2.4 Web-design
The tourism industry has been greatly impacted by the rise of information technology, especially the Internet (Ho & Lee, 2007). It has influenced the way information is sought by travellers and how travel arrangements can be done (Li & Law, 2007; Jensen, 2012). Studies call great attention to the ease of use of a website to affect the way in which a website is regarded in terms of quality and trustworthiness (Martín & Camarero, 2008; Lee & Lin, 2005). Accordingly, Kuan, Bock and Vathanophas (2008) bring forward that consumers abandon websites due to low quality. Jeong & Lambert (2001), refer to a study conducted by White and Manning (1998) who identify website attributes that contribute to a higher likelihood of making online purchases; amongst these attributes, a website ought to be useful, clear, attractive and informative. According to Szymanski & Hise (2000), customers’ perception of a website’s design may affect satisfaction, which in turn influences purchase intention (Taylor & Baker, 1994; LaBarbera & Mazursky, 1983; Udo, Bagchi, & Kirs, 2010). Hausman and Siekpe (2009) argue that disorganised, malfunctioning and visually inadequate websites would diminish the likelihood of returning customers and purchase intentions (Kuan, Bock, & Vathanophas, 2008).
A plethora of researchers have identified attributes to which online customers attach great importance (Ho & Lee, 2007; Martín & Camarero, 2008; Lee & Lin, 2005; Song & Zahedi, 2005; Kuan, Bock, & Vathanophas, 2008). Albeit somewhat fragmented throughout most research, Lee & Kozar (2006) researched the factors to online success for online travel customers. Their list ranks, in respective order, relevant information, navigability, price savings, security and tele-presence as top five influential factors for the success of a travel website. Other attributes for online customers according to their research are reputation, reliability, understandability, responsiveness, and response time.
H5: Easier to use and visually appealing websites yield greater purchasing intentions by students for booking trips online.
2.5 Reference Groups
Word-of-mouth (WOM), in its traditional (offline) sense and contemporary Electronic-WOM (E-WOM) are both of high importance for influencing the buying decision of consumers (Hennig-Thurau, Gwinner, Walsh, & Gremler, 2004). These researchers also state that word-of-mouth provides consumers with information about the quality of firms, products and/or services.
The traditional WOM made consumers seek advice from friends, relatives and other third parties (King, Racherla, & Bush, 2014). This has changed with the advent of the Internet that made it possible for consumers to interact with one another via computer-mediated conversations (King, Racherla, & Bush, 2014). Therefore, E-WOM can be defined as “any positive or negative statement made by potential, actual, or former customer about a product or company, which is made available to a multitude of people and institutions via the Internet” (Bearden & Etzel, 1982). Maxhamm (2001) states that the level of satisfaction influences positive WOM. The same study also puts forwards that positive WOM is one of the factors that positively influences purchase intentions (Maxhamm III, 2001).
Moreover, Bearden and Etzel (1982) categorised four types of products that might have a high/low influence on reference groups. In respective order, the list is as follows; public luxuries (high influence of reference groups), private luxuries (high influence of reference groups), public necessities (low influence of reference groups) and private necessities (low influence of reference groups) (Bearden & Etzel, 1982).
H6: High sensitivity for social influence (reference groups) will influence the intention of students to purchase travel online.
2.6 Promotion
According to Chandon, Wansink, & Laurent (2000), it is not only the monetary incentives (e.g. discounts) that make consumers respond to sales promotions but also the nonmonetary incentives (e.g. bonus packs).
Both incentives metrics provide consumers with hedonic benefits (opportunities for value expression, exploration, and entertainment) as well as utilitarian benefits (discounts, higher product quality, and enhanced shopping convenience) (Chandon, Wansink, & Laurent, 2000). These incentives subsequently influence consumers’ purchase intentions (Mela, Jedidi, & Bowman, 1998).
With reference to the web-based (i.e. online) domain, consumers may also search for hedonic and/or utilitarian benefits. However, research concerning web-based hedonic and utilitarian shopping objectives found that providing consumers with utilitarian benefits instead of hedonic value might boost online purchases (Bridges & Florsheim, 2008). Suggesting the importance of providing the aforementioned utilitarian benefits such as discounts (monetary incentives), enhanced shopping convenience (nonmonetary incentives), etc.
H7: The higher the level of consumers with utilitarian benefits, the greater students’ buying intention to purchase travel online.
2.7 Conceptual Model
Please find the conceptual model (i.e. descriptive representation) that depicts the relationships between the relevant independent (i.e. predictor) variables, mediator and dependent (i.e. outcome) variable below. This conceptual model is subjected to the iterative process of evaluating the listed independent variables based on literature study. In addition, the conceptual model indicates positive (+) or negative (-) relations between the variables Chapter 3: Research Design
This chapter will elaborate on the research methodology utilized. An intricate description on the design and research procedures will be touched upon at first. Subsequently, an elucidation will clarify the operationalization of the selected variables found in the conceptual model. Lastly, sample- size and technique are made intelligible.
3.1 Designs and Procedure
Collection of data was done through the dissemination of questionnaires tailored to the set hypotheses. This quantitative method of data collection accommodates the gathering of information profoundly from individuals. More importantly, accumulating data through questionnaires is considered to be a rather inexpensive tool that enables quick results.
The questionnaire has been widely spread by means of a traditional paper version and a contemporary digital version. The former was mostly used to approach students who were present in open spaces of Tilburg University, such as the Library and waiting area of the Student Desk. The latter was distributed amongst peer-students of the researchers by email and social networks. This digital version was a PDF-file – a mere copy of the paper version – that was equipped with clickable 7-point likert scales and other selectable options. Since both designs are identical, a single copy of the questionnaire can be found in Appendix 3, table 3.1.
The data collection timeframe, as shown in the Gantt-chart (Appendix 1, figure 1.1) specified that all data would be gathered at a single point in time. To be more precise, a 7-day period was devoted for gathering data. Stemming from this particular method, the study can be defined as cross-sectional.
3.2 Operationalization
As alluded in preceding paragraphs, the questionnaire’s intent is to examine the presence of relationships between the independent variables, mediator and the dependent variable; intention of students to purchase travel online as depicted in the conceptual model. Various existing measurement scales have been put in place to ensure reliable and valid results. Furthermore, the mediating variable (perceived value) was identically incorporated into the survey. Predominantly, 7-point likert scales were used to ascertain the relationships of the variables.
The following paragraphs will explain the measurements per variable and the numerous scales that were used as foundation of the survey. Appendix 3, table 3.2, contains a detailed overview per variable and scale. Subsequent to the clarifications of measurements, a table will present all Cronbach’s Alpha’s per construct. Also some wordings of construct items had to be changed in order to fully match the researchers’ intentions; these can also be found in Appendix 3, table 3.2.
Price for instance was measured by earlier work of Gotlieb and Sarel (1991). Albeit, one item has a certain degree of similarity with a measure created by Peterson and Wilson (1985). Notwithstanding, the four item 7-point likert scale was exerted because it would indicate a respondent’s belief of price-quality relationships. Furthermore, a second (three item) scale was used as additional foundation for the variable price. This scale regards pricing issues for air travel and was originally developed by Bruning, Kovacic, and Oberdick (1985). It entails the concern of respondents to high or low priced travel. This is insightful as it provides information on price sensitivity of University students.
To determine the presence of any kind of relationship in regards to trust, the scale “Consumers trust in Internet shopping” was used. Its origins, derives from a study by Chueng and Lee (2000), contains four items and is measured by means of a 7-point likert scale. These four items enable the drawing of conclusions on students’ trust in Internet shopping and its influences on online travel purchase intentions.
The variable convenience has multiple scales to determine its significance. The first (two item, 3-point likert) scale used regards whether one believes to choose the first product/service or to shop around. Secondly, the (six item, 5-point likert) scale original to Saegert, Hoover and Hilger (1985) entails the identification of store features that consumers regard important. Thirdly, Szymanski and Hise (2000) developed a three item, 7-point likert scale to measure the extent to which one thinks that Internet stores are easier to shop at and are time-saving compared to traditional retail stores. Lastly, Donthu and Gilliland (1996) have likely constructed a three item, 5-point likert scale that measures the degree to which a person expresses a desire for convenience and simplicity, particularly as it relates to gathering information about products and shopping activity. It was decided to standardize all scales to 7-point for consistency reasons.
The fourth independent variable; web-design, encompasses a (four item) likert scale by Lynch, Kent, and Srinivasan (2001), and a three item, 7-point likert scale by Szymanski and Hise (2000). The former provides insight into the attitude of a person towards the helpfulness of a website and its ease of use. Whereas the former provides insight into the influence of Internet store fronts towards a person’s perception of a website.
For reference groups, the fifth independent variable tested, a four item, 5-point likert scale by Moschis (1981) was utilized. This particular scale investigates the validity of the cognitive development approach to socialization to predict a wider variety of consumer-related cognitions. Imperative as students belong to a classification of people who might be more socially pressured than others. Pertaining the 5-point likert scale, the researchers have standardized it into a 7-point likert scale.
The last independent variable examined is promotion and especially consumer’s tendency to purchase the brands that are offered on sale. This particular 7-point likert scale originally belongs to Lichtenstein, Ridgway, and Netemeyer (1993).
The mediating variable perceived value presented in the conceptual model is subject to a scale most likely original to Ailawadi, Neslin and Gedenk (2001). The scale, composed out of three, 5-point likert scales, indicates that a high perceived value implies a greater intention for students to purchase travel online. Also this 5-point likert scale has been modified to a 7-point likert scale.
The dependent variable of this study the ‘intention of students to purchase travel online’, will be tested by a three item, 11-point summation scale of probabilities. This scale measures the purchase intention, in this case the purchase intention amongst students.
3.3 Sample Strategy and Size
The target population for this research regards Dutch students within the higher education system. However, due the whys and wherefores explained in the demarcation, only Dutch students from the Tilburg University have been selected to be included within the survey, especially between the age of 18-25 as this group is thought to be distinctive and well versed with technology (Prensky, 2001). This enabled the researchers to adopt a sampling frame of e-mail addresses (for digital surveys) and official University Cards (for traditional paper surveys). A non-probability sampling strategy has been selected since representativeness for the entire population cannot be met, since only students from Tilburg University are included.
The sampling procedure entails convenience sampling and is by Sekeran and Bougie (2013, p. 252) referred to as “the collection of information from members of the population who are conveniently available to provide it”. This procedure is a quick and inexpensive way to obtain information.
The sample size is determined by the rule of thumb, which infers that a sample size larger than 30 and less than 500 is appropriate (Sekaran & Bougie, 2013). Since the research is of multivariate nature and the rule of thumb indicates that for each variable at least ten or more respondents are preferred in order to conduct a multiple regression analysis (Sekaran & Bougie, 2013), a minimum amount of 80 respondents is required to ensure the information gathered is reliable and valid.
3.4 Respondent Characteristics
The number of respondents has surpassed the set minimum amount of 80 as explained in the preceding section. Thereupon distribution, 120 respondents, of which all were students, took part in the survey. Examining the respondents’ level of education revealed that 47.5% of all respondents are Pre-master students, 33.33% are Bachelor students, and 19.17% are Master students. Appendix 3, figure 3.1, depicts the distinction of male (55%) and female (45%) respondents, of which all (100%) have, at least once, booked a trip online. Subsequent to investigating the age of respondents, it became evident that the average respondent was 22.1 years old with a standard deviation of 1.7604.
3.5 Reliability Analysis
For each construct in this study the Cronbach’s Alpha’s have been checked. The Cronbach’s Alpha’s of each construct are greater than the critical value of 0.60 as can be seen in table 3.1 below and from the tables in Appendix 3, table 3.2. Values above the critical value indicate that there is sufficient internal consistency within the items of each measurement scale (Sekaran & Bougie, 2013).
Variable Cronbach’s Alpha standardized items* Cronbach’s Alpha N of Items
Price 0.85, 0.78 0.695 6
Price M No examination 0.844 3
Trust 0.86 0.843 4
Convenience 0.61, 0.73, 0.69, 0.71 0.691 15
Web design 0.78, 0.85, 0.72 0.801 7
Reference groups 0.85, 0.74 0.867 4
Promotion 0.88, 0.86 0.862 6
Y 0.90, 0.92 0.753 3
Table 3.1 | Cronbach’s alpha
*are respective in order to the operationalization section.
Chapter 4: Data Analysis
This chapter analyses the preliminairy data collected by means of administering questionnaires. First of all, the descriptive statistics quantitatively describe the main features of the data collection. Secondly, all variables are tested for non-normality. Furthermore, this chapter covers the correlation matrix, statistical vadility, face vadility, hypothesis testing, Durbin-Watson, mullitocllinearity and ends with the main conclusions.
4.1 Descriptive Statistics
4.1.1 Descriptive Statistics on Item and Variable Level
In this subsection, all variables in the conceptual model are further explained by taking into account the number of items that form each and every construct, a measure of central tendency namely the mean (i.e. average score), and several measures of dispersion including the standard deviation, range, and minimum and maximum scores. Furthermore, a reproduction of the conceptual model with the means of all variables can be found in the Appendix (figure 4.1 and tables 4.1 till 4.9).
As aforementioned, all independent- and the mediating variable have been measured by a 7-point Likert scale (e.g. 1 = very unimportant to 7 = very important etc.) (mean is 3.5). The dependent variable, however, has been measured with 11 answer possibilities (mean is 5.5).
Variable Scale Type Items Mean of Scale Mean by Data
Price 7-point Likert 6 items 3.5 4.3569
Trust 7-point Likert 4 items 3.5 4.8256
Convenience 7-point Likert 15 items 3.5 5.2589
Web-design 7-point Likert 7 items 3.5 5.3313
Reference groups 7-point Likert 4 items 3.5 4.2625
Promotion 7-point Likert 6 items 3.5 4.7928
Perceived value 7-point Likert 3 items 3.5 4.3583
Purchase Intention 11-point Likert scale 3-items 5.5 6.7472
Table 4.1 | Descriptive Statistics

The table above shows that all mean values of all variables (mean by data) are higher than the means of the given scale (mean of scale), albeit some score slightly higher than average and other more substantially.
The standard deviation, which gives an index of the spread of a distribution or the variability in the collected data, the range, and the minimum and maximum scores of each and every item have been checked for any discrepancies. Just a few standard deviations are ≤ 1, but at the same time, also close to 1. However, it can be stated that a considerable portion of all standard deviations are ≥ 1, which means that there is sufficient variability in the data. For more profound details on all items described above, please go to the original tables in Appendix 4, figure 4.1 and tables 4.1 till 4.9.
4.1.2 Testing for Non-Normality
Usually, p-values < 0.05 are wanted as they enable the researcher(s) to reject the null hypothesis (i.e. there is no difference) and conclude that there is a difference. However, when testing for normality it is the other way around; p-values > 0.05 are wanted as they allow the researcher(s) to state that there is no difference. In other words, when the p-values are > 0.05 it can be concluded that the data is not different compared to normally distributed data.
In order to test whether or not the mean variables incorporated in the conceptual model are normally distributed, the statistical Kolmogorov-Smirnov method has been utilised. Unfortunately, the results of the Kolmogorov-Smirnov tests show for all variables p-values < 0.05 (see Appendix 4, table 4.10 - table 4.17). Although the variable price has the highest p-value (p-value is 0.036) of all variables, it is still < 0.05 which means that also this variable is not normally distributed. Histograms (incl. the best fitting normal curve) are provided for all variables to visualise the above and improve understanding (see Appendix 4, figure 4.2 - 4.9).
Taking into account the fact that all variables are not normally distributed, one could suggest that no further tests should be undertaken. However, Thomas, Paula, Scott, and Lu (2002) investigated the importance of the normality assumption and concluded that "it is widely but incorrectly believed that the t-test and linear regression are valid only for normally distributed outcomes". For this reason, it is allowed to move on and undertake the other statistical tests (incl. multiple linear regression analysis) in order to investigate the factors affecting the intention of students to purchase travel online, even though the "normality requirement" has not been satisfied.
4.1.3 Correlation Matrix (Independent Variables on Dependent Variable)
As can be seen in the correlation matrix (Appendix 4, table 4.18) there has not been any outstanding result detected, i.e. values above 0.60. In addition, the findings of the correlation matrix are all in line with the expectations. Which was that the X variables would have a positive influence on the Y.
Variable Type Pearson Correlation Sig. 2-tailed Desired Alpha
Price Independent 0.110 0.231