How do research shoppers differ? : an analysis on the basis of psychographics, benefits sought, demographics and product categories for showroomers, webroomers and mobile-assisted shoppers

(1)

How do research shoppers differ?

An analysis on the basis of psychographics, benefits sought,

demographics and product categories for showroomers, webroomers

and mobile-assisted shoppers

Course:

Master’s Thesis

Program:

MSc Business Administration – Marketing Track

Student:

Phanwadee Seetanet (10656626)

Thesis Supervisor:

Prof. Umut Konuş

Submission Date:

July 22, 2016

(2)

How do research shoppers differ? 1

STATEMENT OF ORIGINALITY

This document is written by Phanwadee Seetanet who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

ABSTRACT

As shopping channels have proliferated, consumer purchasing pathway is also becoming increasingly complex and fragmented, creating both challenges and opportunities for firms. One of the most important challenges in the multichannel environment is the “research shopping phenomenon”– consumers’ propensity to search for products on one channel, but buy on another channel. It is extremely crucial that marketers gain a deeper understanding of the topic as this phenomenon is two-sided. On the one hand, research shopping behavior could cause customers to switch to another retailer during the shopping process, resulting in the loss of customers and potential revenue. On the other hand, if the research shopping behavior is well understood and managed effectively, it could potentially create an opportunity for cross-selling which enhances customer satisfaction, revenue and long-term customer loyalty. This study investigates different psychographics, expected benefits, and demographics that characterize different types of research shopping behavior: showrooming (search offline, buy online), webrooming (search online, buy offline) and mobile-assisted shopping (simultaneous search on mobile devices while in-stores, but purchase either offline or online), integrating multiple product categories namely clothing and apparel, consumer electronics and personal care items. The findings yield relevant academic and managerial implications that help

marketers understand and manage their customers more effectively, leading to cross-channel synergies which are beneficial for businesses.

(4)

1. INTRODUCTION

The rapid technological advances have resulted in the proliferation of channels through which customers can interact with businesses. Nowadays, consumers have countless ways to shop and communicate with firms such as the Internet, stores, mobile shopping, catalogs and call centers. Consequently, customers no longer rely only on one channel in their shopping process but have become multichannel shoppers who are engaged in various channels across different phases in the shopping journey, including information search, product purchase and after-sale services (Balasubramanian, Raghunathan, and Mahajan 2005; Neslin et al. 2006; Konus, Verhoef and Neslin 2008). This multichannel environment represents both opportunities and challenges for firms (Neslin et al., 2006). One of the most important challenges is the “research shopping phenomenon” – the behavior of searching for products on one channel, but buying on another channel (Verhoef, Neslin and Vroomen 2007). For instance, a study by Harris Poll (2013) shows that 40% of US adults look for products in-store and purchase online for better prices.

Research shopping can be beneficial to a firm as long as customers search and purchase across channels within the same firm. As suggested by Kumar and Venkatesan (2005),

customers who use multiple channels from the same firm tend to spend more than single-channel buyers, leading to increased revenue and higher share of wallet for firms. Nevertheless, research shoppers could pose a serious threat to a firm if customers search at one firm but purchase at another, causing the firm to lose customers and potential revenue during the shopping process (Nunes & Cespedes, 2003; Verhoef et al., 2007). For this reason, the ability of

(7)

firms to understand the varying dynamics and motivations behind consumers’ research shopping behavior is extremely crucial in devising an effective cross-channel strategy.

From the conventional point of view, research shoppers can be categorized into two main forms which are offline-to-online and online-to-offline research shoppers. “Offline-to-online research shopping” or “showrooming” refers to the behavior of customers who check out products in physical stores before making the purchase online (Luo et al., 2014).

Conversely, “online-to-offline research shopping” or “webrooming” (also called “reverse

showrooming”) occurs when customers research products online but complete the purchase in-store. Although showrooming used to be a critical concern for traditional brick-and mortar retailers in the past due to the loss of potential revenue and customers to online competitors, recent studies have shown that webrooming is becoming more prevalent among research shoppers, providing retailers with an opportunity to attract customers back in stores (Taylor & Hunter, 2015; Merchant Warehouse, 2014; Nielsen, 2014). For example, in the Irish car

insurance industry, 66% of consumers use the Internet for research while only 32% purchase online (Consumer Barometer, 2014).

Further complicating the matter is the emergence of a new type of research shoppers who browse products in-store while simultaneously using their smartphones or tablets to assist in their purchase decision (e.g., price comparison and product information search), but then purchase either online or offline (Quint, Rogers & Ferguson, 2013). This so-called “mobile-assisted-shoppers” or “M-Shoppers” raised even more concerns for firms to understand and cope with these evolving and complex dynamics of shoppers in an effective manner.

(8)

In response, this study aims to provide a better understanding of the aforementioned three types of research shoppers (showroomers, webroomers and mobile-assisted shoppers) by answering the crucial question: “whether and how different types of research shopping

behavior are driven by psychographic, benefits sought and demographic characteristics across different product categories?”. More specifically, the present study is intended to achieve the following objectives:

1. identify and compare different psychographic, expected benefits and demographic characteristics that drive showrooming, webrooming and mobile-assisted shopping behavior;

2. investigate whether and how the three types of research shopping behavior might differ across product categories, namely apparel, consumer electronics and personal care.

In pursuing these objectives, it is expected that this comparative study will provide multiple theoretical and managerial contributions. From the academic perspective, the present study will enrich the existing literature in the multichannel marketing field by deepening our understanding about the psychographic, benefits sought and demographic drivers of three types of research shopping behavior, incorporating multiple product categories. In addition, inclusion of mobile-assisted shopping as a newly emerged type of research shopping behavior is expected to result in relevant and interesting implications for researchers and practitioners. Moreover, the research yields important managerial relevance as it provides marketers with useful insights regarding the drivers of research shopping behavior. This helps marketers allocate their resources more effectively across different phases in the shopping and decision journey, thereby creating a more appropriate multichannel strategy and improving

(9)

cross-How do research shoppers differ? 8

channel synergies. In particular, managers can tailor specific strategies once understanding which elements should be emphasized for particular channels and types of research shoppers. Managing research shoppers strategically will lead firms to respond to specific needs of

customers in a proper manner, which consequently enhances customer satisfaction, profitability and long-term customer loyalty.

In order to achieve the stated objectives, the research paper is structured as follows. Firstly, prior literature regarding research shopping behavior will be reviewed, followed by a brief discussion of expected academic and managerial contributions, and the identification of the research gaps and research question. Secondly, the conceptual framework along with the psychographic, benefits sought and demographic predictor variables that are hypothesized to characterize different types of research shopping behavior will be visualized and elaborated. Thirdly, the research strategies and methodology will be explained. Subsequently, the results and analysis of the findings will be presented. The research ends with detailed discussion of results, managerial implications, limitations and suggestions for future research. Finally, a list of references and appendices are also provided at the end of this thesis document.

(10)

2. LITERATURE REVIEW

This chapter provides a detailed review of the important concepts that are used in this study, and discusses previous research associated with the topic and the relevant key

constructs. Firstly, the background regarding multichannel marketing and multichannel shoppers is introduced. Secondly, the research shopping phenomenon, the two conventional types of research shopping behavior and the rise of webrooming, are explained. Next, the newly emerged type of research shoppers, mobile-assisted shoppers are elaborated. Lastly, prior studies regarding consumer decision journey in the multichannel context and research shopping behavior are discussed, followed by an identification of the research gaps, the research question and expected theoretical and managerial contributions.

2.1 Multichannel Marketing and Multichannel Shoppers

Due to the rapid growth of the Internet and mobile technologies, the channels through which customers can interact with firms have proliferated. Apart from the traditional brick-and-mortar stores, consumers are nowadays engaged with various channels as part of their

shopping and decision journey, for instance, online shopping, catalogs, call centers, direct marketing, home shopping networks, ATMs, and mobile shopping (Neslin et al., 2006). As a consequence, the concept of multichannel marketing has been developed and become a widespread phenomenon in various industries such as consumer goods, B2B companies, retailing and services (Neslin & Shankar, 2009).

(11)

Figure 1. Customer Shopping and Decision Journey

According to Kumar and Venkatesan (2005), multichannel shoppers refer to customers who have used more than one channels in different phases of their shopping process (Figure 1), including information search, product purchase and post-purchase (out of the scope of this study). Many findings have shown that customers are becoming increasingly multichannel. Based on a study by PWC (2011) 86% of US customers use at least two channels as part of their shopping journey, whereas 25% of global customers and 21% of US customers shop across four to five channels. Additionally, a 2014 study by A.T. Kearney reveals that two thirds of customers who buy online use a physical store in various stages of their shopping journey such as before or after the purchase.

As aforementioned, the multichannel environment represents both opportunities and challenges for firms. As shown by prior findings, multichannel shoppers are a more lucrative

(12)

customer segment than single-channel buyers, as they spend 82 percent more per transaction than customers who only shop in stores, yield higher revenue and share of wallet, and are more likely to be active than other customers (Kumar & Venkatesan, 2005 Deloitte; 2010). However, some evidence reveals the negative side of multichannel shoppers, including channel

cannibalization and loss of sales and customers to e-commerce competitors (Nunes &

Cespedes, 2003; PWC, 2011; Heitz-Spahn, 2013). For this reason, firms, especially store-only, catalog-based and internet-only businesses, are faced with the challenges of how to manage their multichannel strategies effectively and whether to integrate multiple channels into their existing business models, or to remain single-channeled and risk becoming obsolete and outperformed by other multichannel competitors (Schoenbachler & Gordon, 2002).

2.2 Research Shopping Phenomenon

Within the field of multichannel customer management, an important issue that has received much attention from practitioners and researchers, is the “research shopping

phenomenon”, whereby customers research for product information or compare prices on one channel but buy on another channel (Verhoef et al., 2007). Provided that a proper

cross-channel strategy exists and that customers switch between different cross-channels within a single firm, research shopping could increase customer satisfaction, loyalty and firms’ revenue. However, this phenomenon may pose a serious threat to firms if customers search at one firm, but purchase from a competitor (Neslin & Shankar, 2009), thereby causing the loss of revenue and potential customers during purchasing processes, and eroding customer loyalty. For instance, Heitz-Spahn (2013) reveals that consumers who adopt multiple channels are more likely to use one retailer’s channel during the search phase and subsequently switch to another

(13)

retailer’s channel when buying the products, which can drastically harm the firm’s profit margins. Figure 2 summarizes the three different types of research shopping behavior and the different channels used in the search and purchase phases of the consumer decision journey, which will be elaborated in the following sections.

2.2.1 Conventional Types of Research Shopping Behavior: Showrooming vs. Webrooming

From the conventional perspective, research shopping behavior can be categorized into two types which are “showrooming” and “webrooming”. Traditionally, there had been growing concerns regarding the threat of “showrooming” to brick and mortar stores in various media. In the past, showrooming (offline-to-online research shopping) represented a dominant form of research shopping scenario, which can be defined as the phenomenon whereby customers visit a store to check out the products and obtain information before making their purchase online

Showrooming (offline-to-online)

Webrooming (online-to-offline)

Mobile-Assisted Shopping Offline & Online Search Online Search

Offline Search Online Purchase

Offline Purchase

Online Purchase

Offline Purchase

Figure 2. Types of Research Shopping Behavior and Channels Used in Different Shopping Phases

SHOPPING PHASES TYPES OF RESEARCH

SHOPPING BEHAVIOR

Purchase Search

(14)

from the same store or from another (Luo et al., 2014). For instance, a study by Accenture (2013) illustrates that 63% US consumers plan to browse products in a physical store and then purchase online for better prices. Showrooming is shown to have detrimental impacts on brick-and-mortar stores as consumers may potentially use physical stores for merely browsing the products and then switch to an online retailer that offers a more competitive price. For instance, a recent study demonstrated that a US-based clothing retailer called JC Penney had suffered a decline of 32% in its store sales due to the impacts of showroomers who visit a store more frequently than the average consumers (Business Insider, 2013).

On the contrary, a pure e-commerce player like Amazon.com appears to benefit from the practice of showrooming at the costs of other brick and mortar stores. A study by Placed (2013) illustrates that Amazon customers visit physical stores such as Walmart, Target and Best Buy before making the purchase on Amazon’s website (Figure 3). Nonetheless, consumer shopping behavior has altered abruptly in recent years due to the prevalence of “webrooming” or the so-called “reverse showrooming”

(research online and purchase offline), which occurs when customers research products online and then visit a physical store to complete their purchase. The webrooming phenomenon will be further elaborated in the next section.

(15)

2.2.2 The Rise of Webrooming

Several studies have confirmed the prevalence of webrooming in the evolving retail and multichannel landscape. According to a research shopping study by DoubleClick (2004),

searching on the internet and buying in-store represents the most common form of research shopping. In accordance with this finding, a study by Nielsen (2014) reveals that 60% of shoppers browse online and make a purchase store, whereas 51% check out products in-store before buying online. More recently, a study by PWC (2015) shows that webrooming is as important as showrooming since 68% of respondents reported exhibiting showrooming

behavior while 70% indicated that they have showroomed (Figure 4). As such, these findings indicate that showrooming may take a backseat while webrooming is becoming increasingly prevalent among research shoppers, representing an opportunity for retailers to gain the momentum and bring shoppers back in physical stores (Taylor & Hunter, 2015).

(16)

More importantly, a study by Merchant Warehouse (2014) has revealed that “while 60% of webroomers have showroomed, nearly 90% of showroomers have webroomed”, implying that customers often make their purchases in physical stores rather than online, and that showroomers in one product category may simultaneously be webroomers when purchasing products from other categories, and vice versa. As it turns out that research shoppers are increasingly purchasing products in physical stores, the previous assumptions that customers prefer to buy online mainly due to cheaper offerings or convenience may no longer be valid, and should therefore be further investigated in the current context.

2.2.3 Emergence of New Type of Research Shopping Behavior: Mobile-Assisted Shopping

With the rapid penetration and developments of smartphones in recent years,

consumers are becoming much more empowered in their purchasing process. Apart from the Internet, new technology-based formats such as smartphones and tablets have become an integral part of consumers’ everyday lives. Due to the ubiquitous nature of these mobile devices, consumers are able to search for product information, compare prices, read product reviews, and buy products anywhere and anytime, even while they are browsing products in a store. Evidence suggests that consumers are increasingly using their mobile devices while in-store and that these mobile devices are influencing a significant proportion of sales in the retail businesses (Deloitte, 2013). Research by Infographic (2014) shows that 50% of consumers with smartphones use their devices for price comparison while shopping in-store. In accordance, a study by Razorfish (2015) reveals that 56% of US Millennials and 23% of US Gen Xers utilize mobile devices as a significant tool assisting them while shopping.

(17)

According to Quint, Rogers and Ferguson (2013), this type of research shoppers can be referred to as “mobile-assisted shoppers” or “M-shoppers”. More specifically, mobile-assisted shoppers browse products in-store while simultaneously use their mobile devices (e.g.,

smartphones and tablets) for price comparison or product information search, and then buy the items either online or in the store. The use of mobile devices at the point of purchase raises some serious concerns among practitioners as some evidence suggest that price comparison is the major reason for the usage of mobile devices while shopping in-store (Broeckelmann & Groeppel-Klein, 2008), which can potentially cause customers to switch to another retailer once a better price is discovered elsewhere.

Some studies suggest that the use of mobile devices in-store may become an issue to brick-and-mortar stores. Based on a study by Akamai (2013), 37% of customers who researched on their smartphones while shopping purchased in-store, whereas 63% purchased on

smartphones, laptop and tablets. In a similar way, JiWire's Mobile Audience Insights Report (2013) shows that the majority of customers who browse in-store still purchase from the store, while 37% purchase on their mobile devices after visiting a store. Nevertheless, many findings suggest that in-store mobile usage of customers is not necessarily a threat provided that firms understand differing motivations of their behavior and provide customers with the best experience, services and assistance possible (e.g., coupons for in-store purchase, product barcode scanner). Table 1 provides an overview of the types and definitions of research shopping behavior in this study.

(18)

How do research shoppers differ? 17 Table 1. Research of Shopping Behavior: Types and Definitions

Types Behavior

Showrooming Search for product information or compare prices in-store and subsequently buy the product online (PC/laptop) or on a mobile device (smartphone/tablet)

Webrooming Search for product information or compare prices online (PC/laptop) or on a mobile device (smartphone/tablet) and subsequently buy the product in a store

Mobile-Assisted Shopping Use mobile devices (smartphone or tablet) to search for product information or compare prices while shopping or searching for the product in the store and subsequently buy either online or offline

2.3 Consumer Decision Journey and Research Shopping Studies

Given the complexity of the customer decision journey in the multichannel environment, several studies have examined consumer behavior in various stages of the shopping process. According to Verhoef et al. (2007), the majority of these research either focussed on the search or purchase decision for single or multiple channels, whereas research that explored the interdependencies between the search and purchase phases remains relatively limited. More importantly, most of these findings do not have a specific focus on research shopping (Verhoef et al., 2007). For instance, Balasubramanian, Raghunathan and Mahajan (2005) examined the perceived product utilities that influence consumer channel choice across multiple stages, including forming a consideration set, product selection and purchase phase. However, the interdependencies between these stages were not addressed.

A few studies have investigated the issue of research shopping from various perspectives. Verhoef, Neslin and Vroomen (2007) developed a framework with three important underlying mechanisms that drive online-to-offline research shopping behavior, including attribute-based decision making, channel lock-in, and cross-channel synergy. From the

(19)

salesperson’s perspective, Rapp et al. (2005) has examined the impacts of perceived customer showrooming in the retail environment and found negative relationships between perceived showrooming and salesperson self-efficacy and sales performance. Heitz-Spahn (2013) and Baal & Dach (2005) investigated the issue of cross-channel free-riding behavior whereby customers switch to another retailer during the decision making process. In addition, Verhoef, Kanan and Inman (2015) touched upon the issue of research shopping briefly in their paper regarding special issue on multichannel retailing and point out that further research on research shopping and its drivers remain an important issue under the domain of retail mix across channels. Specifically, one of the research questions that they put forward is “What is driving

showrooming behavior of shoppers, and how can retailers push back against this behavior or benefit from it?” (See table 2 on p.20 for an overview of past research shopping literature).

2.4 Research Gaps and Research Question

Despite the growing importance and the tremendous impacts of research shoppers in various business sectors, the absence of academic literature with the main emphasis on this issue is quite surprising. For this reason, it is interesting to further examine the issue of research shopping and identify whether and how research shoppers are driven by different consumer characteristics, so as to help firms understand and manage them effectively. As pointed out by Neslin et al., (2006), understanding customer behavior in the multichannel environment

remains one of the main issues to be tackled. Understanding the differences between different types of research shopping behaviors will help firms to effectively allocate their resources across channels, leading to synergies across different sales channels. A proper management of

(20)

research shoppers will subsequently help firms to retain, attract and grow their customer base in an effective manner.

Table 2 provides an overview of past literature regarding research shopping behavior

(also based on alternative terms such as “research online, purchase offline” or “research offline, purchase online”).Thus far, there has not been any specific study focussing on comparing the three types of research shopping behavior. Additionally, multichannel literature that integrates the mobile channel as part of the studies remain relatively scarce. Accordingly, the present study aims to fill these existing gaps in the multichannel literature by investigating the different drivers of the three types of research shopping behavior: showrooming (offline-to-online), webrooming (online-to-offline) and mobile-assisted shoppers (simultaneous search in-store and on mobile devices and purchase either online or offline), incorporating psychographic, expected benefits and variables across three different product categories, namely clothing and apparel, consumer electronics and personal care items. This has led to the formulation of the research question “Whether and how research shopping behavior are driven by psychographic, benefits

(21)

How do research shoppers differ? 20 Authors & (Year) Channels Empirical/

Theoretical

Motivations

Psychographic Benefit Demographic

Mobile-Assisted Shopper

Multiple Categories Shim, Eastlick & Lotz

(2004)

Online, offline Empirical - √ - - √ Choi & Park (2006) Online, offline Empirical √ - √ - - Verhoef, Neslin &

Vroomen (2007)

Online, offline Empirical - - - - - Kollmann, Kuckertz &

Kayser (2012)

Online, offline, mobile

Empirical √ - √ - √ Elliott, Fu & Speck

(2012)

Online, offline Empirical √ √ √ - - Heitz-Spahn (2013) Online, offline Empirical √ √ √ - √ Quint, Rogers &

Ferguson, (2013)

Mobile, online, store Practical - √ √ √ - Wolny & Charoensuksai

(2014)

Mobile, online, offline

Empirical √ - - - - Frasquet, Molla & Ruiz

(2015)

Online, offline Empirical √ √ √ - √ This study Online, offline,

mobile

Empirical √ √ √ √ √

(22)

2.5 Theoretical Contributions

From the theoretical perspective, it is expected that this research will contribute

significantly to the multichannel marketing field in several ways. First, the present study will be the first to compare the three types of research shopping behavior in terms of their

psychographic, benefits sought and demographic drivers. Second, it investigates the research shopping behavior in multiple product categories, in which specific managerial implications for marketers in various industries can be derived. Third, this study will enrich previous academic literature on research shopping behavior, which is an important but infrequently researched area in the academic literature. Fourth, the inclusion of mobile-assisted shopping, the newly emerged type of research shopping behavior, will demonstrate how disruptions in mobile innovation can influence consumer behavior in today’s retail environment, thereby yielding interesting implications for both researchers and practitioners.

2.6 Managerial Contributions

Apart from theoretical contributions, this research provides managerial relevance as it offers managers important insights into the drivers of different types of research shopping behavior. Understanding these varying psychographic, benefits sought and demographic drivers enables managers to learn about their customers on a deeper level and tailor specific strategies so as to retain existing customers, attract new customers and grow the customer base. As aforementioned, managers can develop specific strategies to cope with different types of research shoppers once it is understood which elements need to be emphasized or

(23)

insights obtained from this research to allocate resources more effectively across different channels, which leads to cross-channel synergies for the business.

For instance, if the firm attempts to shift their customers online in order to save costs on physical stores, they might do so by focussing on providing clear and simple return policies or providing online discounts, given that given that these aspects are considered important by the target market. Alternatively, if the firm wishes to drive customers to buy in physical stores, they can focus on elements that webroomers find important, which might be offering in-store discounts and treat its website mainly as a research tool. In case the firm wants to use their varying channels complementarily, the knowledge gained from this study can be utilized to help firms allocate their marketing resources across customer channel segments effectively. In addition, properly managing research shoppers will help prevent customers from shifting to another retailer, decrease the risk of channel cannibalization, improve customer satisfaction, profitability and long-term customer loyalty. Last but not least, this research provides insights that are useful for managers from various industries or managers who manage multiple product categories, including clothing and apparel, consumer electronics and personal care items.

(24)

3. CONCEPTUAL FRAMEWORK AND HYPOTHESES

In this chapter, the conceptual framework and predictor variables hypothesized to identify different types of research shopping behavior will be elaborated. First, the conceptual model will be explained, followed by the development of hypotheses based on psychographic, benefits sought and demographic variables, and the expectations regarding product category differences.

Psychographic

Price-Consciousness, Time Pressure, Shopping Enjoyment, Variety-Seeking

Benefits Sought

Convenience Orientation, Information Privacy Concern, Ease of Return Policies,

Demographic Age, Gender, Education Phase I. Showrooming (offline-to-online) II. Webrooming (online-to-offline) III. Mobile-Assisted Shopping

Offline & Online Search Online Search

Offline Search Online Purchase

Offline Purchase

Online Purchase

Offline Purchase

Predictor Variables

Figure 5. Conceptual framework

(25)

3.1 The Framework

Figure 5 provides an overview of the conceptual framework of this research paper. The

framework illustrates the relationships between the different types of research shopping

behavior and their psychographic, benefits sought and demographic variables across two stages of the purchase process, pre-purchase and purchase. The post-purchase phase is out of the scope of this study. Three distinct types of research shopping behavior can be seen in the model, including showrooming (offline-to-online) which occurs when consumers search in-store and then buy online using their laptop/PC or mobile devices, webrooming (online-to-offline) which happens when consumers use their PC/laptop or mobile devices for search and subsequently purchase the product in a store, and mobile-assisted shopping (simultaneous mobile and offline search while in a store but buy on either online or offline channel), which is drawn from the Aimia shopper research report by Quint, Rogers & Ferguson (2013). The inclusion of the last type of research shopping behavior, mobile-assisted shopping, as part of this study will enrich our understanding in the multichannel environment by illustrating the influence of the mobile technology disruptions on the shopping process.

It is expected that showrooming, webrooming and mobile-assisted shopping behavior might be driven by various psychographic (price-consciousness, time pressure, shopping enjoyment, variety-seeking), benefits sought (convenience orientation, information privacy concern, ease of return policies) and demographic (age, gender and education) characteristics of consumers. In the next part, several research hypotheses have been developed in accordance with this framework.

(26)

3.2 Predictor Variables and Hypotheses Development

To identify whether and how different types of research shopping behavior are driven by psychographics, and benefits sought and demographics across different product categories will be used as predictor variables in this study. Due to the fact that these variables are widely used as important motives in various consumer behavior and multichannel literature, it is crucial that we understand how they might influence distinct types of research shopping behavior and subsequently provide managers relevant implications on how to manage consumers who exhibit these different types of research shopping behavior more effectively. The following section will explain these predictor variables in relation to the hypotheses development.

3.2.1 Psychographics

As pointed out by Konus et al. (2008), psychographic variables have better explanatory power than demographics and are meaningful in explaining consumer behavior. The following four psychographic predictors: price-consciousness, time pressure, shopping enjoyment, and variety-seeking, which are hypothesized to drive different types of research shopping behavior will be discussed in more detail.

3.2.1.1 Price-Consciousness

Price consciousness refers to the degree to which consumers focus on buying products at the lowest price or getting the best deals for the money spent (Lichtenstein, Netemeyer and Burton, 1990; Brown, Pope & Voges, 2003; Konus et al., 2008). According to Elliott, Fu and Speck (2012), these customers are more likely to engage in price search activities and prefer to

(27)

buy from the channel that offers the lowest price possible. Thus, their channel choice behavior can be influenced by the particular perceptions they hold about prices in specific channels (Baker et al., 2002; Montoya-Weiss, Voss, and Grewal, 2003; Verhoef et al., 2007; Konus et al., 2008). Since the online channel provides consumers with ample opportunity to explore various options and the best deals offered, some evidence suggests that a better price is the dominant reason that consumers first browse in-store or use their mobile devices in-store, but then purchase the products online (Broeckelmann & Groeppel-Klein, 2008). As such, the following hypothesis can be formulated:

H1a: Price-consciousness is positively associated with showrooming and mobile-assisted shopping behavior, but not with webrooming behavior.

3.2.1.2 Time Pressure

Time pressure can be defined as the extent to which consumers consider time as a scarce resource that should be spent cautiously (Konus et al., 2008 & Mollá and Ruiz, 2015). Variables such as time of the day, urgency of the purchase and immediate possession of the product represent situational factors that can influence customer channel choice behavior across shopping phases (Konus et al., 2008 & Elliott et al., 2012). Although several findings conclude that time pressure drives customers to select the online channel for purchasing in order to save time (Alreck and Settle, 2002), a more recent study by Alreck et al. (2009)

demonstrates that, as the matter of fact, consumers rarely shop online to save time, due to the delay caused by shipping and overpriced options for returning the product. In addition, Quint, Rogers & Ferguson (2013) reveal that a significant proportion of mobile-assisted shoppers are

(28)

more likely to buy online and will purchase in-store mainly because of the need to obtain the product immediately, implying that time-pressured customers tend to use their mobile devices in-store, so as to make the purchase decision more quickly and may also buy in the store due to time pressure. In line with these prior findings, the following hypothesis can be developed:

H1b: Time-pressure is positively associated with webrooming and mobile-assisted shopping behavior, but not with showrooming behavior.

3.2.1.3 Shopping Enjoyment

Shopping enjoyment refers to the extent the consumer derives emotional benefits such as fun and pleasure from using particular channels in the purchase process, including both pre-purchase and pre-purchase phases (Konus et al., 2008; Nijboer 2015; Mollá and Ruiz, 2015). This can include, for instance, the excitement from trying new experiences, custom designing and co-creating the products, shopping with friends and learning new procedures (Nicholson, Clarke & Blakemore 2002; Forsythe et al. 2006; Konus et al. 2008; Mollá & Ruiz, 2015). Quint et al. (2013), illustrates that shopping pleasure can be gained from in-store experience while

browsing for products (Quint et al., 2013). In contrast, Verhoef et al. (2007) find that perceived enjoyment can influence consumer attitudes toward the online channel for both the search and purchase phases. Additionally, some consumers may find searching and purchasing products on mobile devices a pleasant experience. Despite the limited and relatively mixed insights of shopping enjoyment in the multichannel context, it is expected that shopping enjoyment might influence all the three types of research shopping behavior due to that shopping pleasure might

(29)

be gained in both search and purchase phases and in multiple channels. Thus, the following hypothesis can be formed:

H1c: Shopping enjoyment is positively associated with showrooming, webrooming and mobile-assisted shopping behavior.

3.2.1.4 Variety-Seeking

Heitz-Spahn (2013) refers to variety-seeking as the need to maintain an ideal level of simulation through novelty, complexity or change. Therefore, variety-seeking consumers tend to seek innovations and look for opportunities to try new product offerings and new ways to shop, which is comparable to the definition of innovativeness used by Konus et al. (2008) and Ailawadi et al. (2001). As several authors indicate, variety-seekers engage in multiple channels in order to gain exposure for a broader range of product assortment (Kumar and Venkatesan, 2005; Konus et al., 2008). As a result, they might search in one or several channels (online, offline or both) for the sake of variety, but then purchase either online or offline. Apart from seeking for variety in-store, consumers may also exhibit variety-seeking behavior when using the online channel, as they could be stimulated by the ability of comparison shopping engines that are now widely available and accessible on the Internet (Rohm & Swaminathanb, 2004). Thus, based on prior findings, it is expected that:

H1d: Variety-seeking is positively associated with showrooming, webrooming and mobile-assisted shopping behavior.

(30)

How do research shoppers differ? 29 3.2.2 Benefits Sought

As suggested by Bhatnagar & Ghose (2004), benefits sought is a more robust variable, providing more diagnostic information and insights in comparison with demographics which are merely descriptive. For this reason, past research has incorporated benefits sought in order to identify shopper typologies for the purpose of customer profiling. Tauber (1972) assert that personal motives (e.g., sensory pleasure and self-satisfaction) and social motives (e.g., social experiences and peer group attraction) can also be reasons for customers to shop. Heitz-Spahn (2013) and Konus et al. (2008) indicate that customers assess the benefits and costs derived from search and purchase in the multichannel context and select the channels that can provide them the best value. These benefits can be associated with customers’ hedonic or utilitarian needs (Heitz-Spahn, 2013). In this study, several benefits sought variables are drawn from Bhatnagar & Ghose (2004) and Quint et al. (2013), which will be discussed in more detail in the next section.

3.2.2.1 Convenience Orientation

Various authors have attempted to define convenience in different contexts. In this study, the construct “convenience orientation”, which refers to consumers’ attitudes or motives to save time or effort when it comes to planning, searching or buying products (Rohm & Swaminathan, 2004), is investigated. Consumers may be motivated by only time- or effort-related dimensions or both of these dimensions at the same time (Brown, Pope, & Voges, 2003). Past findings suggest that convenience is highly associated with the benefits for online shopping since customers can access the Internet anywhere and anytime, especially due to the advent of smartphones. However, recent evidence shows contradictory results. Choi and Park

(31)

(2006) show that online channels may provide customer with convenience in the sense that they allow for prompt updates for new products and help customers gain exposure to a wide variety of assortment and product information. Forsythe and Shi (2003) demonstrate that consumers might hesitate to use the online channel for purchase due to the inconvenience caused by delays in receiving the products despite their intention to purchase. Based on these recent findings, it is expected that the Internet and mobile devices might be perceived as a convenient means for search due to its dominant attribute in gathering information (Verhoef et al., 2007), whereas the store might be the convenient way to purchase, allowing consumers to possess the products immediately. Thus, the following hypothesis can be formulated:

H2a: Convenience orientation is a significant driver of consumers exhibiting

webrooming and mobile-assisted shopping behavior, whereas this effect is not expected for showrooming behavior.

3.2.2.2 Information Privacy Concern

Information privacy concern is one of the most frequently discussed topics in the current digital age. In this study, information privacy concern can be defined as concerns or unwillingness to provide personal information in an online vendor for the purpose of conducting an online transaction (Albersa, 2007). Privacy issues related to the use of online channel may demote customers’ desire or intention to buy online due to the fear that the personal information provided might be used by third parties. For instance, consumers might be reluctant to provide credit card information due to the threat of cybercrime. Considering privacy concerns, it is possible that consumers will use the Internet merely as a research tool, but eventually buy in-store due to the perception that it is more secure not to provide personal

(32)

information in an online transaction. Based on these findings, the following hypothesis can be developed:

H2b: Information privacy concern is a significant driver of consumers exhibiting webrooming and mobile-assisted shopping behavior, whereas this effect is not expected for showrooming behavior.

3.2.2.3 Ease of Return Policies

Several findings suggest that easiness of return policies is amongst one of the most important factors that retain consumers’ trust in the firm’s retail channels. Consumers expect to be able to return the merchandise bought without a hassle if the product is defective or turns out to be unsatisfactory (Hahn and Kim, 2009). Gordon (1999) indicates that return and exchange policies is a top common type of complaints for Internet purchase. For instance, some online vendors require that the return shipping is at the expense of the buyers. A study by Quint et al. (2013) confirms that some consumers expect better return policies at the store and therefore choose to buy in-store instead of online. Accordingly, it is expected that:

H2c:Ease of return policies is a significant driver of consumers exhibiting webrooming and mobile-assisted shopping behavior, whereas this effect is not expected for showrooming behavior.

3.2.3 Demographics

Apart from psychographics and expected benefits, demographic factors such as age, gender and education, are another most widely used type of variables in marketing, consumer

(33)

characteristics that might influence customer decision-making and buying behavior in different contexts. Yet, the results remain contradictory and inconclusive (Girard, Korgaonkar &

Silverblatt, 2003). Given the importance of demographics in better understanding factors that drive consumer channel choice behavior, it is important to take the following demographic variables into consideration.

3.2.3.1 Age

There are some evidence that age can be a significant predictor of consumer behavior and purchasing process. Research by McGoldrick & Collins (2007) indicate that multichannel customers tend to skew more towards younger than older consumer groups. In addition, they also find that older customers are more likely to be store-prone whereas younger counterparts tend to be more internet-prone. Conversely, Konus et al. (2008) and Gupta et al. (2004) indicate no effects of demographic factors like age in their attempt to profile multichannel customers, and investigate factors that influence consumer channel switching behavior. Based on the mixed results of previous findings, it is expected that:

H3a:Age is not a significant predictor of any type of research shopping behavior.

3.2.3.2 Gender

Several findings in the multichannel domain suggest that the difference in gender is generally not substantial in explaining consumer behavior (Konus et al., 2008; McGoldrick & Collins, 2007; Quint et al., 2013). For instance, Konus et al., (2008) find no significant impacts of gender in their multichannel shopper segments. In addition, Quint et al. (2013) find that mobile assisted shoppers are distributed quite equally among male and female consumers.

(34)

Nevertheless, some evidence shows contradictory results. Girard, Korgaonkar & Silverblatt (2003) demonstrate that gender plays an important role in predicting preferences for channel choice behavior in relation to varying product types. In their study, males had a strong

preference to shop online for search goods (e.g., books and computer), while their female counterparts preferred to engage in online purchase for experience products such as apparel and fragrances. Due to the contradictory results, it is not expected that gender will have a significant impact in determining research shopping behavior:

H3b:Gender is not a significant predictor of any type of research shopping behavior.

3.2.3.3 Education

Education is another demographic variable used in many shopping behavior and

customer segmentation studies. Yet, inconsistency of results can be seen across findings. Konus et al., (2008) find no significant differences in levels of education among their identified

multichannel customer segments. Elliott, Fu and Speck (2012) demonstrate that education is not associated with their proposed information search segments. Richa (2010) shows that education has no impact on online shopping behavior. On the contrary, Li, Kuo and Rusell (2006) argue that education is a robust predictor of the frequency of online buying behavior. In accordance with this study, Choi and Park (2006) find significant differences in channel choice behavior among single-channel and multichannel buyers. Due to the contradictory results on the findings discussed, it is expected that demographic factors do not differ significantly among different types of research shoppers:

(35)

4. METHODOLOGY

In order to examine whether and how different types of research shopping behavior are driven by psychographic, benefits sought and demographic variables, quantitative data was collected. This chapter discusses the research design used in this study, followed by the sample and survey design. Subsequently, the procedure in conducting the pilot and the main study will be explained. Finally, the measurement and operationalization of the dependent and predictor variables in this study will be clarified. To investigate the drivers of research shopping behavior, the collected data will be further analyzed using binary logistic regression since the dependent variables being investigated are binary (0,1) and the predictors consist of continuous and categorical variables.

4.1 Research Design

The deductive approach was applied in this study to test the hypotheses formulated on the basis of prior literature regarding shopping behavior and multichannel marketing. The research design was quantitative design utilizing cross-sectional, self-administered survey methodology. The purpose of the design was to collect data so as to examine the relationships between the predictor and dependent variables at a particular time, using the combination of descriptive and explanatory research. More specifically, this study investigated whether and how

demographic, psychographic and benefits sought variables have the ability to predict different types of research shopping behaviors. Accordingly, the predictor variables in this study were demographics (age, gender and education level), psychographics (price-consciousness, time pressure, shopping enjoyment and variety-seeking) and benefits sought

(36)

(convenience-How do research shoppers differ? 35

orientation, information privacy concern and ease of return policies), whereas the dependent variables consisted of showrooming, webrooming and mobile-assisted shopping behaviors.

Through the use of online questionnaires, the collection of structured data from a large amount of respondents within a relatively limited time frame was possible (Gillham, 2000; Saunders & Lewis, 2012). Subsequently, the collected data could be analyzed using the binary logistic regression technique in order to measure whether the results were consistent with the hypotheses and whether the theory could be verified (Saunders, Lewis and Thornhill, 2014). As the dependent variables in this study are dichotomous (0,1), binary logistic regression

technique is appropriate for analyzing the data.

4.2 Data Collection and Sample

The sample was drawn from the shopping population in the Netherlands with access to the Internet, which amounts to approximately 10 million people (CBS, 2015). Since the sampling frame was unknown due to the large population and the sample was not selected at random, a non-probability sampling technique had been used (Saunders, Lewis & Thornhill, 2014). The aim of the data collection was to obtain as many respondents as possible, so as to increase the chance that the sample would be representative of the population. To further improve the chance of having a representative sample, it was ensured that potential participants from various age groups including young adults (aged 18 and older), middle-aged adults (aged 30+) and older adults (aged 50+) were included. It was expected that at least a minimum of two hundred and fifty to three hundred respondents would be achieved to ensure an adequate sample size for the analysis.

(37)

By the end of the data collection on May 10, 2016, the number of respondents who entered the survey was 486, whereas 392 respondents had completed the survey, representing a

completion rate of 83.76%. Out of 392 responses, three respondents with missing values were excluded from the analysis, using the listwise deletion method. As a result, a total of 389

complete surveys (N=389) were subsequently included further in the analysis. A brief discussion regarding the handling of missing data can be found in the analysis chapter of this research.

The overview of sample characteristics are reported in Table 3. The sample consisted of 70.9% female and 29.1% male respondents. Participants aged 18-29 represented 83.7% of the sample, whereas older participants aged 30-49 and 50+ contributed 11.8% and 1.8% to the total sample respectively. The main reason that the average age of respondents is skewed towards the younger age group (M = 25.4, SD = 6.7) was that the survey was distributed in the researcher’s personal network which contains mostly young adults. Overall, the education level of participants is high. Only 8.4% indicated that they had completed or were currently following an education lower than the bachelor level, resulting in too many disperse groups with low number of observations. For this reason, the education level was eventually recoded into two groups (Bachelor’s Degree or lower = 0 and Master’s Degree or above = 1).

Table 3. Sample Characteristics (N=389)

Gender % Education %

Female 70.9 Bachelor’s Degree or below 60.7

Male 29.1 Master’s Degree or above 39.3

Age %

18-29 years 83.7

30-49 years 11.8 50+ years 1.8

(38)

4.3 Survey Design

The standardized online questionnaire consisted of three parts. The first part included questions regarding respondents’ actual past purchase behavior. The respondents were asked to recall their shopping behavior for the search and purchase phases in the past six months. Questions were asked separately for each product category to reveal respondents’ channel choice behavior for the following product categories: clothing and apparel, consumer electronics and personal care. These product categories were selected based on their differences in terms of complexity, tangibility, perceived monetary value and frequency of purchases. The incorporation of these different product categories is extremely important as it is expected to yield interesting results that confirm prior findings regarding the impacts of product category differences on consumer channel choice behavior (Kushwaha & Shankar, 2013; Konus et al., 2008; Balasubramanian, Raghunathan & Mahajan, 2005; Levin & Levin, 2003; Schoenbachler & Gordon, 2002). The answers regarding channel choice behavior from this part were subsequently used to categorize whether respondents exhibited showrooming, webrooming or mobile-assisted shopping behaviors in each respective product category.

The second part involved questions regarding psychographic and benefits sought motives that were hypothesized to drive research shopping behavior. In the final part, respondents were asked to indicate their demographic profiles, including age, gender and education level, which might characterize consumers exhibiting different types of research shopping behavior.

(39)

4.4 Procedure

To test the hypotheses, an online survey was created on Qualtrics in English. Prior to the data collection, a pilot study was conducted to pretest the survey design, and ensure the reliability and validity of the questionnaire (Radhakrishna, 2007).

4.4.1 Pilot Study

A pilot study was carried out in a small scale of seven participants in order to test the comprehensibility and feasibility of the questionnaire. This pilot study was crucial as it helped to test the adequacy of the research instruments and highlight unexpected issues that might occur when distributing the survey to a larger group of respondents (Van Teijlingen & Hundley, 2002). The online survey was administered to one older adult, two working professionals, two master students and two bachelor students, who were asked to critically assess the clarity of the questions and suggest points for improvement. The time taken the survey among

participants varied from five to seven minutes, which was not too long. Several questions were deemed unclear and misinterpreted. As a consequence, they were revised based on the

feedback from the participants before the start of the actual data collection process.

4.4.2 Main Study

The online survey was launched on April 29, 2016 and active for the period of one and a half week. The survey was distributed mainly in the researcher’s personal network, specifically via social media sites which were Facebook, LinkedIn and e-mails. The convenience sampling technique was used due to the time constraints and difficulty in identifying members in the population (Saunders, 2011). To ensure a wider reach of respondents, the snowball technique

(40)

was also utilized by asking the respondents to forward the survey to others in their personal network. The limitation might be that this resulted in having a homogenous sample as initial sample members were likely to identify participants who are similar to them (Saunders and Lewis, 2012). In order to reach many respondents as possible, the link to the survey was also posted on online forums and various Facebook groups, thereby giving potential participants an opportunity to self-select whether they would like to take part in the study. Since the snowball sampling and the self-selection sampling techniques were used, it was not possible to identify how many respondents had been reached. Consequently, the response rate could not be calculated. Nevertheless, the completion rate of 83.76% could be calculated based on the number of completed surveys (N=486) divided by the number of respondents who entered the survey (N=392) (Kviz, 1977). Three out of 392 responses contained missing values and were excluded from the analysis by listwise deletion, yielding a final sample of 389 respondents.

4.5 Measurement and Operationalization

In the following section, different constructs from the conceptual model are elaborated in detail. Table 4 summarizes the predictor and dependent variables in this study and illustrates how they are operationalized.

(41)

How do research shoppers differ? 40 Table 4. Measurement and Operationalization

Type of Variable

Dimensions Researched Variables Operationalization Type of Scale Dependent Variables Research shopping behavior Showrooming Webrooming Mobile-assisted shopping

(0=NO; 1=YES) Nominal

Predictor Variables Demographics Age Gender Education (18+ years) (0=Male; 1=Female) (0=Bachelor’s Degree or below; 1=Master’s Degree or above) Ratio Nominal Nominal Psychographics Price-consciousness Time pressure Shopping enjoyment Variety-seeking (1=strongly disagree; 2=disagree; 3=neither disagree nor agree; 4= agree; 5=strongly agree)

Ordinal

(5-point Likert scales)

Benefits sought Convenience-orientation Information privacy concern Ease of return policies

(1=strongly disagree; 2=disagree;

3=neither disagree nor agree; 4= agree; 5=strongly agree)

Ordinal

(5-point Likert scales)

4.5.1 Dependent Variables

To measure respondents’ actual channel choice behavior during the search and

purchase phases, a categorical multiple answer scale borrowed from Reis and Judd (2000) was used. Participants were asked to consider their shopping behavior for each respective product category (clothing, consumer electronics and personal care items) for the past six months. Firstly, they were asked to report which channel(s) they had used for the search and purchase phases for each product category. Respondents could select one or multiple answers out of three possible answers: stores, online (laptop/PC) and mobile devices (smartphone/tablet). If

(42)

How do research shoppers differ? 41 participants selected more than one channels for either or both phases, they were subsequently asked to indicate the channel that they used the most.

These measures were later used to identify whether respondents belonged to consumer groups exhibiting “showrooming behavior” or “webrooming behavior” for each product

category. Respondents who used stores for search and online (laptop/PC) or mobile devices (smartphone/tablet) for purchase were considered exhibiting “showrooming behavior”. On the contrary, respondents who used online (laptop/PC) and mobile devices (smartphone/tablet) for search and stores for purchase were considered exhibiting “webrooming behavior”. In the analysis section, these measures would be created as two different groups of showrooming and webrooming variables in order to be used as dichotomous dependent variables (0 = no; 1 = yes) in binary logistic regression.

Also, respondents were asked to report whether they used their smartphone or tablet while searching or shopping for products in store for each product category. This was to identify whether participants could be considered exhibiting mobile-assisted shopping behavior based on their categorical (yes/no) answers.

4.5.2 Predictor Variables

In the subsequent set of questions, psychographic and benefits sought variables were measured based on a five-point Likert scale anchored by 1 (strongly disagree) and 5 (strongly agree), which were either derived or developed based upon prior marketing literature. This part consisted of 25 items that capture respondents’ shopping motives and expected benefits that are hypothesized to drive different types of research shopping behavior. Respondents were

(43)

asked to rate their degree of agreement with the statements on a five-point scale measuring the following constructs: price-consciousness, time pressure, shopping enjoyment, variety-seeking, convenience-orientation, information privacy concern and ease of return policies.

Five price-consciousness items were adapted from Lichtenstein et al. (1990), Sproles & Sproles (1990), Heitz-Spahn (2013) and Chandon, Wansink & Laurent (2000). Statements that used to measure this construct were, for instance, “It is important for me to have the best price for the product” (Lichtenstein et al., 1990; Sproles & Sproles, 1990) and “Comparing prices is important while shopping” (Heitz-Spahn, 2013). Two-item scale for time-pressure was derived from Srinivasan & Ratchford (1991). Five-item shopping enjoyment scale based upon the previous work by Babin et al. (1994), Dawson, Bloch & Ridgway (1990) and Heitz-Spahn (2013) was used to assess respondents’ shopping enjoyment motive. Four-item scale for variety-seeking was adapted from scales employed in prior studies by Heitz-Spahn (2013) and Rohn & Swaminathan (2004). Three convenience-orientation items were borrowed from Heitz-Spahn (2013) and Kollmann et al. (2012). The Cronbach's Alpha Reliability Coefficients of the multi-item scales were all above 0.6, which was considered acceptable and internally consistent (Sijtsma, 2009). Therefore, these measures should be reliable for their intended constructs (Malhotra, 1993). As for the last two constructs, information privacy concern and ease of return policies, the instruments were self-developed based on previous work from Kau, Tang, & Ghose (2003) and Bhatnagar & Ghose (2004). The reliability of these two multi-item scales were tested for validity and reliability, using the data collected from the pilot study. The reliability

(44)

acceptable and valid (Sijtsma, 2009). An overview of the reliability analysis of the items measuring psychographics and benefits sought used in this study can be found in Appendix C.

Lastly, respondents’ demographics including gender (nominal variable), age (ratio variable) and education level (ordinal variable) were also assessed in the final section of the survey. Gender and education level were asked in the form of a multiple-choice question, whereas respondents could indicate their age in the numeric format. With regard to education level, respondents could select one of the answers ranging from Secondary or High School (VMBO, HAVO, VWO), Tertiary Education (MBO), Bachelor’s Degree (HBO, WO), Master’s Degree or MBA, Doctoral Degree and Other. However, since the sample is relatively highly educated, the data collected for this variable resulted in too many disperse groups with low number of observations. For this reason, the variable was simply recoded into two groups (Master’s Degree or above = 1; Bachelor’s Degree or below = 0) to be used in further analysis.

(45)

5. RESULTS AND ANALYSIS

In this chapter, data preparation and preliminary analysis were performed to check the reliability and validity of the data collected. Additionally, binary logistic regression was

conducted to test the hypotheses and investigate the effects of psychographic, benefits sought and demographic variables on different types of research shopping behavior.

5.1 Data Preparation and Preliminary Analysis

Several steps were taken to ensure that the data was properly prepared for the analysis. Firstly, a frequency check was performed to examine whether there were any errors or missing values for the variables being investigated. No errors were detected in the data, however, some missing values were found and excluded, using the listwise deletion method throughout the analysis. Therefore, this yielded a final sample of 389 respondents (N=389) which was usable for further analysis. To prevent mistakes, missing values were also indicated with a discrete value of 999.00, and all system missing values were recoded to zero (=SYSMIS). Since the scales did not include any counter-indicative items, it was unnecessary to recode them.

5.1.1 Reliability Analysis

Next, reliability analyses were executed to ensure the validity and internal consistency of the measurements. Reliability checks were run for all the scale items representing

psychographic (Price-Consciousness, Time Pressure, Shopping Enjoyment and Variety-Seeking) and benefits sought variables (Convenience-Orientation, Information Privacy Concern and Ease of Return Policies). All the scales, except for Convenience-Orientation, yielded high internal consistency as Cronbach’s Alpha for all the items were above 0.7 (Price-Consciousness α=0.760;

How do research shoppers differ? : an analysis on the basis of psychographics, benefits sought, demographics and product categories for showroomers, webroomers and mobile-assisted shoppers