1
Simplifying the decision-process:
The moderating role of product type on
the relationship between retail
technologies and customer behavior
Name: Denise Deniz (Didi) de Bruin Student number: 10102892
Submitted: 22 jun. 18 Master Thesis
MSc. Business Administration – Digital Business Track University of Amsterdam
2
Statement of Originality
This document is written by Didi de Bruin who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are 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 Table of Contents Statement of Originality ... 2 Abstract ... 5 1. Introduction ... 6 2. Literature Review ... 9 2.1 Customer experience ... 9 2.1.1 Customer Satisfaction ... 9 2.1.2 Value Perception ...11 2.1.3 Technology Acceptance ...12 2.2 Behavioral Intentions ...12 2.3 Research question ...13
2.4 Retail Technology Types ...14
2.4.1 Self-service technologies ...14 2.4.2 Interactive technologies ...16 3. Conceptual framework ...17 4. Methodology ...21 4.1 Research Design ...21 4.2 Set-up ...21 4.2.1 Pretest ...22 4.3 Variables ...23
5. Analysis and Results ...25
5.1 Preliminary Analysis ...25
5.1.1 Missing Values ...25
5.1.2 Sample ...25
5.1.3 Recoded variables ...26
5.1.4 Computing Scale Means ...26
5.1.5 Reliability and Validity ...27
5.2 Correlation Matrix ...28
5.3 Findings ...29
5.3.1 Mediation Analysis...29
6. Discussion & Future Research ...34
6.1 Discussion ...34
6.2 Implications for Theory ...35
6.3 Practical Implications and Contributions ...37
4
6.5 Future Research ...40
Literature ...42
Appendices ...49
7.1 Appendix A – Instructions of the survey ...49
7.2 Appendix B - Scenarios ...50 7.3 Appendix C – Variables ...55 7.3.1 Value Perception ...55 7.3.2 Satisfaction ...55 7.3.3 Behavioral Intentions ...56 7.4 Appendix D – Output ...57
5
Abstract
The emergence of technological innovations has made a noteworthy impact on the retailing industry. Nowadays, retailers are faced with an array of possible shopper-facing technologies to implement in brick-and-mortar stores and this does not make the decision on whether or not to adopt technologies any easier. Bot academic as well as managerial literature widely discuss this decision-process, including the effects of retail technologies on shopper’s perceptions and reactions. This study attempts to shed some light into the discussion and introduce a new approach to the relationship, by including the moderating role of product category type. Self-service technologies and interactive technologies are selected to be the subject of this study. The utilitarian product type is estimated to enhance the positive effect of self-service technologies on customer outcomes, whereas the hedonic product type is
estimated to enhance the relationship between augmented reality technologies and customer outcomes. The findings of this studied indicate that there is no significant effect of retail technologies on either customer’s perceived value of the retailer and customer’s behavioral intentions. Moreover, marginal support is found by this study to support the claim that satisfaction mediates the relationship between using both types of technologies and
behavioral intentions, and that this effect is conditioned by hedonic product type categories.
Keywords: retail technologies, customer satisfaction, value perception, behavioral intentions, product type
6
1. Introduction
Technological innovation has a substantial impact on the retailing industry in several ways (Padgett & Mulvey, 2007). Purchase decisions are still driven by consumers' needs, but the evolution of new technologies, new business models, and big/data analytics in the retail landscape indicates that the retail environment is changing rapidly (Grewal, Roggeveen, & Nordfält, 2017). These turbulent changes are demonstrated by the emergence of new ways of shopping (i.e. online retailing), as well as by the implementation of a variety of new
technology-based innovations in the “brick-and-mortar” stores (i.e. self-service checkouts, virtual or augmented reality) (Bernstein, Song, & Zheng, 2008; Verhoef et al., 2009).
Consumers have changed their shopping behavior significantly (Grewal, Roggeveen, Compeau, & Levy, 2012). Grewal, Roggeveen, and Nordfält (2017) state that technology enables consumers "to make better-informed decisions about which products or services to consume” (p 1.). Digital technology has made it possible to research, browse, and purchase goods better than before (Yasav, 2015). Moreover, technology is playing a central role in the positioning strategy of retailing firms in the market (Padgett & Mulvey, 2007). Previous studies have emphasized that retailers must fully understand what the adoption of new technologies involves and how it will affect their businesses (Clodfelter, 2010). Grewal et al. (2012) suggest that when retailers monitor their environment, identify and respond to changes in the industry, they can enhance their value offering to the customers. There are several ways in how the adoption of retail technologies can offer value to retailers. Devices like mobile phones, computers, and kiosks have become a central transaction mode for customers, where they replace the traditional interaction with employees (Padgett & Mulvey, 2007). These technologies are thus able to decrease firm's cost, e.g., by offloading labor to shoppers or through automation, since expenses on labor still encompass the largest costs for retailers
7 (Demirkan & Spohrer, 2014). Subsequently, a recent study conducted by Inman and
Nikolova (2017) concluded that new technologies provide value for the retailer by
"increasing revenue through drawing new shoppers, increasing share of volume from existing shoppers, extracting greater consumer surplus (e.g., charging higher prices to shoppers who have a higher willingness to pay), or increasing supplier payments" (Inman & Nikolova, 2017, p. 23).
Nowadays, retailers are faced with an expanding array of potential technologies used for the physical point of sale (i.e., brick-and-mortar stores), each providing benefits as well as disadvantages for retailers and customers. Retailers are continuously trying to make decisions on whether or not and which technologies to adopt (Renko & Druzijanic, 2014). This
decision-making process is often discussed in the academic literature, as well as in
managerial practice (Inman & Nikolova, 2017; Yasav, 2015). It signifies the importance of the subject because retailers are trying to keep a grip on the fast moving changes in the industry (Grewal et al., 2012). Fast, strategic response to the rapid changes is necessary for offline retailing to keep existing (Pantano & Timmermans, 2014).
This quantitative research aims to shed some light into the discussion and facilitate easier decision-making on this subject, by examining the relationship between the use of retail technologies on customer perceptions and customer reactions. Though this relationship has already been studied in prior research (Demirci & Kara, 2014; Inman & Nikolova, 2017; Poushneh, Vasquez-Parraga, van Noort, Voorveld, & van Reijmersdal, 2017), the effect of a moderating role has yet to be investigated. The benefits that different types of technologies offer to the customer, hence the needs they fulfill, causes them to provide distinct forms of value, e.g., functional or enjoyment (Willems, Smolders, Brengman, Luyten, & Schöning, 2017). By conducting an experiment, this study tries to find out w the value provided by the retailer matches the benefits provided by the technologies, by investigating the moderating
8 role of product type on the relationship between retail technologies and customer’s actions, through their perceptions. Therefore, this research will contribute to the academic literature by introducing a different approach to this relationship. Moreover, the managerial practice can benefit from this research as well because this study will present new solutions to the decision-making process regarding retail technologies.
This paper is structured as followed: first, the literature on the subject of retail technologies and its effect on consumer outcomes is discussed, followed by the proposed research question. Hereafter, the conceptual framework is presented including the hypotheses of this study. Subsequently, the hypotheses will be tested followed by a description of the analysis of the results. This research concludes with a discussion chapter, including limitations and ideas for future research.
9
2. Literature Review
2.1 Customer experience
When adopting new technologies, retailers are possibly motivated by firm
performance outcomes such as reduced costs, efficiency, and productivity (Demirci & Kara, 2014). However, as technology is becoming an integral part of the retail environment, retailers should take into account the impact on the customer experience (Demirci & Kara, 2014; Verhoef et al., 2009). Verhoef et al. (2009) define the customer experience as “holistic in nature and involving the customer’s cognitive, affective, emotional, social and physical responses to the retailer” (Verhoef et al., 2009, p. 32). Understanding the customer
experience has been an important area for academic research, as well as managerial practice, as retailers realize that their firm performance and profitability is determined by the quality of their customer experience (Grewal, Levy, & Kumar, 2009). The customer experience
consists, on the one hand, of factors the retailer can control (e.g., store atmospherics, service quality) and on the other hand factors that cannot be controlled by the retailer (e.g. contextual factors, shopping goal) (Verhoef et al., 2009). Prior research indicates that the use of in-store technologies is part of these factors. For instance, Foroudi, Jin, Gupta, Melewar, and Foroudi (2016) have argued that the quality of customers' experiences depend amongst other things on the capability of the retailer to use technology. Subsequently, Blázquez (2014) considers in-store technologies to be part of in-store atmospherics.
2.1.1 Customer Satisfaction
There is a considerable amount of literature discussing the determinants and key drivers of customer experience. Although some authors have advocated the need for a new measurement scale to measure the richness of the customer experience, in its holistic nature, (Klaus & Maklan, 2012, 2013), retailing firms still tend to measure particular aspects of the
10 customer experience (Lemon & Verhoef, 2016). One of those aspects is customer
satisfaction, which is defined as “the result of a post consumption or post usage evaluation, containing both cognitive and affective elements” (Homburg, Koschate, & Hoyer, 2005, p. 85). In other words, it is the customer’s judgment of a service or product that is related to a pleasurable level of fulfillment (Demirci & Kara, 2014). Satisfaction is considered to be a essential consumer outcome because it has a definite (positive) relationship with purchase intention (Homburg et al., 2005).
Previous studies on retail technologies have shown that the use of in-store technologies has a positive effect on customer’s satisfaction (Fuentes-Blasco, Moliner-Velázquez, Servera-Francés, & Gil-Saura, 2017; Inman & Nikolova, 2017; Pantano & Naccarato, 2010). Fuentes-Blasco et al. (2017) suggest that retail firms should innovate because the introduction of new technologies improves customer’s satisfaction with the store. According to the Renko and Druzijanic (2014) “higher technological equipment of the store indicates higher quality and speed of serviced offered in the store” (p. 841). Their research also highlights some of the perceived benefits in-store new technology for the customer, including short lines at checkouts and the avoidance of pricing errors, which is confirmed by other studies as well (Weijters, Rangarajan, Falk, & Schillewaert, 2007). These perceived benefits will eventually result in higher levels of customer satisfaction (Renko & Druzijanic, 2014).
In their study, Roy, Balaji, Sadeque, Nguyen, and Melewar (2017) describe the concept of "smart customer experience as a component of smart retailing which focuses specifically on the technology-mediated retailing experience" (p. 258) (e.g., Augmented Reality or Internet of Things). Smart customer experience subsists of smart technology, smart objects with wireless connections that interact with each other. Their findings suggest that
11 smart customer experiences, enhanced by (smart) technology, directly increases satisfaction towards the using smart technology (Roy et al., 2017).
2.1.2 Value Perception
Following customer satisfaction, the consumer's value perception of the retailer is considered to be an additional important determinant of their shopping behaviors (Inman & Nikolova, 2017). Perceived value can be defined as “the consumer’s overall assessment of the utility of a product based on perceptions of what is received and what is given” (Zeithaml, 1988, p. 14). In general, the perception of value results from the customer's consideration of the benefits and sacrifices (e.g., time, effort or money) that is associated with the
consumption of the retailer's product or service (Fuentes-Blasco et al., 2017; Inman & Nikolova, 2017; Willems et al., 2017).
The concept of shopping value, which is the evaluation of the shopping experience by the customer, enriches the definition of value perception. The shopping experience can provide value along two dimensions: hedonic outcomes and utilitarian outcomes (Babin, Darden, Griffin, Darden, & Griffin, 1994; Willems et al., 2017). The former reflects the shopping’s “entertainment and emotional worth” (Babin et al., 1994, p. 646), while the latter reflects a functional, or task-related intention. In other words, hedonic shopping value provides value to the shopping experience by adding entertaining and exciting elements. In contrast, utilitarian shopping value provides a more functional value, e.g., cost savings or reduced waiting time. The role of utilitarian or hedonic value for retail technologies so far has had little attention within academic literature. Some scholars have found that retailers tend to focus more on the utilitarian benefits of technologies, rather than hedonic benefits, which enhance a different part of the shopping experience (Pantano & Viassone, 2014; Willems et al., 2017). However, further research on this subject is required.
12 2.1.3 Technology Acceptance
Acceptance of technology is a crucial aspect in determining both consumer attitude and behavioral intentions (Lee, Fiore, & Kim, 2006). The Technology Acceptance Model (TAM) developed by Davis (1989) to measure this construct, consists of the following components: perceived usefulness, which is defined as “the degree to which a person believes that using a particular system would enhance job performance” (Davis, 1989, p. 320), and perceived ease of use, which refers to “the degree to which a person believes that using a particular system would be free of effort” (Davis, 1989, p. 320). In other words, perceived usefulness is related to utilitarian outcomes of usage or consumption, whereas perceived ease of use reflects the user's judgment on how well he or she is capable of performing a specific action concerning the prospected outcome.
2.2 Behavioral Intentions
Previous sections dealt with the shopper’s perceptions of the retailer (i.e., value perception and satisfaction) and their perceptions of the technology (i.e., perceived usefulness and ease of use). It is important to discuss whether or not these perceptions turn into
shopper’s reactions or actions, that can create favorable outcomes for the retailer. The literature review on this subject shows that, in general, customer's satisfaction with the retailer and perceived value are considered to be essential determinants of the customer experience because increased levels of satisfaction and value perception encourage behavioral intentions (Cronin, Brady, & Hult, 2000; van Birgelen, de Jong, & de Ruyter, 2006). According to Cronin et al. (2000), behavioral intentions are associated with loyalty and recommending intentions, retail patronage intentions, and (re-) purchase intentions. The literature review on service marketing points out a significant link between satisfaction and behavioral intentions, also in technological contexts (Demirci & Kara, 2014). Specifically, Inman and Nikolova (2017) found that shopper’s reactions like satisfaction and perceived
13 value mediate the relationship between the use of retail technologies by the retailer and customer’s behavioral intentions. Furthermore, the role of technology acceptance in the relationship between retail technologies and attitude or behavioral intentions has also been discussed by several authors in literature. Lee, Fiore, and Kim (2006) apply the TAM, supplemented with perceived enjoyment, to examine the factors that influence the relationship between interactive technologies and consumer attitude towards the retailer. Subsequently, Weijters et al., (2007) study the determinants of self-service technology use in the retailing industry and have included components of the TAM into the model. The authors have identified these components as the key drivers of self-service technology use, followed by customer satisfaction and subsequently behavioral intentions (Weijters et al., 2007).
To conclude, high levels of customer satisfaction and value perception will cause higher levels of behavioral intentions, which means that satisfied customers will purchase more and they will be more loyal to the retailer. This will eventually result into increase the retailer’s sales.
2.3 Research question
Retailers are faced with an array of potential technological innovations, and they continuously need to make decisions on whether to adopt new technologies or not (Inman & Nikolova, 2017; Renko & Druzijanic, 2014). The effect of retail technologies on certain parts of consumer behavior like satisfaction and behavioral intentions has been thoroughly examined, but prior research does not provide sufficient information for retailers who are deciding which retail technologies they should adopt. For instance, as stated before in section 2.1.2, the role technologies play in providing shopping value is not presented in the literature. Moreover, the type of product category the retailer sells has not been taken into account for
14 yet. This research thus attempts to present a new approach to the relationship between the use of retail technologies in stores and customer outcomes.
The experience of consuming different types of product categories differ in a way in which the consumption of hedonic products is considered to be more fun, amusing and exciting (Dhar & Wertenbroch, 2000, p. 60). Utilitarian products, on the other hand, are mainly considered to be instrumental and functional (Dhar & Wertenbroch, 2000), serving maximum utility to the consumer’s wants and needs (Hirschman & Holbrook, 1982). These product features address certain characteristics that are similar to the two types of shopping values. Furthermore, Chitturi, Raghunathan, and Mahajan (2008) found that the benefits offered by either type of product have a significant effect on the customer experience, including either delight/dissatisfaction or satisfaction/anger. The role of product type on the relationship between retail technologies and customer outcomes has yet to be investigated.
To address the research gap that is described above, this study will focus on the following research question:
What is the impact of the use of shopper-oriented retail technologies on customer's perceptions and reactions and how does type of product moderate this relationship?
2.4 Retail Technology Types
For this research, two types of retail technologies were selected to be the subject of research; self-service technologies (self-service checkouts) and interactive technologies (augmented reality). A short review of the literature on these technologies is discussed below.
2.4.1 Self-service technologies
15 to produce a service independent of direct service employee involvement" (Meuter, Ostrom, Roundtree, & Bitner, 2000, p. 50). There have been numerous studies to investigate the role of self-service technologies in retailing. Dabholkar and Bagozzi (2002) found several relevant situational factors and consumer traits that emerge when implementing self-service
technologies. These traits include novelty seeking, efficacy towards technology, self-consciousness and also the need for human interaction (e.g., with a store employee) (Dabholkar & Bagozzi, 2002). Subsequently, Dabholkar et al. (2003) researched possible reasons for consumers to either use or avoid self-service technologies. Their findings suggest that consumer's motivation related to these technologies is influenced by several factors, including the attributes of the technology, consumer differences, and situational factors. Collier, Moore, Horky, and Moore (2015) emphasized the situational factors even more. In their paper, they have conceptualized a model and tested the influence of situational
circumstances on customer's attitude towards self-service technologies. They found that situational variables like order size, waiting-time tolerance, location convenience and
employee presence, have a strong influence on customer's decision on using or avoiding self-service technologies (Collier et al., 2015).
Customers can benefit from self-service technologies, for example, because of reduction of waiting times, convenience, and limited pricing errors. However, dissatisfaction with self-service technologies can take place as well. Customers could experience a negative attitude towards these technologies because of missing human interaction (Verhoef et al., 2009). Or rather, when the technology fails to perform, the customer could experience service failure (Zhu, Nakata, Sivakumar, & Grewal, 2013).
16
2.4.2 Interactive technologies
In contrast to self-service technologies, the literature on the subject of other
technologies, those that enhance the customer's shopping experience regarding pleasure and amusement, is less extensive.
Augmented reality (AR) technologies mostly consist of interactive elements. These technologies enhance the perception of reality, by "integrating computer generated virtual information into the user's real world" (Poushneh, 2018, p. 169), thus changing the perception of reality (Poushneh & Vasquez-Parraga, 2017).
A small number of studies have highlighted the importance of enhancing customer experience with interactive and enjoyable elements. A study conducted by Poncin and Ben Mimoun (2014) focuses on two store technologies (i.e., magic mirror with augmented reality and interactive game terminals), both with playful and interactive elements. Their results show that perceived store atmosphere, value perception, and behavioral intentions are all positively affected by these technologies (Poncin & Ben Mimoun, 2014). Pantano and Laria (2012) also found that using virtual reality technologies in the store to create a more
enjoyable and fun shopping experience can positively affect the customer's shopping behavior (Pantano & Laria, 2012).
17
3. Conceptual framework
In this research, the impact of certain types of retail technologies on customer's satisfaction, perceived value and behavioral intentions is investigated. Behavioral intentions include customer loyalty (e.g., WOM), retailer patronage intentions, and (re-)purchase intentions (Cronin et al., 2000; Inman & Nikolova, 2017). Prior literature indicates that there is an effect of the use of retail technologies on behavioral intentions. Therefore, the following hypothesis was developed:
H1: The use of in-store retail technology(ies) positively affects customer’s behavioral intentions towards the store, in comparison with the absence of retail technologies.
For the purpose of this paper, two types of technologies were chosen to be the subject of research:
- Self-service technologies, which involve “technological interfaces that enable customers to produce a service independent of direct service employee involvement.” (Meuter et
18 al., 2000, p. 50). Specifically, self-service checkouts will function as the subject of the questionnaire.
- Augmented reality technology, which can be defined as technologies that enhance the in-store experience and are depicted as interactive technologies (Huang & Liao, 2015; Poushneh & Vasquez-Parraga, 2017). They distinguish themselves primarily from self-service technologies by providing entertainment (Pantano & Naccarato, 2010),
excitement (C. P. Lin & Bhattacherjee, 2010; Poncin & Ben Mimoun, 2014) and more enjoyable elements to the shopping experiences (Blázquez, 2014; Pantano &
Timmermans, 2014).
The benefits that either of these two types of technologies offer and the needs that they fulfill match with specific shopping values: experiential or hedonic value on the one side and functional or utilitarian value on the other (Babin et al., 1994; Willems et al., 2017). Self-service technologies are known to make the shopping experience more convenient. In contrast, augmented reality technologies are considered to be more interactive and
experiential (Poushneh & Vasquez-Parraga, 2017). The degree to which these specific retail technology positively affect behavioral intentions is therefore estimated to depend on the shopping goal or orientation of the customer, i.e., whether the customer shops for hedonic or utilitarian products. In this study, it is therefore estimated that the (positive) effect of the use of self-service technologies on behavioral intentions is larger when customers shop for
utilitarian products than for hedonic products. Subsequently, the (positive) effect of the use of augmented reality technologies on behavioral intentions is larger when customers shop for hedonic products than for utilitarian products. The following hypothesis was developed to test this relationship:
19 H2: Type of product moderates the relationship between retail technologies and behavioral intentions.
Many studies have stated that high levels of shopper’s perceptions, such as
satisfaction or perceived value, lead to high levels of shopper’s reactions (Chih‐Hung Wang, 2012; Inman & Nikolova, 2017; Roy et al., 2017). Shopper's perceptions consist of two parts: perception of the retailer and perception of the technology. Technology perception is
considered to be an important determining factor for customer's perceptions of the retailer (Renko & Druzijanic, 2014) and is often measured by using components of the TAM (F. D. Davis, 1989). However, measuring technology acceptance requires a more extensive experimental research setting, as will be explained in the following chapter. Unfortunately, due to time and resources constraints, this study is not able to conduct such a method of research. Therefore, it was decided not to measure components of the TAM, but rather the components of customer perception of the retailer. In prior literature, multiple researchers have stated that implementing new technologies in the store improves customer’s value perception of the retailer (Inman & Nikolova, 2017; Willems et al., 2017). Value perception has a positive relationship with behavioral intentions (Cronin et al., 2000) and mediates the relationship between the use of in-store retail technologies and behavioral intentions. Therefore, the following hypotheses were developed:
H3: The positive relationship between the use of in-store retail technologies and customer’s behavioral intentions is mediated by the customer’s value perception of the retailer.
H3a: The use of in-store retail technologies positively affects customer’s perception of the retailer, in terms of perceived value.
20 Satisfaction is considered to be another essential customer outcome because it can generate profit by stimulating purchases or purchase intentions (Homburg et al., 2005), but also customer loyalty behavior, including word of mouth and recommending (Fuentes-Blasco et al., 2017; Inman & Nikolova, 2017). These findings generate the following hypothesis:
H4: The positive relationship between the use of in-store retail technologies and customer’s behavioral intentions is mediated by customer’s satisfaction with the retailer.
H4a: The use of in-store retail technologies positively affects customer’s perception of the retailer, in terms of satisfaction.
H4b: Customer’s satisfaction with the retailer positively influences behavioral intentions.
Finally, the conceptual model designed for this research will test conditional indirect effects, otherwise known as moderated mediation effects (Preacher, Rucker, & Hayes, 2007). It shows that the relationship between retail technologies (predictor variable) and behavioral intentions (outcome variable), via value perception and satisfaction (mediating variables) will depend on levels of product type (moderating variable). The primary cause for choosing this particular model is because of the possibility that the indirect effect, or rather the strength of it, of using retail technologies on behavioral intentions, might depend on the level of product type. The final hypotheses are therefore developed as followed:
H5: Type of product moderates the relationship between retail technologies and customer’s value perception of the retailer.
H6: Type of product moderates the relationship between retail technologies and customer’s satisfaction the retailer.
21
4. Methodology
4.1 Research Design
This research is quantitative, and the research design consists of an experimental vignette methodology (EVM). This methodology presents “participants with carefully constructed and realistic scenarios to assess dependent variables including intentions, attitudes, and behaviors.” (Aguinis & Bradley, 2014, p. 352). It allows researchers to
manipulate variables to enhance experimental realism. For this research, the treatment groups of each vignette contained either one of the levels of the Retail Technology variable (self-service, Augmented Reality, both or none) and either one of the Type of Product variable (hedonic or utilitarian). Thus, this research is a 4x2 between-subjects factorial design with eight conditions.
4.2 Set-up
The method for this research was an online questionnaire, containing fictive situations. As stated in the previous chapter, the components of the TAM could not be
measured using such a method. To accurately measure the components of the TAM, it is best if participants would make use of the technology in a real-life setting, whereafter they would answer questions about their perceptions of the technology. However, since at least 240 participants were needed to form a representative sample for this study, it was not feasible for this research to conduct the experiment in such a way, due to time and resources constraints. Therefore, an online questionnaire was designed using survey software developed by
22 Each vignette contained two scenarios of two different product categories1, following a set of questions measuring the variables. Participants were first instructed to read
descriptions of self-service checkouts and augmented reality technology. After this, they were asked to read two scenarios, which each contained a situation in which a fictive retailer is selling a particular product and adopted one, both or none of the technologies mentioned before in their store. Participants were asked to imagine themselves being in these situations. The product categories that were the subject of the scenarios both belonged to each of the product types (hedonic or utilitarian). The scenarios can be found in appendix B of this paper. At the end of the survey, they were asked to answer demographic questions about their gender, age, and level of education.
4.2.1 Pretest
Before sending out the survey, a pretest was conducted to assess which product categories were going to be the subject of the scenarios. A short survey was created, and first respondents were required to read definitions of utilitarian and hedonic products (Hirschman & Holbrook, 1982; Voss, Spangenberg, & Grohmann, 2003). Hereafter, they were asked to rate twelve product categories, including six hedonic and six utilitarian product categories, according to their judgment of the product category's hedonic or utilitarian value, using a 7-point bipolar matrix. Two product categories with the highest mean would function as the hedonic products, and two product categories with the lowest mean would function as the utilitarian product. The results of the pre-test reported that fashion products (M = 6.58) and sports products (M = 6.56) were rated as the most hedonic products, whereas grocery products (M = 2.52) and furniture (M = 2.16) were rated as the most utilitarian products. Therefore, these product categories were selected to be the subject for the vignette scenarios.
1. The reason for choosing to use two different product categories per vignette, instead of one,
23 4.3 Variables
This study consists of eight treatment groups, which contain either one of the following levels of the predictor variable Retail Technology: self-service technology (SST), augmented reality technology (AR), both technologies, and none of the technologies (control group). Also, the treatment groups are conditioned with either one of the two levels of the moderating variable Product Type, i.e., hedonic or utilitarian product.
Behavioral Intentions is selected to function as the outcome variable of this study. This consist of loyalty intentions, (re)purchase intentions and willingness to pay. To measure this construct, two items from the three-item scale developed by Cronin et al. (2000) was used supplemented with items regarding willingness to purchase and willingness to visit. These additional items were adapted from Inman and Nikolova (2017).
The conceptual model designed for this study consists of two mediators. The first one is Perceived Value. In the previous sections, it is already stated that value can consist of
multiple levels. For this research, perceived value is measured according to the cost-sacrifice definition (Willems et al., 2017; Zeithaml, 1988). To measure this concept of value
perception, the two-item scale developed by Cronin et al. (2000) was used (α = 0.88). This scale is initially developed to measure service quality, but the items are sufficient for
measuring value perception according to the cost-sacrifice definition (an example of an item includes: ‘Compared to what I had to give up, the overall ability of this facility to satisfy my
wants and needs is…’).
The second mediator selected for this study is Satisfaction. This measured by the three-item SAT2scale, adapted from Cronin et al. (2002) (α = 0.85). This scale is based on an
24 added, as is done in research of Inman and Nikolova (2017) who adapted this item from Maxham and Netemeyer (2003) (e.g., ‘My overall satisfaction with the retailer is’).
For the mediating and outcome variables, all items are measured on a seven-point Likert scale. Several control variables were included in the research. The findings of Weijters et al. (2007) highlighted that gender and education play a vital role in the customer's decision to use technology. Furthermore, prior literature has yet to explain the relationship between age and attitude or behavior towards technology, so age will be added as a control variable to this study. A full overview of all items per scale can be found in appendix C.
25
5. Analysis and Results
5.1 Preliminary Analysis
5.1.1 Missing Values
A frequency test was run to check for missing values for all predictor and outcome variables in the model. The amount of missing values was < 23 %. Also, the time duration per case was checked. Participants who spent less than 100 seconds on the questionnaire were deleted from the data. After eliminating these cases, together with the missing values of the predictor and outcome variables, this resulted in a total of 253 valid questionnaires ready for data analysis. Furthermore, these cases were distributed over the eight different treatment groups, with Self-service x Hedonic containing the largest sample (N = 36), and Both x Utilitarian containing the smallest sample (N = 27). Distribution of the vignettes can be found in table 1.
5.1.2 Sample
The population for this study consists of consumers over the age of 18. The size of this population is very large, and the sampling frame is unknown. Participants for this study were collected through a combination of two methods. First, availability sampling was used to collect participants through social media platforms (Facebook) and personal e-mail. This sampling technique causes it difficult to predict the response rate (Deutskens, Ruyter,
Wetzels, & Oosterveld, 2004). Second, participants were recruited through Mechanical Turk, where they completed the survey in exchange for a small compensation. Approximately 45 % of the respondents were recruited via Mechanical Turk. The sample group has an equal gender distribution (49,4 % male, 48,4 % female, remaining did not disclose). The most frequent age group of the sample was 25-34 years old (37.9 %) followed by 18-24 years old (35.2 %). The less frequent age group was 65 years or older (2.8%). Most of the respondents
26 included in the sample group were of Dutch nationality, with a total of 37,5%. Other frequent nationalities include American (N=54) and Indian (N=41). More than 50 % of the participants indicated that they obtained a Bachelor’s degree at a University (54.2 %).
Table 1
Retail Technologies and product Type (Sample sizes)
Product Type Retail Technology Hedonic Utilitarian Total
None 31 29 60 Self-service 36 33 69 Augmented Reality 34 32 66 Both 31 27 58 Total 132 121 253 5.1.3 Recoded variables
No counter-indicative items were used in any of the scales. A couple of dummy variables were created. The variable ‘Random’, which indicated which treatment group the participants were assigned to, was separated into two dummy variables; first the variable ‘RetailTech’ was created for the retail technology condition, containing the values 1
(‘None’), 2 (‘SST’), 3 (‘AR’), and 4 (‘Both’). The second variable ‘ProdType’ was created to depict the product type condition and contained the values 0 (‘Hedonic’) and 1 (‘Utilitarian’). Covariate ‘Gender’ was recoded into a 0/1 dummy variable, containing values 0 for ‘Male’ and 1 for ‘Female’. Finally, covariates education level and age group were recoded, were the former variable was recoded into three 0/1 dummy variables (no degree, university/college degree, other degrees) and the latter was recoded into six 0/1 variables.
5.1.4 Computing Scale Means
As stated before, the scenarios in each vignette contained two product categories of the product type that was part of the condition. Computing the scale means was done
27 b was computed for all single items. Next, the mean was calculated for all items of every vignette to describe a variable. Means and standard deviations of all variables are depicted in table 4.
5.1.5 Reliability and Validity
Because this is a survey-based study, reliability checks were needed for variables. This was done in two parts, using Cronbach's alpha. This test represents the estimator of internal consistency (Cronbach, 1951). First, Cronbach's alpha's for three variables per each of the vignettes were tested. Second, the reliability of the scales for all vignettes was tested. To do so, the data for every case of each variable was copied to one single vignette. The Cronbach's alpha for every variable, and also for every single vignette was > 0.7, which is very good. The variables and their corresponding Cronbach's alpha are exhibited in table 2 and 3. It should be noted that multiple reliability coefficients of the scales within the
vignettes, and additionally the coefficient of the general Behavioral Intentions scale, are very high (> 0.95). It means that the items on this scale have a very high degree of internal
consistency (Streiner, 2003); however, this can also mean that the items on this scale are considered to be “overly redundant and the construct measured too specific” (Briggs & Cheek, 1986, p. 114). Because no clear standards exist in academic literature, and therefore no decision could be made on which item to delete, it was decided to keep all the items on the scales.
Table 2
Reliability Coefficients
Variable Cronbach´s Alpha All Vignettes
Satisfaction 0.89
Perceived Value 0.89
Behavioral Intentions 0.95
28 Table 3
Reliability Coefficients
Variable Cronbach's Alpha Vignette 1 Cronbach's Alpha Vignette 2 Cronbach's Alpha Vignette 3 Cronbach's Alpha Vignette 4 Satisfaction 0.79 0.89 0.91 0.96 Perceived Value 0.9 0.78 0.87 0.95 Behavioral Intentions 0.93 0.94 0.94 0.97
Variable Cronbach's Alpha Vignette 5 Cronbach's Alpha Vignette 6 Cronbach's Alpha Vignette 7 Cronbach's Alpha Vignette 8 Satisfaction 0.91 0.97 0.9 0.88 Perceived Value 0.95 0.98 0.87 0.94 Behavioral Intentions 0.96 0.98 0.92 0.96 5.2 Correlation Matrix
The correlation matrix exhibited in table 4 shows that there is statistical evidence for several correlations between the variables. The Pearson r correlation statistic was used for continuous variables. Since categorical variables are included in this analysis, some correlation values were calculated using Spearman rank correlation (Field, 2014). The Pearson correlation statistic produced the following results: first, in line with theory, behavioral intentions is positively correlated with variables satisfaction (0.91, p <0.01) and perceived value (0.86, p <0.01). These values are > 0.50 which indicates that the correlation effect size is very strong (Cohen, 1988). Second, satisfaction is also positively correlated with variables Perceived Value (0.87, p <0.01). Again, this correlation effect size is strong. The results of the correlation matrix indicate that Age Group is the only one covariate that has correlations with other variables. Age Group has a negative correlation with variables behavioral intentions (-0.14, p <.05) and perceived value (-0.19, p <0.01). However, the effect sizes can be considered small, since the values are < 0.30 (Cohen, 1988).
29 Table 4
Means, Standard Deviations and (Non-Parametric) Correlations
Variables M SD 1 2 3 4 5 6 7 1. Behavioral Intentions 4,98 1,24 (.95) 2. Satisfaction 4.95 1.1 .906** (.89) 3. Perceived Value 4.77 1.12 .858** .869** (.89) 4. Retail Technology 2.48 1.09 -.028 -.073 .009 - 5. Product Type 1.48 .5 .021 .026 .064 -.010 - 6. Gender 0.49 .5 -.049 .009 .010 -.017 .039 - 7. Age Group 3.13 1.26 -.138* -.098 -.19** -.075 -.223** -.226* - 8. Education Level 4.88 1.23 .034 .052 .063 .007 .043 -.008 .104
**Correlation is significant at the 0.01 level (2-tailed) *Correlation is significant at the 0.05 level (2-tailed)
5.3 Findings
5.3.1 Mediation Analysis
Testing of the hypotheses was done using IBM SPSS, a computer software program used of statistical analysis (Field, 2014), in combination with the SPSS Process Macro developed by Hayes (2012), to conduct a moderated mediation analysis. Model 8 was used for this analysis because this model consists of one predictor (X), one outcome (Y) and one moderating variable (W), and this model allows up to 10 mediators (M) operating in parallel (Hayes, 2012). Covariates gender, age, and education level were added to the model2.
Because the predictor variable Retail Technology (X) is categorical of nature, dummy variables were created. The arrangement of the dummy variables is depicted in table 5. For this research, hypotheses were tested at a significance level of <.05. Results of the analysis are presented in table 6 and table 7. Also, tables containing the results including control variables can be found in appendix D.
2As a consequence of using multiple dummy variables for age and education level, too much
(near) linear dependency existed in the data and the analysis. Therefore, it was decided to exclude the sixth age group dummy variable (65 years or older) and the third education level dummy variable (other degrees) from the analysis.
30 Table 5
Arrangement of dummy variables for retail technologies
Dummy Variables
SST 1 0 0
AR 0 1 0
Both 0 0 1
The first hypothesis (H1) stated that the use of retail technologies affects behavioral intentions. The results indicate that there is no significant direct effect found between the use of retail technologies and behavioral intentions. This applies for all groups, including self-service technologies (c’1 = 0.04, p = 0.71), augmented reality technologies (c’1 = 0.13, p =
0.29), as well as using both technologies (c’1 = 0.03, p = .78). Therefore, H1 is rejected .
Furthermore, H2 stated that the direct effect of retail technologies is moderated by product type. As depicted in table 6, this effect is not significant (c’4 = 0.02, p = 0.9 for self-service
technologies, c’4 = .02, p = 0.9 for augmented reality technologies, and c’4 = 0.05, p = 0.76 for both technologies), so H2 is also rejected. Concerning the covariates that are
depicted in appendix D, gender and certain age group have a significant direct effect on behavioral intentions. The results indicate that females (Gender = -0.17, p = 0.007) score significantly lower on behavioral intentions. This also counts for people between the age of 25 and 54 years old (Age Group 25-34 y/o = -0.38, p = 0.041, Age Group 35-44 y/o = -0.4, p = 0.046, and Age Group 45-54 y/o = -0.53, p = 0.012).
31 Table 6
Consequent
Behavioral Intentions (Y)
Antecedent Coeff. SE p Retail Technology (X) c'1 SST .043 .115 .710 AR .127 .119 .288 Both .033 .119 .781 Perceived Value b1 .307 .057 <.01 Satisfaction b2 .694 .058 <.01 Product Type (W) c'2 -.067 .122 .581 Retail Technology x Product Type (XW) c'4 SST x Product Type .020 .166 .904 AR x Product Type .021 .168 .899
Both x Product Type .052 .172 .763
Constant i2 .375 .061 .202
R2 = .8563
F(17, 229) = 80.2727, p < 0.1
Because of the categorical nature of predictor variable retail technologies, and therefore three dummy variables were created, table 7 shows three interaction terms for each of the mediators. As can be seen in the results, for perceived value the moderated mediation effect does not take place for all levels of retail technology (a3 for SST x Product Type = 0.39, p = 0.32, a3 for AR x Product Type = -0.02, p = 0.96, and a3 for Both x Product Type = 0.45, p
= 0.26) Therefore, both H3 and H5 are rejected. Equally to behavioral intentions, certain age groups have a direct effect on the perception of value as well. As can be seen in the tables presented in the appendix D, people with the age of 18 till 44 years old score significantly higher on value perception (Age Group 18-24 y/o = 1.46, p = 0,008, Age Group 25-34 y/o = 1.31, p = 0.003, and Age Group 35-44 y/o = 1.17, p = 0.013). Subsequently, people with no university or college degree score significantly lower on value perception of the retailer (No degree = -1.01, p = 0.028).
32 Table 7
Consequent
Perceived Value (M1) Satisfaction (M2)
Antecedent Coeff. SE p Coeff. SE p
Retail Technology (X) a1 a1 SST .006 .272 .981 -.047 .271 .863 AR -.009 .280 .976 -.224 .280 .423 Both -.187 .278 .498 -.516 .278 .064 Product Type (W) a2 -.097 .286 .735 a2 -.323 .285 .258 Retail Technology x Product Type (XW) a3 a3 SST x Product Type .392 .390 .317 .487 .389 .212 AR x Product Type -.020 .398 .960 .057 .397 .887
Both x Product Type .453 .404 .264 .702 .403 .083
Constant i1 4.030 .616 <.01 i1 4.790 .614 <.01
R2 =.1215 R2 = .1111
F(15, 231) = 2.1307, p < .01 F(15, 231) = 1.9247, p < 0.5
The results furthermore indicate a marginally significant interaction effect (a3 for Both x Product Type = 0.7, p = 0.08) of using both technologies on mediating variable
satisfaction. Moreover, the results point out that there is also marginally support for a
conditional effect of retail technologies on satisfaction when product type = 0 (i.e., hedonic), a1 for Both = -0.52, p = 0.06). These results indicate marginally support for H4a. The second
part of H4 (H4b) states that there is a direct effect of satisfaction on behavioral intentions, and according to the results depicted in table 6, this relationship is significant (p < 0.01). Therefore, marginally evidence is found for the indirect effect of using both retail
technologies on behavioral intentions, via customer satisfaction (H4). The final hypothesis, H6, stated that product type moderates the relationship between the use of retail technologies and satisfaction. The results of the analysis show that the indirect effect is conditioned by product type, in the sense that the effect is only significantly present for the hedonic product type (effect = -0.36, SE = 0.18, LLCI = -0.74, ULCI = -0.02). Thus, there is marginally support for H6. In other words, when customers shop for hedonic products and use both
self-33 service as well as augmented reality technologies, their behavioral intentions are estimated to decrease, and this is partly due to the decrease in their level of satisfaction.
Table 8
Mediator: Perceived Value
Product Type Unstandardized Boot Effects
Boot SE Boot LLCI
Boot ULCI
Conditional indirect effect at Behavioral Intentions for the levels of Product Type
SST Hedonic -.002 .078 -.153 .160 Utilitarian .122 .095 -.050 .325 AR Hedonic -.003 .081 -.159 .163 Utilitarian -.009 .093 -.204 .168 Both Hedonic -.058 .078 -.219 .091 Utilitarian .081 .101 -.114 .281 Table 9 Mediator: Satisfaction
Product Type Unstandardized Boot Effects Boot SE Boot LLCI Boot ULCI
Conditional indirect effect at Behavioral Intentions for the levels of Product Type
SST Hedonic -.033 .180 -.399 .301 Utilitarian .306 .194 -.071 .682 AR Hedonic -.156 .168 -.504 .151 Utilitarian -.116 .195 -.522 .252 Both Hedonic -.358 .182 -.739 -.019 Utilitarian .129 .215 -.307 .543
Finally, equal to value perception, covariates age group, and education level have a significant direct effect on satisfaction. Similar to value perception, people with the age of 18 till 44 years old score significantly higher on satisfaction (Age Group 18-24 y/o = 1.13, p = 0,009, Age Group 25-34 y/o = 1.02, p = 0.019, and Age Group 35-44 y/o = 1.22, p = 0.01), whereas people with no college or university degree score significantly lower (No degree = -0.1, p = 0.026).
34
6. Discussion & Future Research
6.1 Discussion
In this research, the impact of certain retail technologies on customer outcomes, such as behavioral intentions, was investigated. Additionally, the moderating role of type of product and the mediating roles of customer's value perception and customer's satisfaction with the retailer was studied.
The results of this study do not fully support the proposed conceptual model. The first hypothesis (H1) stated that the use of specific in-store technologies (self-service and
augmented reality technologies) positively affects customer's behavioral intentions. There was found no statistical evidence to support the claim that there is a direct effect between retail technologies and behavioral intentions.
Regarding the second hypothesis (H2), the role of product type (hedonic versus utilitarian) as a moderating variable was tested. There was no statistical evidence found to support an interaction effect between retail technologies and product type on behavioral intentions. The findings of this study, therefore, reject H2.
The third and fourth hypotheses (H3 and H4) stated that respectively customer's value perception and customer's satisfaction of the retailer both functioned as mediators in the relationship between retail technologies and behavioral intentions. Also, the two final
hypotheses (H5 and H6) stated that product type moderate these mediating relationships. For perceived value, no statistical evidence was found to support the claim that retail technologies influence the relationship between perceived value and subsequently, it was not statistically proven that product type moderates this relationship. This rejects H3, as well as H5.
35 However, the results showed that an interaction effect of product type on satisfaction was visible. The findings of this study indicate that for hedonic products, the use of both technologies (self-service as well as augmented reality) negatively affects the level of customer's behavioral intentions, through customer's satisfaction. Following these results, some evidence was found to support H4 as well as H6, though it should be noted that the significance was marginal.
6.2 Implications for Theory
As stated in the previous section, the conducted study does not support the conceptual model that was designed for this research. First of all, there was no direct effect found
between of all levels of retail technologies (self-service, augmented reality, and both) and customer’s behavioral intentions. Second, the mediation effect of satisfaction did only take place for the relationship between using both technologies and behavioral intentions. Third, no mediation effect of value perception was found at all. This does not concur with prior studies. Several studies have proven that in-store technologies do increase levels of both satisfaction and value perception. For instance, the study of Fuentes-Blasco et al. (2017) stated that introduction of in-store technologies improves consumer value perception and satisfaction with the store. Inman and Nikolova (2017) found that the use of retail
technologies, including self-service checkouts, did positively influence retail patronage intentions, which are part of behavioral intentions. Also, they found that this relationship is mediated by several shopper's perceptions, including value perception and satisfaction. Subsequently, Roy et al. (2017) studied the role of augmented reality as part of the smart customer experience and found evidence that smart customer experience significantly improves customer's satisfaction. Other researchers highlighted the importance of
36 impact it has on satisfaction and behavioral intentions (Pantano & Laria, 2012; Poncin & Ben Mimoun, 2014; Poushneh & Vasquez-Parraga, 2017). There are several possible reasons why this part of this study's findings does not concur with prior literature. Previous studies have shown that the customer's satisfaction level could depend on a couple of factors. First of all, self-service technologies (e.g., self-service checkouts) could negatively affect the
customer experience when there is a need for human interaction (Grewal et al., 2009; Verhoef et al., 2009). Second, technology acceptance, including perceived ease of use and perceived usefulness (F. Davis, 1998), often can be a determining factor of shopper’s perceptions (Renko & Druzijanic, 2014; Vrechopoulos, O’Keefe, Doukidis, & Siomkos, 2004). Finally, the quality of the service can be considered as a determinant of the customer's satisfaction or value perception. These factors are not included in the model, but might be determinant for the outcomes.
For the second part of this study, the moderating role of product type was tested in the model. So far known, the moderating role of hedonic products versus utilitarian products was yet to investigate in prior literature. The results of this study indicate a negative indirect relationship between making use of both technologies and the level of customer's behavioral intentions, through customer's satisfaction with the retailer's store, and this is only the case when the retailer sells hedonic products. It should, however, be noted that the first part of the indirect effect was only marginally significant. Unfortunately, this study cannot reveal any other significant moderating effects of product type on the other retail technology levels. It cannot be ignored that this perhaps has something to do with factors of technology
acceptance as well. As previously reported by the study of Lee et al. (2006), perceived usefulness and perceived ease of use both reflect the aspects of utilitarian shopping behavior, whereas perceived enjoyment reflects aspects of hedonic shopping behavior. These variables have not been studied in this research. It may be assumed that these variables are inevitably
37 necessary to study in order to understand the relationship between retail technologies, product type, and the customer experience (including satisfaction, value perception, and behavioral intentions).
Returning to the research question posed at the beginning of this study, it is now possible to state that:
There is found marginally significant evidence that only for hedonic products, the use of both self-service, as well as augmented reality technologies together, negatively affects customer's behavioral intentions, via customer's satisfaction with the retailer.
6.3 Practical Implications and Contributions
As stated in the introduction, this research attempted to shed some light for retailers in the present-day turbulent retail environment. Moreover, as retailers are embracing in-store retail technologies and start experimenting with them, this research tried to simplify the decision-making process for retailers regarding whether or not and which retail technologies to implement in the brick-and-mortar stores.
The results of this study implicate that first of all, when retailers are deciding on which technologies to adopt, they should carefully consider the type of product range they offer, as well as customer’s demographics. This research’s findings suggest that
demographics like age, education level, and gender potentially could influence customer’s behavior in the relationship with retail technologies.
Second, according to the results, retailers should also carefully consider not to implement too many technologies, since this can harm customer outcomes like satisfaction.
38 6.4 Limitations
Some limitations of this research have to be acknowledged. The first part of the limitations regards the methodology of research. This study was conducted using a quantitative research methodology, by means of a questionnaire. This method involves a structured questionnaire containing limited options of responses, which are selected by the researcher. Limited outcomes are thus considered to be a limitation of this research method because the complexity of human experience is often unable to be revealed.
Moreover, due to a limited time frame, this was a cross-sectional study. It means that the variables were measured at a single point in time. In contrast with using longitudinal methods, this can cause low internal validity, since the data obtained cannot be used to determine causal relationships. The use of self-reported measures is a limitation as well because it can create common-method bias (Bono & McNamara, 2011).
Non-probability sampling was used as a sampling technique. Using such a technique, representativeness of the population cannot be guaranteed because non-probability sampling is very sensitive to self-selection bias, so subjects are excluded that may differ from the ones that are included. Regarding the size of the sample, the total amount of respondents to the questionnaire was sufficient. However, due to the vignette method that was applied to this study, participants were divided into eight different groups. The sample size of each group did not include more than 36 participants, which can be considered as too small.
A vignette study is a valuable method for research because it enables the researcher to combine ideas from standard experimental and survey methodology. However, there are still some limitations. The situations as described in each vignette are unable to simulate reality completely. With vignette studies, it cannot be guaranteed that the decision-making process that is modeled can be applied to what will happen in real life. Even though this is not
39 necessarily the intention of the vignette method, it can be argued if the results of the research can model general behavior. A second common limitation for this method involves the lack of a contextual reference point. Because participants were assigned to only one of the vignettes, instead of multiple, they were not able to make a comparison of their given
situational factors with one another. This might have caused them to inaccurately reflect their judgment on the situation, which is a possible explanation for why no significant differences between the groups were found.
Concerning the proposed conceptual model, some limitations have to be considered as well. First of all, the scale that is used for measuring value perception only contained two items. This might not be sufficient enough to measure this variable. Besides this, the concept of value consists of more levels. For this research, merely the cost-sacrifice definition of value, as stated by Zeithaml (1988), is used to prevent redundancy and too much complexity in this study but a broader definition might be needed to understand the concept of the construct fully. Second, as has been stated in the methodology chapter, the reliability coefficient of the scale to measure behavioral intentions was very high, which can indicate too much internal consistency (Briggs & Cheek, 1986). The items of this scale were selected and composed together aiming to measure the full construct of behavioral intentions, but according to the reliability analysis, it is possible that the information measured with this scale was too redundant.
Finally, some variables were not taken into the model. Technology acceptance, including perceived usefulness and perceived ease of use, and perceived enjoyment are factors that have been widely studied with regards to the relationship between retail technologies and shopper's perceptions/reactions. These factors were not included into the model for this particular study, because the limited time frame did not allow for this research to develop a particular kind of set-up in which these variables could be adequately tested.
40 6.5 Future Research
Because this study was unable to find sufficient support for the conceptual model, it is highly recommended that future research should be undertaken. Regarding the proposed limitations, future research could generate different outcomes.
First of all, the direct effects of satisfaction and value perception on behavioral intentions are confirmed in this study. However, no mediating role value perception has been found with retail technologies as a predictor variable. One explanation for this to happen is that the concept of value perception possibly was not clearly defined in this research. The concept of value perception might be extended for future research, by adding the shopping value dimensions, to understand the relationship between retail technologies and perceived value better.
Second, significant direct effects of covariates age and education level were found for value perception as well as satisfaction. Subsequently, the results of this study found a
significant direct effect of gender on behavioral intentions. Since the relationship between retail technologies and demographical factors like age, gender, and educational background have yet to be investigated in academic literature, future research can possibly clarify this relationship.
As stated before, some variables were not included in the conceptual model. Prior research indicates that technology acceptance is an important factor for customer’s perception of retailers implementing technologies (F. Davis, 1998). When time and resources are
sufficient, it might be an interesting topic for future research to examine the moderating role of product type on the relationship with factors like perceived usefulness and perceived enjoyment.
41 Another variable that was not included in the model is service quality. Multiple
studies reported that service quality can be an important determinant for customer’s
satisfaction or value perception level (Lin & Hsieh, 2011; Parasuraman, Zeithaml, & Berry, 1988). Also, Cronin et al. (2000) consider service quality, next to value and satisfaction, as a significant predictor for behavioral intentions. Future research is needed to investigate a potential mediating role for service quality on the relationship between retail technologies and behavioral intentions.
Finally, as stated in the literature review, retail technologies affect the customer experience. The customer experience is a multidimensional construct, involving “cognitive, emotional, behavioral, sensorial, and social components” (Lemon & Verhoef, 2016, p. 74). Because of the complexity of this concept, it remains difficult to measure. Over the past years, researchers have attempted to develop a robust and reliable measurement scale (Lemon & Verhoef, 2016). However, a validated scale has yet to be accepted and implemented in general scientific (and managerial) literature. Up to present day, researchers and managers try to measure (parts of) the customer experience by using validated metrics like satisfaction, loyalty and service quality. Future research on how to ultimately measure the customer experience and its dynamics is still needed, to investigate the role that technologies have in it.
42
Literature
Aguinis, H., & Bradley, K. J. (2014). Best Practice Recommendations for Designing and
Implementing Experimental Vignette Methodology Studies. Organizational Research Methods,
17(4), 351–371. https://doi.org/10.1177/1094428114547952
Babin, B. J., Darden, W. R., Griffin, M., Darden, W. R., & Griffin, M. (1994). Work and / or Fun : Measuring Hedonic and Utilitarian Shopping Value Work and / or Fun : Measuring Hedonic and Utilitarian Shopping Value. Journal of Consumer Research, 20(4), 644–656.
Bernstein, F., Song, J. S., & Zheng, X. (2008). “Bricks-and-mortar” vs. “clicks-and-mortar”: An equilibrium analysis. European Journal of Operational Research, 187(3), 671–690. https://doi.org/10.1016/j.ejor.2006.04.047
Blázquez, M. (2014). Fashion Shopping in Multichannel Retail: The Role of Technology in
Enhancing the Customer Experience. International Journal of Electronic Commerce, 18(4), 97– 116. https://doi.org/10.2753/JEC1086-4415180404
Bono, J. E., & McNamara, G. (2011). Publishing in “AMJ” - Part 2: Research Design. The Academy
of Management Journal, 54(4), 657–660.
Briggs, S. R., & Cheek, J. M. (1986). The role of factor analysis in the development and evaluation of personality scales. Journal of Personality, 54(1), 106–148. https://doi.org/10.1111/j.1467-6494.1986.tb00391.x
Chih‐Hung Wang, M. (2012). Determinants and consequences of consumer satisfaction with self‐ service technology in a retail setting. Managing Service Quality: An International Journal,
22(2), 128–144. https://doi.org/10.1108/09604521211218945
Chitturi, R., Raghunathan, R., & Mahajan, V. (2008). Delight by Design: The Role of Hedonic Versus Utilitarian Benefits. Journal of Marketing, 72(3), 48–63. https://doi.org/10.1509/jmkg.72.3.48 Clodfelter, R. (2010). Biometric technology in retailing: Will consumers accept fingerprint
43 authentication? Journal of Retailing and Consumer Services, 17(3), 181–188.
https://doi.org/10.1016/j.jretconser.2010.03.007
Cohen, J. (1988). Statistical Power Analysis for The Behavioral Sciences (2nd ed.). Hillsdale, New Jersey: Lawrence Erlbaum.
Collier, J. E., Moore, R. S., Horky, A., & Moore, M. L. (2015). Why the little things matter: Exploring situational influences on customers’ self-service technology decisions. Journal of
Business Research, 68(3), 703–710. https://doi.org/10.1016/j.jbusres.2014.08.001
Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16(3), 297–334. https://doi.org/10.1007/BF02310555
Cronin, J. J., Brady, M. K., & Hult, G. T. M. (2000). Assessing the effects of quality, value, and customer satisfaction on consumer behavioral intentions in service environments. Journal of
Retailing, 76(2), 193–218. https://doi.org/10.1016/S0022-4359(00)00028-2
Dabholkar, P. A., & Bagozzi, R. P. (2002). An Attitudinal Model of Technology - Based Self - Service : Moderating Effects of Consumer Traits and Situational Factors. Journal of the
Academy of Marketing Science, 30(3), 184–201.
Dabholkar, P. A., Michelle Bobbitt, L., & Lee, E. (2003). Understanding consumer motivation and
behavior related to self‐scanning in retailing. International Journal of Service Industry Management (Vol. 14). https://doi.org/10.1108/09564230310465994
Davis, F. (1998). Perceived Usefulness , Perceived Ease of Use , and User Acceptance of.
Management Information Systems Quarterly, 13(3), 319–340.
Davis, F. D. (1989). Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology. Information Technology MIS Quarterly, 13(3), 319–340. https://doi.org/10.2307/249008
