Who are the Showroomers? Socio-Demographic Factors Behind the Showrooming Behavior on Mobile Devices

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ Who are the Showroomers? Socio-Demographic Factors Behind the Showrooming Behavior on Mobile Devices © 2022 the Authors Published version Holkkola, Matilda; Nyrhinen, Jussi; Makkonen, Markus; Frank, Lauri; Karjaluoto, Heikki; Wilska, Terhi-Anna Holkkola, M., Nyrhinen, J., Makkonen, M., Frank, L., Karjaluoto, H., & Wilska, T.-A. (2022). Who are the Showroomers? Socio-Demographic Factors Behind the Showrooming Behavior on Mobile Devices. In A. Pucihar, M. Kljajić Borštnar, R. Bons, A. Sheombar, G. Ongena, & D. Vidmar (Eds.), 35th Bled eConference : Digital Restructuring and Human (Re)action (pp. 113128). University of Maribor. https://doi.org/10.18690/um.fov.4.2022.7 2022 DOI https://doi.org/10.18690/um.fov.4.2022.7 ISBN 978-961-286-616-7 WHO ARE THE SHOWROOMERS? SOCIODEMOGRAPHIC FACTORS BEHIND THE SHOWROOMING BEHAVIOR ON MOBILE DEVICES Keywords: showrooming, omnichannel, consumer behavior, mobile shopping, sociodemographics. M ATILDA H OLKKOLA ,1 J USSI N YRHINEN ,2 M ARKUS MAKKONEN,1 LAURI FRANK,1 HEIKKI KARJALUOTO3 & T ERHI -A NNA W ILSKA2 1 University of Jyvaskyla, Faculty of Information Technology, Jyvaskyla, Finland. E-mail: [email protected], [email protected] 2 University of Jyvaskyla, Faculty of Humanities and Social Sciences, Jyvaskyla, Finland. E-mail: terhi-[email protected]i, , jussi.nyrhine[email protected]i 3 University of Jyvaskyla, Jyvaskyla University School of Business and Economics, Jyvaskyla, Finland. E-mail: [email protected] Abstract This quantitative study focuses on socio-demographic variables and their associations with different forms of showrooming behavior. The purpose of this study is to find which consumer groups based on age, gender, and income level are demographically the most probable showroomers, and how much each of these variables explain showrooming. The data used is a structured online survey from 1,028 Finnish omnichannel consumers aged between 18 and 75 years. We compare the means of demographic groups’ shares on different aspects of showrooming, and t hen use partial least squares structural equation modeling with confirmatory factor analysis to see how much each of the variables explain showrooming. The findings show that showrooming behavior is explained most by age, and that the most probable showroo mers are younger consumers, higher income consumers and female consumers. The findings also show that finding information and better prices for the products are the most typical forms of showrooming. 114 35TH BLED ECONFERENCE DIGITAL RESTRUCTURING AND HUMAN (RE)ACTION 1 Introduction In today’s omnichannel shopping environment, where the seamless usage of all the networked channels is possible (Rigby, 2011; Verhoef et al., 2015; Srinivasan et al., 2016), using mobile devices for shopping has become popular among many consumer groups. With the emergence of these mobile channels, cross-channel behavior has been increasing (Xu et al. 2014; Srinivasan et al. 2016). This means, for example, the use of one channel for information search and another channel for purchasing the product. Showrooming, i.e. the visiting of offline stores before purchasing online, and/or using the mobile channel while visiting offline stores, is one form of cross-channel behavior. It can thus be considered as a part of omnichannel consumer behavior, where consumers integrate the use of various channels of consumption (Rigby 2011; Verhoef et al., 2015). Although different forms of cross-channel behavior have been studied during the past decades, including, for example, webrooming (Kleinlercher et al., 2020), showrooming has not gained much attention from earlier studies. For example, Burns et al. (2019) call for future research on how demographic factors affect the probability to engage in showrooming behavior. Thus, in this study, we aim to contribute to this call for further research with an aim to describe and compare different demographic consumer groups’ probabilities in engaging in showrooming behavior. We also inspect the prevalence of different forms of showrooming behavior, thus providing a more nuanced insight on different consumer groups’ different behaviors. In the pursuit for this aim, we use quantitative survey data from Finnish consumers collected in 2021. The consumers are reviewed based on their age, gender and income. Our contribution to the omnichannel literature increases the understanding of the associations of demographic factors with showrooming behavior and its different forms. Additionally, the results will help business management to notice the preferences and tendencies of different consumer groups in showrooming behavior. In the second section, we first introduce the key concepts and theories related to this study. In the third section of our paper, we introduce our research data and methods. Next, in the fourth section, we test our hypothesis and analyze the results of this. Finally, we conclude with the fifth section by providing conclusions and further research suggestions having emerged from our study. M. Holkkola, J. Nyrhinen, M. Makkonen, L. Frank, H. Karjaluoto & T.-A. Wilska: Who are the Showroomers? Socio-Demographic Factors Behind the Showrooming Behavior on Mobile Devices 115 2 Showrooming in an Omnichannel Context 2.1 The Concept of Showrooming Showrooming means ”a practice whereby consumers visit a brick-and-mortar retail store to (1) evaluate products/services first-hand and (2) use mobile technology while in-store to compare products for potential purchase via any number of channels” (Rapp et al., 2015). In other words, in showrooming, a consumer gathers information offline but purchases the product online, with a physical store serving as a showroom for online products (Mehra et al., 2013; Brynjolfsson et al., 2013). According to statistics, 57 % of respondents living in the USA and the UK have engaged in showrooming (JRNI, 2019), and 21% of Finnish people, 50% of Swedish people and 43 % of Norwegian people showroomed during the year 2018 (Statista, 2019). The prior research on showrooming has concentrated mainly on asking why consumers are showrooming; what are the drivers for engaging in it (Rapp et al., 2015; Daunt & Harris, 2017; Gensler et al., 2017). Commonly researched customerled drivers for showrooming include, for example, perceived risk, uncertainty, and consumer involvement (Sahu et al., 2021; Balakrishnan et al., 2014). The results, thereby, suggest that the drivers behind showrooming are more complex than just the desire for lower prices in online stores (Gensler et al., 2017). In addition to these drivers, the literature on showrooming has also emphasized the challenges that offline retailers face due to this phenomenon (Fassnacht et al., 2019). According to Rapp et al. (2015), showrooming leads to offline retailers facing “severe consequences”, since the shoppers who are going cross-channel are often noted being irrespective of the change of retailer (Grewal et al., 2016). In addition to the potential sales losses, showrooming has also been shown to negatively influence salesperson self-efficacy and performance (Rapp et al., 2015). Therefore, it becomes important to know who the most probable showroomers are demographically. 116 35TH BLED ECONFERENCE DIGITAL RESTRUCTURING AND HUMAN (RE)ACTION 2.2 Demographic Factors Affecting Showrooming Behavior: Age, Gender, and Income Age. The prior research on age and showrooming has considered age mainly as a control variable. Dahana et al. (2018) found that showrooming frequency was affected negatively by age. Consequently, they found that younger people engaged in showrooming more often than older people. However, their hypothesis of age affecting showrooming probability was not supported. Also, when studying crosschannel free-riding in general, Heitz-Spahn (2013) found that age did not affect the likelihood in these phenomena. However, consistent with Dahana et al.’s (2018) showrooming frequency results, Donnelly and Scaff (2013) found that young adults engage in showrooming more than any other age group. Young showroomers are also suggested to be more driven by mobile and to purchase more via mobile than older showroomers (Schneider & Zielke, 2020). The association of age and the utilization of mobile technologies can also be affected by potential generational differences, which divide consumers into those who have grown up with such technologies and those who have not (Prensky, 2001a; Prensky, 2001b; Fischer et al., 2017). Gilleard et al. (2015) and Madden (2010) have used 50 years’ age as a threshold in comparing the use of mobile technologies of younger and older people. Based on the above, we hypothesize: H1: The older the consumer, the less there is showrooming behavior. Gender. The effect of gender on showrooming has not been studied extensively. Dahana et al. (2018) did not find gender having a statistically significant effect on showrooming. With the wider omnichannel perspective, no statistically significant relationship between gender and cross-channel free-riding (Heitz-Spahn, 2013) nor gender and multi-channel shopping (Jo et al., 2020) has been found. In spite of behavior, when surveying the attitudes towards showrooming, Burns et al. (2019) found that men regarded showrooming as more ethical than did women. Consistently, Schneider and Zielke (2020) found that women showroomers are more loyal than men showroomers, and stick more with one retailer when switching from an offline to an online channel. However, in omnichannel fashion shopping women were found to belong more often to the category of omnichannel shopping enthusiasts and men to the category of omnichannel reluctants (Mosquera et al., 2019). Thus, we hypothesize that: M. Holkkola, J. Nyrhinen, M. Makkonen, L. Frank, H. Karjaluoto & T.-A. Wilska: Who are the Showroomers? Socio-Demographic Factors Behind the Showrooming Behavior on Mobile Devices 117 H2: Women showroom more than men. Income. Similarly with showrooming and age, investigating showrooming and income has generated both statistically significant and not significant results. In the US study by Gallup (2013), 40% of the respondents with lower incomes reported having showroomed at least once, while the percentage climbed to 53% for those with higher incomes. This would suggest that consumers with higher incomes showroom more than those with lower incomes. Similarly, Schneider and Zielke (2020) found that respondents with lower incomes engaged less in showrooming behavior and, when doing so, they preferred online purchasing with stationary devices over mobile purchasing. Lower income adults are also generally suggested to be less likely to utilize internet technologies (Kutner et al., 2006; Schmeida & McNeal, 2007). On the other hand, Jo et al. (2020) found no statistically significant relationship with multichannel shopping and annual income. The contradictory results on whether income has a positive or statistically not significant effect make this hypothesis worth testing. Accordingly, we hypothesize that: H3: The higher the income, the more consumers showroom. 3 Methodology 3.1 Sample and Data Collection Using a structured online survey, we collected data from 1,028 Finnish omnichannel consumers aged between 18 and 75 years. The respondents were selected from a large panel with random sampling. The criteria for selecting the respondents were that they had visited both the online and brick-and-mortar store of the same retailer. The response rate of the invited panelists was 36%. Non-response bias was assessed by comparing the sample to the gender and age distributions of the Finnish adult population. The sample was found representative of the adult population in Finland with respect to gender and age, and the distribution of the other socio-demographic variables was in line with the demographics of the Finnish population. Thus, it can be considered as representative (OSF, 2021a; OSF, 2021b). 118 35TH BLED ECONFERENCE DIGITAL RESTRUCTURING AND HUMAN (RE)ACTION 3.2 Measurements and Data Analysis The respondents of the survey rated three statements measuring showrooming behavior with a 7-point standard Likert scale (ranging from 1=strongly disagree to 7=strongly agree). Respondent were also allowed to not give a rating or leave the questions about their background information unanswered. The statements were: “I often use mobile devices to find more information about products in the store”, “I use mobile devices to find better prices for products online”, and “I use mobile devices to look for information about products while still in the store”. We adopted this established scale from Li et al. (2018), which is consistent with the definition of showrooming by Rapp et al. (2015). The scale was made to fit in the context of this research. Age and annual personal taxable income were measured as ordinal variables with six age and income groups, whereas gender was measures as a binomial variable. These were all used as predictors for showrooming behavior. Next, we first use Welch’s analysis of variance (ANOVA) and independent samples t-tests to examine the differences in the mean ratings of the aforementioned three statements between men and women and across the six age and income groups. If statistically significant differences were found in Welch’s ANOVA, the pairwise differences between the age or income groups were examined in more detail by using the Games-Howell post-hoc tests. After that, we use partial least squares structural equation modeling (PLS-SEM) conducted with the SmartPLS 3.2.7 software (Ringle et al., 2015) to examine how these socio-demographic variables together explain showroom behavior. 4 Results 4.1 Mean Comparisons Generally, the respondents moderately agreed with the statements “I often use mobile devices to find more information about products in the store” (mean 5.00), and “I use mobile devices to find better prices for products online” (mean 4.93). However, the respondents were rather indifferent with the statement “I use mobile devices to look for information about products while still in the store” (mean 4.15). We report the results of the mean comparisons by age groups, gender, and income groups below. M. Holkkola, J. Nyrhinen, M. Makkonen, L. Frank, H. Karjaluoto & T.-A. Wilska: Who are the Showroomers? Socio-Demographic Factors Behind the Showrooming Behavior on Mobile Devices 119 Age. In terms of age, the results of Welch’s ANOVA (Table 1) indicated that the respondents had statistically significant differences across age groups in the use of mobile devices to find more information about products in the store (F(5, 332.905)=19.600***), in the use of mobile devices to find better prices for products online (F(5, 333.505)=20.678***), and in the use mobile devices to look for information about products while still in the store (F(5, 336.378)=39.549***). However, the post-hoc tests indicated that these differences mainly existed only between the respondents aged under 50 years and 50 years or over, with the former group agreeing more and the latter group agreeing less with the statements. Thus, in our further analyses in Section 4.2, we focus only on the differences between these two age groups. This is also consistent with prior literature, in which the age threshold of 50 years has been used, for example, when studying the differences in the use of mobile technologies between younger and older people (Gilleard et al., 2015; Madden, 2010). Table 1: Age and showrooming behavior Age N Mean SD I often use mobile devices to find more information about products in the store. F(5, 332.905)=19.600*** 18–29 y. 30–39 y. 40–49 y. 50–59 y. 60–69 y. 191 213 194 196 179 5.70 5.42 5.28 4.36 4.40 1.49 1.45 1.58 2.04 2.11 ≥ 70 y. 53 4.15 2.09 I use mobile devices to find better prices for products online. F(5, 333.505)=20.678*** 18–29 y. 30–39 y. 40–49 y. 50–59 y. 60–69 y. 191 212 194 198 179 5.66 5.34 5.22 4.38 4.25 1.40 1.49 1.62 2.09 2.16 ≥ 70 y. 53 3.98 2.05 I use mobile devices to look for information about products while still in the store. F(5, 336.378)=39.549*** 18–29 y. 30–39 y. 40–49 y. 50–59 y. 60–69 y. 188 212 193 197 178 5.11 4.88 4.40 3.46 3.04 1.65 1.75 1.77 1.97 1.92 ≥ 70 y. 53 3.13 1.97 Notes: ns=non-significant, *=p<0.05, **=p<0.01, ***=p<0.001 120 35TH BLED ECONFERENCE DIGITAL RESTRUCTURING AND HUMAN (RE)ACTION Gender. In terms of gender, the results of Welch’s t-tests (Table 2) indicated that the respondents had statistically significant differences between men and women in the use of mobile devices to find more information about products in the store (t(1,015.119)=4.428***) and in the use of mobile devices to find better prices for products online (t(1,017.444)=3.142**) but not in the use mobile devices to look for information about products while still in the store (t(1,014.929)=1.461ns). In the case of using mobile devices to find more information about products in the store and using mobile devices to find better prices for products online, women agreed more with the statements than men. Table 2: Gender and showrooming behavior Gender N Mean SD I often use mobile devices to find more information about products in the store. t(1,015.119)=4.428*** Male 496 4.74 1.84 Female 526 5.24 1.82 I use mobile devices to find better prices for products online. t(1,017.444)=3.142** Male 496 4.74 1.86 Female 527 5.11 1.86 I use mobile devices to look for information about products while still in the store. t(1,014.929)=1.461ns Male 497 4.05 1.94 Female 520 4.24 2.04 Notes: ns=non-significant, *=p<0.05, **=p<0.01, ***=p<0.001 Table 3: Income and showrooming behavior Income N Mean SD I often use mobile devices to find more information about products in the store. F(5, 358.955)=1.620ns < 10 k€ 10 k€ – < 20 k€ 20 k€ – < 30 k€ 30 k€ – < 40 k€ 40k€ – < 50 k€ 87 218 180 169 130 4.84 4.77 5.17 5.18 5.18 1.99 1.97 1.74 1.74 1.79 ≥ 50 k€ 107 5.14 1.69 I use mobile devices to find better prices for products online. F(5, 359.338)=1.448ns < 10 k€ 10 k€ – < 20 k€ 20 k€ – < 30 k€ 30 k€– < 40 k€ 40 k€ – < 50 k€ 87 218 180 168 130 4.62 4.76 5.09 5.01 5.03 2.00 1.95 1.87 1.84 1.77 M. Holkkola, J. Nyrhinen, M. Makkonen, L. Frank, H. Karjaluoto & T.-A. Wilska: Who are the Showroomers? Socio-Demographic Factors Behind the Showrooming Behavior on Mobile Devices 127 Prensky, M. (2001b). Digital natives, digital immigrants part 2: do they really think differently? On the Horizon 9(6), 1–6. Rapp, A., Baker, T. L., Bachrach, D. G., Ogilvie, J., Beitelspacher, L. S. (2015). Perceived customer showrooming behavior and the effect on retail salesperson self-efficacy and performance. Journal of Retailing, 91(2), 358–369. Rigby, D. (2011). The Future of Shopping. Harvard Business Review, 89(12), 64–75. Ringle, C. M., Wende, S., Becker, J.-M. (2015). SmartPLS 3. Boenningstedt: SmartPLS GmbH. http://www.smartpls.com, last accessed 5.2.2022. Sahu, K.C., Naved Khan, M., Gupta, K.D. (2021). Determinants of Webrooming and Showrooming Behavior: A Systematic Literature Review. Journal of Internet Commerce, 20, 137-166. Schneider, P. J., Zielke, S. (2020). Searching Offline and Buying Online: An Analysis of Showrooming Forms and Segments. Journal of Retailing and Consumer Services, 52. https://doi.org/10.1016/j.jretconser.2019.101919 Srinivasan, S., Rutz, O. J., Pauwels, K. (2016). Paths to and off purchase: quantifying the impact of traditional marketing and online consumer activity. Journal of the Academy of Marketing Science, 44(4), 440-453. Statista. (2019). Showrooming penetration in the Nordic countries in 2018. https://www.statista.com/statistics/317473/nordic-countries-showrooming/, last accessed 18.2.2022. Verhoef, P. C., Kannan, P. K., Inman, J. J. (2015). From Multi-Channel Retailing to Omni-Channel Retailing. Journal of Retailing, 91(2), 174–181. Xu, J., Forman, C., Kim, J. B., Van Ittersum, K. (2014). News media channels: Complements or substitutes? Evidence from mobile phone usage. Journal of Marketing, 78(4), 97–112. 128 35TH BLED ECONFERENCE DIGITAL RESTRUCTURING AND HUMAN (RE)ACTION Appendix 1: Data description of the respondents. Gender N % Male 497 48.5 Female 527 51.5 Age N % 18–49 years 595 58.1 50–75 years 429 41.9 Annual personal taxable income (€) N % Under 20,000 € 304 34.3 20,000–39,999 € 349 39.3 40,000 € or over 234 26.4 Missing 137 –