Bricks or Clicks? Consumer Attitudes Toward Traditional Stores and Online Stores

 

Abstract

Determining what consumers value, and how online stores compare to traditional stores on valued attributes is a necessary first step in understanding the relative benefits of e-commerce.  In this paper, we measure consumers’ valuation of online stores compared to traditional stores by measuring their perceptions of the performance of online stores on 18 attributes, as well as the importance of each of those attributes.  These individual perceptions and preferences from a survey (both web- and paper-based) of 224 shoppers are combined in a self-explicated multiattribute attitude model.  We find that all product categories in our survey of online stores are less acceptable overall than traditional stores. Online stores are perceived to have competitive disadvantages with respect to shipping and handling charges, exchange-refund policy for returns, providing an interesting social or family experience, helpfulness of salespeople, post-purchase service, and uncertainty about getting the right item.  These disadvantages are not entirely overcome by online stores’ advantages in brand-selection/variety and ease of browsing.

Keywords: retailing, online, attitudes

JEL codes: L81, M31, D12

“If a man ...  makes a better mousetrap than his neighbor, tho’ he builds his house in the woods, the world will make a path to his door” — Ralph Waldo Emerson (attributed)

1. Introduction

Do consumers prefer bricks to clicks While, the U.S. Census Bureau reports that retail

e-commerce sales continue to grow, they still represented 4.7% of total retail sales (U.S. Census Bureau, 2013). So what is the future of e-commerce? What is the future of e-commerce?  Do consumers really prefer to buy from traditional retail stores, or do they prefer to shop online?  The answer to this question has significant implications for manufacturers and retailers seeking to establish an e-business, for firms that want to expand their market potential by tapping into customer segments that otherwise would not buy, or for manufacturers who are strategically contemplating dual supply chains (Chiang, Chhajed, and Hess 2003).

 Online stores sell goods and services where the buyer places an order over an internet, extranet, electronic data interchange network, electronic mail, or other online system.  It has been suggested that online retailing is a more convenient shopping channel for consumers because online stores offer greater time-savings (Szymanski and Hise 2000).  Consumers can more easily find merchants, products, and product information by browsing the web, reducing search costs, and eliminating the need to travel.  Thus, consumers may prefer the convenience of online stores compared to traditional stores.  In 2005, however, conventional stores rang up 97.5% of all retail sales compared to e-commerce’s 2.5% share (U.S. Census Bureau 2007a), so certainly convenience is not the only factor influencing consumers’ decisions of whether to buy online or at a traditional store.  Some costs of buying from an online store such as shipping and handling charges, or delayed consumption during the delivery period exceed those costs associated with buying from a traditional store (see Liang and Huang 1998).  The Wall Street Journal (Wingfield 2002) reported that, “Online buyers cite shipping discounts as more likely than any other promotion to encourage them to purchase goods.  Amazon credits free shipping as a key factor in boosting its growth.”  For the 2002 holiday shopping season, 144 merchants on BizRate.com, an online comparison shopping site, offered free shipping to buyers an increase of 31% from the number of online retailers in 2001 (Zimmerman, Merrick, and Tkacik 2002).

Understanding consumer’s acceptance level of online stores appears crucial in a business-to-consumer e-business context.  Determining what consumers value, and how online stores compare to traditional stores on valued attributes is a necessary first step in resolving the bricks or clicks question. 

In this paper, we measure consumers’ valuation of online stores compared to traditional stores by taking into account their perceptions of the performance of online stores on several different attributes, as well as the importance of each of those attributes.  These individual perceptions and preferences are then combined to form what psychologists call a self-explicated multiattribute attitude model (Fishbein 1963, 1967, Meyer and Johnson 1995) or what Keeney (1999) calls a value model.  We then investigate in what ways this online attitude measure varies across the population.

2. Prior Research

            Keeney (1999) interviewed consumers about the pros and cons of Internet commerce and qualitatively categorized their responses into objectives (attributes) such as maximize product quality, minimize cost, minimize time to receive the product, maximize convenience, and maximize shopping enjoyment.  Such “voice of the customer” interviews (Griffin and Hauser 1993) are valuable in identifying the attributes upon which customers distinguish one store-type from another.  Keeney (1999) did not measure consumers’ perceptions of attributes for online and traditional stores nor did he measure the importance of each attribute, but he recognized that consumer attitudes (what he calls values) are critical to understanding online shopping:

The values of prospective customers are a key element in essentially all the major decisions facing any organization involved in or considering being involved in Internet commerce…[A] useful research project associated with quantifying customer values… is an applied research project to develop a sample of customer values for a specific category of products… Then the objectives would be quantified and combined with the quantification of prospective customer objectives.  This would allow the company to simultaneously investigate the implications of proposed… delivery decisions on both the value proposition to the customer and on the achievement of fundamental company objectives (Keeney 1999, pp. 541-542).

As suggested by Keeney, measuring and quantifying customer values is the fundamental issue for companies considering whether to establish an online retail presence.  Our paper addresses this issue in a suitably empirical approach.

            Several studies recently published seek to explain consumers’ acceptance of online shopping.  In an empirical study of consumer willingness to buy from online retailers, Liang and Huang’s (1998) respondents stated that they preferred to buy some products (shoes, toothpaste, microwave oven) from traditional stores and other products (books and flowers) from online stores (although only 28 of the 86 student respondents had online shopping experience).  The authors explained this acceptance of online buying using consumer perceptions of transaction-costs associated with shopping (composed of seven indicators: search, comparison, examination, negotiation, payment method, delivery, and post-service costs), uncertainty (product and process indicators), and asset specificity (site, human, special, temporal, and brand asset indicators).  Missing from their structural equation model analysis are any direct measures of the relative importance of each of these indicators.  Moreover, the structure of their model of online acceptance is under-identified (Fisher 1966, Hess 2002), so their empirical results do not necessarily measure the intended relationships.

            Szymanski and Hise (2000) investigated consumers’ satisfaction with Internet shopping.  They found that greater satisfaction with online shopping is positively correlated with consumer perceptions of the convenience, product offerings, product information, site design, and financial security of an online store relative to traditional stores.  The authors did not experimentally manipulate perception levels, so this correlational study cannot impute causation.  The question of whether perceptions of convenience cause satisfaction or satisfaction causes perception of convenience is left unanswered.  Their survey also does not attempt to measure differences in satisfaction across product categories, nor does it measure consumers’ overall attitude toward online stores compared to traditional stores.  Further, their survey of consumers’ satisfaction with online shopping necessarily excluded people who shop only at traditional stores.

            Degeratu, Rangaswamy and Wu (2000) studied the decision of individuals to use Peapod online grocery shopping.  They gathered a sample of Peapod online buyers and a matching sample of individuals who did their grocery shopping in traditional supermarkets.  As part of their broader study of brand preferences, their random utility model specified an indirect utility function for online versus offline shopping that depended only on the income of individuals.  Perceptions of online grocers versus traditional grocery stores were not measured.  While demographic measures are valuable in describing differences between online versus traditional grocery store buyers, such variables do not address Kenney’s (1999) call to understand and quantify customer values.  A single demographic measure, in contrast to measures of a variety of attribute perceptions, does not provide a very rich answer to the question of why some people shop online and others in a traditional store.

            Bellman, Lohse, and Johnson (1999) analyzed the responses of over 8000 participants in the Wharton Virtual Test Market who completed an initial survey about online buying and attitudes.  Their logistic regression model found that online experience (i.e., web browsing) was the dominant predictor of whether or not the respondent had ever bought anything online.  The survey did not measure respondents’ perceptions or the importance of attribute differences between online and traditional stores.

            Kwak, Fox, and Zinkhan (2002) surveyed chatroom participants via email to discover whether these consumers had bought any of nine products online.  Four broad independent constructs (attitudes toward the Internet, experience with the Internet, demographics, and personality type) explained Internet purchases of these products in logistic regressions.  Unfortunately, four distinct single-variable logit models were estimated rather than a single multivariate logit model with all four variables, resulting in biased coefficient estimates (see Judge et al. 1988, p. 842). 

            All five of these empirical studies are forms of what Urban and Hauser (1980) call “preference regressions” and all share the same problem: the data from all respondents are pooled together and the estimated preference coefficients are assumed equal for all individuals.  Other preference measurement methods have been intensely studied over the past two decades.  Whether a conjoint or self-explication approach is chosen (Srinivasan and Park 1997), or a logit choice model is estimated, heterogeneity must be recognized by allowing the preference coefficients to vary within the population (Andrews, Ansari, and Currim 2002; Andrews, Ainslie, and Currim 2002). 

A study by Levin, Levin and Weller and his colleagues (2005) allowed for heterogeneity among respondents who were surveyed about their shopping preferences for five products.  Their multi-attribute analysis of consumers’ perceptions of nine online and traditional store attributes, and their ratings of the importance of these attributes, found that in general online stores were perceived to be better on the attributes “shop quickly,” “large selection,” and “best price” while traditional stores were rated more highly on “see-touch-the-product,” “speedy delivery,” and “no hassle exchange.”  The attributes “best price,” “no hassle exchange,” “large selection,” and “speedy delivery” were rated as more important than “enjoying the shopping experience,” and “see-touch-the-product.”  There were some differences among the attribute ratings depending upon the product (books, electronic entertainment products, clothing, and computer products).  Attributes contribute heavily to the overall attitude if their perceptions weighted by their importances are large. Unfortunately, these multiplicative combinations were not reported making it impossible to directly compare consumers’ evaluations of traditional stores and online stores by attribute, or to judge relative performance of each type of store on each attribute.

In our study, each respondent’s valuation of online stores is compared to traditional stores by taking into account both their perceptions of the performance of online stores in delivering eighteen attributes, and also the importance of each of those attributes.  Our multiattribute attitude model allows us to measure differences in perceptions (beliefs about the extent to which a store type possesses an attribute) and preferences (the importance of an attribute) among respondents, and to compare consumers’ evaluations of traditional and online stores on each attribute in order to better understand consumers’ acceptance (or lack of acceptance) of online retail stores. 

Specifically, our research addresses the following questions: Do consumers accept online stores as they do traditional stores?  If not, are consumers willing to pay more for products at traditional brick-and-mortar stores than at online stores?  What are consumers’ perceptions of online stores compared to traditional brick-and-mortar stores for a variety of product types?  How do various factors such as product search costs, ability to inspect the product before purchase, shipping and handling charges, or delivery waiting time affect consumer preferences for store type?  When compared to traditional brick-and-mortar stores, what are the relative advantages of online stores?  How do these perceptions and preferences vary within the population?

Our study fits nicely into the Online Shopping Acceptance Model (OSAM) proposed by Zhou, Dai, and Zhang (2007).  Their survey of the literature prompted them to design a conceptual model to explain consumer acceptance of online shopping.  Consumer attitudes, the central concept of their model, directly affect online shopping intentions which lead to online buying.