Exploring Choice Overload,Internet Shopping Anxiety,Variety Seeking and Online Shopping Adoption Relationship: Evidence from Online Fashion Stores

Introduction

In this age of technological advancement, an increasing number of people are logging on to the Internet and as such, an increasing amount of business moves online as well. The number of Internet users in the world has increased more than eight times since the beginning of the twenty-first century and has reached to 2.3 billion users (Internet World Stats, 2012). Retail e-commerce transaction volume in the USA increased more than five times since the year 2000 and reached 165.4 billion USD in 2010 (U.S. Commerce Department, 2012). In 2013 alone, 191.1 million US citizens were engaged in online shopping and had at least once browsed products, compared prices or bought merchandise online (Statista, 2014). Even today, Internet shopping continues to grow as the Internet penetration and usage rates increase around the world. Across the globe, China and Indonesia ranked first and second, respectively, as the countries with the highest overall annual online retail growth rates (Statista, 2014), while global business-to-consumer (B2C) e-commerce sales are expected to reach 1.77 trillion US dollars by 2015 (Statista, 2014). Given the increased reach of the Internet, online shopping is penetrating the lives of people as a new shopping channel. According to Kim, Park and Pookulangara (2005), 78 per cent of US shoppers purchase from both online and offline stores.

Although more than 85 per cent of world’s online population has ordered goods over the Internet during recent years and online shoppers accounted for over 40 per cent of adult population in Japan, Norway, the UK, South Korea, the Netherlands, Denmark and Australia (ACNielsen, 2008; OECD, 2008), yet online retail sales still account for a small percentage of total retail sales even in these top online shopping countries. Attractiveness, control, efficiency and helpfulness are the most important factors for a pleasurable online shopping experience. Furthermore, features that enhance a sense of freedom and control, such as email notifications, saving information and tracking a purchase, also motivate a purchase and therefore patronage (Yang, 2011). Since Internet is fast becoming such a powerful medium of shopping, the need to study just what affects the consumers’ intentions to patronize e-stores demands attention.

Even though Internet is fast becoming the new channel of shopping, it is not without its weaknesses. The anxiety faced by those who are not computer-savvy restricts the growth of online shoppers. An overload of information regarding products to be bought online may affect a shopper’s intention to buy online. Therefore, Internet shopping anxiety related with online shopping is a significant factor that needs further investigation. Although researchers have looked into different aspects related with online shopping, a lot still needs to be explored before conclusive results can be drawn. Therefore, the present study undertakes to examine information overload, anxiety and variety-seeking behaviour of shoppers as important constructs that need investigation in order to understand consumers’ online shopping behaviour which may further lead to patronage intentions thereby making it an emerging and relevant area for future research.

Review of Literature

Theoretical Framework of the Study

A large number of studies in the e-tailing literature target at exploring and explaining the relationships between the risks, benefits, technology usage-related factors and consumers’ shopping intentions associated with online marketing (Chang & Wu, 2012; Rishi, 2010; Turan, 2012). Considering that India has recently joined the list of nations using Internet for shopping, little is known about consumers’ behaviour in adopting this new shopping channel and the factors that influence this behaviour. Consequently, in the present study, we specifically attempt to understand the relationships and linkages among consumers’ variety-seeking behaviour, choice overload, online shopping anxiety and e-store patronage intentions.

Donthu and Garcia (1999) showed that the average Internet shopper seeks convenience, is innovative, is impulsive and seeks more variety. Variety-seeking shopping behaviour is when a consumer tries a different brand in a category simply for the ‘utility inherent in variation’, that is, not because the new brand has better attributes, but because the change feels good (Van Trijp, Hoyer & Inman, 1996) and since variety-seeking sometimes leads to permanent brand switching, it has become an area of concern for the managers. Van Trijp et al. (1996) point out that true variety-seeking behaviour is intrinsically motivated, not driven by external pleasures (Hoyer & Ridgway, 1984; McAlister & Pessemier, 1982). Psychologically, variety-seeking behaviour is positively valued by consumers because of its contribution to the underlying processes of relief from boredom with the choice task, relief of attribute satiation and satisfaction of curiosity (Van Trijp et al., 1996). Again, the customer follows a specific plan, based on the need for something different, and as a consequence makes a new or different purchase.

When making buying decisions, consumers are often unable to evaluate all available alternatives and sometimes use a two-stage process to make the decisions. At the first stage, consumers typically screen a large set of available products and identify a subset of the most promising alternatives. Subsequently, they evaluate the latter in more depth, perform relative comparisons across products on important attributes and make a buying decision (Haubl & Trifts, 2000). The major purpose of such information search is to support decision-making and product choice in a sense that information search could enhance the quality of choice outcomes (Bettman and Michel, 1979). Standard microeconomic theory implies that more choice cannot make rational people worse off, especially if they have different tastes. Among other things, more variety offers the possibility of better matches and greater flexibility (Chernev, 2011). Bawa (1990) and Seetharaman and Chintagunta (1998) found that people display variety-seeking in frequently purchased products. It is thus natural to think that firms must provide greater variety as a strategy to take advantage of consumers’ hybrid purchase behaviour and online shopping is a perfect way in which firms can offer a wider variety of products that can be attractive to consumers.

Although ‘more is more’ has been a long-held opinion in traditional economic theory, however, recent studies have shown that it may no longer be true when the choice set becomes too large, known as choice overload (Iyengar & Lepper, 2000; Schwartz, 2004). In this sense, if enough varieties are offered in a certain brand to satisfy consumers’ variety-seeking desire, will the consumers then be less likely to switch to other unknown brands for novelty or would they experience choice overload?

Evidence has been found in psychology, economics and marketing literature that too many choices lead to less satisfaction with the decision (Schwartz, 2004), reducing the likelihood of decision (Anderson, 2003), lowering the quality of decision (Besedes, Deck, Sarangi & Shor, 2012; Kaiser, 2011) and do not necessarily result in higher consumer surplus (Lucarelli, Prince & Simon, 2012). While the human desire for choice is infinite, human ability to manage more and more choices is limited. Although it is commonly believed in marketing that a large assortment is beneficial for consumers because it provides for a better match between the consumers’ own preferences and the product offering, recent research findings suggest that increasing the size of choice set may have adverse consequences on the strength of preferences because it can confuse consumers, increasing the probability of delaying their choice or not choosing at all (Iyengar & Lepper, 2000). As a result, the decision-maker cannot make as decision rationally (Simon, 1997).

We argue that when a consumer is confronted with a large number of choices to choose from, initially the consumer may enjoy the availability of choices as it tends to satisfy the variety seekers legitimate curiosity, but eventually the consumer seems to get confused while taking the final decision not for want of ability to search effectively or lack of comprehension, but for the presence of too much information. Information overload has been studied in a variety of contexts, beginning arguably with Miller (1956) who concluded that short-term memory has the capacity to process about seven items. For anything beyond that amount, a coping strategy would need to be employed. Bettman (1975) proposed that ‘bounded rationality’ leads to an information-processing approach. The bounds include limited memory and computational capabilities that keep consumers from accessing a predetermined set of preferences. Instead, consumers ‘construct’ their preferences when confronted with the alternatives, picking, to some extent, the best option for that moment, and thus challenging a consumer’s confidence in the decision.

Jacoby, Speller and Berning (1974), while sparking some of that early debate, raised the concern that too much information on product packages, along with too many brand choices, could cause consumer confusion, especially in time-pressed situations. Malhotra (1982) too provided empirical evidence that ‘dysfunctional’ outcomes occur in situations with too much information, whether in the form of choices or attributes of those choices leading to choice overload—a mental state in which the amount of choice information that needs to be processed exceeds the committed cognitive capacity of the decision-maker.

Based on the above discussion, we further argue that experiencing choice overload is more likely to be encountered by someone who has been exposed to a lot of choices and therefore has a lot of information that needs to be processed before taking a decision. By virtue of being a variety seeker, such a consumer will experience choice overload as against someone who is not a variety seeker and is therefore excused from too many choices to choose from. Consequently, we expect that variety-seeking shoppers are likely to exhibit choice overload when confronted with too many choices (H1).

Although Internet offers consumers with more variety and convenience, risks, anxiety and drawbacks concerning Internet shopping have taken their place in consumer minds (Toa, Liaob & Linc, 2007). In online shopping, the fearful traditional mindsets of Indian shoppers hold the myth that what you see may not be what you get as the ‘touch–feel–hear’ experience is attached with trust, security and privacy concerns (Dutta, 2012; Mann & Sahni, 2011). To add to it is the problem with online grievance redressal and complaint follow-ups in case of delivery of wrong or defective items ordered online or delayed delivery creating a sense of negative emotion towards the whole process. Studies regarding consumers’ emotions, especially negative emotions including anger, hate, resistance, fear and anxiety, have developed for a long time (Menon & Dubé, 2007). Among these, anxiety is the most widely studied (Norris, Pauli & Bray, 2007). Anxiety is considered to be either a state of mind representing an individual’s short-lived negative emotional response to a stimulus (or a situation) or an individual trait referring to an individual’s constant condition of being under tension regardless of a stimulus (or a situation) (Gilbert, Lee-Kelley & Barton, 2003).

Although anxiety has often been discussed in literature on consumer behaviour (Viswanathan, Rosa & Harris, 2005), computer use (Durndell & Haag, 2002) and Internet use (Joiner, Brosnan, Duffield, Gavin & Maras, 2007), it has been relatively understudied in the context of e-tailing service. Research has proved that online shoppers tend to suffer from purchase anxiety more than offline shoppers (Gehl, 2007). After all, when you buy something over the Internet, it is often a product you have never seen before sold by a person you have never met before. In this context, several researchers have examined negative emotions associated with using technology or computer (Rosen & Weil, 1995; Tseng, Tiplady, Macleod & Wright, 1998). Among these negative emotions, computer anxiety is the most widely discussed (Brosnan, 1998; Meier, 1985; Norris et al., 2007). Computer anxiety is defined as a transitory condition of being fearful, apprehensive, intimidated, uneasy and aggressive when interacting or considering interaction with the functional (software) and mechanical (hardware) aspects of computers (Brosnan, 1998). Most prior research considered causes of computer anxiety in relation to computer use only (Brosnan, 1998; Meier, 1985; Norris et al., 2007), while some researches indicated that computer anxiety is a direct determinant of consumers’ attitudes towards using and intentions to use a computer (Hackbarth, Grover & Yi, 2003; Wu & Li, 2007).

Similar to the concept of computer anxiety are the concepts of Internet anxiety and Internet shopping anxiety which may seem similar; however, we draw a distinction between the two. Internet anxiety refers to the negative emotions that result from an individual’s general experiences while using the Internet and web-related services only, whereas Internet shopping anxiety refers to the negative emotions that result from an individual’s experiences with online shopping in particular arising from any component (either web related or non-web related) of the online shopping process. For example, if consumers are afraid that they will not receive products that they purchased online, this may increase their Internet shopping anxiety. Further, Internet anxiety develops across time and as a result of multiple situations, and thus can be measured in terms other than specific usage experiences. In contrast, Internet shopping anxiety develops as a result of online shopping transactions and thus can be measured after having such experiences. Also online shopping is an economic activity, and if something goes wrong in the process, consumers stand to lose time and money (Çelik, 2011; Kim & Shim, 2002). A number of studies, however, suggest that usage of the Internet as a shopping medium creates additional uncertainty and risk perceptions due to its intangible nature, which further heightens online shopping anxiety (Kim & Forsythe, 2008). Since the existing legal regulations and procedures concerning e-tailing are not sufficient to protect the customers against the risks attached to making transactions with virtual vendors in the world, customers may exhibit a high level of anxiety towards online shopping which, in turn, decreases the salience of their assessments about how easy it is to use. When the user experiences anxiety, the frequency of online shopping will decrease which in turn would reduce their e-store patronage (Asakawa & Okano, 2009).

However, Mochon (2013) documented single-choice aversion and stated that searching through several choices can reduce anxiety by making consumers feel like they have searched exhaustively and hence are ready to buy. Researchers suggest unique characteristics of the process of online search, such as greater opportunity for consumers to realize their purchase fantasies (Venkatesh, 1998), thereby reducing anxiety related to lost opportunities (Alba et al., 1997). Therefore, we expect that the variety-seeking consumer is less likely to experience shopping anxiety (H2).

The success of a web store as a feasible sales channel depends on whether it helps in attaining a significant number of potential customers who are willing to make a purchase online. It would be inaccurate to say that every individual who visits a store will feel the same way towards the shopping experience in general and web-based shopping in particular. Therefore, a customer’s willingness or intention to patronize a store will depend, in part, on the shopping experience. Mathwick, Malhotra and Rigdon (2001) refer to patronage intention as ‘the customer’s willingness to consider, recommend, or purchase from a retailer in the future’. Early conceptualizations of behavioural intentions towards a store/brand were mainly focused on willingness to buy (Dodds, Monroe & Grewal, 1991). Later studies, however, viewed patronage intention as a multidimensional construct including willingness to recommend, willingness to buy and shopping likelihood as dimensions of store patronage intentions (Baker et al., 2002). Similar to offline store patronage intentions, online continuance shopping or e-store patronage intention too is characterized by the same three dimensions.

Also, consistent with the concept of e-store patronage, stickiness, a recent term that is specifically related to websites, refers to anything about a website that encourages a visitor to stay longer (Lin, 2007). A sticky website not only gets visitors to stay longer but also encourages them to frequently return to the same site and therefore, for an online retailer, it is associated with repeat purchase behaviour and ultimately with patronage.

However, with the number of retail websites growing daily and new consumers discovering Internet shopping every day, it is becoming more difficult for retailers to retain customers online. While in the traditional retail settings switching brands may cost consumers time, effort and money, in online shopping the cost to consumers for switching brands is very low as consumers have access to a selection of stores far beyond their local shopping area. In order to accomplish this, whereas before a customer had to physically travel to different stores to look for variety, now the Internet offers a variety of options at the click of a mouse. Accordingly, statistics suggest that 87 per cent of Internet shoppers visited multiple websites before making a purchase (Corbin, 2008).

Since the Internet facilitates search for variety without extra effort or spending extra money, it is only likely for the consumers to exhibit variety-seeking behaviour (Hoyer & Ridgway, 1984; Van Trijp & Steenkamp, 1992). Although such behaviour may benefit the consumers, as they can compare and contrast the same product on several different sites which may not be possible in an offline environment, retailers stand to lose business as research suggests that variety seeking may have a negative effect on patronage (Berné, Múgica & Yagüe, 2001; Oliver, 1999). Variety seekers get bored with products very easily and tend to switch to alternative offerings or try new ones (Trivedi & Morgan, 2003; Van Trijp & Steenkamp, 1992). Given the consumers’ enhanced ability to access and compare multiple offerings by different providers on the Internet (Rohm & Swaminanthan, 2004), there is increased possibility of a variety seeker to purchase online (Donthu & Garcia, 1999), resulting in increased e-store patronage. We therefore expect that variety-seeking behaviour will have a significant positive relationship with e-store patronage (H3).

However, availability of a lot of variety may have its own downside. A considerable amount of research has focused on whether and when consumers will experience the negative consequences of choosing from large assortments (Scheibehenne, Greifeneder & Todd, 2010). Extant research about the effect of assortment size on consumer decision-making has resulted in contradictory empirical findings. One body of research indicates that consumers prefer large assortments particularly due to the increased likelihood of finding a preference match from a larger assortment when compared to a smaller set (Hoch, Bradlow & Wansink, 1998). On the contrary, existing research in the domain focuses on the negative consequences of ‘too much choice’, such as the perceived increase in decision difficulty, reduction in satisfaction levels and increased levels of potential regret due to large assortment sizes (Chernev, 2011; Turri, 2012). A pioneering ‘Jam Study’ (Iyengar & Lepper, 2000) has brought to light the negative consequences of increased assortment sizes on final decision-making such that the purchase likelihood among consumers decreases with an increase in the number of options to choose from, leading to choice overload.

Given the strong desire that consumers have for more choice (Chernev, 2011), it may not be possible or desirable to reduce the number of options consumers consider. However, as consumers process these additional options in their consideration set, they are likely to experience an increase in decision difficulty. An increase in the number of e-commerce sites further results in choice overload for potential buyers, making it difficult for buyers to select appropriate sites for making purchases (Afuah & Tucci, 2000). In other words, increased availability of a large number of products on websites leads to choice overload making it difficult for customers to make purchases online (Lee, Jin & Choi, 2011). Consequently, due to an increase in the consideration sets, the final decision stage becomes more difficult resulting in decreased frequency of online shopping.

However, despite the traction gained by this counter-intuitive idea in the academic arena in the past decade, theoretical explanations about the phenomenon are still nascent and the empirical evidence mixed. Researchers have been unable to demonstrate that the phenomenon occurs in all contexts. Further, the debate about the role of choice set size has been reignited by the recent meta-analysis done (Scheibehenne et al., 2010), whose results question the generalizability of the phenomenon. Therefore, we argue that specifically in the context of online stores, despite the availability of innumerable choices to choose from, the online shopper has the choice of stopping the search process at anytime and resume the search process if one is unable to take a quick decision. This is possible because of the convenience factor inherent in online stores due to anytime, anywhere shopping, a factor missing in offline stores. In other words, since consumers prefer large assortments to choose from, availability of choice (Asakawa & Okano, 2009) in online stores may act as a significant factor resulting in increased patronage of online stores. Consequently, we expect that choice overload will exhibit a significant positive relationship with e-store patronage (H4).

Furthermore, despite the dazzling growth of Internet as the latest sales channel in retail segment due to its distinct advantages for both consumers and retailers, statistics show that there is a huge gap between the number of Internet users and those using Internet as a shopping channel. One of the biggest reasons for this inconsistency is that a large number of consumers and Internet users experience hesitation while using this new channel for purchase purposes (Hannah & Lybecker, 2010; Passyn, Diriker & Settle, 2011). Growing fraudulent activities on the net combined with the absence of personal contact, physical product (service) evaluation and transaction security/privacy protection in online shopping environment have contributed towards the increased hesitation among consumers. Researchers suggest that since the usage of Internet as a shopping medium creates additional uncertainty and risk perceptions due to its intangible nature, it further heightens online shopping anxiety (Kim & Forsythe, 2008). Since individuals with higher levels of anxiety have lower levels of self-efficiency to overcome the system’s operation hurdles, anxiety increases the mental effort expended on Internet usage (Igbaria, Pavri & Huff, 1989). We therefore expect that anxiety associated with Internet shopping will have a negative effect on patronage intentions (H5).

Objectives and Scope of the Study

Based on the extensive review of literature on variety-seeking behaviour and its relationship with choice overload, online shopping anxiety and e-store patronage intentions, the present study undertakes a synoptic view of online shopping with an intention to understand the relationship among the constructs within the ambit of the Indian online shoppers. In this regard, undertaking the present study is justified in order to fill the existing gap in literature by examining the interaction among these variables. Therefore, the study seeks to fulfil the objective of examining the relationship between variety-seeking behaviour and choice overload, online shopping anxiety and e-store patronage intentions. The study is confined to the study of online shopping behaviour of fashion products by the Indian consumers and aims at analyzing the relationship between e-store patronage with choice overload and online shopping anxiety.

Research Methodology

Our method of research consists of two major parts: First, a theoretical analysis based on literature and articles from experts and researchers. Second, we moved towards a more pragmatic approach by gathering accurate information about the constructs and their underlying factors. In order to get a true insight of the customer’s perspective, we completed a consumer survey. Data were analyzed using structural equation modelling (SEM) approach to verify the hypotheses formed.

Data Source and Sample Frame

A sample of 250 respondents based on convenience was surveyed. We selected respondents by asking them a filter question: ‘Have you ever shopped online?’ Only those respondents who replied yes were selected.

Data Analysis

A total of 189 valid responses were analyzed to study the impact of choice overload and Internet shopping anxiety on online shopping adoption. The demographic profile of the respondents can be seen in Table 1. Most of the respondents (about 43 per cent) were in the age group of 20–30 years. The rest were in the age group of 30–40 years (about 37 per cent) and 40–50 years (about 18 per cent). Also majority of the respondents were females (about 57 per cent) and 71 per cent of the respondents were postgraduates.

The Results of the Measurement Model

Structural equation modelling of this study examined the two levels of analysis, the measurement model and the structure model, and their results are shown below. The means, standard deviations and correlation matrix are shown in Table 2. In Table 2, there are positive correlations among patronage intention, choice overload, shopping anxiety and variety-seeking behaviour.

OPEN IN VIEWER

Table 1.

Demographic Characteristics of Respondents

S. No.	Respondent’s Characteristics	% of Respondents
1	Age Group
	20–30	43.21
	30–40	37.04
	40–50	19.75
2	Gender
	Male	43.2
	Female	56.8
3	Education
	Graduate	29
	Postgraduate	71

Source: Authors’ findings.

OPEN IN VIEWER

Table 2.

Means, Standard Deviation and Correlations of the Constructs

Constructs	Mean	Standard Deviation	A	B	C	D
A. Patronage intention	4.89	1.19	1
B. Choice overload	4.41	1.38	0.576**	1
C. Shopping anxiety	3.79	1.48	0.416**	0.616**	1
D. Variety-seeking behaviour	5.16	1.29	0.582**	0.379**	0.390**	1

Source: Authors’ findings.

Note: **p < 0.01.

The factor analysis of the four constructs is shown in Table 3. Every construct in this study can be classified into only one factor. The study referred to the previous studies to design questionnaire items. Before administering the questionnaire to the respondents, this study employed two pre-tests for the questionnaire revisions. Therefore, the measurement of this study is acceptable in content validity. Besides, there are two measurements to confirm the reliability of the constructs. First, one measure of the reliability is to examine the loadings of each constructs’ individual items. With respect to the quality of the measurement model for the sample, the loadings of all items of the four constructs listed in Table 3 are significant. Second, Cronbach’s alpha is the other measure of the reliability. Table 4 lists Cronbach’s alpha for the constructs. In general, the minimum requirement of Cronbach’s alpha coefficient is 0.7 (Hair et al., 1998). It can be observed that the Cronbach’s alpha coefficient of ‘patronage intention’ is 0.744; that of ‘choice overload’ is 0.724; that of ‘shopping anxiety’ is 0.768; and that of ‘variety-seeking behaviour’ is 0.837. Because the Cronbach’s alpha coefficients of all four constructs are more than 0.7, the measurement of this study is acceptable in reliability.

In addition, it is also important to verify whether the validity of the measurement in this study is acceptable. There are two measurements to confirm the validity of the constructs. First, this study applied Fornell and Larcker’s measure of average variance extracted (AVE) to access the discriminative validity of the measurement (Fornell & Larcker, 1981). The AVE measures the amount of variance captured by the construct through its items relative to the amount of variance due to the measurement error.

In order to satisfy the requirement of the discriminative validity, the square root of a construct’s AVE must be greater than the correlations between the construct and other constructs in the model. For example, the square roots of the AVEs for the two constructs, patronage intention and choice overload, are 0.930 and 0.791, respectively, in Table 4 which are more than the correlation between them, 0.576, in Table 2. This demonstrates that there was adequate discriminative validity between the two constructs. The square roots of all constructs’ AVEs in Table 4 of this study are all greater than the correlations among all constructs in Table 2. Therefore, the discriminative validity of the measurement in this study is acceptable. Second, if the AVE of a construct is greater than 0.5, then it means that there is convergent validity for the construct. As shown in Table 4, the AVEs of the four constructs are 0.865, 0.791, 0.854 and 0.918, respectively, which are all greater than 0.5. It indicates that there is convergent validity in this study. Thus, the measurement of this study is acceptable in discriminative validity and convergent validity. According to several tests of reliability and validity, it demonstrates that there is adequate reliability and validity in this study.

OPEN IN VIEWER

Table 3.

Factor Analysis

Constructs	Number of Items	Number of Factors	Accumulation Percentage of Explained Variance
A. Patronage intention	7	1	21.6
B. Choice overload	5	1	40.5
C. Shopping anxiety	3	1	56.6
D. Variety-seeking behaviour	3	1	69.1

Source: Authors’ findings.

OPEN IN VIEWER

Table 4.

Item Loadings and the Construct’s Cronbach’s Alpha and AVEs

Constructs	Items	Item Loadings	Cronbach’s Alpha	AVE	Square Root of AVE
A. Patronage intention	PI1PI2PI3PI4PI5PI6PI7	0.7730.7010.6490.6140.6080.5160.511	0.744	0.865	0.930
B. Choice overload	CO1CO2CO3CO4CO5	0.6910.6800.5910.5890.548	0.724	0.627	0.791
C. Shopping anxiety	SA1SA2SA3	0.6860.6470.566	0.768	0.731	0.854
D. Variety-seeking behaviour	VSB1VSB2VSB3	0.6940.6560.542	0.837	0.844	0.918

Source: Authors’ findings.

Test of the Proposed Model

A SEM technique was used to test the model. AMOS Ver.17 was employed for this purpose. The observed variables used to predict the latent variables in SEM were obtained by processing the data in the instrument. Results of SEM analysis indicate that the model offers a good fit to the data.

Seven fit indexes which are commonly used in the literature (chi-square/degrees of freedom, goodness-of-fit index [GFI], adjusted goodness-of-fit index [AGFI], non-normed fit index [NNFI], comparative fit index [CFI], root mean square residual [RMSR], root mean square error of approximation [RMSEA]) were employed to test the model fit. The commonly used measures of model fit, based on results from an analysis of the structural model, are summarized in Table 5. According to Schumacker and Lomax (2004), chi-square/degrees of freedom less than 3, GFI, NNFI, CFI greater than 0.9, an AGFI greater than 0.8, RMSR less than 0.1 and RMSEA less than 0.06 or 0.08 are considered indicators of good fit. As seen in Table 5, all goodness-of-fit statistics are in the acceptable ranges.

OPEN IN VIEWER

Table 5.

Summary Statistics of Model Fit

Fit Index	Recommended Values	Observed Values
Chi-square/degrees of freedom	≤3.00	2.35
GFI	≥0.90	0.94
AGFI	≥0.80	0.83
NNFI	≥0.90	0.91
CFI	≥0.90	0.90
RMSR	≤0.10	0.08
RMSEA	≤0.08	0.07

Source: Schumacker and Lomax (2004).

Note: GFI—goodness-of-fit index; AGFI—adjusted goodness-of-fit index; NNFI—non-normed fit index; CFI—comparative fit index; RMSR—root mean square residual; and RMSEA—root mean square error of approximation.

The graphical presentation of results is shown in Figure 1 along with the standardized path coefficients. Figure 1 illustrates the significant relationships among the study variables. All five paths estimated are significant. Hypothesis 1 postulates that variety-seeking behaviour has a significant positive relationship with choice overload. The direct path from variety-seeking behaviour to choice overload is significant since the regression coefficient is 0.41 with t = 9.34 and p < 0.05. Therefore, the hypothesis that increased variety-seeking behaviour will have a positive impact on choice overload is supported. Also the direct path from variety-seeking behaviour to online shopping anxiety is significant since the regression coefficient is –0.56 with t = 3.02 and p < 0.05. The second hypothesis that variety seeking-behaviour has a significant negative relationship with online shopping anxiety is accepted. The third hypothesis is also accepted because the direct path from variety seeking to e-store patronage is significant since the regression coefficient is 0.47 with t = 2.91 and p < 0.05. Therefore, the hypothesis that variety-seeking behaviour has a significant positive relationship with e-store patronage is accepted.

Further, the fourth hypothesis is supported in that the direct path from choice overload to e-store patronage is significant since the regression coefficient is 0.42 with p < 0.05. This indicates that choice overload leads to increased patronage intentions. Finally, the fifth hypothesis that shopping anxiety has a significant negative relationship with e-store patronage is supported since the path from shopping anxiety to e-store patronage is significant with regression coefficient equal to –0.44 with p < 0.05. The negative sign indicates that higher levels of shopping anxiety lead to lower levels of patronage intentions.

OPEN IN VIEWER

Figure 1.

Source: Authors’ findings.

Conclusion

E-stores, where buyers place orders over the Internet, have emerged to become a popular shopping channel. The revolutionary way in which the Internet has changed the way in which firms market and sell their products has affected not only the marketers but also the consumers. While the firms are thinking of new and evolutionary ways of presenting their offerings to their customers, the consumers are also exhibiting purchase behaviour never before seen in traditional ways of shopping. Speculations have arisen regarding the success of these new channels of distribution and whether shoppers will accept a new way of shopping online.

As one of the first attempts to study online shopping in the context of e-store patronage behaviour, we developed a model to show that variety-seeking behaviour can be used as a key criterion when explaining choice overload and Internet shopping anxiety, which in turn affects patronage intentions of online shoppers. In this context, the purpose of the study was to examine the extent to which consumers seeking variety while shopping online would experience an overload of the innumerable choices offered to them and whether the availability of large assortment of choices will have a significant effect on the patronage of e-stores providing choice in their product offerings. The study also hoped to examine the anxiety experienced by consumers while shopping online and the resultant effect of shopping anxiety on consumers’ intention to patronize web stores.

Results of our study reveal that the variety-seeking variable is a significant predictor of choice overload and shopping anxiety experienced by online shoppers. One can attribute a number of reasons for this result to be true. One of the motivations for consumers to shop online is the convenience with which one can jump from one e-tailer to the other in search for variety. Although this may help in achieving the motive of convenience at the doorstep for a variety-seeking consumer, it will also result in the consumers making a more difficult decision of making a selection from among too many options. Also, since consumer memory has limited capacity to process information (Miller, 1956), it is assumed that beyond a specific number, even a variety-seeking consumer is likely to experience information overload and hence choice overload. Therefore, the more the number of choices made available to shoppers, the more likely are they to experience choice overload, and finally after making a purchase decision a consumer is likely to regret it for want of computational capacities that consumers need for assessing their preferences. The direct implication of the finding is that online retailers need to devise strategies such that a limited choice is made available in a particular product category for a limited time so that consumers can make informed choices and at the same time are satisfied with the variety made available. In such a case, the variety-seeking shopper will be exposed to variety in small assortment sizes which will be made available over a period of time instead of a lot of variety, all at the same time. Therefore, both the objectives, of making variety available and reducing choice overload, can be achieved.

Furthermore, although a number of studies suggest that usage of Internet as a shopping medium is likely to increase the level of anxiety, results of the present study suggest that consumers exhibiting variety-seeking behaviour while using Internet as the shopping medium are less likely to feel anxiety while shopping online. The answer to this is possibly rooted in the fact that customers who are variety seekers have more experience of the online shopping technology and therefore may also have less anxiety towards using it for their shopping tasks. Additionally, variety seekers are likely to be more involved with this medium of shopping, subjecting it to greater elaboration than those who are not variety seeking and therefore have less technology interface causing anxiety while using it.

It is also argued that variety-seeking behaviour of consumers will result in increased levels of their patronage towards e-stores. Consistent with the findings of Donthu and Garcia (1999), the present research suggests that the relationship between variety-seeking behaviour and e-store patronage behaviour is positive such that a variety-seeking consumer will buy from and recommend web stores. There is repeated empirical evidence that consumers are convenience oriented and the unprecedented convenience offered by the Internet, allowing anytime, anywhere shopping, makes online shopping a very powerful engine especially for those who seek variety. Therefore, an online shopping medium would encourage a larger number of Internet users to shop online resulting in increased patronage intentions.

Furthermore, findings of the present research indicate that the availability of large consideration sets in an online shopping environment helps in improving the quality of decisions made by increasing the likelihood of finding the right match. Finally, Internet shopping anxiety is negatively associated with e-store patronage intentions. This is consistent with the findings of Chaudhuri (2002) that suggest that anxiety is a negative emotion which is associated with the perception of greater risk. Since online shopping deals with money, a possibility of financial loss increases the level of anxiety resulting in a decreased desire to patronize e-stores for shopping purposes. Unlike the brick-and-mortar stores where a consumer can touch, feel and experience the products before making a final purchase, shopping from online stores does not allow such authorization. However, it is too early to make a generalization of this finding to all product categories since in case of some of the product categories such as books, the risks involved are less when compared to making a purchase of fashion products, as was the case in the present study.

Limitations and Future Research

Due to the respondents negligence in filling the questionnaire, many responses had to be dropped. Moreover, due to time and money constraint only some areas in India were selected. In the future, studies may be carried out expanding the areas and also increasing the number of respondents. Moreover, an experiment-based study may be carried out to see the impact of increased number of choices on consumers and how it affects patronage intentions. Also other variables may be included, such as hedonic and utilitarian shopping orientation, complaint follow-ups in case of delivery of wrong or defective items ordered online or delayed delivery and privacy concerns.