This document discusses how perceived product risk influences consumers' preferences for different types of online retailers, or "e-tailers". It hypothesizes that consumers will prefer retailers with lower perceived product risk and that prestigious store-based e-tailers (like Bloomingdales.com) will be preferred over value-oriented store-based (like Walmart.com) and pure online retailers. The study aims to provide insights into how perceived economic and psychosocial product risks affect patronage of different e-tailer formats.

1. The Influence of Perceived Product Risk on Consumers’ e-Tailer
Shopping Preference
Pradeep A. Korgaonkar Æ Eric J. Karson
Published online: 18 May 2007
Springer Science+Business Media, LLC 2007
Abstract Increasingly, retailers are combining Internet
and store based operations to become ‘‘multi-channel’’ as
they attempt to attract and retain customers. This study
investigates how the type and level of perceived product
risk (specifically economic and psychosocial risk) influence
patronage preference for shopping from three types of
e-tailers. The e-tailer formats studied are: pure play
e-tailers (e.g., Overstock.com), value-oriented store based
e-tailers (e.g., Wal-Mart.com), and prestigious store based
e-tailers (e.g., Bloomingdales.com). The hypotheses,
based upon prior research in the area of perceived product
risk, show that type and level of risk do matter. Further,
e-tailers linked with prestigious stores have an advantage
over both other e-tailer types. Results also show an inter-action
between perceived product risk and the e-tail format.
Based on samples from the Northeast and Southeast
USA, the results are found to be similar in these diverse
regions, improving the generalizability of the findings.
Keywords Risk Internet e-Tailing Online shopping
Store patronage
Introduction
The Internet has clearly revolutionized the way consumers
acquire and process, and marketers disseminate, informa-tion.
As online retail sales continue to increase at a slower
pace than expected, practitioners, and academics alike are
still searching for factors that influence consumer prefer-ence
for shopping on the Internet. Although published re-search
exists related to consumer Internet shopping, little is
known about how consumers shop from stores that have
added web sites to their ‘‘brick and mortar’’ retailing
(e.g., Jarvenpaa Todd, 1996–97; Jones Vijayasarathy,
1998). As technology increases consumer shopping alter-natives,
research is needed to uncover how the web sites of
multi-channel retailers such as Eddie Bauer compare vis-a-vis
pure play Internet retailers such as Shopzilla.com.
Specifically, this study attempts to provide insight into
which products are preferred by consumers using a par-ticular
e-tailer format.
Research to date suggests that perceived risk is likely to
be useful in understanding a variety of online consumer
behaviors, including e-tailing patronage (Donthu Garcia,
1999; Ha, 2002). Still, little is known about how risk per-ceptions
influence patronage among the major variants of
Internet store formats e.g., pure play Internet retailers such
as ShopNBC.com, value oriented discount store based
‘‘click and mortar’’ retailers such as Target.com, or pres-tigious
department store based ‘‘click and mortar’’ retailers
such as Saksfifthavenue.com, henceforth called pure play,
value CM, and prestigious CM, respectively. Although,
past studies have investigated product categories that are
best suited for Internet retailing in general (e.g., Cheskin
Research and Studio Archtype/Sapient, 1999; Girard,
Silverblatt, Korgaonkar, 2002; Peterson, Balasubrama-nian,
Bronnenberg, 1997), published research on the
topic of different types of e-tailers is scant. Taking
advantage of the rich perceived risk paradigm literature,
this study empirically tests whether value and prestigious
CM e-tailers have an advantage over strictly pure play
P. A. Korgaonkar
College of Business, Florida Atlantic University, University
Tower, 220 S.E. 2nd Avenue, Fort Lauderdale, FL 33301, USA
e-mail: Korgaonk@fau.edu
E. J. Karson ()
Department of Marketing, Villanova School of Business,
Villanova University, Villanova, PA 19085-1678, USA
e-mail: eric.karson@villanova.edu
123
J Bus Psychol (2007) 22:55–64
DOI 10.1007/s10869-007-9044-y

2. 56 J Bus Psychol (2007) 22:55–64
e-tailers depending on the product risks perceived by on-line
shoppers for each e-tailer type.
Specifically, this study first tests whether consumers’
overall preference for online shopping differs based on the
perceived product risk. Second, we hypothesize that overall
shopping preference is highest for prestigious CM
e-tailers followed by value CM e-tailers, and the lowest
for pure play e-tailers, regardless of the product risk. Third,
hypotheses about the interaction effects of online e-tailer
types and perceived product risk types (e.g., economic risk
and psychosocial risk) on preference for shopping online
are formed and tested.
While we acknowledge that many factors are likely to
influence online shopping behavior, our focus on risk ex-tends
previous research on Internet retailing as risk is likely
to serve as a ‘‘catch-all’’ for consumers’ reservations to-wards
Web shopping, or their preference for one type of
e-tailer over another. As the perceived product risk concept
has been fruitful in explaining consumers’ choice of
products, retailers, catalog/telephone shopping, and inter-net
shopping in the past, we feel the extension of this
simple and useful concept will aid in understanding and
explaining patronage preferences between the three types
of e-tailers as well.
Product risk
Past studies suggest that the usefulness of the Internet as a
shopping medium is closely linked to the product that
consumers intend to purchase. For example, Rosen and
Howard (2000) propose what they term as e-potential for
different products to be sold on the Internet. Proponents of
the transaction cost paradigm suggest that product features
will influence transaction costs and, as such, play a key role
in e-tailer selection (e.g., Benjamin Wigand, 1995). We
emphasize the perceived risk paradigm in our study as
the focal point of discussion as, for many consumers,
buying from the Internet is a new way of buying. In fact, a
May 2003 study by International Demographics, Inc.,
shows only 22.5% of US households were regular Internet
purchasers in 2002, making over four purchases. As a result
many consumers who buy online are often insecure and
perceive risk. This risk has two main sources: (a) risk re-lated
to the types of product purchased, and (b) the risk
associated with the type of online merchant they are pur-chasing
from.
Since Bauer’s (1960) seminal work, several studies in
marketing have explored the concept of perceived risk to-wards
understanding patronage behavior. The concept of
perceived risk has been used to explain and predict tradi-tional
store based shopping preferences as well as in-home
shopping behavior (e.g., Akaah Korgaonkar, 1998;
Spence, Engel, Blackwell, 1970). Studies by Dowling
and Staelin (1994), Shrimp and Bearden (1982), White and
Truly (1989), among others, suggest that perceived risk
toward a product category is inversely related to purchase
intentions. The literature also strongly suggests that con-sumers
are reluctant to patronize a retail store when they
are uncertain of the risks associated with purchase (Prasad,
1975).
There are different types of risks perceived by con-sumers.
Based on the early work of Roselius (1971) and
Jacoby and Kaplan (1972) the literature generally identifies
six different types of risks: financial, performance, physi-cal,
psychological, social, and time loss. However, rarely
are all six different types of perceived risk incorporated
into a single study. Given that Stone and Gronhaug (1993),
in their study of various dimensions of perceived risk,
suggest that the financial and psychosocial dimensions of
risks captured the majority of the overall risk perceptions
(compared to the time, performance and physical dimen-sions
of risks), and in line with past research (e.g.,
Korgaonkar, 1982; Prasad, 1975), we have selected the two
types of perceived product risk most likely to influence
behavior in an e-tail situation: economic and psychosocial.
Each is defined as follows: Economic risk refers to how the
choice of a product will affect the individual shopper’s
ability to make other purchases. Thus, it varies with the
financial considerations of price in relation to factors such
as the shopper’s income, ability to pay, and alternative uses
of money. Psychosocial risk relates to how the purchase
decision will affect the opinions other people hold of the
shopper. Thus, it varies with such factors as the social
conspicuousness and social relevance of the product.
In addition to the fact that economic and psychosocial
risks are reported to be more relevant than other types, we
maintain that the risk dimensions of time, product perfor-mance,
and physical dimensions of the product remain
largely invariant across the three e-tailer formats. In other
words, regardless of the type of outlet selling a product, the
features, performance, and physical dimensions of the
product do not change. Similarly, given the widespread
availability of overnight and express delivery and tracking
systems from companies such as Federal Express, UPS,
and USPS, except for rare situations, product acquisition
time also remains fairly consistent. Based on past research,
we suggest:
H1: Consumers will prefer low risk versus high-risk
products when shopping online. Hence,
(a) Consumers will prefer products of low levels of
psychosocial risk versus products of high psychoso-cial
risk when shopping online; independent of eco-nomic
risk.
123

3. J Bus Psychol (2007) 22:55–64 57
(b) Consumers will prefer products of low levels of
economic risk versus high economic risk when
shopping online; independent of psychosocial risk.
Literature suggests that, in addition to the level of risk,
the type of risk perceptions also influence shopping pref-erences.
Studies such as Perry and Hamm (1969), Peter and
Tarpey (1975), and Prasad (1975) suggest that the type of
product risk influences purchase decisions. Derbaix (1983)
and Mitchell (1999), studying food, cars, and TVs, show
that economic risks play an important role in influencing
purchasing decisions, while for clothing they suggest that
psychosocial risks play an important role in buying deci-sions.
Based on these past studies we hypothesize that:
H2: Consumers patronage preferences will be influenced
by the type of risk (economic and psychosocial) regardless
of the level of risks.
Online retail store type
Although a limited number of empirical studies exist on the
role consumer risk perceptions play in the selection of
traditional and non-store shopping channels, little attention
has been devoted to the role of perceived risks in online
shopping format preference. Due to this limited research,
the hypotheses of this study are based on the extant
empirical findings on the level of consumers’ perceived
risk in traditional in-store shopping channels (i.e., depart-ment,
specialty and discount retail stores, catalog show-rooms)
and non-in-store shopping channels (i.e., catalog
orders by mail, telephone, or internet. Namely: Bhatnagar,
Misra, Rao, (2000); Cox Rich, (1964); Festervand,
Snyder, Tsalikis, (1986); Hisrich, Dornoff,
Kernan, (1972); Korgaonkar, (1982); Korgaonkar
Moschis, (1989); Miyazaki Fernandez, (2001); Prasad,
(1975); Spence et al., (1970)). Results from non-online
retail studies indicate that the ‘‘perceived risk of a product
is transferable to the store that sells the product’’
(Korgaonkar, 1982, p 78). Previous findings also suggest
that in-store shopping is perceived as less risky than tele-phone
or mail catalog order shopping (Cox Rich, 1964;
Festervand et al., 1986; Spence et al., 1970).
We propose that the work of Bettman (1973) provides a
theoretical base for prior findings. Bettman posits that risk
has two components: inherent risk that is endemic to a
product class, and handled risk, a clear derivative of
inherent risk, that varies with the amount of additional
information available about the purchase. When looking at
different retailer types (e.g., department stores, specialty
stores, discount stores, and non-in store retailing) each
certainly presents different types and amounts of infor-mation
to consumers. In the literature reviewed, generally,
the amount and type of information was greater for in-store
than ‘‘at-home’’ retailers, likely explaining the drop in
consumers’ perceptions of handled risk and increase in
shopping preference for those stores with a physical pres-ence.
The same is expected to hold true for e-commerce. In
general, we hypothesize that e-tailers without a physical
presence will be perceived as riskier places to shop com-pared
to clicks and mortar e-tailers.
Further, while Internet shopping does allow for 24/7 ac-cess,
easier price comparisons, and the ability to find rare
products, along with many other benefits, these advantages
are offset by a number of concerns. Among these concerns
are: privacy and security of the medium (e.g., Korgaonkar
Wolin, 1999; Liebermann Stashevsky, 2002), lack of
familiarity or experience with certain online retailers, and
generally, the risks associated with the intangible nature of
online shopping. Patronage of a pure play e-tailer such as
eBay poses the additional risks of getting a defective/
damaged product, delayed product arrival, the products not
matching descriptions posted on the seller’s Web site, etc.
Conversely, a store-based internet operation such as
Sears.com allows the consumer to physically check the
merchandise prior to purchase, or easily exchange or return
the merchandise to the store after purchase. Additionally a
physical presence provides a variety of available tangible cues
such as product displays supplemented with POP material, as
well as quality cues to help reduce perceived risk prior to and
post-purchase. Thus the store based CM e-tailers are able to
offer the best of both worlds and reduce the risks associated
with shopping from a pure play e-tailer. Therefore:
H3: Consumer’s shopping preferences will be the lowest
for pure play e-tailers compared to store based CM
e-tailers when shopping online, independent of product
risk.
Based on prior literature (Grewal, Iyer, Levy, 2004),
we further speculate that among store-based CM
e-tailers, the prestige CM e-tailers, with their stronger
brand reputations, will be perceived as less risky than value
CM e-tailers, as their brand equity communicates a better
selection of quality merchandise, as well as superior cus-tomer
service versus discount stores, again reducing han-dled
risk. Given this, our study predicts that online
shoppers will perceive the lowest risk for shopping from a
prestigious CM Web site, medium amounts of risk
shopping from a value CM Web site, and the highest risk
shopping from a pure play e-tailer.
The following hypotheses, drawing on the increased
levels of handled risk different e-tail formats allow, suggest
that consumers’ online shopping preference will vary by
the type of Internet store independent of the type of per-ceived
product risk. Specifically,
123

4. 58 J Bus Psychol (2007) 22:55–64
H4: Consumers will prefer prestige CM e-tailers over
value CM e-tailers when shopping online, independent of
perceived product risks.
H5: Consumers will prefer value CM e-tailers over
pure play e-tailers when shopping online, independent of
perceived product risks.
Interaction effects between product risk and online
store-type risks on online shopping preferences
Two way interactions
Past studies in traditional, as well as online, retailing sug-gest
that retailer type and product type have important
influences in determining retail patronage (e.g., Darden,
1979; Jones Vijayasarathy, 1998; Kling Palmer, 1997;
Sheth, 1983). As expected, the interaction between product
and retailer type is supported in studies incorporating the
congruency concept in e-tailing (De Figueiredo, 2000;
Jahng, Jain, Ramamurthy, 2000), as patronage should
decrease as perceptions of risk increase. Previous research
has suggested that perceived product risks affect prefer-ences
not only for retail store selection, but between
product categories as well (Bhatnagar et al., 2000;
Miyazaki Fernandez, 2001).
Based on these studies, we predict that online e-tail store
type and product risk interact as consumers choose the type
of Internet retailer they prefer to shop from. Simply stated,
for different levels of perceived risk in varied product cate-gories,
consumers will prefer different types of e-tailers, with
store type, and the information this signals, being more
important the greater inherent risk a product class has.
Compared to store based e-tailers, pure play e-tailers are
likely to pose higher economic, as well as psychosocial,
risks because of the limited information consumers can get
about these stores through physical inspection. Because
consumers are unable to personally experience/evaluate the
product or service prior to purchase, products that are high
in economic and/or psychosocial risk will be least preferred
by shoppers on pure play sites. Stated another way, pure
play e-tailers will have higher shopping preferences only
when risks are perceived to be low. Formally:
H6: Online shopping preference will be the highest for
pure play e-tailers for products with low inherent (eco-nomic
as well as psychosocial) risk.
Turning to the effects of economic risk on retailer
preference, online shopping from the Web site of value
CM e-tailer should be preferred when it lowers economic
risk. In other words, value CMs’ positioning helps
‘‘handle’’ economic risk. Specifically, shoppers for high
economic risk products will view discounter’s ‘‘value ap-peal,’’
a common discount/value store strategy, as lowering
the economic risk. This is evidenced as the online suc-cesses
of value CM Web sites such as Bestbuy.com and
Wal-Mart.com, etc. are partly attributable to their capacity
to offer low prices, especially for expensive products. This
leads to the following hypothesis:
H7: For high economic risk products, online shopping
preference will be the highest for value oriented discount
store e-tailers.
Finally, the small number of research studies that have
investigated the role of perceived product risk in the selec-tion
of a shopping channel (Bhatnagar et al., 2000; Forsythe
Shi, 2003; Hisrich et al., 1972; Korgaonkar, 1982;
Korgaonkar Moschis, 1989; Prasad, 1975) indicate that
when shopping for high social risk products, consumers
perceive a lower amount of risk for department and specialty
stores versus discount stores. We expect similar relation-ships
in the context of online retailing. Online shopping from
the Web site of prestigious CM e-tailers will be appealing
to shoppers for high psychosocial risk products as the
prestigious stores: offer more desirable brands, enable an
authentic view of the merchandise, provide higher security,
and are of superior graphic quality. These, and other po-tential
information cues, should reduce handled risk over
that of value CM e-tailers. Thus we propose that:
H8: For high psychosocial risk products, online shopping
preference will be the highest for prestige CM e-etailers.
Methodology
Pretests
Given our interest on risk perceived across different types
of Internet retailers, and the types of risks (economic and
psychosocial) with different product types as well, pre-testing
was done to establish several categories of products
that would satisfy all four combinations of high or low
economic and psychosocial risk. First, a group of 36 stu-dents
in a public southeastern university were given the
definitions of economic and psychosocial risk. They were
given a four quadrant diagram with high and low category
on one axis and economic and psychosocial risk on the
other axis. Then they were asked to develop a list of
products and/or services that would fit the four possible
categories. That task yielded 42 unique high economic and
high psychosocial risk products or services (henceforth
products), 56 low economic and high psychosocial risk
products, 77 high economic and low psychosocial
123

5. J Bus Psychol (2007) 22:55–64 59
risk products, and 96 low economic and low psychosocial
risk products. After combining like items (e.g., ‘‘cleaning
products,’’ ‘‘cleaning supplies,’’ and ‘‘glass cleaner’’) 34,
40, 45, and 50 major categories of each product, respec-tively,
remained.
Products such as cars, personal services, and tattoos that
appeared in more than one classification were eliminated.
Next, the authors selected 16 products thought to best
typify products in each of the four categories of interest.
This list, shown in Table 1, was then cross validated by
asking 99 respondents from a private northeastern univer-sity
to classify each of the 16 product categories as being
high or low on economic and psychosocial risk. These
ratings supported the classifications of product categories
used for the main study.
Main study
In order to ensure a more diverse sample, two groups of
Internet shoppers were used, one from the southeast (SE),
and one from the northeast (NE). The data were collected
by having undergraduate marketing students conduct face-to-
face survey interviews. The students were told to dis-tribute
surveys to ‘‘persons 18 years or older who are
regular internet users.’’ Students received extra-credit for
distributing the surveys. Two hundred and forty completed
surveys were gathered in the SE, while 276 surveys were
gathered in the NE. While this is, admittedly, a conve-nience
sample, the divergent populations and pre-selection
of Internet users is appropriate given the objectives of the
study. As can be seen from Table 2, the two samples were
only statistically similar on gender. Chi-square tests reveal
(p .05, df = 5) that the NE sample was older (38.5%
were over 44, while in the SE only 16.6% were), and had
much higher income (41.1% of the NE sample recorded
income over $100,000, while only 10.4% of the SE sample
did). As for Internet buying, when asked if they had bought
on the Web in the last 6 months, 85.5% of the NE sample
had versus 74.6% of the SE sample. Further, the NE sample
was more satisfied with their ‘‘most recent online pur-chase’’
reporting a mean of 4.26 (on a scale of 1–5), versus
the SE (3.6).
Survey instrument
In the main study, four different survey versions were
prepared. Not only did this allow for four different
groupings of product categories from Table 1, but coun-terbalancing
the ordering in which product risk categories
were presented. Prior to answering any questions, defini-tions
of e-tailer types in the study, and definitions of the
types of risk, were provided. During the survey respondents
were presented with each of the three e-tailer types one by
one. For each e-tailer type four examples of products for
the four risk combinations we presented and subjects were
asked to indicate their shopping preference for each of the
four types of products on a 1–5 scale (anchored by ‘‘may
never buy’’ and ‘‘may prefer buying’’). The statistical
design was a 3 (Type of e-tailer: Pure play, Value CM
e-tailer, Prestige CM e-tailer) · 2 (Perceived Level of
Economic Risk: Low, High) · 2 (Perceived Level of Psy-chosocial
Risk: Low, High) within subject design. Analysis
of variance was used to test the specific research hypoth-esis.
Table 3 shows the means and standard deviations for
each of the 12 cells for the combined sample, while Fig. 1
shows the overall preferences between e-tail type.
Table 1 Product service
classifications used in final
study
Ninety nine respondents in pre-test
to verify
Product category Economic risk Psychosocial risk
Accessories (friendship bracelet, watch $40, costume jewellery) Low High
Personal grooming (deodorant, cologne, hair care) Low High
Apparel (fabric gloves, plastic sunglasses) Low High
Sundries (bottled water, greeting cards, wallets) Low High
Personal care items (soap, shampoo, toothbrush) Low Low
Office/school supplies (pen, pencil, notebooks) Low Low
Household products (cleaning supplies, detergent, napkins) Low Low
Toiletries (toothpaste, chap stick) Low Low
Home furnishings (floor covering, bedding) High Low
Home appliances (refrigerator, washing machine) High Low
Home entertainment (TV, stereo system, DVD player) High Low
Electronics (digital camera, camcorder, computer) High Low
Online services (education, health care) High High
Formal/dress apparel (dress shirts/blouses, shoes, suit) High High
Jewellery (diamond rings, formal wristwatch) High High
Durables (furniture, cars, boats) High High
123

6. 60 J Bus Psychol (2007) 22:55–64
Analysis and results
Manipulation checks
To verify our manipulations for each of the four surveys,
means were tested for all combinations of high economic
risk (e.g., products with high economic risk and low psy-chosocial
risk summed with products with high economic
risk and high psychosocial risk) versus combinations of
low economic risk, as were high psychosocial risks versus
low psychosocial risks across the three e-tailer types. In all
cases there were significant differences across all com-parisons
of high and low economic and psychosocial risks
at the p .001 level (t-values [economic risks]: 6.439,
8.900, 9.156, 12.352; [psychosocial risks]: 4.795, 19.399,
6.750, 6.197, with between 120 and 140 degrees of
freedom). These tests confirm our manipulations for the 12
product categories used in Table 1.
H1: shopping preference influenced by level of risk
Looking at the significant economic risk by type of e-tailer
(p .001), and psychosocial risk by type of e-tailer
(p .05) interaction in Fig. 2, one can clearly see that
across e-tailer type, products with low psychosocial risks
are preferred over those of high risk (supporting H1a), and
products of low economic risk are preferred over those of
high economic risk, supporting H1b (for prestigious CM
e-tailers there is no significant difference at the p .05
level in the difference between low and high economic risk
products). The direction of the relationship is suggested in
the table of means for the combined sample (Table 3).
3.2
3.1
3
2.9
2.8
2.7
2.6
2.5
2.4
Pure Play
Shopping Preference
Value CM Prestigious CM
Fig. 1 Store preferences, combined sample
The two samples were also analyzed independently for
cross validation (see Tables 4 and 5). As expected, some
differences are noted in the two regions, however, there is
more convergence than divergence among the sample results.
In both samples we find significant main effects for both
psychosocial risk, although its interaction with e-tailer type is
only significant in the NE sample, supporting H1a. On the
other hand, themain effect of economic risk is only significant
in SE sample (p .05), although the NE sample is direc-tionally
correct (mean = 2.38 low, 2.76 high). Again, the
interaction between economic risk e-tailer type is significant
(p .01) in the both samples. These results support H1b.
For H1 we are looking for the main effect of psycho-social
risk and economic risk that occurred in both sam-ples.
Thus, the results of the combined and separate
samples provide support for Hypotheses 1 (level of risks).
H2: shopping preference influenced by type of risk
Figure 3 displays the psychosocial · economic risk inter-action.
Coupled with the significant main effects just
discussed, we find support for H2, that the type of risk does
have an effect on shopping preference, with the possible
exception of economic risk in the NE sample.
H3–H5: difference in store preferences
The ANOVA analysis for both the combined and regional
data shows that the type of e-tailer is significantly related to
shopping preference (p .01), with the sample means also
Table 2 Sample characteristics
Descriptor NE SE
Bought OL in last 6 months 85.50% 74.60%
Satisfaction with last OL buying
experience (1–5)
4.26 3.6
Gender (percent female) 50.40% 52.10%
Household income $50–74,999 $35–49,999
Age 25–34 20–24
N 275 240
Numbers in parenthesis indicate range of categorical answers
Table 3 Shopping preference
means (standard deviations in
parenthesis): combined sample
Cells report mean, standard
deviation
n = 511
Perceived product risk Pure play
e-tailer
Value CM
e-tailer
Prestige CM
e-tailer
Row mean
Low psychosocial, low economic 2.58 (1.351) 3.08 (1.366) 2.96 (1.398) 2.87
High psychosocial, low economic 2.64 (1.284) 2.99 (1.358) 3.14 (1.315) 2.92
High psychosocial, high economic 2.03 (1.215) 2.56 (1.371) 2.91 (1.355) 2.50
Low psychosocial, high economic 2.60 (1.247) 3.12 (1.241) 3.48 (1.180) 3.07
Column mean 2.46 2.94 3.12 2.84
123

7. J Bus Psychol (2007) 22:55–64 61
3.4
3.2
3
2.8
2.6
2.4
supporting Hypothesis 3 about the relative preference for
e-tailer type. For the combined data, the overall shopping
preference mean for pure play e-tailers is 2.46, while it is
2.937, and 3.122 for value CM and prestige CM
retailers respectively. Further, consumer’s shopping pref-erences
are significantly (p .01) lower for pure play
e-tailers compared to clicks and mortar stores. Adjusting
for multiple comparisons, all e-tailers have different levels
of shopping preferences from each other (p .05).
The data also show support for the proposition that the
overall preference for prestige CM e-tailers is higher than
the overall preference for value CM e-tailers regardless
of type of risks (H4) for the combined as well as regional
data (p .01). Similarly, support is found for Hypothesis 5
stating that overall shopping preference is higher for value
CM e-tailers than for pure play e-tailers in all three data
sets (p .01). Figure 1 clearly reflects the overall store
preferences hypothesized.
Hypothesis 6: pure plays preferred under low risk
The hypothesis that higher preferences will be demon-strated
for pure play e-tailers with low economic and/or
low psychosocial risk products is not supported. As seen in
Fig. 2 the results for the total sample show that the pref-erence
for low economic risk products was the lowest for
pure play (2.61), versus prestige CM e-tailers (3.05), and
value CM e-tailers (3.10), largely mirroring overall
preferences just reported. Similarly, for psychosocial risk
the mean preference score for pure plays (2.59) is lowest in
comparison to both value CM e-tailers (3.04) and
2.2
Pure Play
Shopping Preference
Lo Econ
Hi Econ
Lo psychosoc
Hi psychosoc
Value CM Prestigious CM
Fig. 2 Risk · type of retailer interaction, combined sample
Table 4 Analysis of variance
(ANOVA) of patronage
preference by type of perceived
risk (High–Low), and type of
e-tailer (Northeast sample)
a p .001
** p .01
* p .05
Source of variance Sum of
squares
Degrees of
freedom
Mean
square
F
Main effects 4
Economic risk 4.718 1 4.719 1.346
Psychosocial risk 68.123 1 68.123 38.817a
Type of e-tailer 205.276 2 102.638 56.065a
2-way effects 5
Econ risk · Psychosocial Risk 114.198 1 114.198 63.160a
Econ Risk · Type of e-tailer 56.330 2 28.165 32.105a
Psychosocial Risk · Type of e-tailer 4.892 2 2.446 5.123**
3-way effect 2
Econ Risk · Psychosocial Risk · Type of
e-tailer
5.886 2 2.943 5.273**
Table 5 Analysis of variance
(ANOVA) of patronage
preference by type of perceived
risk (High–Low), and type of
e-tailer. (Southeast sample)
a p .001
** p .01
* p .05
Source of variance Sum of squares Degrees of freedom Mean square F
Main effects 4
Economic risk 16.336 1 16.336 4.013*
PsychoSocial risk 34.084 1 34.084 20.761a
Type of e-tailer 303.292 2 151.645 87.672a
2-way Effects 5
Econ risk · Psychosocial risk 37.211 1 37.211 21.219a
Econ risk · Type of e-tailer 9.015 2 4.507 6.786**
Psychosocial risk · Type of e-tailer 1.338 2 .669 1.266
3-way effect 2
Econ risk · Psychosocial risk · Type
of e-tailer
.597 2 .299 .527
123

8. 62 J Bus Psychol (2007) 22:55–64
3.4
3.2
3
2.8
2.6
2.4
prestige CM e-tailers (3.22). A pairwise t-test shows that
the preference scores for each type of risks is significantly
lower (p .01) for the pure play versus the CM
e-tailers. Similar results are found for each region. Thus,
the results are opposite our stated hypothesis, suggesting
that even for low risk products consumers are reluctant to
patronize pure play e-tailers over both clicks and mortar
formats.
Hypothesis 7: higher preference for value CM
e-tailers with high economic risk
The hypothesis suggesting that value CM e-tailers will
be most preferred for high economic risk products was
partially supported. The mean scores of preferences for
high economic risk products for the total sample show that
value CM e-tailers are preferred over pure play e-tailers
(2.78 vs. 2.31) at p .001, but are less preferred over the
prestige CM e-tailers (3.20) at p .001. Similar results
are seen for each of the two regions. In the SE, for high
economic risk products, value CM e-tailers are preferred
over pure plays (2.96 vs. 2.29), but less preferable to
prestige CM e-tailers (3.06, p .01). In the NE, the
prestige CM e-tailers are, once again, most preferred
(3.00) followed by value CM e-tailers (2.61) and the
pure play (2.37, all p .001). Thus, overall, we see that
for high economic risk products, value CM e-tailers are
preferred over pure plays, but not the prestige CM
format.
Hypothesis 8: High psychosocial risks raises shopping
preference of prestige CM e-tailers
Our last hypothesis states that for high psychosocial risk
products, shopping preference will be highest for presti-gious
CM e-tailers. The results for the total sample, as
well as two regions, support this hypothesis. For the total
sample we find that prestige CM e-tailers have the
highest mean score for high psychosocial risk products,
3.02, followed by a mean preference of 2.84 for value
CM e-tailers, and 2.33 for pure plays. Pairwise tests show
the prestige CM e-tailer’s preference is higher than other
e-tailers at p .001. Similarly, in the SE, preference for
prestige CM e-tailers is highest at 3.13 and marginally
higher than the preference for value CM e-tailers (3.00,
p .10), and higher than pure plays (2.28, p .001). Fi-nally,
the NE preference is significantly higher for prestige
CM e-tailers (p .001) than the other two with the mean
preference scores of 3.23 for prestige, 2.69 for value, and
2.35 for pure play, respectively.
Discussion
Although the number and type of firms who sell products
online continues to increase, relatively small numbers of
consumers have embraced the e-tailing alternative (Cheung
Lee, 2001). While e-commerce retail sales in the third
quarter of 2003 reached $13.3 billion, an increase of about
7% over the previous quarter, this still only accounted for
1.5% of total retail sales (U.S. Census Bureau, 2003).
While the press is enamored with the success of Ama-zon.
com, a pure play e-tailer, many other pure plays (e.g.,
eToys.com, Pets.com, Streamline.com and Webvan) have
met with failure and, some would say, helped fuel the
dot.com bust of the early 2000s.
Recognizing these difficulties, pure play e-tailers
increasingly opt for hybrid clicks and mortar approaches in
several product categories such as general merchandise
(Target at Amazon), clothing (with Sears’ acquisition of
Land’s End), travel (Marriott.com), electronics (Best Buy),
etc. It seems that multi-channel retailing is here to stay.
However, few published studies exist exploring which
e-tail format is suitable for various kinds of products. Al-though
scholars have suggested which products are best
suited for selling on the Internet (e.g., Rosen Howard,
2000), little published information is available to e-tailers
of the three formats studied here. Our results suggest that,
overall, pure play e-tailers will continue to have a signifi-cant
disadvantage in comparison to the clicks and mortar
e-tailers, almost regardless of the type or level of inherent
risk. In this study, for each of the four categories of
products surveyed in each of two regions, the preference
for pure play e-tailers was always the lowest. This suggests
that pure play e-tailers have yet to fully earn the trust of
consumers. Additionally, our findings demonstrate the
substantial advantages that brand equity, visibility, and
multi-channel consumer options hold for CM e-tailers
over pure play e-tailers.
2.2
Low Psychosocial
Shopping Preference
Low Econ
High Econ
High Psychosocial
Fig. 3 Psychosocial risk · economic risk, combined sample
123

9. J Bus Psychol (2007) 22:55–64 63
The literature suggests that trust is essential for the
development of e-tailing. At the most basic level, trust
helps address concerns over factors such as privacy and
security essential to online transactions (Cheskin Research
Studio Archtype/Sapient, 1999). Also, trust can mini-mize
feelings of risk and lack of control that are often
characteristic of e-tailing transactions (Bhattacherjee,
2000). Trust becomes especially pivotal in selecting
products or services that are already perceived as risky
(Mayer et al., 1995). As our results clearly demonstrate,
pure-play e-tailers need to overcome these trust issues to
reduce risk, and draw even with CM e-tailers.
Several studies already suggest perceived risks as an
antecedent to trust (e.g., Corbitt, Thanasankit, Yi, 2003;
Tan, 1999). One of the innovative ways of reducing risk
and building trust for pure play e-tailers is by providing
online sales help similar to live salespeople in retail envi-ronments.
Instead of just offering chat rooms as an option
to shoppers, a few e-tailers are monitoring Web shoppers
on their site, looking for opportunities to open a chat
window on the shopper’s screen to ask if they need any
help (Higgins, 2004). Another way of reducing the risk of
purchasing products from pure play e-tailers is to carry
well known brand names. Brands can communicate
valuable information to consumers, especially in online
environments where it is harder to physically inspect
products. Consumers may have personal experience or
knowledge about well known brands, lowering the risk of
purchasing them from a pure play e-tailer. A third way of
reducing risk, if possible, is to build the brand of the
e-tailer itself, either through heavy promotion or creation
of a very large e-tailer, such as Amazon. Generally, con-sumers
are less apprehensive purchasing products from
well known and/or large organizations. Finally, a seal of
approval from an organization such as eTrust may also go a
long way in alleviating consumers risk perceptions of
shopping from pure play e-tailers.
Limitations and future research
While statistics on Internet shopping vary widely, estimates
are that some 60–80% of all US adults are online, with 30–
50% of them buying online (ABC News Poll, 2003;
Pastore, 2001). This is, of course, US adults only, and our
sample, while diverse, does not represent all US shoppers.
Perhaps more significantly, if one looks at statistics
worldwide, one can see that online shopping penetration is
much lower than in the US. EMarketer (2004) reports that
only 16% of Internet users in the EU-15 buy online. Clearly
more representative samples in both the US and worldwide
are called for, with particular attention to the drastically
lower shopping rates in other countries.
Additionally, as this study identifies the challenges pure
play e-tailers face, branding—of either goods or sites, as
suggested—is likely to overcome many of these chal-lenges.
Studies on the effect of well-known versus less-known
brands’ ability to mitigate various risks are certainly
needed, and should provide useful insight as to additional
antecedents of online shopping risks.
Finally, much as catalogers removed perceived risk with
‘‘satisfaction guaranteed’’ pledges years ago, e-tailers must
fully understand all the risks perceived by potential online
shoppers, and how to address them. Once these risks, and
their interactions, are fully understood, consumer seg-mentation
based on online shopping risk perceptions is
possible, as well as insight into how to overcome these risk
perceptions. Given the potential for growth in online
shopping those firms that most fully recognize and address
consumers’ concerns will likely reap great benefits.
Our results show that the prestige clicks and bricks have
an advantage over other e-tail formats for three out of four
product categories, while value CM e-tailers have an edge
over the other two online formats for products with low
economic risks. Thus, the results are largely supportive of
our study hypotheses in two different regions of the country.
This study shows that level and type of perceived risk
provides a good explanation for the congruity approach, and
the importance of handled risk provided by prestige CM
e-tailers. This research helps suggest which types of prod-ucts
are most suitable for the three e-tail formats.
References
ABC News Poll. (2003). 2003 Holiday shopping season: More
shoppers plan to go online. Retrieved November 23, 2005, from
abcnews.go.com/images/pdf/883a36Holiday Spending.pdf
Akaah, I. P., Korgaonkar, P. K. (1988). Conjoint investigation of the
relative importance of risk relievers in direct marketing. Journal
of Advertising Research, 28(August/September), 38–44.
Bauer, R. A. (1960). Consumer behavior as risk taking. In R. Hancock
(Ed.), Dynamic marketing for a changing world: Proceedings of
the 43rd conference (pp. 389–398). Chicago, IL: American
Marketing Association.
Benjamin, R., Wigand, W. (1995). Electronic markets and virtual
value chains on the information superhighway. Sloan Manage-ment
Review, 36(Winter), 62–72.
Bettman, J. R. (1973). Perceived risk and its components: A model
and empirical test. Journal of Marketing Research, 19(May),
184–190.
Bhatnagar, A., Misra, S., Rao, R. H. (2000). On risk, convenience,
and internet shopping behavior. Communications of the ACM,
43(November), 98–105.
Bhattacherjee, A. (2000). Acceptance of e-commerce services: The
case of electronic brokerages. IEEE Transactions on Systems,
Man, and Cybernetics: Part A, 32(July), 411-420.
Cheskin Research and Studio Archtype/Sapient. (1999). E-commerce
trust study. Retrieved from http://www.Sapient.com/cheskin
Cheung, C. M. K., Lee, M. K. O. (2001). Trust in internet
shopping: Instrument development and validation through
123

10. 64 J Bus Psychol (2007) 22:55–64
classical and modern approaches. Journal of Global Information
Management, 9(July–September), 23–36.
Corbitt, B. J., Thanasanki, T., Yi, H. (2003). Trust and e-commerce:
A study of consumer perceptions. Electronic Com-merce
Research and Applications, 2(Autumn), 203–215.
Cox, D. F., Rich, S. U. (1964). Perceived risk and consumer
decision making—the case of telephone shopping. Journal of
Marketing Research, 1(November), 32-39.
Darden, W. R. (1979). A patronage model of consumer behavior. In
R. Stampfl, E. C. Hirschman (Eds.), Competitive structure in
retail markets: The department store perspective (pp. 43–52)
Chicago, IL: American Marketing Association.
Derbaix, C. (1983). Perceived risk and risk relievers: An empirical
investigation. Journal of Economic Psychology, 3(March),
19–38.
De Figueiredo, J. M. (2000). Finding sustainable profitability in
electronic commerce. Sloan Management Review, 41(Summer),
41–52.
Donthu, N., Garcia, A. (1999). The internet shopper. Journal of
Advertising Research, 39(May–June), 52–58.
Dowling, G. R., Staelin, R. (1994). A model of perceived risk and
intended risk-handling activity. Journal of Consumer Research,
21(June), 119–135.
eMarketer. (2004). Internet users in the EU-15 who have ever
purchased products and/or services online by country, September
2003, (March 1), Retrieved from http://www.emarketer.com
Festervand, T. A., Snyder, D. R., Tsalikis, J. D. (1986). Influence of
catalog vs. store shopping and prior satisfaction on perceived
risk. Academy of Marketing Science, 14(Winter), 28–36.
Forsythe, S. M., Shi, B. (2003). Consumer patronage and risk
perceptions in internet shopping. Journal of Business Research,
55(November), 867–875.
Girard, T. R. S., P. K. Korgaonkar (2002). Influence of product
class on preference for shopping on the internet. Journal of
Computer-Mediated Communication, 8 (October), Retrieved
from ascusc.org/jcmc/.
Grewal, D., Iyer, G. R., Levy, M. (2004). Internet retailing:
Enablers, limiters and market consequences. Journal of Business
Research, 57(July), 703–713.
Ha, H. (2002). The effects of consumer risk perceptions on pre-purchase
information in online auctions: Brand, word-of-mouth
and customized information. Journal of Consumer Mediated
Communication, 8(October), Retrieved from ascusc.org/jcmc/
Higgins, M. (2004). Pop-up sales clerks: Web sites try the hard sell.
The Wall Street Journal, (April 15) D1–D2.
Hisrich, R. D., Dornoff, R. J., Kernan, J. B. (1972). Perceived risk
in store selection. Journal of Marketing Research, 9(November),
453–459.
Jahng, J., Jain, H., Ramamurthy, K. (2000). Effective design of
electronic commerce environments: A proposed theory of
congruence and an illustration. IEEE Transactions on Systems,
Man, and Cybernetics: Part A, 30(July), 456–471.
Jarvenpaa, S. L., Todd, P. A. (1996–97). Consumer reactions to
electronic shopping on the world wide web. International
Journal of Electronic Commerce, 1(Winter), 59–88.
Jacoby, J., Kaplan, L. B. (1972). The components of perceived risk.
In M. Venkatesan (Ed.), Proceedings, 3rd annual conference
association for consumer research (pp. 382–393). Chicago, IL:
Association for Consumer Research.
Jones, J. M., Vijayasarathy, L. (1998). Internet consumer catalog
shopping: Findings from an exploratory study and directions for
future research. Journal of Internet Research: Electronic
Networking Applications and Policy, 8(4), 322–330.
Kling, R., Palmer, J. W. (1997). Electronic commerce in retailing:
Differences across retail formats. The Information Society,
13(January–March), 75–91.
Korgaonkar, P. K. (1982). Consumer preferences for catalog show-rooms
and discount stores: The moderating role of product risk.
Journal of Retailing, 58(Fall), 76–88.
Korgaonkar, P. K., Moschis, G. P. (1989). The effects of perceived
risk and social class on consumer preferences for distribution
outlets. In P. Bloom, R. Winer, H. H. Kassarjian, D. L.
Scammon, B. Weitz, R. Spekman, V. Mahajan M. Levy
(Eds.), American marketing association educators’ proceedings:
Enhancing knowledge development in marketing (pp. 39–43).
Chicago: AMA.
Korgaonkar, P. K., Wolin, L. D. (1999). A multivariate analysis of
web usage. Journal of Advertising Research, 39(March/April),
53–68.
Liebermann, Y., Stashevsky, S. (2002). Perceived risk as barriers to
internet and e-commerce usage. Qualitative Market Research,
5(November), 291–300.
Mayer, R. C., Davis, J. H., Schoorman, D. F. (1995). An integrative
model of organizational trust. Academy of Management Review,
20(July), 709–734.
Mitchell, V. (1999). Consumer perceived risk: Conceptualizations
and models. European Journal of Marketing, 33½., 163–195.
Miyazaki, A. D., Fernandez, A. (2001). Consumer perceptions of
privacy and security risks for online shopping. The Journal of
Consumer Affairs, 35(Summer), 27–44.
Pastore, M. (2001). Half of American adults now shop online, (April
24), Retrieved from http://www.clickz.com/stats/market/retail-ing/
article.php/6061_751021
Perry, M., Hamm, C. (1969). Canonical analysis of relations
between socio-economic risk and personal influence in purchase
decision. Journal of Marketing Research, 6(August), 351–354.
Peter, J. P., Tarpey, Sr. L. X. (1975). A comparative analysis of
three consumer strategies. Journal of Consumer Research,
2(June), 29–38.
Peterson, R. T., Balasubramanian, S., Bronnenberg, B. J. (1997).
Exploring the implications of the internet for consumer market-ing.
Journal of the Academy of Marketing Science, 25(Fall),
329–346.
Prasad, K. V. (1975). Socioeconomic product risk and patronage
preferences of retail shoppers. Journal of Marketing, 39(July),
42–47.
Roselius, R. (1971). Consumer rankings of risk reduction methods.
Journal of Marketing, 35(January), 56–61.
Rosen, K. T., Howard, A. L. (2000). E-retail: Gold rush or fool’s
gold? California Management Review, 42(Spring), 72–100.
Sheth, J. N. (1983). An integrative theory of patronage preference and
behavior. In W. R. Darden R. F. Lusch (Eds.), Patronage
behavior and retail management (pp. 9–28). North-Holland, NY,
NY: Elsevier Science Publicator Company.
Shimp, T. A., Bearden, W. O. (1982). Warranty and other extrinsic
cue effects on consumers’ risk perceptions. Journal of Consumer
Research, 9(June), 938–946.
Spence, H. E., Engle, J. F., Blackwell, R. D. (1970). Perceived risk
in mail-order and retail store buying. Journal of Marketing
Research, 8(August), 364–369.
Stone, R. N., Gronhaug, K. (1993). Perceived risk: Further
considerations for the marketing discipline. European Journal of
Marketing, 27(3), 372–394.
Tan, S. J. (1999). Strategies for reducing consumers’ risk aversion in
internet shopping. The Journal of Consumer Marketing, 16(2),
163–180.
U.S Census Bureau, Retail 3Q. (2003). E-commerce report (Novem-ber
21).
White, J. D., Truly, E. L. (1989). Price-quality integration in
warranty premium evaluation: A preliminary test of alternative
models of risk assessment. Journal of Business Research,
19(September), 109–125.
123

11. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.