The relationship between consumer characteristics

and attitude toward online shopping

Shwu-lng Wu
Department of Business Administration, National Chin-Vi Institute of Technology,
Taichung, Taiwan, ROC

(Ann strong and Kotler, 2000). Because

Keyword.
Electronic commerce, I Introduction attitudes are dilllcult to change, to understand
Consumer behaviour.
The Internet has been widely cited in the consumer attitudes toward online shopping

--
Consumer attitudes, Internet.
Shopping popular press (Miyazaki and Fernandez, can help marketing managers predict the
2001). Reports in 2000 stated that over half of online shopping rate and evaluate the future
all American adults used the Internet (Sefton, growth of online commerce. However,
The purpose of the experiment
reported here was to e'(amine 2000). Moreover. approximately half of the attitudes are developed from personal.
Internet user concerns and current Internet users have purchased [Link] and learning with reality, as well
perceptions of online shopping. products or services online (Sefton, 20(0). as from infonnation, from mends, salespeople
The attitude o11ntemet users
Ernst & Young (2000) reported that 79 percent and news mOOia. They are also deriVed from
toward online shopping was
of non-buyers planed to purchase via the both direct and indirect experiences in Ufe
measored using the Fishbein
model. The relative factors Internet, resulting in increasing online sales. (Loudon and Delli, Bitta, 1993).
influencing user attitudes toward Unlike traditional. media, the Internet It is thus important to recognize that
online shopping and the
encompasses the entire sales process. numerous factors precede attitude fonnation
relationship between the attitude
Marketing campaigns can create awareness and change. Consmner background
and the influence factors were
explored. The results show that then drive consumers all the way through the characteristics are the innately stable
the Fishbein modd can effectively process to actually making a purchase online characteristics of a consumelJs life based
measure consumer attitudes and
(Goodwin, 1999). on the con..'i\lIIreI"s cultural. background.
the examined consumer values and demographics, psychological.,
charSl;teristics were important Attitudes, perceptions and motivations are
itrnuencc fectors on cOn!.iumer not apparent from clicks on banners or online and social. attitudes (Wells and Prensky, 1996).
attitUdes and online shopping purchases, but are an important part of the The object of this r~ch was to explore
decisions.
success or failure of online marketing the consumer characteristics intluencing
strategies (Goodwin, 1999). A person's buying consumer attitudes toward online shopping.
choices are further influenced by fow' major The results from investigating the
psychological. factors: motivation. perception. relationships among online shopping levels,
learning and beliefs and attitude (Annsirong attitude and the relative intluence factors are
and Kotler. 20(0). This is central. to a buyer's presented.
This study begins with an [Link] of
purchase behavior process. These are the tools
Internet User concerns and perceptions
people use to recognize their feelings. gather
regarding online shopping. The attitudes of
and analyze infonnation, fannulate thoughts
Intenlet users toward online shopping are
alld opinions and take action (Wells and
then measured. The relative factors
Prensky, 1996). That means that, through
influencing consumer attitudes toward
motivation, perception and learning, attitudes
online shopping are then explored. The
are fonned and consumers make decisions.
relationship between the attitudes and
Thus, attitudes directly influence decision
intluence factors are discussed with in
making. Attitudes serve as the bridge between conceli wlth the policy issues that sun-ound
consumers' background characteri<rtics and
these attitude jnfluence factors.
the consumption that satisfies their needs.
Attitudes deSClibe a person's relatively
consistent evaluaUons, feelings and tendencies
toward an object or idea. Attitudes put people
I Attitude measure method
Marketing Intelligence & into a frame of mind for liking or disliking Attitude surveys are widely used throughout
Planning things, for moving toward or away from them marketing today. One of the most influential
21/1 [2000J 37-44
,t, MCB UP Limited
[ISSN 0263-4.I03J The Emer~d Rese.,(h Register for this ioomal h ava,abIe at I&:!\ The cu"",1 hsue iIfId lui text archive of this joomal ~ available at
[DO! 10.1I0811l263450031 O458135J 0
[37J
Shwu-lng Wu and widely researched models in the introduced by more factors. Fishbein
The relatjonship between literature is Fishbein's behavioral model responded with the behavioral intentions
consumer characteristics and
attitude toward online
(Bumkrant and Page, 1982). Many marketers models. Many researchers used this model to
shopping and consumer behaviorists have given more measlU-e and accolmt for COIlSumer behavior.
Marketing Intelligence & attention to the Fishbein model (e.g. Harrell and Bennett (1974) reported on a
Planning Woodside and Clokey, 19;4; Bass and comprehensive physician drug prescribing
21/1 [2003) 37·44 behavior study using a national sample of
Talarzyk, 1972; Mazis et ai., 1975; Etter, 1975).
Etter (1975) examined the relationship private practicing physicians. The Fishbein
between Fishbein's attitude model and behavioral inTentions mod ..] was tested and
decision theory models. Lutz (1977) reported cross-validated in this study. Evans (1977)
on two laboratory-type experiments designed applied Fishbein's behavioral intention
to investigate the causal relationships within model and path a.'[Link] to the subject of
the Fishbein model. Fishbein's position was message content assessment. The results
that people form attitudes toward obje..-ts on indicated that the evaluative message was
the basis of their beliefs (perceptions and the most effective and had the most
know)E'.dge) about these objects. Fishbein's influence. BlU-nkrant and Page (1982)
model was construL'ted so that a person's empirically examined issues relevant to the
overall attitude toward some object could be construct validity of Fishbein's behavioral
derived from his beliefs and feelings about intention model The results sUPpolted a
various attitudes about the object. The model in which a single attitude construct
Fishbein model can thus be used as a multi- and single normaiive construct were
attribute attitude measurement model. antecedents of intention.
Fishbein's attitude model can be expressed Because mle of the attitude models stated
in equation form as (Fishbein, 1967a. b): that the conceptual. foundation for marketing
n STUdies is the Ftshbein model (Ahtola, 1975;
Ao = Lbiei, Fishbein, 1963, 196.5, 1967a, b; Fishbein and
icd
Raven, 1962), this study measured the
where: altitudes ofInternet users toward online
Ao = the person's overall attitude toward shopping using the Fishbein model and
object o. explored the relative factors which iniluenced
bi = the strength of his belief that the object consumer attitude toward online shopping.
is related to attribute i (such as the The attitude model proposed by Fi<;.'lbein is
strength of the belief that online somewhat similar to the subjCL'tively expected
shopping is convenience). utility modcls. In this modcl the attitude is a
e, = evaluation or intensity of feelings sum of the evaluative reaction to a salient
property of the product or service and the
toward attribute 1.
n = the number of relevant beliefs for that strength of tIle belief connecting this property
person. to the object or event across all salient
properties (Ahtola, 1975): for example, if in a
The strength (b ,) of each belief can be pilot study some suhJects say: "online
measured on a scale such as the following: shopping is very safe". The problem with this
Online shopping i.. convenience approach is that when the bellef strength (b,)
Measure Likely _ _ _ _ _ Unlikely
is measured by [Link] like: "probable-
(True) 5 4 3 2 1 (False) improbable", "true-false", or "likely-unlikely",
12) (1) (Oli~I)I~21
the evaluaiion (ei) of this is measured by
After obtaining the belief score, the scales like: "good-bad", "very important-not
consumer would be asked to indicate their important at all". In this model the overall
evaluation (ei) of each product or service attitude is a sum of the evaluative reaL'tion to
attribute for which a salient bellef exists. a salient property and the strength of belief
TIle convenience of online shopping is: connecting that property to the attitude,
Good _ _ _ _ Bad object or event across all salient properties.
(Important) 5 4 3 2 1 (Unimportant) When the [Link]· overall attiltldes are
(2) (I) (Ol(~I)(~2) acquired, the relationship hetween the
consumer's characteristics and the attitudes
Each of the consumer's belief scores (b,) is
toward online shopping can be e"lllored.
now multiplied by its respective evaluation
score (ei) and all of the scores for the product
or service attributes are then added, .
producing the consumer's overall attiiude I Consumer characteristics and
regarding this product or service. attitude
New modeling efforts were necessary to Consumer purchases are influenced strongly
account for the additional complexity by cultural, social, personal and
[38)
Shwu-lng Wu psychological characteristics. For the most benefit perceptions are the sum of online
The relationship between part, marketers cannot control such factors, shopping advantages or satisfactions that
consumer characteristics and
attitude toward online but they must take them into account meet an individual's needs or wants.
shopping (Armstrong and Kotler, 20(0). The external Constuner lifestyle is defined as a person's
Marketing Intelligence & influences upon conSlUner behavior include pattern of living. It involves measuring the
Planning demographic, economic, social, situational consumers' major .0.10 dinlensions:
21/1 [2003] 37·44 and technological factors. The internal activities, interests and opiniollS. These four
factors, such as beliefs and attitudes, areas are all important influence factors
learning, motives and needs, personality, toward a consumer's attitude and purchase
perception, and values are involved. The decisions.
lifestyle is between the external and internal
influences on consumer buying behavior,
because it truly mvolves elements of both.
Although external factors have a substantiae
I Methodology
effect on the behavior of buyers, no Jess FrameworK
important are the internal factors (Keegan The conceptualization of the relationship
et aI., 1992). Wells and Prellsky (1996) divided construct L, shown in Figure 1. This
these tmderlying factors into two broad pro-ticular framework for collSlUner attitude
components that Were the key parts of the is affected by the consumer characteristics
framework for consmner analysis: consumer and directly affeets the shopping decision.
backgrolUld characterL<tics, and behavioral According to this framework, the four
processes. Consumer background prois of constuner characteristic.< extend
characteristics are an innate part of a influenL'C upon the consumer's attitude
consumer's makeup. These are the things toward online shopping and direct consumer
that consumers are - the way that purchases. These consumer c1uu:acteristics
individuals describe themselves and the way have a significant relationship with tile
they label others. These cbantcteristics are attitude toward online shopping and the
stable aspects of a consumer's life that cannot attitude toward online shopping has a
be changed. Demographic characteristics, significant relationship with the online
such as gender, age, or ethnic background, shopping rare. The following hypotheses are
are exan1ples of background charac-teristics. offered with assumed collStllI1er' evaluations
Behavioral processes are the motivational, of the relationship between the attitude
perceptual, le:ul1ing, attitude fonnation, and toward online shopping and the other
decision'making tools consumers use to influence variables:
complete the activities that satisfy their Hl. The attitude toward online shopping is
needs. Unlike background characteristics, significantly different based on the
behavioral processes can be affected by a various consumer demographics.
person's environment because they are H2. TIle attitude toward online shopping is
applied on specific occasions. The significantly different based on the Hypo
background characteristics are the influence various consmner ptu-chase preference.
factors of behavioral processes. Marketers HS. The attitude toward online shopping has
and public policy actors are particularly a significant relationship with the
interested in tJlese processes because they eonstllI1ers' benefit perception.
offer opportunities for them to e,xert their H4. The attitude toward online shopping has
influence over constllliers. Because attitudes a significant relationship with the
are easier to change than beliefs or values, consumer lifestyle.
they are often the focus of marketing etfOl-ts H5. The attitude toward online shopping is
to get consmners to buy. significantly different based on the
According to the above theory and various ollline shopping rates.
viewpoint, this study arranged and combined
the influence factors of attitude and Measure
identified the consumer characterL,tics using An initial focus group with ten potential
four areas that are: consumer demographics, customers was conducted to collect original
consmner purchase preference, cOllSumer consumer needs and attitudes associated
benefit perception, and consumer lifestyle. with online shopping. About ISO descriptions
Consumer demographics are the extenla! of benefit needs were COllected. All possible
influence factors that include the consumer's and non-redundant needs obtained from the
gender, age, occupation, education, income, first focus group were recorded as primary
interest, and living area, etc. Consumer needs. The secane! lOCus group, with ten
purchase preferences belong to intemal customers, was used to combine and reduce
factors that include the consumer's purchase the number of prtmary needs. The results
motivation and preference. The consumer generated 40 representative items about the
[39)
 Measure
Shwu-lng Wu respondents' benefit need perceptions and OnJiUf> shopping is cheap
The relationship between True _ _ _ _ _ False
attitudes toward online shopping. A third
consumer characteristics and
focus group, composed of ten Internet u.",r8. 5 4 321
attitude toward online Important _ _ _ _ _ Unimportant
shopping was used to verify the descriptions in arder 5 4 321
Marketing Intelligence & to design a questionnaire concerning the
Planning benefit needs and attitudes for online Onlioe shopping is elfective
21/1 [2003] 37-44 True _ _ _ _ _ False
shopping. FinallY,3B items concerning
benefit needs and attitudes were obtained 5 4 3 2 1
Important _ _ _ _ _ Unimportant
and put into a questionnaire for a random 5 4 321
sampling survey. The SRI value and lifestyles
(VALS) Program (Piirto, 1991) was used to Respondent evaluation scores were
design and acquire 26 lifestyle questions. multiplied by each of the consumer belief
Lifestyle was deImed as a person's pattern of scores and all3B items were added, producing
living. It involves measuring consumers' the consumer's overall attitude.
major AlO dimensions: activities, interests
and opinions. In this study, consumer benefit Sample
needs were measnred using a five-point The primary data from this research were
semantic difference scale, lifestyle were collected using a survey of 600 Internet users
collected using a five-point Likert scale. through personal interviews, Members were
Consumer purchase preferences, randOlnly selected in Taiwan, Because some
responses were not usable, the final sample
demographic data and the online
was 539 for an effective response rate of
shopping rate were assessed using a nominal
89.83 percent. Respondent ages ranged from
scale.
IS to 40 years old. Gender was almost equally
Attitude beliefs toward online shopping
balanced (49,9 percent male, 50.1 percent
were measured using 3B semantie ,lifference
female). EducatiOllleveJs ranged from junior
items on benefit needs and attitUdes that high school to graduate degree. Monthly
respondents evaluated with true/false individual gross income ranged from US$O to
responses on a five-point scale. Attitude US$.'i50 to over US$2,2"L Respondent
toward online shopping in general Was occupations were engineers (32.84 percent) or
meas\U'ed with 38 five-point semantic students (32,10 percent) followed by
differential items requesting respondents to businessman (11.1,0 percent). Their interests
evaluate whether their attitude toward were variant and most lived in the city (58.07
online shopping was impOltantiunimportant percent) followed by villages (30.06 percent).
(Craig et aI., 1994). For example, each subjed These demographic characteristics were
rated profIles with the fonn: similar to those of fnternet users.

Analysis of data
This study used analysis of variance to
FI~re i provide evidence that consumer attitudes
Consumer characteristics, attitude and online shopping
toward online shopping had [Link]
('nn~wncr chanlctclisric~
differences based on consumer
demographics, The results showed that the
Consumer
attihlde toward online shopping had
demographics
significant differences in all the items of the
consumer demographics (p < 0.(15), as shown
in Table 1. This supports Hl.
Using analysis of variance, the online
Consumer shopping attitude Was shown to have
pun:hasc significant differences on two items in the
rrcJcrcm:c consumer pnrchase preference (p < 0.05).
The attitude Online Ii was shown that the attitude toward online
toward onJ inc ~h()ppiag shopping had a SIgnificant relationship with
shopping r.H.: these two consumer purchase items, number
Consumer
benefit of times and payment method. This supports
H2 partly, as shown in Table II.
pcn:~pli\)11
The consumer benefit needs data were
submitted to a principal component factor
analysis with a varimax rotation, Using an
Consumer Conceptual Map eigenvalue greater than I as a selection
lifc:-;Iyfc criterion, nine factors ...merged. These
factors Were:
[40]
ShwlHng Wu Table I
The relationship between Analysis of variance for consumer demographics
consumer characlerislics and
attitude toward online Conlumer demographics Item At\!!ude mean F p Scheffe teat
shoppIng
Marketing Intelligence & Gender
Planning 1. Female (SO.l percent) 446.84 8.972 0.003*"
21/1 [2003J 37-44 2. Mal. (49.9 peresnt) 477.44
Age (yeNS)
1. 15-20 (21.0 p~l 480.93 8.920 0.000" (1,3)(1.4)
2. 21·25 (18.0 pefCeIrt) 487.16 (2.3)(2.41
3, 26030 (19.0 percent) 428.41 (3,1)(3.2)13,5)
4. 31-35 (21.0 percent) 423.07 (4.1)(4.2)(4,5)
5. 36-40 (21.0 petCetrt) 490.68 15,3)(5,4)
Education
1. Junior high school (2D.41 percent) 470.56 3.425 0.017' (1,4)
2. [Link].. high school (32.1 percent) 463.99 (2.4)
3. Colleg. (42,67 percent) 464.91 (3,4)
4. Gladuate Schoo/{4.82 percent) 390.19 (4,1)(4,2)(4,3)
OccupatJon
1. Student (32.1 percent) 484.57 2.283 0.027'
2. Soldier (1.86 percent) 500.80
3. Government emjlloy_ (4.46 percent) 464.38
4. Flnandal worker (5.57 peIW1\l 443.23
5. Engineer. (32.84 percent) 444.88
&. Buolnesamen (11.50 percent) 477.29
7. Housewtfe (8.53 percent) 440.44
8. Others (3.15 percent) 425.35
Income monthly (US$)
1. below 550 (34.51 percent) 483.88 5.314 0.000" (1.2)
2. 551·1,150 (42.49 percent) 437.84 (2,1)(2,3)
3. 1,151·1,700 (9,46 percent) 498.88 (3,21
4.1,701·2,250 (7.42 percent) 455,20
5••bove 2,251 (6.12 percent) 460.27
Int..,est
1. Sport. (19.65 percent) 480.37 14.122 0.000'*" (1.2)
2. Reading (1&.14 percent) 389.13 (2 .lli 2 .4)(2.5)(2 .61
3. Music (22.82 percent) 437.30 (3.6)
4. Internet (ll.69 percent) 476.62 (4.2)
5. Travel (15.96 percent) 488.83 (5.21
6. Watching TV (13.54 percent) 520.53 (6.2)(6.3)
Living area
1. City (58.07 percent) 448.43 5.023 0.007" (1.3)
2. Suburban (u.s7 percent) 478.73
3. VIllage (30.06 percent) 482.15 (3.1)

Notes: • p < 0.05; ,. p < 0.01

1 effectiveness and modenl; Table ill. E.'[Link] the correlation between

2 purchase convenience; attitude and the nine factors of benefit
3 infonnation ablmdance; perception tested H3. As shown in Table IV,
4 multifoml and safety; there was positive association In every case
5 service quality; (p < 0.05). This supports H3. Thus, all of the
6 delivery speed; consumer benefit perception factors were
7 homepage design; shown to have a positive influence on
8 selection freedom; and attihlde toward online shopping.
9 company name familiarity. There were 26 lifestyle variables employed
in a principal component factor analysis with
These nine factors accounted for 71.05 varimax rotation, using eigenvalues greater
percent of the variance. Cronbach's " of all than 1 as the criterion. Eight lifestyle factors
factors Was greater than 0.52. as shown in Were successfully retained. The eight
[ 41]
Shwu-lng Wu Table II
The reiationsMp between Analysis of variance for consumer purchase preference
consumer characteristics and
attitude toward online Purchase behavior item Attitude me.., F p Sch.~ test
shopping
Marketing Intelligence & Purchase times
Planning 1. Once a month (20.76 percent) 547.00 3.422 0.024' (1.4)
21/1 [2003J 37-44 2. Once every three montI1S (24.53 percent) 523.15
3. Once every six months (26.41 percent) 503.64
4. Once a year (28.30 percent) 434.53 (4.1)
Payment method
1. Credit card (36.18 percent) 509.33 25.083 0.000'- 11,2)(1.3)(1.4)
2. CilSh (23.19 pereent) 442.06 (2.1)(2,4)
3. Transfer account (36.36 percent) 443.43 (3.1113,4)
4. Check (4.27 percent) 331.23 14,1)(4.2)(4.3)
Delivery
1. Sent to home (73.47 percent) 468.74 2.304 0.101
2. Take at store (8.53 percent) 447.63
3. Mall (18.0 percent) 442.21
Not ••: • p < 0.05; .. p < 0.01

Taiiieiii 1 leadership;
Factor analysis and reliability for benefit perception 2 actively;
3 knowledge searcher;
Cumulatlw percentage Cronbach's 4 like computer;
BeneIIt factoJ ElJ:envalue of vllllance alpha 5 fashion;
1. Effectiveness iUICf modern 15.2807 39.95 0.929& 6 attach to appearance;
2. Purchase convenience 2.2481 45.87 0.8460 7 spend time at home; and
3. Informatlon abundance 1.9507 51.00 0.8138 8 regular life.
4. MWtlform and safety 1.7709 55.66 0.8191 Examining the cOlTeJatiOl1 between attitude
S. SeNlce qualty 1.3427 59.19 0.8079 and the eight factors oflifesryle tested HI. As
6. Delivery speed 1.2481 62.48 0.6373
shown in Table VI, there was positive
7. Homepage design 1.1377 65.47 0.5181
association in three cases (p < 0.05). This
8. SelectIon freedom 1.0780 68.31 0.7044
supports H-I partly. Thus, it shows
9. Company name famlUarity 1.0420 71.05 0.6084
consumel~s lifestyle factors "like computer".
"attach to appearance", and "regular life"
Were a positive influence on attitude toward
Table IV
online shopping.
Correlation analysis between attitude and
Through analysis of variance, this study
consumer benefit perception
confIrmed that consumer attitude toward
AttItude online shopping showed significant
(Pearson differences based on the various consumer
comolatlon online shopping rate (p < 0.05). Consumers
Ben.1It factor coeftIdent) P shopping on online who had a significantly
1. Efleetlveness and modem 0.537 0.000-' higher attitude mean score tban consumers
2. Purchase convenience 0.244 0.000-' Who were not shopping online, as shown in
3. Information abundance 0.246 0.000" Table VIT. Thls supports H5.
4. Multllonn and safety 0.182 0.000"
S. ServIce quality 0.110 0.010-
S. Dehvery speed 0.129 0.003" I Results
7, Homepage doIgn 0.242 0.000' •
The results of this study supported nearly all
8. Selection freedom 0.216 0.000-'
of the hypotheses. It was shown that
9. Company name familiarity 0.302 0.000"
cOllSumers who shop online have higher
Notes: • P < 0.05; .. P < 0.01 attHude scores and this higher attitude
score ls directly r!.'lated to onlin!.' purchase
common factors account fOl" 04.19 percent of decisions. The group with the higher
the total variance. Based on the attitude score should be the target market.
corresponding factor loads for each variable, The cOllSumer demographic items all had
the eight lifesryle factors are named in a significant relationship with the attitude
Table V. They are: toward online shopping. The mean attitude
(42)
ShwlHng Wu score for males was significantly bjgher benefits of online shopping, effectiveness
The relationship between than that for females_ Consumers 36 to 40 and modern, company name familiarity,
consumer characteristics and
attitUde toward online years old had the highest atutude scores. purchase convenience, infOl1nation
shopping Consumers with a junior high school abundance and selection freedom etc., then
Marketing Intelligence & education and the following occupations: design an excellent homepage to catch the
Planning soldiers, student, who like to watch TV. attention of consumers and meet the
21/1 [2003) 37-44 with a monthly income from US$l,151 to consumer's information needs.
US$1,7oo and live in villages, have higher
attitude scores. Consumers who like
computers, are attached to their appearance I Conclusion
and have regular life activities have higher
The purpose of the experiment was to
attitude scores. Thus, the group '-'lith the
examine Internet user concerns and
above consumer characteristics is a target
perceptions of online shopping and measure
segment for online shopping. Marketing
the attitude of Internet users toward online
managers could design a marketing
shopping using the Fishbein model The
strategy to focus on this group. The
relative influences fadors on attitude
marketing strategist must emphasize the
toward online shopping and were explored
and the relationship between the attitude and
TaiiieV the influence factors was presented. The
Factor analysis and reliability for lifestyle results showed that the Fishbein model could
effectively measure COllSillner attitudes and
Cumulative percentage
the important consumer characteristics that
Ufestyle factor Eigenvalue 01 variance CrOllb;!eh'S alpha
influence online shopping attitude and
1. leadership 4.9407 19.00 0.7051 shopping decisions. Future research can use
2, Actively 2.7998 29.77 0.5341 the Fishbein intention model to account for
3. Knowledge searchef 2.0005 37.47 0.7094 the additional complexity introduced by
4. Uke computer 1.6255 43.72 0.6344 morc factors. It should be a more effective
S. FashlOll 1.5153 49.55 0.6337 re,earch tool to measure and account for
8. Attach to _ance 1.4273 55.04 0.4527 consumer behavior.
7. Spend time at home 1.3298 60.15 0.5364
8, Regular life 1.0504 64.19 0.4784 Reference
Ahtola, O.T. (1975), "The vector model of
preferences: an alternative to the Fishbein
model", Joumni of Markfling RiSl!arch,
Table VI
Vol. XlI, pp. 5Z-9.
Correlation analysis between attitude and Armstrong, G. and Kotler. P. (2fJOO), Marlu!ting,
lifestyle factor 5th ed., Prentice-Hall. Englewood Cliffs, NJ.
Attitude pp. 15.>-4.
(Pearson Bass. F M. and Talarzyk. W.W. (197'2). "An
COf1'eIatkln attitude model for the study of brand
pl't'[Link]""", Joul7lnl of Marlu!ting Research.
Uf..tyIe factor coelIlclent) p
[Link].~.
1. lAadership 0.014 0.753 Burnkrallt, RE. and Page, T.J. Jr (1982) "An
2. Actively -0.059 0.173 examination of tbe ('~nve..gent, discriminant,
3. KnOl!o1edge searcher -0.074 0.088 and predictive validity of Fishbein's
4. Uke compute, 0.110 0.010' behavioral intention model", Journal of
S. FashiOll 0.058 0.179 Mark£ting R£search, Vol XIX, November,
6. Attach to appearance 0.171 0.000" pp. 55!ki1.
7. Spend tbM at home 0.046 0.284 L'raig, A.J., Durvasula, S. and Netemeyer, RG.
(1994), "Testing the cross-national
8. Regular life 0.123 0.004"'*
applicability onJS and Russian advertising
Not..: • p < 0.05; .. p < 0.01 belief and attitude measures", Journal of
Advertising, Vol 23 No_ 1. pp. 71-115_
Ernst & Young (2fJOO), "Global online retailing: a
special report", available at: [Link]/
Table VII [Link]
AnalY5is of variance for online shopping GlobaLOnlineJ{etailinlL-_RCP (accessed
Item Attitude moen F p August 20. 2(00).
Etter, WL (1975), .. Attitude theory and decision
Online shopping theory: wbere is the common ground')',
1. Yea (9.83 percent) 497.87 5.289 0.022* Journal of Marketing Research, Vol. XII,
2. No (90.17 percent) 458.27 pp.48Hl.
Evans, RH. (1977), "Planning public service
Note: • p < 0.05 advertising messages: an application of the
[43 J
Shwtt-lng Wu Fishbein model and path analysis". JaumaJ of Keegan, W., Moriarty, S. and Duncan, T. (1992),
The relationship between Advertising, Vol. 7 No.1, pp. 2&34. MarkeJing, Prentice-Hall, Englewood Clllfs,
consumer characterjstlcs and Fishbein, M. (1963), "An investigation of tlle NJ, pp. 191-205.
attitude toward online Loudon, OJ... and Della Bitta, AJ. (1993),
shopping relationships between beliefs about an object
and the attitude toward that object", Human ConsU/7/£r Behavior, 4th ed., McGraw-Hill,
Marketing Intelligence &
Planning Relations, Vo!. 16 No_ 3, pp. Z33-40_ New York, NY, p. 425.
21/1 [2003J 37-44 Ftshbein, M. (1965), "A consideration "fbeliefs, Lntz, RJ. (1977), "An experim<mtal investigation
attitudes, and their relationship", in of causal relations among cognitiOns.
Stemer, l.D. and Ftshbein, M. (Eds). CUrrent affect, and behavioral intention-, Jourru:tl
Studies in Social Psychology, Holt, Rinehart of CoTlSU1rJS!r Resrorch, Vol 3 No. 4,
and Winston, New York, NY, Pl'. 107-20. pp. 197-208.
Fishbein, M. (1967a), "A consideration of beliefs Maris, M.B.. Ahtola, O.T. and Klippel, R.E. (1975),
and their role in attitude measurement", in "A comparison of four multi-attribute models
Fishbem, M. (Ed.), Riadings in Attitude in the prediction of consumer attitudes",
JoumaJ of Co,"U1IID Research, Vol. 2,
Theory and MeasurerrumJ, John Wiley & Sons,
pp.38-52.
New York, NY, Pl'. 257~.
~nyazaki, A.D. and Femarulez, A. (2001),
Fishbein, M. (1967b), "A behavior theory
"Consumer perceptions of privacy and
approach to the relations between beliefs
seL'lIrity risks for online shOWing", The
about an object and the attitude toward the JourTUll of Consumer Affairs, Vol. 35 No.1.
object", in Fishbein, M. (Ed.), Reading tn.
pp.27-44.
Auitude Theory and MeasUYl!TTU!1lt, John Piirto, R. (1991), "VALS the second time",
Wiley & Sons, New York, NY, W. 389-400. American Demographics, .Tuly, p. 6.
Fishbei}l, M. and Raven, RH. (1962), "The AB Sefton, O. (2000), "Daily Web surfLng }lOW the
scales: an operational definition of belle! and nonn", USA Today. June 7, available at:
attitude", Human Relations, Vol. IS, pp. 3&44. [Link]/life/cyber/techJ
Goodwin, T. (1999), "Measuring the effectiveness [Link]
of online marketing", Journal of the MarkeJ Wells, W.o. and Prensky, D. (1996), Co11Sumer
Research SocWy, Vol. 41 No.4, Pl'. 403-7. Behavior, John Wiley & Sons, New York. NY,
Harrell, G.O. and Bennett, P.O. (1974), "An pp.44-6.
eValuation of the expectancy value model of Woodside, A.G. and Clokey, J.D. (1974). "Multi-
attitude measurement for physidan attribute/multi-brand models", Journal of
prescribing behavior", Journal of Marketing Advertising R=ch, Vol. 14, October,
Research. Vol. XI, pp. 269-78. pp.33-4D.

[441
 An executive summary for managers can he
Consumers' perceptions found at the end of this article.

of e-shopping Intt:met shopping is o':"c,,-":nL"'1.g an acccpt~_d \\"ay ':.c:

purchaSe various rypes of goods and sen iCe:'
characteristics: an (Doniliu, 1999). In 2001, ,,)njinc saies we're $48.3
expectancy-value billion, representing an annual growth rare of 45.9
p~rc~nt.1 and online sales a~ expected t\.") grO\\' to
approach $108 billion by 2003 (Shim et ,;I,. :'00]). Tnrough
a computer-mediated shopping d)'..'ironment)
online retailers have attraned con'£lwners by
Heejirt Lim and offering a reduL---rion in search costS for pr0-ducts
Alan J. Dubinsky and product-related inlormation (jan" .. n and
Moraga, 2000; Shankar et <11., 19(9).
Attendam with the explosion in Internet
shopping is trenlendously incrc2s,lng intereH in
"'[100".'1- e-commerce [Link]~ particularly with respect to
e-shopping attribute},. For inst<lnce) prevll'u . . .
Heejin lim is a Doctoral Candidate and Alan J. Dubinsky is
Professor, both in the Department of Consumer Sciences and researchers have eXaTIxined e-store characteristic::,
Retaiiing, Puo:iue University, West Lafayette, Indiana, USA. as predictors of online consunle-rs' intention (Shit:l
el aI., 200 I), satisfaction ([Link] and Hise,
GiR·,,'; 2000}, and acceptance of new technology
Electronic commerce, Internet, ShopPing, EXpeCtiltlon, (:\\orriwn and ROD"rts, 1998). In these snIdi"s,
Value analysis, Altitudes e-store characteristics were developed fu.'m either
qualitative ",search (e.g. !vlorrison and Roberts.
1998; Szymansky and Hise, 2000; Yoo and
An increasing number of COOsumer.; are turning to the Interoello Donthu, 2001) or a literaru", review (e,g. Shim
<It aI., 2(01).
make their purrhases, Yet, many Hailers are going out of
business or retrenching. If e-tailer.; hope to attract and retain Notwithsranding the exwnt [Link]-Jrtlre, there are
satisfied online shoppers, they need to know what evaluative limitations in previous studies that demand
aiteria consumers use when selecting an Haller Past research attention. First, there has. been discordance in
has provided some insight into what characteristics shopper.; categorizing e-shopping attributes. For example.
assess in cyberspace outlets, The extilnt work though, has not some srudies have included the navigation functioE
been without its Jimitillions, Coosequently, the present study (e.g. acce&s to the W'eb site:> loc<:"lting an item iln the
utilizes a literature review, qualitative research, and quantitative
Web &1te) with convcn~e_nce chaIT.lcterist1c', (e.g.
research to identify the undenying €-Store choice dimensioos of
l\lorrison and Roberts:: 1998; Shim a al.:: 2001~':
shoppt'''- In addition, results of multiple regression analysiS
show that merchandise and interactivity Web attributfs art? yet others have classified tht"~.e [Link] ~s two
predictOf'; of consumers' attitude toward online ShOpplOg independent criteria (e.g. Lohse and Spiller,.
ImpiJcations for e-store managers and future research are 3150 1998).
provided. Another problem with prior work on e-store
dil1k"n:-iions lies in th..:- inc{tIJsi""rent' re:;,earch
Electronic ac~ess 111ethod:;,. For cxatT~pit:' .. Loh~e and Spllier (1993
The Emerald Research Register for this journal is auenlpted to identify attrihute~ of online ret3il
available at stores in terms of Il1erchanJise. !:>ervice,
[Link].(om/researchregister promotion. cOll\"enience. and nuv1s<ltion. The;r
findings, howe-i.·cr., wen:'- pro:dic2tC'd on a survey 1.1f
The current issue and full text archive of this journal is
srores ratht:.r than COIlSun.1.t:'rs - thus-~ critical
available at
[Link]/[Link] consurncr input regarding c-taiI~r \l?cb site
charaderi..;,riCS \vas ov"Crlooked. As a result, they
analyzed only descriptive attribures of e-srores
(e.g. merchandise~ service._ promotion,
convenience" navigation) but did not comidcr
(nher attributes that are reported to be import3nr
factors <liTecting cyber 5h(tJ"lpi!r5' online
tran:;,actions~ such as security and privacy poiicic:>
Journal of Servkes Marketing
Volume 1B . Number 7 . 2004 . W 500,513 The authors gratefully acknowledge L~C ednor ::wd
l~' bnerald Group Publishing limited· iSSN 0887-6045 reviewers for their valuable input and
DOl to.ll08108876040410561839 e-ncourage-fnc-nL
500
 fh.'fji:1 Lim and Alan 1. Dubinsky

(e!\!arketer, 200 I b, c), as well as download waitin~ (1) a [Link]\ "tangJ~lt (~r functional quaErie~" "t'_.;;.
time (Dellaert and K1hn. 1999: \'(!einberg, 20(0). rnerchandjse sde.:riofl) price range~~ credit
Th~ foregOl..flg v;~ak:!:lesses in previous rese2fch policl(''S) store layou(; and
indicate additional empirical work is needed tu (2) "intangible or p:-;ydlological altributc-5<-'-' (c.g. a
identify evaluative crireria consumen: consider sense- [Link] bdonging. the- feeling of warrrlth (lr
when selecting a cyberspace store. By knowing friendliness.' a feeiing of excitement or
these criteria, e-tailers should be able to enhance inrerc-,st)
the design of their c01l1mercial sites and quality of "Attribute!'." repreSl"'nt [nt:· cO'::ll::·~ncd CdnCi.:'pt (\1'
service fultilimerH in urder CO increa~e custumc:r.:,' functional and psychologicaI factors thnt ('xist in :~
positive attitude about a given e-tailer. StOre. \\'hen fi13king 2 qore cn\)ice dC'ci"i . . 1O~
Consequently, a study was designed to explnre [Link] evalwltt' ~:()rt:' :::ftcrn::1ti\'cs on ~~ nur:1h('!,
consunters' perceptions of e-shopping attributes. of ~!on:- arnlburt:"s (Lindli'Jj5,t, 1974). PMrons 8n~:
[Link] \):,'eb 5ite design and service fulf1ltn1enL non'T'atr-oa:" h3vc diff,.:TcDt p..:rc("ptiol".rs Qf <} stnrc- ';:;
iJif-J-7./L'i: online Sh(lPping attitude. \\~e did so irnaJ;e. As such, rewile-~ need to ensure [Link]
lltili7ing an expectancy-value approach~ as dirnen..'iions that the-ir loyal CUst()1:1eD- view .15 'Dein;
promulgated by Fishbein and Ajzen (I975). irnportant are de,s,lgllC'd t() be attractive to then1-
Although previous studies categorized e-shoppiog SiIllibrly, e-store inl<.ige is likely to have 3 I1l3jO[
attributes using [Link] a literature r~\-'ie\v or inl1l'.ence on onEn.-: [Link][(lITh::r-.; \\'hen d1ey
qualitative research, the present investigation de-f;.:'nnine- frQra \vh!ch e--tal1er to buy. E-srol"::
identifies e-shoPPlng arrrlbures through a inlage~ t'n,)ugh) ".-iH likelY be defined differe-ntly

literature re\';c\v (i.e. traditional retalI stores~ frorn hrick~-and-rnorlar st(")r~ in1age. After all, tni:.'
home-based shopping such as TV and paper \\'3Y in v.'nich consumers ~hop in e-tail \"enut's is

catalogs) online shopping), qualitative research, different fn'n1 how they shop In ~-l physical st()re~
and quantitative research. Exploratory factor owing to the absence I,)f a physical store milieu.
analysis and co~mnnarory factor analysis are C()ncei\'alJI~.'_~ tht'D, consurnerS st.''::I!iinglY'l.'iU likdJ

conducted to create profile categories ':Iharc:d by asseSs s:ornc" un.;.quc MDre an.-ributcs in GaLine
multiple attributes. Then, multiple regre"ion shopping \'i:-'<l-V~':; rhose utilized in physical srnre
analy~js is employed to [Link] the jmpact of
shopping.
Arguably, t'-stores do share SO'll!:' con1r:non
these underlyL."Jg e-store dimensions on
features \\ith a phY'slcal store in [enDS of
consumers' attitude tClwards i.mlir1e purcha~t:.
merchancEse, se-rvice-~ 2nd pn.1lTIot:ion. 'There :s
An e-~tore- can be defined at'> a commerClal \XTc:b
31~o s.:m:lt: ~in1ilarjty b-etv;een tr3dit10n7.J Z'.ode:;. cf
site on Whlch consumers can shop and make a
l.:l-honJt' ~hopping, such 3S TV and c.:l[alug
purchase. E-stores can be operated by either a pure
shopping" and online shopping. Owing td fnt'
player (i.e. a [Link] that has only an online outlet)
narure of [Link] COTl1mUnlc-,;:-rion.
or u traditional retailer (i.e. a retailer that l)\\Tj:)
hOW(",t'f, on lint.' rewil stores h::l';c nnique {emufe-'-
both brick-and-mortar slOres as well as an online
that do nor e'_xist in either the rhy:;,ical <.:tnre Or
"utlet). In this study, attributes of general e-,rores in··-hD_me shopping. Prior to developing: :."-5!:-;r:s'
(i.e. both pure players and traditional r~taikrs) are unique aurin-utes._ thOSe ;,)f the pI'(y~ic:&l :"J'!"C iin,j
considered. ESSentiaJly., then) the objecri\'e~ of rhe in-horne shopping ':T',/ and catalogs) are
study '[Link] tn: disc:u:;st'd. Femu!'.:" I.A~ ~3ch :-;hopping uircrnar;v('
(I) determioe the underlying dinlensions of are identified ;:.!lfOLg;.'1 J re\..-iew of ErerJ ture
o)nsumers' perceptions of e-shopping pertaining t.o '-,f(lre iraage and conSUn1i.'r qore
attributes; and ch()~cc<
(2) explore their eifect on consmners' attitude
toward online purchasing.
Evaluative criteria of physical retail stores
Sheth (1983) t::xpandcd dt't~rrninant.s of s.t~}[c
choic...'" br c1as.si1:}'ing COnSUi!lt'r.<.,· shopp~ng [Link]
int() tv,'o leyds: fun~tiC'rwl and non-funcbon8L
Literature review Functional D1mj,:e-s involve tang~ble fe3[ure~ i:such
3~ price~ con\-'enien\:e~ and merch:andis.e:
Lindquist (I 974) has underlined the importance 35sortment); n,.:m-funct:on::ll motives involve
of store image as a pr~dictor of consumers' store inlangih1e- features (such as :;;1 ore atmo~phereJ sale~
choice. A p~on lS hehavlor is not only a function pen,onn~l s-erYlce, and p~.ychoh)gica1 rea~-A)nS for
of knowledge and inf()rmarion but also is shopping). By evaluating iur.ulon31 and n~m-
predicated on the cOlbulner's ilnage of a pr-oduCl fuw..Liollal qualiri~ of a n:taii store slInuhancously,.
or store. From a marketees "iewpoint, store image conSUIners form their store inlage (Lindquist,
is characterized by two dements: 1974). Consun1cfs ultimately CllOO::".t a Slt>re thar
501
 (CnsU.--nHs' p{',ceptiOilS of ('-shopping characteristics
fieej{n Um and Alan 1 Dubinsky

Inaximizes their satisfaction with these perceived high incidence of COlhUr:Jt-f conlJJlaints abQut b3d
qualities (Sheth, ]<)83). Retail physical store qualit:y and poor dehn.'ry w-ith TV shoPPlng
characteristics identified hy rrcyious [Link]!"s (B~ter1Jd and SlO., 1993) _ A!so~ cons,umers often
are presented in Table 1. complain about out-of-.c,tock merchandise
(Consumer Reports Bf{Villg Guide, 20(0). For thest"
reasoDS, consumerS likely ree] impelled to pay heed
Evaluative criteria of in-hotne shopping to shjpping and handling jnformation, s3tJ&faction
Traditional in-home shopping venues ha,'e gu8rantees, and a;-ailah;:;ry of a roll-free rhone
included chiefly TV and catalog shopping. Dun)ber to m.1rumize dissaris13CI"jon from .home-
Shopping via TV affords consumers the ba~d shopping. A summary of in-hmne shoppeD;
opportunity to t'xperience convenience through attributes is presenred in Table II. (It is ba:"ed
reduced shopping costs vis-a-vis physical eft"et. ,'. soldy \,.)fl p2per carah)g !;,}1\.)rp~ng re~carch,
disrinct feature of TV shopping over cawlog hov.'ever~ as prior w()rk r.a5- not in\TSrigatc:d 1"\:'"
shopping is the role of the hostfhostess. Also, the shopping charact~tjstics,>
entertainment aspect of TV shopping appean to
be an important factor for senior citizen~ (U,)'.~l
']ilda:v Magaz;ne. 19(7), Catalog shopping has Evaluative criteria of e-t,[Link]
curried consumers' favor with enhanced On11De reran store'S h..1ve [Link] slmilar fe-Mures [0
fl1t::l'chandlr-,e variety, as wen as tJ1e rdiabiEty and phY5..1cal retail St\..We:; and catall'g~. For e-xampk,
security that can be garnered from established online retailers offer e-mail addresses of salts
conlpanies. [Link]~ consumers seem to like catalog associates or trequendy asked questions (FAQ)
shopping owing to its ease ofU'e: products tend to sections to communicate with their customers) just
he clearly portrayed, and product information as physical stores have sales pe!"SIJnneL Also. they
provides rapid comparisons . .Nloreover, telephone [Link] common attrib~,jt.-;:.t with paper catal-(\g~. by
asso(.iates arc available to help an.... Wef consumers' providing consumc-rs wi(h (he con\'enience of ht-
questions about products and services (Consumer home shopping and purc:h.2se ddivery. And like
Repom Buying Guide, 2000). caralogs, retail ,x·reb $it~ typically pro\ide a [011-
In-home shopping, however, can present free tekph~Jne nU1!1ber thrQugh which their
certain disadvantages, such as the inrangibility of custoIllen; alay contact sale~ as~oc:iatc~ for further
products and relatively high shipping and handling lnformarioD. Compilted 1O l'mer retail fi)D"!latS.
fe·c:s (thw.i- in"2rea&ing the catalog'S eff~ctive cost to hO\l"c.Y~r, lnany onhn-c r;:t3i~ stor~s ha\-~ the
the consumer). For example, one study found a advantage of ~eclnlng1y unlirnited IYl>..~rchandisl'

Table r Rflevant attributes of physical retail stores

Berry Lindquist Tigert McDaniel and Bumett
Factors Attributes (1969) (1974) \1983) (1990)

...
......
Merchandise Wide selection
Numerous brands
Well-known brands
Availability in stoci. ...'"
Price v v
Product quality Value for money V
Convenience Locationa! convenience
Parting ...
I'
'"
Moving through a Store
location of Items
Exchange
Acceptance of credit cards
Physical facilities Store attractiveness v v v
Sales personnel Friendl in€ss}eourtesy .... v ...
Service
Infonnatjon service
Ease of retums
Delivery service
......
yo
v
...v...
v

Promotions Silles promotion v

Advertising .... ...
Institutional factors Reputiltion v
Reliability ...
Clientele of a store Social class app"al v
Seihmage congruency ...
502
 Consumers' pl:':ceptions of e-shopping ch.::racte,istics
h'eejfn Urn and A/an 1 Dubinsky

Table 1/ Relevant attnbutes of home-based shOjJping

Eastlick McDonald Seaver and Simpson
Catalog factors Attributes (1989) (1993) (1995)
Merchandise Quality
Assortment
Style
Price
Uniqueness
Availability of merchandISe in stock
Convenience Accessibiiity
Time-saving
Elfort-saving
Ease of order placement
Method of payment
Ease of C3ta!09 use Easy to find merchandise
Well-displayed merchamJise
Easy to read and understand
Home environment Comfort at home
SeIVjce Guaranl€es
Ease of mer<:handise return
Delivery service
Promotions Clearance
Reputation Recommendation by friends
Well-known Mtional brands
Trust company's merchanoise

and product information. Furthermore, e-tailer E-shopping attributes pre'>ented in Table III are
store design and layout have distinct features now discussed_
compared to those found in ph)"ical stores and
paper catalogs (Spiller and Lohse, 1998)_ Aferchandi,e dtaraclen-stics
TIle e-shopping attributes presented in Table III ~"lerchandjse can be defined as either goods or
were drawn from an analysis of literature sen';ces offered by a retail ,tore (EastEck, 1989;
pertaining to physical retail stores, paper catalogs, Lindquist, 1(74)- Because of the unique nature of
and e-tailen.. lIowen:r, store dimensions of a the Internet-mediated "hopping environment,
physical store that m'e not applicable J(Jr online consumers' evaluation criteria for e-taller
outlets (e.g. clientele of the store, physical merchandise might he somewhat ditTerent {rpm
facililies, store atmosphere) were excluded_ tho~e for traditi0nal retailers. For instance, unlike

Table III Summary of e-shopping attnbutes used in previous sludies

E-store factors Attributes Examples
Merchandise Product infonmation The pemlved depth of prO'Jua miotTnation
Brand selection Well-known national bramh
Price Merchandise price
Convenieoce Timely delivery Delivery on time, delivery options
Ease of ordering Fast ch<>ck-out, order coniirmltion by t'-tTlail
Product display Product lists with both d"x butt""s and pictures
fnteractivity Customer support Sohware dDwn!oading, e-form inquiry_ Oider status checking, (ustOfi'l-t'f (ommen! and
feedbild
Personal-choICe helper Keyword search, 'mproved search flioctioo
Surfer postings (us/"{Jffim' reviEW of product/SErvice experieOCE
Reliability Reputation Company [Link] ion
Security Information on transaction s€<urity
Privary Privacy policies for personal 1Olormal!on
Promotions Promotion on the cybermall C!£arance, free shipping, frequent buyer incentives, prile for [Link]
home page
Navigation Time to gel to bome Pdges The time taken from ads 00 other sites to horne pag€s
Expected waiting time The perceived duration of the tlrr-J€ to tiovmload pages on the site
Waiting [Link] Dura~\oo information at the [Link] of the wa~t countcO'.Ivn infQW,1-.a~;oo

503
 Heejin Om dna Alan 1. DubinsKy

a phys-icaJ ~tore, e-tailers can provide customers the Ll~ability ::md percei\'ed depth of onEne
with as much variety as they want without physical Jnfc\rrn~ti\)n (Lynch and Aridy, 2000; Shankar
space restrictions. AI$<-\ consumers can compare <;'! :11. ~ 1 f)9Q).
product prices more ~asily than eVer bet()[~.
E-tailer dimensions traditionally ascribed to
merchandise-related aspects include product Con'i:"enience is a key motin:. behind in-horne
infonnation, brand selection, and price. shopping (Eastlick and Feinherg. 1994).
As in catakg shopping, accurate reproduction Convenience is mea..;;ured by efft1rt savings (e.g.
of descriptive and experiential pmdua inj()rmariDl1 "::<lS':: of a Jocating a product in a store) and

is a critical factor [Link] conSUlners.' choice- in. ioeational convenience (e.g. '.."'~'..Se 0f10catng a stQre
electronic shopping because consumers cannot and finding a parking spa,:~) (Lindquisr~ 19(4). In
touch or see products (Alba et al., 1997; Lohse and onHne shopping~ convenl.t'l1cc. includes timely
Spiller, 1998; Lynch and Ariely, 2000; \X'ard and deli\'ery, ease of ordering. and product [Link]
Lee, 2000). Interestingly, despite the advantage of (L()hse and Spiller, 19(8).
the lower C()~t in delivering text and images Lohse and Spiller (1998) discerned that ,everal
through the Internet versus paper cal:.::ilubTS, Inure factors can be subsumed under the converuence
~ttnbute 1..11' enline. ~hopp~ng: m. m"'~ner of links ~ntc,
than 50 poercent of e-tailer sites provide kwer than
[he site, nUlnh~r and type of dilferen£ shopping
three lines of text describing each product (Lohse
mode~., average nlLrnher of :tems. per preducr menu
and Spiller, 1998).
Previous studies about Sh)re attributes have Ii:; ring, number of [Link] that require scrolling,
ShO\.Hl thaI: merchandise sckction has an [Link]
presence of price infonnatl\.)n in prodU<.t. listings,
on consunlers' store choice (Berry, 1969; and type of product bsr::;.. AlTIOng these anributes~
Lindquist, 1974; McDaniel and Burnett, 1990; they found [Link] pmdU(l displt{1' has a .signincanr
il1'tpact (Ja f..itc vi;;.it!'l and :-,-..L:Cfo. SpecificaU)\
Tigert, 1983). The vest number of product
displaying product lists using both dick buttons
alternati,·es is a key benefit for orJine retailers.
and pictures leads to ITIOft po~iti\"e reactions. frt)In
HO\\'t'ver, Alba .;'1 al. (1997) argue that consunlel~
u)nsmners than simply disphlying a product list
might l'ecome tired and stressed by examining
u.',ing only a button or picture..', -in online catalog~.
infonnation on hundreds of products. Lohse and
East' of ordtttt"ng appears. (Q infh.1-c-nce hnme-
Spiller (1998) dispute the importance of
shopp{'rs' huying (.-'k··cision:; (Eastl!ck~ 19R.9;
nlerehandise variety in e-tailing. In particular) their
~\lcl.j\mald, 19 ..}'3:'. Therefore, order proc6s1ng on
work showed that the number of products '
\\'eh .sites iliould he [Link] for cusrome-r& to do,
incrca$.cs c-5torc traffic, but it docs not affect sales.
l\.ioreover, receiving order confirmations Yla
Apparently, "hether or not an e-tailer has a
e-rnai)~ including information about shipping,
specific product 3 customer is looking for is morc
returns., ~nd order tracking numbersJ f2cilJ!ates
important than simply having a large variety of
(}rde.r-pro("es.~ir..g be..~.a"ior .
If order proce~~~ng 1:;'
items ([Link] and Spilkr, 1998). Therdore.. brand
(trne consulning and complicated] customers will
selectLml might well be more likely to affect iikely b~com~ frustrated and giw up pUfcha.,ing
custonleDi' bUying decisions and subsequent from the e-railer (Lohse and Spiller, 1998).
e-store patronage than lnerchandise variety \\·ith in-home shopping, physical stOre
(Doegeratu ", aL, 20(0). Indeed, brand names also dimensions of convenience, such as. geographica.l
appear to affect consumers' buying [Link].~, location and parking. do not exisr. Instead,
especially when they are unfamiliar with an e-tailer in-holne. shoppet'S ~c-ck con\·enlcnc.c through use of
(Ernst & Young, 1998). Further, when conswners mail or phone shopping and through lime~v dclir.':r!ry
have difficulty in scaTching for products on the (to honlt.». A Price WaterhOLL~e Coopers study
Internet, they tend hl rely on brand names (iX'ard r~vealed that "the biggest sow:ces of [Link]~..sfaction
and Lee, 2000). anl0ng e-shoppers had to do -with gifts not arriving
Price is a key attribute for customers when on time for the [Christmas] holidays" (e.\Lrketer,
forming perceptions of retailers (Berry, 1969; 2001d).
Eastlick, 1989; Lindquist, 1974; McDonald,
1993; Tigert, 1983). Online shopping enables IntiTu(./t.z.110' cha,..aCleri...rtics
con!'lumers to reduce search CCStS and compare Interacti\'jty on the Internet refers ro the degree to
product information and prices simultaneously. which customers and retaiIer~ can comnnmicate
This benefit, concomitamly, has accelerated directIy with one another anyv,.,here. any t1me
rctailets~ competition and made e-tailers especiat~y (Blatt berg and Deighton:- 199 I). For e-railet"""$,.~ the
concerned about consumers' increasing price d..::grce of inreracti,,'iry inftu-:n..:.:es the pen.:eiv~d
sensitivity (Shankar ct aJ., 1999; Ward and Lee, quality or the Wen site (Gnn,oe and Doll.. 1998).
20(0). However, previous studies have also t,mnd Ghose and Dou (1998) surveved 101 \'feb sites to
that price sensitivity can be reduced by increasing identify key inleractivlfY faC(ors mar: influence \X'ec
504
 (onsume:rs' p€'Jcepticns of f.-shopping [Link]€ristics
Heejin Lim and Alan 1. Dubinsky

site appeal by usage frequency of each factor. They the purpose of promotional aclivities for particular
found that CUsloff/et· Sllppon was the interactivity products is to encourage consume"" to buy either a
aspect most frequently used by customers. In particular product or some orher product,;. Spiller
addition to customer support, several additional and [Link] (1998) have drdwn analogies among
dunensions can be classified as "interact1v:ity" retail store, paper catalogs, and online catalogs and
characteristics - personal-choice helper, surfer have characterized e-store promotion activities a:;
posting" and promotion. heing special offers, online games and lotteries,
E-tailers provide several types of online sen~ce links to other sites of interest, and appetizers.
that can increase interactivity with customers, such S ubsequendy, they have also discerned [Link] hours
as software downloading, e-form inquiry, order of promotion on the e--stOre entrance appears to
status tracking, customer comnlent, and feedback. increase consunlers' buying decisions (Lohse and
In a physical store, customers interact with sales Spiller, 1998).
personnel; their friendliness and knowledge can
afrect consumers' purcha,ing decision (Berry, Rdia bility charaaerisu~-s
1969; Lindquist, 1974; McDaniel and Burnett, Company reliability i., an important criterion
1990; Tigert, 1983). On the Internet, e-tailers c()nsumers utilize when making a [Link] choice
offer consumers with sales clerk service in different decision (Lindqu~st, 1(74). Consumers mig.~t
forms, such as a toll-free phone number, e-mail wiili !G protect themselves from unreliable e-railers
by paying dost! attention to company infonnatlon.
addresses, FAQs, and customer feedback.
;'.ccording to GVU's \'{"W,\V user surveys
Research has found that [Link] FAQ sections and
(Graphics, Visualization, and Utilization Center.
feedback in~ases e-store ,isits and sales (Lohse
1998), reliability of online companies is the third
and Spiller, 1998). Empirical work about the u,ag~
rnost important attribute consumers consider. In
frequency of customer support functiotl5 (e.g.
addition .. 'Security and privacy are gaining
e-inquiry, comments, and feedback) reveals that
increased c()nccrn among online users (&Ilman
customers prefer two-way communication \vith
"I aI., 1999) and thus merit research attention.
e-tailers ramer than merely being passive
In home-based shopping, a retailer's [Link]
receipieDts of information (Ghose and Dou,
has a significant influence on consumers' purchase
1998).
decisions (McDonald. 1993\ The provision of
Online outlets provide various fOfilS of search
sen,-ice infOrlllation (including conlpany history)
functions for customers to locate items for which
can help a cus-t0t11er feel Inure conlfortabie about
they are searching. Ghose and Dou (1998, p. 32)
dealing ~irh a given firm anj about sending Licdit
define a -[Link] hclpiT as "a funct10n that
card information through the Internet (Lohse and
can make:: rc::lalively ",phislicated Spiller, 1998). So, in-depth company information
reconlnlendarions (lI1 consumen/ choice,; based on
might ahate consumers' uncertainty and perceived
their input of preferences and decision criteria~. risk in dealing wi th e-reraikr •.
This function (such as a keyword search) gives Transacllons in online shopping fend to be
custornen; more refi...f1ed alternatives. For example~ made \\-'ith a tOTedi[ card. Ht)wever, consumers have
multi-layered information assists customers to been warned not to release their credit card
narrow down target items based on their decision information onJine but l<' make a phone order for
criteria (e.g. [Link]; Shankar online purchasing (Purger, 1996). Neari)' two our
etaJ., 1999). of three Americans do nor trust e-wilers, and
Web sites provide customers with interacti\~ty (OnSun1ers are \\'o-rri~d about ti'.te ~_curity of c-n:dir
not only ;>1th e-tailers but also with online card information (Jeffrey, 1999). By info-rming
communitics. Ghose and Dou (1998) found that (Ustonlers about the ;k-'[uri:;y of online tr3n.~3ctions~
online customers frequently use suifer p<JStings, e-tailers can help reduce online risk perc<'c;w'd by
which are customers' reports of their feelings and custOIllen> (Ernst & Young, 1998).
experiences with products and e-tailers. E-tailers Company \"(Teb sites collect a \'3St affiOlillt of
often pro1cide a page of customer reviews (e.g. custm11e-r infi_ Jrnlat1clD through the Internet, \-vhich
\v\\,'[Link])., which gives customers i~ & fundamental 3Sf.,et for companies. C(ln-SUn1er~~
indirect experience with the products and service. ic contrast_, may fed uncomfortable relea&ing their
Conmmer behavior tends to be influenced by personal inJormation (sllch as credit card and
external environments, such as promotion. The social security Dumbers) via the Web (Ernst &
beha\~orist approach in consumer research posits Young, 1998). The top privacy concern of US
that ~the reinforcement of a series [Link] \\ul consumers appears to be .....hether or not a "I);'eb site
gradually bring the consumer to the desired final asks permission to share personal information with
behavior" (\V'ilkie, 1994, p. 271). For instance, a other compani~s (elvtarkerer, 2001 c). A recent
'~clear-ance sale-" sign on a store window can report reveals that almost 65 percent of
stimulate consumer Slore traffic. In ph}"Slcal store.s., respondents gave up online purchasing because of
505
 Consume-iS' pe:rcep-tic..;)s of e.-shopping characteristics
-.-~-.~ --- --- .. ~-------- - - .. - -----.. --- --- - - , - - - -
Heejin Lim and Alan 1 Dubinsky Volume 18· Number 7·2004· 500-513

pri.,,,c), concerns (eMarketer, 200lc). Consumers Attitude toward online shopping

are discomfited when they receive e-mail from a Au:itude is "a psychological tendency that is
company with which they are unfamiliar (Sheehan, expres>ed by evaluating a particular entity with
1999). some degree of famr or disfavor" (BagIyand
Chaiken, 1993, p. I). Attimde has a strong
influence on consumers> bUy-mg ina.--ntion (e.g.
NavigatitJ11 ch-ilractel1stics Ryan, 1982), the [Link]<iate precursor of actual
Lowering search costs for shopping is a key beha,,~or (Fishbein and Ajzen, 1975). Previous
motivation for consumers to shop online. As the studies have observed a poslrh'e a~sociarjon
total number of working hours of households hehveen attinlde and beha\''1oral intenton (e.g.
increases, online shopping has attracted working Chang er al., 1996; Chiou, 2000; Ryan, 1982;
families, thus enabling them to save time by Shimp and Kavas, 1984; Taylor and Todd, 1995),
purchasing products and services in a non- induding III an online ;,hopping context (Shim
traditional way. In physical-store shopping, "t ai, 2001). Applied to the present study, attitud~
consumers seek to lower their search costs (e.g. toward online purchasing i., considered to be a
time and elrons): physical effort is employed when function of the consumer's belief.; about an
going into a store, finding products, and e-:;tore's character~tic5. and the degree- of
comparing alternatives across stores (Bell et ai., subjective importance a conswner attaches to
1998). In online shopping, navigation time and those attributes (Fishbein and Ajzen, 1975). Base-.d
efforts are analogous to the physical elIort on the foregoing prior work, then, an e-tailer's
e'-I'ended to locate items in traditional shopping. failure to foster a favorable 3ttitude t()\,vard its \\':eo
Gupta and Chattcrgec (1997) define search site will likely lead [Link] t,;) [Link] onEne
costs on the Internet as: purchase$ with that particular e-lailcc.
(I) Internet connection time;
(2) actual time and effon taken for the user to
search an e-tailer's site (e.g. use of online
search engines, links from related pages, Method
suggestions from newspapers/magazines!
friend>,); and
Sample and data collection
The data were coUeeted in a cla~<.,rDonl setting
(3) time to do"'nload information fTom an e-,tore
froln a convenienc-.:- :;ampk of students at a large
(which essentially depends on the connection
~1idwest~rn university. In a study of consumers'
s-peed,. usage charges., traffic on the network,
behavior-al intentions- to m,e different retail fOrt112t:S
trafiic at the sitk) and me kind of information
(e.g. retail stores, catalog, Internet), Keen (199l)~,
being obtained).
compared results betw~~n a student sample and
Internet users are not tolerant of the waiting rime to mall shopper sample. '111" [Link] showed no
arrive at a Web site's homepage. GVU's W\V,\" diffe~Dce between [Link] ['.in) sampks in pred:cting
user Burveys showed that consumers are [Link] conswner deci..,;,ion-u13king on the Internet.
by and annoyed with long waits to do'Wnioad an i\ioreoverJ colk·ge students deserve c-rerailcT
e--tailer's homepage rrom Web ads (Graphics, anenrion because of theif significant numbers ris-
Visualization, and Utilization Center, 1998). a-vis the Intem~t. For instance, college students
When do\mloading is delayed, potential sp~nd more than 20 hours per week t'n the
customers are likely to drift to alternative e-tailers Internet, and 81 percent of them have made
or give up online shopping, at least for that time pure-hases online. Furthcrmore_~ coH~ge students
('\i\Teinberg, 2000). are considered brand loyal and are early adopter:;
Consumers are more affected by their perceived of new products (FururePages, 20(2). Snortiy, this
durati<m of download w"iling lime than by the actual group "ill enter the majnsrream of the online
waiting time (Dellaert and Kahn, 1999). That j" consumer group., which ':"';;' charactc-rized a..;;, heing
when the wait to download is shoner than between 34 and 45 years old, highly educated, and
eA-pected, consum~rs' :illtisfaction with the service WeE paid ill> compared to -:.he gc-neral [Link]
increases. Conversely, if the wait is longer than population. Thus, college stud"TIts should be
expected, consumers' satisfaction decreases. considered a key target market in the long-term
Consumers are likely to feel disturbed by the sucCess for lnany online businesses. For these
wait when they are uncertain about the actual reasons, the smdem sample us~d it: th" present
waiting duratior! (\'7einberg, 2000). By prol'iding investigation is. 'Seemi..flgiy appropriak {p_c q1.~t':ry-in.~
'[oailing tZlne infin-1uation (e.g. tinle bar indicator at onlin~ cOf1$umcrs.
the bottom of the Web page), e-tailers might help Among 252 quesrionnajre completed,
consumers become more tolerant of the wait and approximately 75 percent of the respondents were
more favorably disposed toward the site. female. 1ne mean age was 20.9. About 62 percent
506
 Consumm' perceptions 01 ELshcpping characteristics )00,",,1 of Ser.'"", Marketing
-----------------_._-----.- --_.-_.,"------- - .. ----------
lleejin Lim and A/an 1. Dubinsl:y Volume 18· Number 7· 200!' 5(]()'513

of the respondents reponed their income level to that e-store attribute (ec), Respondents were
be less than 1)5,999, Approximately 72 percent of queried about 16 e-shopping attnbutes using
the respondents had purchased a product through se"en-point >eales. The scales of belief strength
the Internet. were anchored from "unlikely" (l) to "likely" (7).
The ,,,'ales of evaluation were anchored from
Measures "unimportant" (I) to "important" (7). An
E-s/u>pping attributes example of each is noted below:
Prior to developing questionnaire items, • (b,) The e-tailer would provide me "1th high
qualitative research was conducted to generate quality product infonnation (unlikely!
important e-shopping attributes, In this likely).
procedure, 29 students were given open-ended • (ei) For me, high quality product information in
questionnaires. These questions were generated online buying is (uninlportantlimportant).
from rhe work of Mathieson (1991). Specifically,
they queried respondents about the advantages
and disadvantages of making a purchase on the Auiume [Link] behavior (AB)
Internet, and their likes and dislikes about online [Link] with the work of Fishbein and Ajzen
shopping. (1975), attitude toward online pUrcl13Sing was
In the qualitative research, interviews with measured by four different statements using seven-
respondents revealed that merchandise variety point semantic differential scales. Scales were
(Table IV) incorporates brand selection (Thble anchored using four different terms: "dis!ikeJlike~,
ITI). TIlerefore, in this stud)T's tinal questiOlmaire, "foolishlwise", "bad/good n , and "unpleasant/
merchandise variety (a physical store feature) was pleasant n • For example:
included. Also, "ease of use", which was referred (Y[) Making a purchase on the Internet is a Chad/
to in our literature review, was combined with good) idea.
"ease of ordering" (Table III), owing to their
overlapping nature, A pretest was conducted to
check clarity of measurement items in the find
study questionnaire. Ambiguous sentences were Data analysis
revised. Data were analyzed using both factor analysis and
Fishbein's expectancy-value measures have confirmatory factm analysis (CFA), Facmr
been used as a person's" evaluative implications of analysis attempts to identifY hypothetical variables
an underlying cognitive structure" (Fishbein and that explain the pattern of correlations within a set
Middlestadt, J995, p. 186), An expectancy-value of observed variables. While exploratory factor
measure is obtained by multiplying a person', analysis attempts to identify the minimum number
behavioral belief that an object possesses a certain of common factors that represent correlations
attribute (b.) by the degree of subjective among the observed variables before de"cioping
importance the person attaches to the object's hypotheses, CFA provides self-validating
attribute (e;) (Fishbein and Ajzen, 1975), Applied infom13tion for a given h}'P"thesis (Kim and
to this study, the degree to which a respondent Mueller, 1(78). That is, the purpose of conducting
belie,'ed e-tailer \\'Teb sites possess a certain CFA is to build a model "assumed to describe,
attribute (b;) was multiplied by the degree of e>..'Plain, or account for the empirical data in terms
subjective importance the respondent attached to of relatively few parameters n (f6reskog and
S6rl>om, 1993, p. 22). CF .. was employed to
IV Beliefs about e"hopping attributes {qualitative research} confirm me c-sTore attribute [Link].
obtained in the factor analysis. In addition to these,
Number mentioning
ltagesldisa/lvantages of online attributes multiple regression analy"" was conducted to
ases (n = 29) Percentage investigate the underl}~ng Web site dimensions
obtained in the CFA on consumers' attitude
19 65.5 toward online shopping.
ty 19 65.5
, delivery 9 31.0
y guarantees 9 31.0
Iity of return and exchanges 7 24.1 Results
ility of a retailer 7 24.1
andise variety 6 20.7 Exploratory factor analysis
:t information 5 17.2 The SPSS computer software extracted lour
oadiproce5s time 5 17.2 factors from 16 observed variables (consumer
t use 4 13.8 evaluative perceptions of e-shopping atttibutes)
me customer service 3 10.3 using principal component analys;:; and Varimax
507
 Ccnsum"fS' pe-rQPtions of e-shcpping characteristics
Heejin Lim and Alan 1 Dubinsky Volume 18 > ~~umbE>f 7· 2004· SOO-ST3

rotation. Factor analysis revealed fimr underlying In essence, CFA identified four distinct
dimensions with eigenvalues greater than 1.0 and a dimensions of e-shopping attributes: merchandise;-
communality of 59.7 percent. The resulting interilcti\'ity~ reliabili!y~ and navigation
factors generally emerged as expected for characteristics. ;\lerdlandist:' consisted of product
e-shopping chara(1:eristics, except for the information C,I = 0.70.7, P < 0.001) and Fariety of
promotion characteristic. Promotion loaded on merchandise (A ~ 0.757,p < 0_001); interactivit,.,
fuctor 2 (i.e. imeractivity characteristics). Factor customer support (A = 0.653., P < 0.001) and
loadings ranged from 0.347 to 0.876, which personal-choice helper (A = 0.610, P < 0.(01);
exceed the threshold value of 0.30 (Kim and reliability, an c-tailer', good reputation
[Link], 1(78). However, delivery, ease of (A = 0.883., P < 0.001), security (A = 0.633,
ordering, and product display on \V'eb pages were P < 0.001), and primey (A~' 0.652, P < 0.001);
excluded in subsequent analyses because of their and tlavigation~ tinle to get to an e-tailer's
crossloading on factors 1 and 2. The final results of homepage (A = 0.S78, P < 0.001) and time to
the exploratory factor analysis are pre..,ented in download Web pages on the Web site (A = 0.687,
Table V. P <: OJ)Ol). TI,e e-shopping attributes confirmed
in the final analysis ,"vere grouped into four
different e-shopping cJmracterisrlc$.,
Confinnatory factor analysis
CFA wa. conducted to confirm the prespecifed Multiple regression analysis
dimensions ootained in the foregoing exploratory Table VII =mari= the results ofmultip!e
factor analysis. In general, the results supported a regression analysis. The VIF index showed no
measurement model that included jOllr overriding significant multicollinearity problem (>:t:ter i!l ai..,
characteristics U~o = 21.43, RMSEA = 0.018, 1996). The lour independent variables (i.e.
GFI = 0.(80). Price, other customers' postings., merchandise, inter-activity, reliability, and
promotions, and provision of waiting infonnation navigation) revealed in the CFA were regressed
were removed from the measurement model owing across consumers' attitude toward online
to reliabilities of less than 0.30 (Bettencourt and shopping. The overall modd is significant
Brown, 1997; Table VI - only final results are (RC = 0.17, F •. 2.17 = 13.01, fJ < 0.(01). Tne
shown). Other than these attributes, all results indicate that consunlers~ attitude toward
standardized fuctor loadings were statistically online purchasing is a functi,)Q (positively) of
significant (p < 0.001). The validity of the merchandL<;e (/3 = 0.25, P < 0.001) and reliability
measures was examined by the index of the ([3= O.19,p < 0.01) attribute". Interestingly,
proportion-of-variance extracted. All proportions t110Ub~1, their attitude is not signific..'ulUy related
in the index were higher than 0.50., which indicates (p > 0.05) to intt-ranivlty or navigation Web site
that the overall amount of variance in e-shopping dimensions.
attributes was captured by the correspondinJ
characteristics reported in Table VI (Hair et aI.,
1998).
Discussion
:esul15 of expiOfatory factor analysis
The findings of this stud, present important
Factor loadings e~~.[Link] [Link] that consumers conl;,idcr in
19 attributes Factor 1 Factor 2 Factor 3 Factor 4
their e-tailer evaluat1ons. For merchanilismg
nformation 0.695 charactt:'ristics, [Link] appear to focus on
f merchandise 0.783 product infonnatioD Whffi they evaluate e-tailers.
:ively low price 0.579 P~vious research suggests that online shoppers
support 0.539 seek detailed information about products and
:hoiee helper 0.474 services rather than SC'llSory attributes) such as
tamers' postings 0.815 visual cues (~ger,Jtu . ?l aL ~ 2000). 'Ibis may be
1$ 0.439 largely attributed. to the nature of Interner
lItation 0.667 shopping in which con.o;,umeTh Glnnot touch or see
0.868
a product. Accordingly, consumers tend to rely on
0.876
product information prO\~ded bye-tailer \\"reb
.t to home pages 0.724
pages. This result lS consjstent with rh()se obtained
)wnload Web pages 0.837
in prt"\rious studies (Lynch and ,t\.riely', 2000~ \\'·ard
formation 0.657
and Lee, 2000). Also, online consumers appear t<>
'a 0.688 0.617 0.610 0.843
seek a variety of merchandise through online
·action method: principal component analysis; rotation method: V?rim2x
shopping; rhI~ finding if) a15.o consistent \vith
norrnalization
con,:;,um~rs' f'~~H..'ti()n:s in physical retail srl"Jre;:; and

508
 ______~~~_;s~~~~_~~_ ~~:~~~~)_~ ~~ctt',i$tks
- ------ ----- -- --- -- --- ----- -
Heejin Lim and Alan 1. Dubinsky Vdume 18, ~JuMber 7,2004, 500--513

Table VI Results of measurement model: confirmatory factor "nolysis

factor Total·item Proportion of
E·shopping attributes loadings t·[Link] reliability' variance extracted
-~.------~------ ----'------.- '"-~~-.------

Merchandise characteristics 071 0.78

Product information 0]07 10.31 0.50
Variety of merchandise 0.757 10.91 357
Interactivity characteristics 0.57 0.70
Customer-support 0.653 8.71 0.43
Personal choke·helper 0.610 8.15 038
Reliability characteristics 0.84 0.76
Good reputation 0.883 12.65 0.78
Security 0.633 9.18 DAD
Privacy 0.652 9.50 0.43

Navigation characteristics 0.76 0.81

Time to get to home pages 0.878 1114 0.77
Time to download Web pages 0.687 9.33 0.47
Indices of goodness-of-fit
Chi-square (d1) 21.43 (20)
RMSEA 0.D18
GFI 0.980
AGfl 0.955
NFl 0.972
Note: 'Italiciled entries in this column represent overall reliability for "ach construct

II Results of multiple regression analysis face communication \-vith saleS asso(.iat'Cs, e-[ail~
Attitude provide a similar service through either personal
Mean SO Standardized coefficients (·value inrer-a<:tiviIY (i.e. call center) or machine
interactivity, such as CUstomer support 0.e.
ndise 30.59 9.71 0.254 3.737' downloading software, e-form inquiry, order
ivity 26.19 8.83 0.189 2.789"
tracking) and ell Olce helpers. The findings of our
ily 33.29 9.68 - 0.004 -[Link]
study confirm that customer support and choice
ion 29.69 9.17 0.073 1.112
helpers an! key intemeti,';t)' dimensions of e..,;rores.
7 (F•. ,., =13.01, p<0.OO1)
Reliability has emerged as a critical infiuence on
1= 252; 'significant at the 0.01 level consumers' purchase decisions in home-based
shopping (McDonald, 1(93). In online
transactions, consumers are likely to release
in-home shopping (such as 1V and paper
personal information as well as credit card
catalogs). In addition, IDe findings of multiple
infonnation. Accordingly, they tend to be mOn!
regression analysis imply that the merchandi:;e
careful than in tf'dditional stOn! shopping by
attribute is positively associated with respondents'
examining e-tailer credibility prior to making a
attitude toward online purchase.
purchase. In determining an e-retailer's n!liability,
In the online shopping litef'dture, the term
consumers appear to evaluate e-tailers' n!putation,
"interacti\~ty" has been used to refer to different
as well as security and pri,'acy policies. Also, the
e-store characteristics:
=nlt of multiple regression analysis n!veals that
(J) speed of interactivity between a person and
reliabilitY is a significant predictor of respondenu;'
related devices (i.e. computers, modem,,) (e.g.
attitude toward online purchase.
Alba et ai., 1997; Novak et al., 2000); and
In online shopping, customers tend to be
(2) internctivity between a customer and e-tailers
sensitive to waiting time (Dellaert and Kahn,
(e.g. Ghose and Dou, 1998).
1(99). Despite its importance in pn!dicting
In our study, interactivity was observed to be a consumer behavior on the In'ernet, many
service that customers can receive from the researchers have not delineated navigation
Internet, just as they can from a salesperson in a chumcteristics dearly. Inste3d, th"y have tended tQ
bricks and mortar store. Through online shopping, employ a single term - "ease of browsing" - and
customers and retailers use different form., of induded it under "convenience" dimensions (e.g.
communication. While the physical store pro\~des Shim et aL, 2001; Szymansky and Hise, 2000).
customers with personal interaction via face-to- The re!~u1ts of our factor analysis, however,. reveal
509
 Ccnsumers' perceptions of e-shcpping charocteristics Jourrkll of Se~ Marketing
- ,--.~,---- ---- ----- ---------.-,----------
Heejin Lim and Alan 1. Dubinsky VOlU;;;18~7:;;004-:-500_51i---~--·-

that navigation dimensions constitute an the present work also highlight the importance
independent construct, which is characterized by: online shoppers place on the variety of
time required to get to the e-tailer's merchandise in an online outlet. By utilizing
homepage; and consumer profiles and sales tracks, e-tailers should
time required to download an e-tailds Web be able to provide the variety of merchandi.~e that
page. meets their customers' needs and preferences.
The findings obt2ined demonstrate that Reliability of e-tailers was found to be a
consumers' attitude toward online shopping is significant factor when cu~tomers evaluate a
commercial s-i[e.~ as well as- influencing COfl![Link].':/
positively related to their perceptions of Web site
merchandise and reliability attributes, but online purchase attitude. Indeed, one coru,umer
surprisingly unrelated to inter-activity or navigation SlIITey showed that about 88 percent of online
characteristics. Specifically, consumers deVelop a consumers make a purchase through the Internet
positive affect toward online shopping to the extent using credit cards. It also reponed that about 60
that th~' perceive a Web site to provide d"tailed percent ofInternet users are concerned about their
product information and merchandise \'3riety, as credit card number being stolen when using a
well as mechanisms that enhance feelings of trust commercial Web "~te (eMarketer, 2001e).
in the e-tailer. Interestingly, though, consumers' Furthermore, privacy has emerged as a critical
affect is seemingly unrelated to a their perceptions concern among online consumers. Although
of an e-tailer's offering customer support and customers' personal information is a crucial asset
personal-choice helpers or Web site navigation for retailers, online shoppers appear to fean; ha,,,
speed. The latter two findings may be a function of about revealing their personal information on the
the state of current 'X'eb sites. Adequate customer Inrernet. To convince online customers that their
support and navigation speed are a necessary, out personal inlormation will not be violated, e-tailers
insufficient, condition for e-tai1er success. Because should provide con.,umers ",ith their privacy
of early Internet shoppers' complaints about poor police" as well as a guarantee that the information
Web site design, interactivity quality, and will not be misused (eMarketer, 2001a).
navigation speed (Dellaen and Kahn, 1999), Although interauivity of e-tailers was found to
perhaps e-tailers have made steadfast efforts to be a signiiicant factor when customers evaluate a
enhance these qualities. If so, online shoppers may commercial site, it was not discerned to influence
have reduced their fonner concerns about these consumers' onlint: purchase anitude. ThIs finding,
two attributes, thus vitiating the impact of these however, does not imply that e-tailers should
two attributes on online shopping attitudes. ignore interactivity issue•. CFA results suggest that
online customers desire the equivalent quality of
seC\~ce that might be pro\'ided in a physical store,
Managerial iInplications
TIu-ough Internet commercial sites, online such as two-way communication [Link].
marketers can collect a plethora of consumer shoppers and salespeople. As such, online
information, such as age, gender, and zip code. shoppers apparently ezpect to experience a high
Also, they can track consumers' interests and degn,e of customer service from e-tailers in the
prelerences. To parlay this data rich advantage of forms of software dm,uloading, e-fornl inquiry,
electronic commerce, online marketers should order status tracking, customer comments., and
understand consumers' online shopping behavior feedback: so, such tools should be provided by
so that they can develop effective marketing [Link].
programs. Online customers appear to be sensitive to the
The ,-unent study eJ>:plored consumers' wairing time required to dO'WTIload 'X'eb pages
evaluative criteria of an online retail outlet. The (Dellaert and Kahn, 1999; Weinberg, 20(0),
findings of this study are helpful for e.-railers as a although it is not asso<."iated ....ith their attitude
general guideline for Web site design. In terms of toward online purchasing. Notwithstanding this
merchandisiug and its positive relationship with result, CFA fi.,dings suggest [hat <-tailers still need
comumers' attitude toward online purchasing, to be circumspect about Web site waiting time
e-tailers should pay acute attention to the contents given that it is an evaluative criteria of online
of their Web pages through which consumers reach shoppers. \Vaiting time on the Internet may vary
purchase decisions without physically touching or because of several factors: the extent of
seeing a product. In particular, online consumers sophistication of a Web page, the number of
appear to place emphasis on quality product graphics on a Web page, and the number of people
intormation. Moreover, effective descriptions of who log on to the Internet. Although e-tailers
the products on an e-tailer's Web site have been cannot control every factor that affects
found to mitigate online customers' price do....n1oading time, they can design their 'Xreb
sensitivity (Shankar et ai" 1999). The findings of pages to be time-efficient. This does not mean Web
510
 CilASlWllefS' perceptions of e-shopping characteristics .!ou:'na.I of SeMces Marketing
lI...jln Lim iJIld AI.n 1. Dubinsky ---~um.18~T 2004· 5<JO.513--- -

designers should forego cutting-edge graphic characteristics of an online store as a predictor of

technology. Rather, they should avoid extensive online shopping behavior m:lY vary depending on
utilization of graphics and animations, as well as the product category - and even, perhaps, owing
consider downloading time when they design to the brand being sold (e.g. manufucturerfservice
e-tailer sites (Dellaert and Kahn, 1999). principal's brand versus an e-tailer's own brand).
Consequently, e-tailers should consider how to Also, there might be other factors that influence
provide customers with detailed product shoppers' evaluations of an e-tailer, such as the
information and increased interactivity without level of their expertise or experience in online
annoying them with excessive downloading time. shopping: so, such variables should be examined in
subsequent work.
Limitations and future research
This study employed a student group to measure
their attitudinal beliefs about online retail outlets.
Although college students account for a major
References
portion of online consumer groups, the sample
may be biased toward those who are younger and Alba, I., lynch, J., Weitz, 8., Janiszewski, C, lutz, R., Sdwyer. A.
more educated compared to the general consumer and Wood, S. (1997), "Interactive horne showJ1g:
population. Also, 75 percent of the sample consumer; retailer, and manufacturer incentives to
consisted of females and 19 percent had not made participate in electronic matketplaces", Journal of
an online purchase: this situation may decrease MirleulI!J, Vol. 61, July. pp. 38-53.
representativeness, Accordingly, the results of the Belt O.R, Ho, I-H. and Tang. (5. (1998), "Determining where to
shop: fixed and variable costs of shopping", Journal of
current work might not generalize to the general
MarItemg Research, Vol. 35, pp. 351-69.
consumer population. Also, the study is Bellman. s.. lobse. G..L and Johnson, U (1999). "Predictors of
geographically limited, because data were online buying behavior", Cnmmunicarions of the ACM
collected from a Midwestern university. Therefore, VoL 41 No. 12, pp. 31·8.
utilizing a random sampling of general consumers Benterud, I and Sw. E. (1993), "TV-shopping in Scandinavia:
nationwide should help overcome these consumer satisfaction, dissatisfaction and complaining
shortcomings. behavior", Journal of COfI5umer Satislaaion,
Dissatisfaction and Complaining 8ehiliJior; Vol. 6,
In characterizing e-store image, previous
pp. 196--203.
researchers have not been lucid about which Berry, l.L (1969), "The COffiponellts of department store image:
dimensions characterize interactivity. Novak et aI. a theoretical and empirical analysis", Journal of Retailing,
(2000) identify interactivity solely in terms of Vol. 45 No.1, pp. 3-20.
speed. They found that Web site design affects BetteocOOl!, LA. and BrOWll, S.W. (1997), "Con1a<:t employees:
speed of interacti vity between online consumers relationslJips among woriqJiace fairness. job satisfaction
and the medium (i.e. computer, related device) in and prosorial service behaviors", Joumal of Retailing.
Vol. 73 No.1, pp. 39-61.
measuring customer experience on the Internet.
BIattbeIy, R.C. and Deightoo. 1. (1991). "Illteractive marketing:
However, in terms of the scales that measure exploring the age of addressability", Sloan Management
interactivity - which is equivalent to navigation Reviev.( Vol. 31 No.1, pp. 5--14.
charactcriJfic.' in our study - Novak ct aI.'s study Chang. Y., Bums. LO. and Noel. (J. (1996). 'Attitudinal versus
had almost identical items as ours: nonnative influence in the putdlase of bland-name casUil!
(1) waiting time between a person's actions and apparel". Family and C()IlSumer Sde1Jces Re>eardl
the computer's response; and Journal. Vol. 25 No. f, pp. 79-109.
(2) Web page dowuloading time. Chiou, J.·S. (2000), "Antecedents and moderators of behaviOl<l'
intention: differences between US and Taiwanese
Thus, the term mteractivily needs further research students", Geneck; SodaJ. and General PsydwIogy
in order to clearly define this characteristic. Monographs, Vol. 116 No.1, pp. 105-24.
This study examined consumers' evaluative COiI5lJ11lff KlflOI75 Buying Guide (2000), "Catalog shopping",
pp. 14-15.
criteria for an e-tailer without specifying a product Oegeratu, A, Rangaswamy. A. and Wu. 1 (2000), "(onsumer
category. Howe,'er, store characteristics that affect dtoice behavior in online and traditional [Link]:
store image may differ across product types and the effects of brand name. price. and omer search
store types (Klein, 1998; Tigert, 1983). For attribures", International Journal of Research i1
instance, Yoo and Donthu (200 I) found different Marketing, Vol. 17, pp. 55·78.
effects of each site quality (i.e. ease of use, aesthetic Dellaert, B. and Kahn, B.E. (1999). 'How toleral>le is Dealy?
design, processing speed, security) QIl consumers' Consumer.;' evaluations of Internet Web sites after
waiting", Journal of Interaan... Marketing. Vol. 13 No.1,
decision-making across different product
pp.41-54.
categories (e.g. apparel, electronics, food and Doothu. N. (1'199), "The IIltemet shopper", Journal of
drink, music., etc.) Therefore, future research Advettising Research, Vol. 39 No.3, pp. 52·8.
$hould seek to develop different sets of e-shopping Eagly; A.H. and Chaiken, S. (1993), The Psychology of AttmJdes,
attnomes within a specific context. The resulting Harcourt Brace Jovanovich, Loodon.
511
 (00;"",,,'" perceptions 01 "'hopping chafacteristics Journal of Servia» lII....ling
----~----~~--

Heejin Lim and Alan 1 Dubinsky Voiume 18· [Link] ,. 2004·500-513

Eastlid, MA (1989), "Shopping motives of the mail-catalog Lynch, 1.G_ and Ariely, D. (2000J. 'Wine ooline: search cost affect
shopper", unpublished doctoral dissertation, Purnue competition on price, quality, and distribution", Markl!ting
University, West lafayette, IN. Science. Vol. 19 No.1, pp.83-103.
Eastiid, MA and Feinberg, RA. (1994), "Gender differences in McDaniel, S.W. and Burnett. JJ. (1990), 'COIlSIlIIleI" religiosity
mail-catalog patronage motives", Journal of Direcl and retail store evaluative critena', Journal of the
Markl!ting, Vol. 8 No.2, pp. 37-44. Academy of Markl!tiJg Science, Vol. 18 No.2, pp. 101-12.
eMarketer (2001 a), ePrivacy & Security Report, New Yorl<. NY. Md)O/IaId, W.J. (1993), 'COflSIlIl1eI" preference structure
eMarketer (2001b). "Pnvacy fears online", available at analysis: a managerial tool for understanding apparel
[Link] catalog maliet competition", Journal of Direct Marketing.
eMalieter (2001 c), "The great online privacy debate: Part 2", Vol. 7 No.1, pp. 20-31.
available at: [Link] Mathieson, K. (1991), "Predicting user interttions: comparing the
eMarketer (2001 d), "Xmas e-tailing: more on-time delivenes. technology accep\aflce model with the theoIy of planned
more dissatisfaction". available at [Link] behavior", Information Systems Researd!, Vol. 2 No.3,
eMarketer (2001 e), "eConsumers get lots of credit", [Link] pp. 173-91.
at [Link] Momso.., P.D. and Roberts, lH. (1998). "Matching electronic
Ernst & Young (1998), "Internet shopping: an Emst & Young distribution channels to product characteristics: the role of
special report", Stores, Suppl. 2, pp. 1-29. congruence in consieeratJon set formation", Journal of
Fishbein, M. and Ajlen, I. (1975), Belief. Attitude, Intention, and Business Research, Vol. 41 No.3, pp. 223-9.
Behavior: An IntrodUc1ion to Theili)' and Researd!. Neter. l., Wasserman, W., Nachtsbeim, CJ. and Kutner. M.H.
Addisoo-Wesley, Reading, MA.
Fishbein, M. and MiddiestadL 5. (1995), "Non<:09I'itive effects
011 attitude fonnation and change: fact or artifact?",
Journal of Consumer Psyd!ology. Vol. 4 No.2, pp. 181-202.
(1996), Applied linear Statistical Models. McGraw-HilV
(!Win, Chicago, IL
Novak, lP., Hoffman, DL an<! 'I'UIlg. Y.F. (1000), "Measuring the
customer experience'in ootine environment a structural
_ ..

Furger; R. (1996), "Shopping online: what you need to know",

PC World. 14 June, pp. 320·1.
modeling approach', MarkEting Sden~ VoI_ 19 No. 1.
pp.22-42_
Ryan, MJ. (1982), 'Behavioral intention lonnation: the
z
« ~
I
FuturePages (2002), "Why is it important to reach college
students?", available at: [Link]- internepeodency of attrtudinal and social influenc€ ..: ~ <III
variables', Journal of lOl15umer Researd!. Vol. 9,
[Link]
Ghose, S. and 000, W. (1998), "Imeractive lunc1ions and their pp_263-78.
:.:; '" >-
"
"j

impacts on the appeal of Internet presence srtes", Journal

Seaver, B.L and Simpson, E. (1995), • Mail order catalog design <C j <
and consumer response behavior: experimentatoo and IZl
of Advertising Researd!, March/Apnl. pp.29-43. -~!
Graphics, Visualization, and Utilization Center (1998), "GVU's
analysis", Journal of Direct Marketing, Vol. 9 No.3,
pp. &-20_
'"'
00 3 <
www user surveys", available at: [Link] Shankar. v., Rangaswamy. A. and Pusateri, M. (1999), "The o ~ ~
Gupta, S. and Chatlergee, R. (1997), • Consumer and corporate
online medium and customer price seJl5itiviW', ",,,",iog .:-1 0
-' .~."
adoptiOll of the World Wide Web as a commercial
I"'per. eB1JSiness Research Center (e8Re), Penn State
medium", in Peterson, RA (Ed.), Eledronic MarkEting and •
the COilSumer. Sage Publications, Thousand Oaks, CA.
pp.123-38.
University, University I'arl<, PA.
Sheehan, K.B. (1999), "An investigation of gender dffferences in
online privacy 0l!\CefJlS and resultant behavior", Joum;,)
•... 11l\•
Hair, J.F. Jr, Anderson, R.E., Tatham, R.l. and Black, W.e. (1998), ;:J
of Interaaive Martering, Vol. 13 No.4, pp.24-38.
Multivariate Data Analysis, Prentice-Hall, Eng~ Sheth, IN. (1983), "An Integrative theoIy of patronage
Cliffs, Nt Jlfl'fetence and behavior", in Darden, W.R_ and l.u<d!, R.f.
Janssen, M. and Moraga, J.L (2000). "Pricing, consumer search, (Eds), Pi1tronage Behavior and Retail Management. North-
and the size oflntemet maricet£", wori<ing paper. available Holland, New York, NY, pp. 9-28.
at httpJle<:Ommerce.m~tedu/papers Shim,S., East/d, MA, lou, Sol and WiIITWlgton, P. (lOOl), "An
Jeffrey; D. (1999), "Study says 2 of 3 consumers stilileel)' of Web ooline Jlfl'purchase intentions model: the role of intention
purchasing", Billboard, Vol. 111 No. 36, p. 85. to search" ,Journal afRerailing. Vol. 71 No.3, pp. 397A16.
Jiireskog. K_ and Siiroom, D. (1993), LlSREL 8: Structural Shimp, TA. and Kavas, A. (1984), "The theory of re"soned action
Equation Modeli"g with the SIMPLIS Command Language. applied to coupon usage", Jotll1lal of COll5umer Researci1.
5SI, Chicago, Il. Vol. 11, p. 1984.
Keen, LN. (1999), 'The attribute structure of Internet shopping: Spiller, P. and Lohse, G_L (1998), "A dassification of Internet
what is important and what tradeoffs are possible retail stores", IntErnational Joumal of Electronic
between internet, retail. and catalog formats?", Commef£E, Vol. 2 No.2, pp. 29-56.
unpublished do-'toral dissertation, Purnue University, West Szymansky, D. and Hise, R. r. (2000). 'E-satisfaction: an initial
lafayette, IN. examinatioo', Journal of II£tailing. Vol. 76 No_ 3,
Kim, 1.-0. and Mueller. C.W. (1978), Factor Analysis: Statistical pp.3og..22.
Method,; and Practical Issues, Sage Publications, Beverly TayJor. S. and Todd. PA. (1m)' "Decomposition and {JI)Ssover
Hills, CA_ effects in the theory of planned behavior. a study of
Klein, LR_ (1998), "[v2Ioaling the potential of interactive media COIISOfTl€r adoption intentions", International Joumal of
through a new lens: search ve"us experience goods', Researd! in Marketing, Voi. 12, pp. 137-55.
Journal of Business Researd!, Vol. 41 No.3, pp. 195-204. Trgerc DJ_ (1983), "Pushing the hot buttons for a successful
lindquist, J.D. (1974), "Meaning of image: a survey of empirical retailing strategy", in Darden, W.R. and Lusch, R.E (Eds),
and hypothetical evidence", Journal of RetailiJlg. Vol. SO Patronage Behavior and Retail Managemen~ North-
No.4, pp- 29-38. Holland, New York, NY, pp. 8<1-113.
Lohse, G.L and Spiller, P. (1998), "Electronic shopping: how do USA TodiJy Mi1gilZirJe (1997), "Changing attitudes towam TV
(Usto"",r interfaces produce sales on the Internet?", shopping", USA TodiJy Magazine, Vol. 126 No. 2627,
Communi",tions of the ACM, Vol. 41 No.7, pp.81-7_ pp.9-10.
512