Ps4

Effects of Multi-ChannelConsumers’ Perceived Retail Attributes

on Purchase Intentions of Clothing Products

Youn-Kyung KimSoo-Hee Park

Sanjukta Pookulangara

ABSTRACT. Consumers are increasingly shopping through multiplechannels, providing opportunities to retailers who can increase profitsby cross-selling products through multi-channels. The objective of thestudy was to examine the effects of multi-channel consumers’ perceivedretail attributes on purchase intentions for (a) brick-and-mortar stores,(b) catalogs, and (c) the Internet. Following a pre-testing, a Computer-Assisted Telephonic Interview (CATI) was utilized for data collectionwith 500 multi-channel consumers. Structural equation modeling re-vealed that multi-channel consumers perceive important retail attributesdifferently across the three channels (i.e., brick-and-mortar stores, cata-logs, and the Internet). Consumers perceived risks in personal security orbuying private merchandise while shopping in stores. Consumersshopped clothing items via the Internet due to access to a variety of itemsand convenience-related attributes. Additionally, consumers who per-

Youn-Kyung Kim is Associate Professor, Retail and Consumer Sciences, 244AJessie Harris Building, University of Tennessee, Knoxville, TN 37996-1911 (E-mail:[email protected]). Soo-Hee Park is Director of Research, Assessment and EvaluationDivision, Tennessee Department of Education, 2730 Island Home Boulevard, Knox-ville, TN 37920 (E-mail: [email protected]). Sanjukta Pookulangara is Assis-tant Professor, Dietetics, Fashion Merchandising, & Hospitality, Knoblauch Hall 140,Western Illinois University, Macomb, IL 61455 (E-mail: [email protected]).

Address correspondence to: Youn-Kyung Kim at the above address.

Journal of Marketing Channels, Vol. 12(4) 2005Available online at http://www.haworthpress.com/web/JMC

© 2005 by The Haworth Press, Inc. All rights reserved.doi:10.1300/J049v12n04_03 23

http://www.haworthpress.com/web/JMC

ceived higher level of cost tended to purchase clothing products throughcatalogs and the Internet. Managerial implications are provided formulti-channel retailers. [Article copies available for a fee from The HaworthDocument Delivery Service: 1-800-HAWORTH. E-mail address: <[email protected]> Website: <http://www.HaworthPress.com> © 2005 by TheHaworth Press, Inc. All rights reserved.]

KEYWORDS. Multi-channel retailing, direct marketing channels,Internet retailing, apparel shopping

The environment where the retail industry is mature and store expansionhas slowed to a crawl has challenged retailers to widen the range of distri-bution channels available in the consumer market. Many retailers are drop-ping single channel approaches such as brick-and-mortar stores orcatalog-only operations and are switching to multi-channel strategies bylinking their store operations with e-commerce and/or catalogs; Internet re-tailers also are venturing offline (Haydock, 2000; Schoenbachler &Gordon, 2002). This multi-channel consumer market is driven by factorssuch as an increasing number of dual-income families; a lack of consum-ers’ time; technological revolutions; and a myriad of shopping choices–notonly among different products and brands but also among diverse retailerformats such as brick-and-mortar stores, print catalogs, and online shop-ping electronic systems (Shim, Eastlick, & Lotz, 2000).

In fact, multi-channel retailing is gaining importance because a multi-channel approach generates more sales and profit for the multi-channel re-tailer than a single-channel, single-consumer approach (Hoover, 2001). Al-though multi-channel retailers face a number of challenges such as channelconflict, customer-retention issues, and the ability to integrate processes,they can take advantage of several opportunities including establishingbrand equity, leveraging advertising and marketing expense, leveragingdistribution and supplier networks, driving cross-traffic to multiple chan-nels, and accessing more customer information (Baker, 1999; Ernst &Young, 2001; Schoenbachler & Gordon, 2002).

In terms of the spending power, U.S. shoppers who bought throughall three channels of brick-and-mortar stores, catalogs and the Internetrepresent 34% of all shoppers and 78% of shoppers purchase from bothWeb sites and the brick-and-mortar stores (“The Multi-Channel RetailReport,” 2001). As such, it appears that the best chance to increaseprofit margin per customer lies with retailers with the broadest channel

24 JOURNAL OF MARKETING CHANNELS

http://www.HaworthPress.com

representation (Haydock, 2000; Reda, 2001; Schoenbachler & Gordon,2002). Already, most revenues in the apparel sector are driven by multi-channel concepts with a strong brand appeal, such as Victoria’s Secret,Lands’ End, J. Crew, Liz Claiborne and L.L. Bean (Chevron, 1999; Hill,2000; Tiernan, 2000).

While retailers recognize that a multi-channel strategy is essential tolong-term viability, the reality is that many lack the knowledge of howto successfully combine online, catalog, and physical retailing to form asuccessful retailing concept (“Integrating Multiple Channels,” 2001).In order to improve customer loyalty and retention rates, thereby in-creasing profits, retailers have to ensure that their customers stay withthem irrespective of the channel of shopping. Then, it seems critical forretailers to assess what retail attributes their customers perceive as im-portant for each channel and relate these attributes to purchase intentionthrough the channel. This understanding will offer the promise of moreprecise market analysis and marketing strategy development for cloth-ing products, which in turn will better meet customer expectations onthe multi-channel environment.

In the following sections, we first summarize the multi-channel con-sumer market for clothing products and then discuss the literature on re-tail attributes and purchase intentions for clothing products. Next, weillustrate the procedures and the results we obtained from testing the ef-fects of multi-channel consumers’ perceived retail attributes on pur-chase intentions for each channel through structural equation modeling.Lastly, we present our managerial implications as well as suggestionsfor future studies.

MULTI-CHANNEL CONSUMER MARKETFOR CLOTHING PRODUCTS

Brick-and-mortar retailers still dominate the apparel market, with on-line retailers making large strides in gaining market share. In 2000, the to-tal U.S. apparel sales in brick-and-mortar stores accounted for 92.9% ofthe total apparel market; catalogs, 3.9%; the Internet, 3.2% (“Retail Ap-parel Sales Statistics & Trends, 2000”). These figures compare to 88.6%for brick-and-mortar stores, 9.4% for catalogs, and 0.6% for the Internetin 1999 (“Retail Apparel Sales Statistics & Trends, 1999-2000”).

The percentage of catalog sales of apparel products seems to continu-ously have declined from 1980s and early 1990s when they enjoyeddouble-digit annual sales growth (Gordon, 1994). Although their de-

Kim, Park, and Pookulangara 25

clining growth rate can be attributed to competition with the Internetsales, much attention is still needed to examining attributes of catalogsthat are important to retain and regain their customers.

The range of products sold through the Internet has been widening.Fast-selling products on the Internet used to be those products aboutwhich the shopper already had sufficient information, such as books,computer products, travel, health and beauty products (Reda, 2001;Schaeffer, 2000). As technology improves, items previously thought tobe saleable only in a touch-and-feel environment (e.g., apparel, jewelry)are enjoying more widespread sales. Online apparel retailers in theUnited States and Europe (e.g., Lands’ End, J.C. Penney, and GalleriesLafayette) have increased profitability by giving consumers access tointeractive try-on sessions such as the “virtual dressing room,” “digitalsupply chain” and “online fit prediction” (Abend, 2001; “Lands’ EndImproves,” 2001). Furthermore, the recent integration of apparel manu-facturers into direct Web selling (e.g., Fabra U Inc., Shawnee GarmentManufacturing), as well as the continuing incursion of traditional retail-ers into the online channel, has fueled the clothing surge. In fact, apparelranks in the top five product categories sold through the Internet in theUnited States (Global Online Retailing Report, 2000). The growing on-line retailing of clothing products means increased consumer buyingthrough multi-channels, because brick-and-mortar stores are alreadythe major media for selling clothing products.

RETAIL ATTRIBUTES

Consumers may patronize or switch channels and/or retailers de-pending on their perceived attributes (Paulins & Geistfeld, 2003; Wilde,Kelly, & Scott, 2004). This channel selection may be based on their par-ticular needs in specific situations. For instance, consumers’ reasons forthe selection of the Internet versus the brick-and-store for their shop-ping can vary for different consumers and in different situations evenfor the same consumer. Some consumers may shop mainly in thebrick-and-mortar stores because they like synchronous human contactfor receiving services in a safe shopping environment, whereas otherconsumers or the same consumers may use the Internet for such reasonsas being able to shop in the comfort of home and fast transaction with-out having to spend time and energy traveling to the store and findingproducts wanted and waiting in check-out lines. Thus, multi-channel re-


tailers have to be prepared to address the unique challenges of servingcustomers through multiple channels.

Our review of literature suggests that the construct of perceived retailattributes encompasses benefit and cost components. From a con-sumer’s point of view, a consumer wants to obtain the greatest possiblesatisfaction from a consumption activity, while he or she seeks to mini-mize costs needed to accomplish a given shopping activity. Thus, we as-sert that the two components, namely, benefit and cost, should beviewed as the building blocks of how we conceptualize the retail attrib-utes. Zeithaml (1988) posited that consumers vary in what they want toget and what they are willing to give or expend. That is, just as the im-portance of benefits varies across consumers (i.e., some may want a va-riety of merchandise, and others low price, convenience, or newinformation), the importance of costs also varies (i.e., some are con-cerned primarily with money, others with time or energy).

A comprehensive analysis of the literature on the benefit componentof retail attributes reveals multifaceted dimensions such as value, as-sortment, service, convenience, confidentiality (i.e., security, privacy),atmosphere, and community involvement (Jarvenpaa & Todd, 1997;Linquist, 1974-1975; Shim et al., 2000). Among these dimensions, value,assortment, service, convenience, and confidentiality are most relevantto the three channels (brick-and-mortar stores, catalogs, and the Internet)that are the scope of this study. Equivalent examples can be provided foreach dimension. For instance, the “Convenience” dimension consists oflayout of the store (or catalogs or the Internet), saving time (e.g., nolines and no traffic for stores; finding the right product/product categoryfor catalogs and the Internet), and up-to-date and unique items. The“Confidentiality” dimension constitutes privacy (e.g., privacy to buy prod-ucts like lingerie, etc.) and security (e.g., personal security for stores, se-cure credit card information for catalogs and the Internet). The otherthree dimensions–Value, Assortment and Service can be applied to allthree channels. “Value” includes good quality and reasonable price; “As-sortment” assures access to a variety of the same kind of products (e.g.,styles, colors, sizes), access to different products, and availability of na-tional or designer brands; and “Service” refers to good customer serviceand easy return of items.

Cost, as another component of retail attributes, includes “money,”“time,” and “energy” (Downs, 1961; Kim & Kang, 1997). “Money” spentto acquire a product is a cost that is applied to any channel. However,catalogs and the Internet involve shipping and handling costs, which arenot present in the case of brick-and-mortar stores that may instead re-


quire transportation cost. “Time” is spent traveling to the store and find-ing a parking space in the case of brick-and-mortar stores. For shoppingvia catalogs or the Internet, time is spent locating products as well ascompleting a transaction. “Energy” expended on brick-and-mortar storesinclude waiting in checkout lines, finding the product, and fighting withtraffic and parking. While shopping on the Internet, energy is expendednavigating through the Web pages to find products and dealing withWeb site malfunctions (e.g., broken links) as well as electronic check-out. While shopping through catalogs, energy may be expended findingthe right product (Kim, 2002).

CLOTHING PRODUCTS: PERCEIVED RETAILATTRIBUTES AND PURCHASE INTENTIONS

Consumer attitudes toward the retail attributes influence purchase in-tentions (Jarvenpaa & Todd, 1997; Kim & Kang, 1997; Shim et al.,2000). As Hoyer and Alpert (1983) pointed out, “consumers will con-clude that certain important (and if consciously processed, salient) at-tributes discriminate well among alternatives while others do not, and itis the discriminating or determinant attributes which play the major rolein producing a choice” (p. 80).

Studies on store attributes that influence purchasing clothing productshave been limited. Among the limited studies, Shim and Kotsiopulos(1992) discovered that store attributes of quality/variety and price/re-turn policies affected patronage behaviour of discount stores; quality/variety, brand/fashion, price/return policies were important attributesinfluencing patronage behaviour of specialty stores. Kim and Kang’s(1997) study, although not limited to clothing products, examined theconsumers’ perception of shopping costs and its relationship with retailtrends. The study highlighted the retail attributes that include both bene-fit and cost components in a brick-and-mortar retail format in the con-text of a shopping mall. They found that all three cost components (i.e.,money, time, and energy) along with economics, service, institutionalimage, convenience/safety, atmosphere, easy return, and selection af-fected consumer purchase intention. More recently, Paulins and Geistfeld(2003) reported that store preference was influenced by type of clothingdesired in stock, outside store appearance, shopping hours, and adver-tising.

Several researchers have identified the important attributes that consum-ers seek from catalogs. In terms of clothing purchases, convenience has re-


peatedly been found to be a principal reason for favoring catalog shoppingover in-store shopping (Eastlick & Feinberg, 1994; Jasper & Lan, 1992;Kwon, Paek, & Arzeni, 1991; Shim & Bickle, 1994). Other benefits con-sumers seek from catalog shopping for clothing include wide product as-sortment (Shim & Drake, 1990), high level of product quality (Eastlick &Feinberg, 1994), low prices and ease of return (Eastlick & Feinberg, 1994;Shim & Drake, 1990), and credit availability (Kwon et al., 1991).

Retail attributes also have been linked to online shopping of clothingproducts. Kunz (1997) found that online, apparel consumers valuedmerchandise quality, merchandise variety, and customer service. Ac-cording to Taylor and Cosenza (2000), when shopping online for cloth-ing, consumers rated the functional attributes such as price, ease ofmovement and ease of return as important.

In relating perceived important retail attributes to purchase intention,Then and Delong (1999) suggested that consumers tend to buy more ap-parel online if they perceive such features as a convenient and secure sys-tem of ordering, return policy, focus on product display, and the offeringof products that have a range of acceptable fits as opposed to a precise fit.According to Shim et al. (2000), for sensory experiential products (e.g.,apparel and accessories), consumers are less likely to be influenced byfunctional attributes such as fast transaction service and speedy shoppingthan they are for cognitive products (e.g., books, computer software, mu-sic and videos). This is supported by Verton’s (2001) argument that a per-sonalized shopping experience via various incentives and virtual imagetechnology is important to encourage apparel consumers to shop online.On the other hand, Watchravesringkan and Shim (2003) found that onlinepurchase intentions for apparel products were predicted by attitudes to-ward secure transaction (e.g., payment security, consumer informationprivacy, return policy, minimal cost and time for return, and productshopping guarantees) and speedy process. Kim, Kim, and Kumar (2003)identified product and convenience (e.g., variety of merchandise, conve-nience, reasonable price, adequate sales information) and service (e.g.,good customer service, easy of payment options, ease of navigation) as af-fecting behavioural intention to purchase clothing online.

The aforementioned studies have been limited to channel-specificanalyses, not comparing across channels. Further, cost component wasnot fully examined in assessing consumer purchase intention, especiallyin the case of catalogs and the Internet. Because studies on multi-channel-ing have been relatively limited, it is not surprising that there exists no de-tailed framework for understanding channel choice. Current trends,however, assert that the reliance on a single channel will probably be an


exception rather than the rule (Black, Lockett, Ennew, Winklhofer, &McKechnie, 2002). By examining what retail attributes are important tomulti-channel shoppers and relating them to their purchase intention ofeach channel, retailers can develop effective strategies for clothing prod-ucts that will better position them against their competitors.

OBJECTIVES

This study provides an empirical understanding of the retail attributesmarketers should consider when they want to attract and retain the multi-channel buyer. The objective of the study was to examine the effects ofmulti-channel consumers’ perceived retail attributes on purchase inten-tions to buy clothing products for (a) brick-and-mortar stores, (b) cata-logs, and (c) the Internet by using a quantitative modeling of primary datawith multi-channel consumers.

METHODS

Pretesting

In order to check content validity and make minor adjustments priorto main data collection, the survey instrument was pretested with con-sumers (n = 115) who had shopped through catalogs and the Internet.These consumers included students, faculty members, and staff of a ma-jor university in the Southwest. Based on the pretest, items were revisedto ensure readability and a logical flow of questions. The survey instru-ment was transcribed for the telephonic interview.

Measures

The measures included retail attributes, purchase intention, and de-mographic information.

Retail attributes. Retail attributes were measured for each of the threeretail channels (i.e., brick-and-mortar store, catalog, and the Internet).The scale of retail attributes encompassed both benefits and costs. Twelveitems reflecting benefits were selected based on the criteria that the bene-fits should be able to be applied to all three channels. They were derivedfrom two studies (Jarvenpaa & Todd, 1997; Shim et al., 2000) and in-cluded “access to a variety of same kind of products (styles, color, sizes),”“access to different products,” “availability of national or designer


brands,” “layout,” “good consumer service,” “good quality of product,”“reasonable price,” “privacy (e.g., privacy to buy products like lingerie,etc.),” “security,” “saving time,” “up-to-date and unique items,” and “easyreturn of items.”

Some of these items were followed by appropriate examples for eachchannel. For instance, layout was specified as “layout of the store andthe product,” “layout of the catalog,” or “layout of the web page andease of navigation” (e.g., clicking links). Security was exemplified as“personal security” for stores; “secure credit card information” for cata-logs and the Internet. Saving time was exemplified as “no lines and notraffic” for stores; “finding the right product/product category” for cata-logs and the Internet. Respondents indicated the level of importance foreach item and each channel using a 5-point rating scale: 1 (very unim-portant) to 5 (very important).

Cost consisting of money, time, and energy also was measured for eachof the three channels based on a 5-point rating scale: 1 (I spend almostnothing) to 5 (I spend far too much). Consumers responded to how muchmoney, time, and energy were spent while shopping through each channel.Appropriate examples were provided for each channel as follows:

Brick-and-Mortar Stores

• The money you spend for product and other shopping related costssuch as gas, parking, and childcare.

• The time you spend traveling to store, parking, checking out atcash register, etc.

• The energy you spend for the trip to the store, finding a parkingspace, and taking care of children while shopping.

Catalogs

• The money you spend for product and other shopping related costssuch as shipping and handling.

• The time you spend flipping the pages of the catalog placing the or-der, waiting for the transaction to get through, etc.

• The energy you spend to flip through the pages, finding the rightproduct.

The Internet

• The money you spend for product and other shopping related costssuch as shipping and handling.


• The time you spend navigating the web-site, waiting for the webpage to load, waiting for the transaction to get through, etc.

• The energy you spend to find the right web-site, finding the product,etc.

Purchase intention. Purchase intention for each of the three channelswas measured as the frequency of a consumer’s purchase intentions ofclothing, jewelry, or accessories in the next 6 months on a 7-point ratingscale: 0 (never) to 6 (6 or more times).

Sample and Data Collection

A Computer-Assisted Telephonic Interview (CATI) was utilized fordata collection. Nationwide telephone numbers of 5,000 multi-channelconsumers who had purchased products from the Internet and catalogswere purchased from a leading marketing firm. Out of randomly se-lected 6,000 numbers by the firm, 4633 numbers were valid numbersand were contacted. However, 167 consumers were not qualified for theinterview and 800 consumers refused to participate. Five calls weremade to each potential respondent until 500 interviews were completed.

As illustrated in Table 1, a demographic profile of the respondents in-dicated that approximately 65% of the respondents were female; about69% of the respondents were between 30 and 59 years of age; 80% ofthe respondents were married; and 92% were Caucasian. Fifty four per-cent of the respondents reported no children living with them, and ap-proximately 27% had 1-2 children. Annual household income had afairly even distribution across the categories with 54.2% reportingincome in the range of $30,001-$80,000.

Data Analyses

To establish an initial measurement model, exploratory factor analy-sis (EFA) was performed. This study adopted maximum likelihood forestimation method, squared multiple correlation for prior communality,and an oblique method for rotation. To evaluate measurement modelsand to investigate relationships among the latent variables, LISREL 8(Joreskog & Sorbom, 1993) was utilized. A weighted least squares (WLS)method with data from polychoric correlation and asymptotic covariancematrices was used in this analysis. The WLS estimation technique withpolychoric correlations was preferred since this study adopted a Likert-type scale with five levels to measure retail attributes. Furthermore, the



TABLE 1. Demographic Profile of Respondents

n %a

Gender

Male 173 34.6

Female 327 65.4

Age

under 20 2 0.4

20-29 52 10.6

30-39 86 17.5

40-49 117 23.7

50-59 137 27.9

60-69 67 13.4

70 or over 32 6.5

Marital Status

Married 398 79.6

Single 99 19.8

Children Living at Home

0 271 54.2

1 72 14.4

2 61 12.2

3 16 3.2

4 2 0.4

Annual Income

$10,000 or less 5 1.0

$10,001-$20,000 14 2.8

$20,001-$30,000 43 8.6

$30,001-$40,000 55 11.0

$40,001-$50,000 62 12.4

$50,001-$60,000 53 10.6

$60,001-$70,000 59 11.8

$70,001-$80,000 42 8.4

$80,001-$90,000 27 5.4

$90,001-$100,000 18 3.6

over $100,000 74 14.8

Ethnicity

Caucasian 461 92.2

African American 11 2.2

Hispanic 3 0.6

Asian 5 1.0

Native American 5 1.0

Other 9 1.8

aNumbers do not total 100% due to the missing data.

WLS technique is desirable because it is an asymptotically distribu-tion-free method and does not require normality in the variables. The in-put data matrices were generated from a sample of 500 participants.Also, this study adopted a two-stage approach to structural equationmodeling (Anderson & Gerbing, 1988). That is, the measurement modelwas evaluated and established, and then the structural models were esti-mated and evaluated.

RESULTS

Measurement Model

The EFA revealed a four-factor structure and factors were Value/Ser-vice, Assortment/Convenience, Confidentiality, and Cost. Based on theliterature review and the EFA result, the final measurement model wasestablished. The results for the measurement models in Figure 1 are pre-sented in Table 2. For the measurement model of retail attributes, onearbitrarily selected observed indicator of each factor was fixed at 1.0 inorder to give the latent variable a referent, while the others were set free.The overall model was evaluated with the goodness-of-fit index (GFI),the adjusted goodness-of-fit index (AGFI), the comparative fit index(CFI), and the root mean square error of approximation (RMSEA). Theresulting goodness-of-fit index for each measurement model wasaround .95, indicating a good model fit. Although all RMSEA weregreater than .05 and less than .066, the values indicated acceptablemodel fit (because of less than .08). The coefficients for latent con-structs were above .05. The Cronbach’s alphas for the latent constructsranged from .65 to .78, suggesting moderate to high levels of reliability.

Structural Models

Figure 2 illustrates the structural equation models and fit indices forstores, catalogs, and the Internet. The indices of goodness-of-fit indi-cated all three models fit the sample data well.

In the store model, the c2 - value of 272.9 was significant (df = 95, p =0.001), and other fit indices were sufficient to accept the proposedmodel (GFI = 0.954, AGFI = 0.935, CFI = .911, and RMSEA = 0.062).The Confidentiality factor had a negative effect on the purchase inten-tion of clothing products in stores (g = � 0.230, p < .05). Value/Service,


Assortment/Convenience, and Cost did not predict purchase intentionof clothing products.

In the catalog model, overall fit statistics of the proposed model indi-cated that the c2 - value of 237.3 was significant (df = 95, p = 0.001), andthat other fit indices suggested a good model fit (i.e., GFI = 0.964, AGFI =0.949, CFI = .947, and RMSEA = 0.056). The model showed a significantrelationship between the Cost factor and purchasing intention of clothingproducts (g = 0.45, p < .001). Significant relationships did not exist for theother three factors: Value/Service, Assortment/Convenience, and Confi-dentiality.

In the Internet model, the c2 - value of 245.2 was significant (df = 95, p =0.001), and other fit indices were sufficient to accept the proposed model(GFI = 0.967, AGFI = 0.953, CFI = .957, and RMSEA = 0.059). Both As-sortment/Convenience (g = .416, p < .01) and Cost (g = .259, p < .05)were significantly related to purchasing intention through the Internet,whereas Value/Service and Confidentiality were not.


X1

X2

X3

X4

X5

X6

X7

X8

X9

X10

X11

X12

X13

X14

X15

Value/Service

Cost

Assortment/Convenience

Confidentiality

X1: Good custom service X2: Good quality of merchandise X3: Reasonable priceX4: Easy return of items X5: Access-same X6: Access-different itemsX7: Availability of national-designer brands X8: Layout X9: Savings timeX10: Up-to-date and unique items X11: Privacy X12: SecurityX13: Money X14: Time X15: Effort

FIGURE 1. Measurement Model for Store, Catalog, and the Internet

TA

BLE

2.F

acto

rLo

adin

gs,C

ronb

ach’

sa,

and

Fit

Indi

ces

for

Mea

sure

men

tMod

els

Item

Des

crip

tio

nS

tore

Cat

alo

gIn

tern

et

Val

ue/

Ser

vice

Ass

ortm

ent/

Con

veni

ence

Con

fiden

tialit

yC

ost

Val

ue/

Ser

vice

Ass

ort

men

t/C

on

ven

ien

ceC

onfid

enta

ility

Cos

tV

alue

/S

ervi

ceA

ssor

tmen

t/C

onve

nien

ceC

onfid

entia

lity

Co

st

1C

usto

mer

serv

ice

0.84

80.

858

0.82

9

2Q

ualit

yof

prod

uct

0.93

30.

892

0.94

7

3R

easo

nabl

epr

ice

0.50

70.

724

0.78

6

4E

asy

retu

rn0.

694

0.86

80.

884

5A

cces

s-sa

me

prod

uct

0.67

90.

848

0.88

5

6A

cces

s-di

ffere

ntpr

oduc

t0.

700

0.80

40.

905

7N

atio

nal/d

esig

nerb

rand

s0.

573

0.72

80.

735

8La

yout

0.66

70.

725

0.73

2

9S

avin

gtim

e0.

603

0.61

10.

698

10U

p-to

-dat

e/un

ique

item

s0.

658

0.76

20.

839

11P

rivac

y0.

863

0.85

30.

822

12S

ecur

ity0.

829

0.91

90.

944

13M

oney

0.53

00.

542

0.62

4

14T

ime

0.76

50.

833

0.90

7

15E

ffort

0.88

20.

824

0.80

4

Cro

nbac

h’s

a0.

650.

670.

670.

690.

770.

770.

720.

660.

750.

780.

690.

73

Mod

elc2

256.

620

3.9

214.

9

Fit

df84

8484

Inde

xR

MS

EA

0.06

50.

054

0.05

7

GF

I0.

955

0.96

80.

969

AG

FI

0.93

50.

954

0.95

6

CF

I0.

911

0.95

30.

960

Not

e:R

MS

EA

=R

ootM

ean

Squ

are

Err

orof

App

roxi

mat

ion;

GF

I=G

oodn

ess

ofF

itIn

dex;

AG

FI=

Adj

uste

dG

FI;

CF

I=C

ompa

rativ

eF

itIn

dex.

All

fact

orlo

adin

gsar

est

atis

tical

lysi

gnifi

cant

atp

<.0

01.

36

DISCUSSIONS

This study was aimed at identifying retail attributes marketers shouldconsider when they want to attract and retain the multi-channel buyer,in an effort to understand consumer channel choice for clothing prod-


Store

Value/Service


PurchaseIntention

Cost Confidentiality

Catalog

Value/Service

PurchaseIntention

Cost Confidentiality

Internet

Value/Service

PurchaseIntention

Cost

Y1

Y1

Y1

1.0a

0.135 (.148) .125 (.155)

.040 (.114) –0.230 (.108)*

1.0a

.054 (.091) .123 (.113)

.432 (.101)*** –.129 (.099)

1.0a

–.307 (.270) .416 (.190)*

.259 (.094)** –.012 (.159)

Fit indices for store:= 272.9,= 95,

RMSEA = .062,GFI = .954,AGFI = .935, andCFI = .911.

c2

df

Note: Values are from the standardized solution.Values in the parentheses are standard errors.* < .05, ** < .01, *** < .001.starting value = 1.0.p p p

a

Fit indices for catalog:= 237.3,

= 95,RMSEA = .056,GFI = .964,AGFI = .949, andCFI = .947.

c2

f


a

Fit indices for Internet:= 245.2,= 95,

RMSEA = .057,GFI = .967,AGFI = .953, andCFI = .957.

c2

df


a



Confidentiality

FIGURE 2. Structural Models for Store, Catalog, and the Internet

ucts. Retail attributes that consumers perceive important and affect theirpurchase intention varied by channel.

The confidentiality factor had a negative effect on the future pur-chase intention of clothing products in stores. This finding suggests thatconsumers perceive risks in personal security or buying private mer-chandise (e.g., intimate clothing products such as lingerie, plus-sizeclothing products) while shopping in stores. In terms of catalogs, con-sumers who perceive higher level of cost from catalog buying tend topurchase clothing products through catalogs. This seems to contradictthe findings of the previous researchers (e.g., Eastlick & Feinberg,1994; Jasper & Lan, 1992; Kwon et al., 1991; Shim & Bickle, 1994) whoidentified convenience as the principal reason for clothing purchasesthrough catalogs. It may be that multi-channel consumers of clothing,jewelry, and accessories do not mind spending money, time, and energyto find the right product through catalogs, and consider hedonic aspects(e.g., aesthetics, social impact) as more important than minimizing ex-penditure of money, time, and energy.

The findings on the Internet indicate that consumers prefer to shopclothing items via the Internet due to access to a variety of items andconvenience-related attributes. Obviously, consumers prefer access tovariety within the same kind of product classifications in styles, colors,and sizes, access to different products, availability of national or de-signer brands, layout of the Internet, saving time (e.g., finding the rightproduct/product category), and up-to-date and unique items. This resultsupports the previous findings on variety of merchandise (Kim et al.,2003; Kuntz, 1997) and convenience (Kim et al., 2003) as important at-tributes in purchasing clothing. The reason that confidentiality did notinfluence purchase intention to buy clothing products through theInternet may be related to the fact that security systems are rapidly im-proving, dispelling the notion that online shopping is a risky business(Han & Maclaurin, 2002).

At the same time, consumers who perceive high levels of cost tend topurchase clothing products via the Internet channel. This result is surpris-ing considering the well-established acknowledgment that the Internetprovides a shopping tool to meet consumers’ expectation of minimizingtime and energy expenditure, as demonstrated by the results on severalimportant attributes in buying clothing products online: ease of move-ment (Taylor & Cosenza, 2000), ease of navigation and payment options(Kim et al., 2003), and minimal cost and time for return and speedy pro-cess (Watchravesringkan & Shim, 2003). However, it somewhat corre-sponds to Shim et al.’s (2000) report that, for sensory experiential products


(e.g., apparel and accessories), consumers are less likely to be influencedby functional attributes such as fast transaction service and speedy shop-ping than they are for cognitive products (e.g., books, computer soft-ware). As in the case of catalogs, online consumers of clothing, jewelry,and accessories may consider emotional or hedonic aspects (e.g., aesthet-ics, social impact) as more important than functional aspects (e.g., mini-mizing expenditure of money, time, and energy). In buying theseproducts, consumers may be willing to spend money, time, and energy insearching for the right features such as color, size, style, and fit.

MANAGERIAL IMPLICATIONS

This study identified significant effects of multi-channel consumers’perceived retail attributes on purchase intentions of clothing, jewelry, andaccessories for each of the three channels (i.e., brick-and-mortar stores,catalogs, and the Internet). The findings indicate that multi-channel con-sumers perceive important retail attributes differently across the threechannels, which provides salient implications for multi-channel retailers.

For brick-and-mortar store retailing, confidentiality negatively influ-enced consumers’ purchase intentions. Hence, retailers need to addressthis need by ensuring the privacy (e.g., designing a store and creating anenvironment for comfortable shopping intimate or plus-size apparel)and security (e.g., placing security guards) of the consumers in the store.As more consumers are insulating themselves from world problemssuch as crime and violence by staying home as much as possible (Solo-mon & Rabolt, 2004), they may want to be assured of security whenthey do shop in brick-and-mortar stores.

For catalogs and the Internet, cost positively affected purchase inten-tion, suggesting that multi-channel shoppers tend to be active shoppersand are not concerned about shopping cost (i.e., money, time, and en-ergy). Interestingly, they are more likely to buy clothing, jewelry, andaccessories when they perceive a higher level of expenditure in money,time, and energy. Multi-channel shoppers may find products that are notavailable from stores (e.g., the Gap company selling maternity clothingonly through the Internet). Also, catalog and online companies may em-phasize selling exclusive or authentic products that are hard to find inbrick-and-mortar stores.

For the Internet channel, Assortment/Convenience also affected pur-chase intention of clothing products. This finding suggests that providingwidth and depth in products, as well as ease of navigation and convenient


Internet layout (e.g., merchandise display and transaction) would increaseconsumers’ intention to purchase. Due to the lack of interaction with“live” salespeople and the resulting “do-it-yourself” mentality that resultsfrom having to rely on one’s own abilities to locate and purchase mer-chandise, adequate (i.e., quantity) and accurate (i.e., quality) amounts ofinformation are key parts of the service that online retailers must provide(Janda, Trocchia, & Gwinner, 2002).

Given the fact that the multi-channel shopper buys more because ofthe channel alternatives, the multi-channel shopper should be able tocross channels easily for information search, purchases, and post-pur-chases. According to Buechner and Szczesny (2002), more than 30% ofSears’ online purchases are made in the store; about one-fifth of theseshoppers end up making unplanned purchases in stores. In this respect,multi-channel retailers need to use all channels to the best advantage.This multi-channel advantage can only be achieved through continuedfocus on the multi-channel customer. For example, the item purchasedonline can be easily returned in the store; the retail store customer ser-vice issue should be handled online or by telephone.

In conclusion, multi-channel retailers need to formulate a strategythat enhances multi-channel consumer shopping experiences in allchannels of operations in order to increase consumer purchases. Thereis a concern that internal competition among the distribution channelsmay potentially cause unnecessary cannibalization in the same com-pany. One consequence of this concern is that multi-channel retailersignore the fact that some channels might be better than others at differ-ent points in the consumer purchase process. Offline stores, for exam-ple, provide direct experience of the product, as well as establishedlogistics systems. On the other hand, catalog and online retailers can of-fer easier price comparisons, around-the-clock operations, completeproduct information, instant inventory status, and effortless communi-cation, with low cost. Therefore, knowing how to exploit the advantagesof every channel is a basic yet powerful task for multi-channel retailers.

Moreover, accurate customer analysis and development of the corre-sponding strategies seems to be crucial for successful multi-channel retail-ers. As mentioned in the report “The Multi-Channel Consumer” by BostonConsulting Group (2001), 88% of all Internet users are browsers and 42%of all Internet users are online purchasers from their sample. Most compa-nies, however, focus only on the latter, overlooking the significant con-sumer segment that does not purchase online but whose offline purchasesmay be influenced by online information (e.g., helping consumers comeclose to a final choice or decide on a specific product). Given the fact that


the Internet plays its role in multi-channel environment, not only as a pur-chase medium but as a guide leading consumers to other channel,multi-channel retailers should build a contingent strategy based on howconsumers, including segments who are an active purchaser for one chan-nel yet a tentative purchaser for another, select each channel.

LIMITATIONS AND FUTURE RESEARCH

This study may not be generalized to the population as a whole be-cause the demographic characteristics of the sample did not follow nor-mal distribution both in terms of ethnicity (i.e., 93% Caucasians) and age(i.e., 63% ages 30 to 59 years). It is suggested that any future study be ex-panded to include ethnic groups as well as other age groups. Includingother product categories/services also warrants comparison studies. Al-though comparing male and female consumers was beyond the scope ofthis study, it might provide rich information to multi-channel retailers inplanning their marketing mix (e.g., product, promotion) for each targetedgender market. Further, the interaction between different shopping bene-fit and cost parameters could be studied to facilitate a better understand-ing of how each parameter eventually affects the purchase intentions.

The findings indicate that the confidentiality factor did not influencepurchase intention of online shopping for clothing products. Althoughthis is somewhat contrary to previous findings (Bhatnagar & Ghose,2004; Miyazaki & Fernandez, 2001) that reported security is a major con-cern for online shopping, the confidentiality factor in this study was com-posed of privacy and security. In future research, these constructs may beseparated to see the impact of each construct on purchase intention.

REFERENCES

Abend, J. (2001). Tapping into a virtual world. Bobbin, 42(6), 38-47.Anderson, J. C., & Gerbing, D. W. (1988, May). Structural equation modeling in prac-

tice: A review and recommended two-step approach. Psychological Bulletin, 103,411-23.

Baker, M. (1999). Multi-channel retailing. ICSC Research Quarterly, 6(3), 13-17.Bhatnagar, A., & Ghose, S. (2004). A latent class segmentation analysis of e-shoppers.

Journal of Business Research, 57(7), 758-767.Buechner, M. M., & Szczesny, J. R. (2002). Recharging Sears. Time Canada, 159(21),

38, 2p, 1c.Chevron, J. (1999, October). Top of mind: Brick-and-mortar retailers not mass, but

class. Brandweek, 25, 30-32.


Downs, A. (1961). A theory of consumer efficiency. Journal of Retailing, 37, 6-12,50-51.

Eastlick, M., & Feinberg, R. (1994). Gender differences in mail-catalog patronage mo-tives. Journal of Direct Marketing, 8(2), 37-44.

Ernst & Young. (2001). The multi-channel imperative. Stores, 83(2), 17-21.Global Online Retailing Report. (2000). Consumer Trends in Online Shopping, pp. 5-9.Gordon, H. (1994). Retailers are ready for relationship marketing. Direct Marketing,

56(9), 38-44.Han, P., & Maclaurin, A. (2002). Do consumers really care about online privacy? Mar-

keting Management, 11(1), 35-38.Haydock, M. (2000, August 21). Multichannel strategies win confidence. DM News, p. 32.Hill, S. (2000). Liz Claiborne: Liz Claiborne is one of the ‘in-line’ performers. Apparel

Industry Magazine, 61(6), 44.Hoover, B. (2001). Multi-channel integration: The future success of e-tailing. Retrieved

March 12, 2001, from www.proxicom.com.Hoyer, W. D., & Alperts, M. I. (1983). Additional theory and data contrasting measures

of attribute importance. In Patrick Murphy (Ed.), American Marketing AssociationEducator’s Conference Proceedings (pp. 78-82).

Integrating Multiple Channels. (2001). Chain Store Age, pp. 24A-25A.Janda, S., Trocchia, P. J., & Gwinner, K. P. (2002). Consumer perceptions of Internet

retail service quality. International Journal of Service Industry Management, 13(5),412-433.

Jarvenpaa, S. L., & Todd, P. A. (1997). Consumer reactions to electronic shopping onthe World Wide Web. International Journal of Electronic Commerce, 1(2), 59-88.

Jasper, C., & Lan, P. (1992). Apparel catalog patronage: Demographic, lifestyle andmotivational factors. Psychology & Marketing, 9(4), 275-296.

Jöreskog, K., & Sörbom, D. (1993). LISREL8: Structural equation modeling withSIMPLIS command language. Chicago: Scientific Software International, Inc.

Kim, Y-K. (2002). Consumer value: An application to mall and Internet shopping. In-ternational Journal of Retail and Distribution Management, 30(12), 595-602.

Kim, Y., & Kang, J. (1997). Consumer perception of shopping costs and its relation-ship with retail trends. Journal of Shopping Center Research, 4(2), 27-62.

Kim, Y-K., Kim, E., & Kumar, S. (2003). Testing the behavioral intentions model ofonline shopping for clothing. Clothing and Textiles Research Journal, 21(1), 32-40.

Kunz, M. (1997), On-line customer: Identifying store, product and consumer attributeswhich influence shopping on the Internet. Unpublished dissertation, University ofTennessee, Knoxville.

Kwon, Y., Paek, S., & Arzeni, M. (1991). Catalog vs. non-catalog shoppers of apparel:Perceived risks, shopping orientations, demographics, and motivations. Clothing andTextiles Research Journal, 10(1), 13-19.

Lands’ End Improves Online Profitability via My Virtual Model Technology. (2001).Direct Marketing, 64(2), 11.

Lindquist (1974-1975, Winter). Meaning of image. Journal of Retailing, 50, 29-38.Miyazaki, A. D., Fernandez, A. (2001). Consumer perceptions of privacy and security

risks for online shopping. Journal of Consumer Affairs, 35(1), 27-44.


The Multichannel Consumer: The Need to Integrate Online and Offline Channels inEurope (2001, July). The Boston Consulting Group.

The Multi-Channel Retail Report (2001). www.shop.org/research/TMCRR.html.Paulins, V. A., & Loren, V. G. (2003). The effect of consumer perceptions of store at-

tributes on apparel store preference. Journal of Fashion Marketing & Management,7(4), 371-385.

Reda, S. (2001). VeriFone and Russell Renolds Associates top 100 Internet retailers.Stores, http://www.store.org.

Retail Apparel Sales Statistics & Trends (1999-2000). http://retailindustry.about.com/library/weekly/aa022200a.htm.

Retail Apparel Sales Statistics & Trends (2000). http://retailindustry.about.com/library/weekly/01/aa010319a.htm.

Schaeffer, P. N. (2000). Listen to the consumer. Global Online Retailing Report,10-11.

Schoenbachler, D. D., & Gordon, G. L. (2002). Multi-channel shopping: Understand-ing what drives channel choice. Journal of Consumer Marketing, 19(1), 42-53.

Shim, S., & Bickle, M. C. (1994). Benefit segments of the female apparel market:Psychographics, shopping orientations, and demographics. Clothing and TextilesResearch Journal, 12(2), 1-12.

Shim, S., & Drake, M. (1990). Consumer intention to purchase apparel by mail order:Beliefs, attitude, and decision process variables. Clothing and Textiles ResearchJournal, 9(1), 18-26.

Shim, S., Eastlick, M. A., & Lotz, S. (2000). Assessing the impact of Internet shoppingon store shopping among mall shoppers and Internet users. Journal of ShoppingCenter Research, 7(2), 7-44.

Shim, S., & Kotsiopulos, A. (1992). Patronage behavior of apparel shopping: Part II.Shopping orientations, store attributes, information sources, and personal charac-teristics. Clothing and Textiles Research Journal, 10(2), 58-64.

Shop.org Report Highlights Multi-Channel “Super Shoppers.” (2001, November).Stores, 16.

Solomon, M. R., & Rabolt, N. J. (2004). Consumer behavior: In fashion. Upper SaddleRiver, NJ: Prentice Hall.

Taylor, S. L., & Cosenza, R. M. (2000). The impact of E-commerce on the merchandis-ing of women’s clothing in traditional shopping centers/mall. Journal of ShoppingCenter Research, 7(2), 46-66.

Then, N. K., & Delong, M. R. (1999). Apparel shopping on the Web. Journal of Familyand Consumer Sciences, 91(3), 65-68.

Tiernan, M. (2000). How we catalogers can win on the Web. Catalog Age, 17(6), 95.Verton, D. (2001). Churn, Computerworld, 35(6), 50.Watchravesringkan, K., & Shim, S. (2003). Information search and shopping intensions

through Internet for apparel products. Clothing and Textiles Research Journal, 21(1), 1-7.Wilde, S. J., Kelly, S. J., & Scott, D. (2004). An exploratory investigation into e-tail im-

age attributes important to repeat, internet savvy customers. Journal of Retailing &Consumer Services, 11(3), 131-139.

Zeithaml, V. A. (1988). Consumer perceptions of price, quality, and value: A means-end model and synthesis of evidence. Journal of Marketing, 52(1), 2-22.


http://www.store.org

http://retailindustry.about.com/

http://retailindustry.about.com/

Ps4

Documents

multichannel retailing

multichannel consumer

formultichannel retailers

internet retailing

pingsingle channel approaches

mortar stores orcatalog

email address

direct marketing channels