physical vs. virtual information search and purchase … · 2015-10-02 · opportunities for...
TRANSCRIPT
Managing Sustainability?Proceedings of the 12th Management International ConferencePortorož, Slovenia, 23–26 November 20112011
PHYSICAL VS. VIRTUAL INFORMATION SEARCH AND PURCHASE
IN BUYING BEHAVIOR OF POLISH YOUNG CONSUMERS Dr. Radosław Mącik, Maria Curie-Skłodowska University in Lublin, Poland
Dr. Dorota Mącik, School of Finance and Management, Warsaw, Poland
ABSTRACT
This paper investigates selected internal and external to the consumer factors influencing
young consumers’ choices of physical or virtual channel for information search and
purchase. Situations of channel lock-in and channel change between mentioned phases of
buying process are described and their antecedents investigated. Perceived channel
characteristics for information search and purchase, consumer decision-making style profile
information technology adoption level are considered to influence consumer choices in this
case. Structural equation modeling has been used to analyze gathered data – 5 models has
been assessed.
Keywords: internet shopping, physical retail, information search, young consumers, Poland
INTRODUCTION
The increasing internet penetration and private usage among consumers created new
opportunities for business-to-consumers (B2C) electronic markets, mainly: online shops,
internet auction platforms, and virtual services, which gave the sellers new possibilities to
reach out to consumers. Many of consumers easily adopted buying in virtual channel,
perceiving it as more comfortable and cheaper than conventional retail in physical
environment. Buying over the Internet has in many cases also substantial disadvantages –
among others the lack of physical contact with product considered to buy, as well as several
financial and personal risks are perceived as main of them.
The Internet, being the largest information source and facilitator of social media, gives the
consumer opportunities to collect, compare and spread shopping information regarding
products and sellers. For consumer typically is much easier to find, compare and choose
product with certain characteristics and features online in comparison to offline information
search using physical channel sources as well as personal contacts. Information about
739
products and sellers gathered online is perceived as quicker, more detailed and sometimes
more reliable, so it is widely used in purchase decision processes.
Generally, better opinions about the Internet as an information source than as a shopping
channel can lead to situation of shopping channel change during consumer decision making
process – for instance seeking information in online channel, and buying in physical retail
(avoiding i.e. waiting for delivery and its costs). Converse situation – offline information
search (i.e. to try on clothes or “feel” keyboard on laptop) – followed with online shopping is
also possible and quite popular.
The main goal of paper is to investigate factors influencing young consumers’ choices of
physical or virtual channel for information search and buying in case of clothing and
consumer electronics. As physical channel, we are treating different formats of physical retail
outlets. Virtual channel includes websites commonly used to get information about products,
including reviews, personal recommendation, price comparison tools, as well as Internet
stores and auctions used by consumers to gather information and to buy – in all cases
accessible via computers and/or mobile devices (through web browser or specialized
application).
Investigation of channel lock-in and channel change between information search and purchase
phases of buying process also has been made, which includes comparison of antecedents of
researching online to buy offline and of searching information offline to buy online.
INTERNET SHOPPING AS A TECHNOLOGY ADOPTION EXAMPLE
The information technology adoption was extensively studied since over 20 years, leading to
different models explaining determinants of IT adoption, both on user and organization levels,
starting from the TAM (Technology Acceptance Model) by Davis (1989), up to UTAUT
(Unified Theory of Acceptance and Use of Technology) by Venkatesh et al. (2003), and
TAM3 by Venkatesh and Bala (2008).
The TAM and its developments became widely used as simple tools to explain and predict
computer-usage behavior as the example of technology adoption, firstly in the work usage
context (Klopping and McKinney, 2004; Venkatesh et al., 2003; Venkatesh and Davis, 2000).
Over the time the TAM model has been used successfully (among others) to explain WWW
740
usage (Lederer et al., 2000; Magal and Mirchandani, 2001), and to predict online shopping
adoption (Lee, Park and Ahn, 2001; Childers et al., 2001; Chen, Gillenson, and Sherrell,
2002; Klopping and McKinney, 2004).
The TAM is theoretically grounded in Fishbein & Ajzen’s (1975) theory of reasoned action
(TRA) which states that beliefs influence attitudes, attitudes lead to intentions, and finally
intentions are converted to behaviors (Malhotra & Galetta, 1999). The main goal of TAM is to
explain the determinants of computer acceptance (Davis, Bagozzi, and Warshaw, 1989). Two
major determinants of technology adoption in TAM: perceived usefulness (PU) and perceived
ease of use (PEoU) are defined by Davis (1989) respectively as: “the degree to which a person
believes that using a particular system would enhance his or her job performance,” (PU) and
“the degree to which a person believes that using a particular system would be free of effort”
(PEoU) (p. 320). In attitude formation toward information technology use perceived
usefulness has been identified as significant in the literature (Davis, 1989; Dishaw & Strong,
1999; Gefen and Keil, 1998; Moon and Kim, 2001; Taylor and Todd, 1995; Venkatesh and
Davis, 2000), but the evidence for perceived ease of use has been inconsistent (Lin and Lu,
2000; Teo, Lim and Lai, 1999; Lee et al., 2001). The TAM and its extensions are also often
criticized for being to obvious and simple (Bagozzi 2007).
For usage in consumer e-commerce context Klopping and McKinney (2004) modified and
simplified the TAM by: dropping PEoU to PU path, dropping attitude toward technology in
order to investigate the direct relationship between perceived usefulness and perceived ease of
use on intention to use adding direct effect of PU on actual use (Klopping and McKinney,
2004, p. 37). The task of online shopping in study of Klopping and McKinney (2004) has
been described as combination of both the purchase and the product information search
activities. That point of view has been used also by authors of this paper in previous studies
(Mącik and Mącik 2008, 2009), extended by investigating consumer decision-making styles
influence on online shopping activities significantly improving simple TAM approach.
CONSUMER DECISION-MAKING STYLE
Consumer decision making-style concept, introduced by Sproles & Kendall (1986), used in
several more contemporary studies (Walsh et al. 2002, Tai 2005,) proved to be useful to
explain outcomes of particular shopping activities and attitudes toward shopping, including
usage of online channel (Mącik & Mącik, 2008, 2009).
741
A consumer decision-making style concept, introduced by Sproles & Kendall (1986), defined
as “a mental orientation characterizing a consumer’s approach to making choices” (p. 268)
Consumer decision-making styles can be perceived as “basic buying-decision making
attitudes that consumers adhere to, even when they are applied to different goods, services or
purchasing decisions” (Walsh et al., 2001, p. 121). They are relatively stable constructs,
connected to consumer personality (Sproles and Kendall, 1986; Lysonsky, Durvasula and
Zotos, 1995), and particular shopping activities and attitudes toward shopping are direct
outcomes of consumer’s decision-making style (Tai, 2005).
Original Sproles & Kendall (1986) work organizes consumer personality in eight following
dimensions (decision-making styles):
1) perfectionism or high-quality consciousness (PERF),
2) brand consciousness or “price equals quality” (BC),
3) novelty-fashion consciousness (NFC),
4) recreational or hedonistic consciousness (RSC),
5) price-value consciousness or “value for money” (PVC),
6) impulsiveness or carelessness (IMP),
7) confusion from overchoice (CO),
8) habitual, brand loyal orientation toward consumption (HBL).
Authors have updated measurement instrument for decision making-style profile assessment,
by adding two new dimensions: tendency to compulsive buying (COMP) (Mącik and Mącik
2008, 2009) and attitude toward “green” consumption (ECO) – latest addition.
It is important is to note, that those styles are not independent – particular person possesses an
individual combination of them, creating personal profile of all styles manifesting itself on
different levels, from those some are more intense or prominent (Sproles & Kendall 1986).
CHANNEL LOCK-IN AND CHANNEL CHANGE
Channel lock-in between information search and product purchase phases means that a
consumer searching for product information through specific channel (e.g., the Internet) buys
a product through the same channel (Young-Hyuck and Hyung-Jin Park 2008). Channel
change situation – so called cross-channel shopping – means that a consumer, who searches
for product information through channel A, buys a product at channel B (Young-Hyuck and
742
Hyung-Jin Park 2008). For this study, there are two situations of cross-channel shopping:
searching information in physical retail to buy over the Internet, and searching information
over the Internet (using computer and/or mobile device) to purchase offline. Channel change
is an effect of realizing by the consumer that channel B has important advantages in terms of
future purchase outcomes, after being informed and educated through channel A. Such
situation may be perceived as complementarities of both channels (Teering & Huizingh
2005).
Perceived channel characteristics are influencing the choice of channel for both phases:
information search, and buying. Mokhtarian and Tang (2009, 2011) are discussing in detail
description of physical and virtual retail channel features, describing eight dimensions of it,
particularly: Convenience, Enjoyment, Efficiency/inertia, Cost-saving, Store brand
independency, Product risk, Financial/identity risk, and Post-purchase satisfaction.
HYPOTHESES
This study is focused on exploring shopping channel perception and previous experiences on
channel choice for information search and buying, so main hypotheses are formulated to
check if physical or virtual channel choice comes from attitude toward particular channel.
There is expected that channel advantages will influence choosing it, as well as channel-
connected risks and its disadvantages will lead to preference of another one (Mokhtarian and
Tang, 2011). According to Klopping & McKinney (2004) it is expected that some TAM
constructs will be connected with channel choice. As long the consumer’s decision-making
style in Sproles & Kendall meaning is important his/her characteristic, and particular
consumer shopping activities are direct outcomes of his/her own consumer decision-making
styles profile, there should exist at least some styles related to channel choice by consumer.
Therefore, for this study the set of seven main hypotheses has been formulated on the base of
literature review and authors’ preliminary qualitative and quantitative studies on small
samples of undergraduate students. The hypotheses are as follows:
H1: Virtual channel perception and experience influences channel choice for information
search and purchasing.
H1a: Perceived virtual channel advantages are positively related with virtual channel
usage for information search.
743
H1b: Perceived virtual channel advantages are positively related with virtual channel
usage for purchasing.
H1c: Perceived risks of virtual channel usage are negatively related with virtual channel
usage for purchasing.
H1d: Perceived risks of virtual channel usage are positively related with physical
channel usage for purchasing.
H1e: Experience with virtual channel shopping is positively related with virtual channel
usage for purchasing.
H2: Physical channel perception and experience influences channel choice for information
search and purchasing.
H2a: Perceived physical channel advantages are positively related with physical channel
usage for information search.
H2b: Perceived physical channel advantages are positively related with physical channel
usage for purchasing.
H2c: Perceived risks of physical channel usage are negatively related with physical
channel usage for purchasing.
H2d: Perceived risks of physical channel usage are positively related with virtual
channel usage for purchasing.
H2e: Experience with physical channel shopping is positively related with physical
channel usage for purchasing.
H3: Usage of particular channel for information search is positively related with its usage
for purchasing (channel lock-in effect).
H3a: Usage of virtual channel for information search is positively related with virtual
channel usage for purchasing.
H3b: Usage of physical channel for information search is positively related with
physical channel usage for purchasing.
H4: Channel change from virtual to physical between information search and purchasing is
influenced by preference to purchase in physical channel.
H4a: Perceived physical channel advantages are positively related with channel change
from virtual to physical between information search and purchasing.
H4b: Perceived physical channel risks are negatively related with channel change from
virtual to physical between information search and purchasing.
H4c: Perceived virtual channel advantages are negatively related with channel change
from virtual to physical between information search and purchasing.
744
H4d: Perceived virtual channel risks are positively related with channel change from
virtual to physical between information search and purchasing.
H5: Channel change from physical to virtual between information search and purchasing is
influenced by preference to purchase in virtual channel.
H5a: Perceived virtual channel advantages are positively related with channel change
from physical to virtual between information search and purchasing.
H5b: Perceived virtual channel risks are negatively related with channel change from
physical to virtual between information search and purchasing.
H5c: Perceived physical channel advantages are negatively related with channel change
from physical to virtual between information search and purchasing.
H5d: Perceived physical channel risks are positively related with channel change from
physical to virtual between information search and purchasing.
H6: Consumer’s decision-making styles profile predict, influencing directly and indirectly
the usage of particular channel for information search and purchasing.
H7: Consumer’s decision-making styles profile predict, influencing directly and indirectly
the channel change between information search and purchasing.
METHODS
Subjects and procedures
The participants of presented study were 201 Polish consumers, aged between 18 and 35 years
old (with largest group being full-time and part-time students – about 81%). Data were
collected using Computer Assisted Web Interview (CAWI) questionnaire (with response
paths between 26 and 35 screens long, lasting about 20-30 minutes) was utilized. Invitations
has been sent by e-mail to 324 participants of previous author research from 2010, who
agreed to participate in next research - response rate was about 62%. For modeling has been
taken responses of 92 participants, who responded completely. From initial 201 participants
incomplete responses were excluded, as well as responses from participants who consequently
responded regardless the proper scaling direction. Effective response rate was about 28,4%.
Effective sample size is rather small, which is considered as this study important limitation.
The choice of mentioned group as study subjects takes into account more frequent buying of
consumer electronics and clothing than average ones, as well as greater involvement with
such purchases and better information technology adoption, resulting with more complicated
buying processes, more often including shopping channel changes. Consumer electronics and
745
clothing choice is influenced by their wide distribution across physical and virtual retail, and
by higher popularity of buying them over the internet in Poland. It is worth to mention that
about 42% of internet users in Poland declaring buying over the internet bought some clothing
this way in 2009, and consumer electronics created at the same time about 39% of virtual
retail turnover (Internet Standard 2010).
Scales development
To measure perceived physical and virtual channel characteristics scales adapted from
Mokhtarian and Tang (2009) used. Scales items have been translated into Polish and adapted
using experimental statements, considering possible cultural and language differences. Used
form has been obtained using reliability analyses made on data from preliminary sample of 48
undergraduates. From original eight dimensions of it, particularly: Convenience, Enjoyment,
Efficiency/inertia, Cost-saving, Store brand independency, Product risk, Financial/identity
risk, and Post-purchase satisfaction, seven constructs has been used – Store brand
independency items has been dropped as not clearly understandable by participants.
During the analysis – regarding low effective sample size – items from those 7 constructs
were grouped into two larger groups: channel advantage (including: Convenience, Enjoyment,
Efficiency, Cost-saving, and Post-purchase satisfaction), and channel risks (including:
Product risk and Financial/identity risk). Channel experience has been measured as declared
shopping frequency in 10 physical retail formats, and 3 virtual ones (including informal sales
on Internet communities and social media).
Scales measuring usage of particular channel for information search and purchasing are
constructed by authors, for virtual channel loosely resembling items from Klopping &
McKinney (2004) version of TAM for Internet shopping, particularly from IU and AU
constructs. Authors own alteration of typical TAM items regarding personal Web usage was
also used in the questionnaire. Reliability analyses were performed on preliminary sample of
48 undergraduates.
The PCS scale by Sproles and Kendall (1986) was reconstructed earlier by authors of this
paper using two samples of 212 and 324 undergraduate students rescpectively. Our polish
adaptation of PCS consists of 30 items selected using factor and reliability analyses in 10
dimensions. Eight dimensions are styles present in original Sproles and Kendall (1986) works.
The ninth dimension – compulsiveness (COMP) – has been added on the base of previous
746
authors research on unplanned buying and its determinants, using Unplanned Buying Scale
(UBS) by Mącik and Mącik (2005), which is a reception of hedonic, impulsive and
compulsive buying tendencies described by Hausman (2000), Rook and Fisher (1995), and
Faber and O’Guinn (1992). At the second stage the tenth dimension – attitude toward “green”
consumption (ECO) has been added.
All mentioned constructs were measured using Likert-type scale of 1-5 with end points
worded “strongly agree” and “strongly disagree” with exceptions in channel experience items
and channel change frequency, using for both frequency descriptions.
Internal consistency and construct validity
Internal consistency of individual scales for measuring particular construct has been assessed
by using Cronbach’s Alpha coefficient (Cronbach, 1951) – see Table 1. Internal consistency
of all constructs meets typical requirements to be above 0.6 for short scales, being in many
cases above usual limit of 0.7 suggesting enough internal consistency, so those scales can be
used for analysis. Intercorrelations (not reported in text) between constructs are rather low
(not exceeding 0.5 in absolute terms), which indicates measuring different constructs, and
suggests good discriminant validity.
For 15 used scales dimensions Alphas are over acceptable minimum for short scales of 0.6,
which signalizes the need to improve their wording.
Next step in assessing construct validity was exploratory factor analysis using principal
component extraction method with varimax rotation. This procedure has been performed to
confirm that all scales are measuring distinct constructs. For all scales constructs desired
factor structures were confirmed, with extracting at least 63% of the total variance, with no
cross loadings >0.50, which indicates good discriminant validity.
Table 1: Internal consistency of used scales
Construct Description # of items Cronbach’s Alpha
VC_ADV perceived virtual channel advantages 17 0.77
VC_RISK perceived virtual channel risk 8 0.74
747
VC_EXP virtual channel experiences 3 0.62
PC_ADV perceived physical channel advantages 17 0.72
PC_RISK perceived physical channel risk 7 0.62
PC_EXP physical channel experiences 10 0.63
VC_IU virtual channel information search usage 7 0.67
VC_PURCH virtual channel purchasing usage 5 0.64
PC_IU virtual channel information search usage 2 0.61
PC_PURCH virtual channel purchasing usage 5 0.65
V2P virtual to physical channel change frequency 1 n/a
P2V physical to virtual channel change frequency 1 n/a
TAM_PU perceived usefulness of private Internet usage 3 0.65
TAM_PEoU perceived ease of use of private Internet usage 3 0.75
TAM_IU intention toward of private Internet usage 3 0.68
TAM_AU actual of private Internet usage 3 0.85
PERF perfectionism or high-quality consciousness 3 0.62
BC brand consciousness or “price equals quality” 3 0.64
NFC novelty-fashion consciousness 3 0.79
RSC recreational or hedonistic consciousness 3 0.78
PVC price-value consciousness 3 0.63
IMP impulsiveness or carelessness 3 0.67
CO confusion from overchoice 3 0.66
HBL habitual, brand loyal orientation 3 0.75
COMP compulsive buying tendency 3 0.61
ECO „green” orientation toward consumption 3 0.85
Data analysis
Presented analyses have been made using SPSS 17.0PL (for preliminary purposes and
descriptive statistics) and AMOS 17.0 structural equation modeling software (to identify and
test causal relationships). So called “path analysis” approach to structural equations modeling
748
has been utilized. “Path analysis is a method of measuring the influence of explanatory
variables along each separate path in a system and finding the degree to which variation of a
given effect is determined by each particular cause” (Teo, Lim & Lai, 1999, p. 30). This
approach allowed to calculate measures of fit for models reported in the paper, and made
comparisons across those models.
RESULTS
Channel lock-in models
First estimated model describes hypothesized relationships between information search and
purchasing at particular channel (model 1a – figure 1). Paths predicted by hypotheses H1
trough H3 are included. As many hypothesized relationships in this model are not significant
– those paths has been excluded to build model 1b. Both models are acceptable in terms of fit
measures, while model 1b performs better (table 2).
a) Model 1a b) Model 1b
R2= ,31
VC_PURCH
R2= ,05
PC_PURCH
R2= ,14
VC_IU
R2= ,22
PC_IU
VC_ADV VC_RISK
PC_ADV PC_RISK
,38*** -,18 (n.s.)
,34***
,17 (n.s.)
,13 (n.s.)
E1 E3
E4E2
-,49***
,19*
,25**
,32***
VC_EXP
PC_EXP
,29**
,13 (n.s.)
-,39***
,41***
,09 (n.s.)
-,03 (n.s.)
,06 (n.s.)
-,14 (n.s.)
-,23*
R2= ,32
VC_PURCH
R2= ,02
PC_PURCH
R2= ,19
VC_IU
R2= ,21
PC_IU
VC_ADV VC_RISK
PC_ADV PC_RISK
,38*** -,21*
,34***
,19*
,14 (n.s.)
E1 E3
E4E2
-,49***
,32***
VC_EXP
-,39***
,41***
-,23 *
,19*
,25**
,20*
,30**
Figure 1: Initial model (model 1a) and improved model 1b structures
Notes: *** p<0,001 ** p<0,01 * p < 0,05 (n.s.) – not significant
749
In model 1a supported are only hypotheses H1a, H1b, H2a, H2d and H2e, also significant is
not hypothesized path between information seeking in virtual channel and physical one
(signalizing channel synergy). For better in terms of fit model 1b, hypotheses H1a, H1c, H1e,
as well as H2a, H2d and H3a are supported. Choosing virtual channel for information search
depends on perceived advantages of virtual channel and perceived risk of physical channel.
Choosing virtual channel is connected positively with: that channel previous experiences,
usage of that channel for information search, perceived risk of physical channel as well as
negatively with perceived risk of virtual channel (those 4 variables are explaining about 32%
of variance of virtual channel usage. Physical channel usage for information search depends
only on that channel perceived advantages, while using physical channel for purchasing is not
significantly connected with that channel usage for information search. So channel lock-in
effect is present on virtual channel, but not in the physical one.
At next step to model 1b constructs measuring consumer decision-making styles has been
added creating model 2 (figure 2) to check hypothesis H5. In terms of fit model 2 is
comparable to model 1b (table 2).
750
R2= ,38
VC_PURCH
R2=,10
PC_PURCH
R2= ,21
VC_IU
R2=,26
PC_IU
R2=,30
VC_ADV R2= ,17
VC_RISK
R2= ,17
PC_ADV
R2= ,15
PC_RISK
,17*
E1 E3
E4E2
RSC
PVC
CO
HBL
COMP-,20*
,39***
-,17*
,24**
,21
,21*
E8
E6 E5
ECO
,25
,27**
-,22*
,22*
,38***
-,39***
,32**
-,32**
,27**
,19* ,31***
,33***
VC_EXP
E9
,25*
-,42***
,29**
-,25*
Notes: *** p<0,001 ** p<0,01 * p < 0,05
Figure2: Model 2 – consumer decision-making styles extension to model 1b
Connections between 6 styles and constructs from model 1 has been found, which supports
hypothesis H6. Buying in virtual channel is positively connected with compulsiveness and
negatively - with confusion form over choice, also indirect influence of recreational shopping
orientation is present (trough virtual channel information usage), as well as ecological
orientation toward consumption (trough perceived risk of physical channel). Information
seeking in physical channel depends on price-value orientation (positively) and compulsive
buying tendency (negatively). Buying in physical channel is positively connected with
habitual brand-loyal orientation.
751
Channel change models
First channel change model describes relationships causing channel change as has been
hypothesized in H4 and H5 (model 3 – figure 3). Paths predicted by hypotheses H4 and H5
are included. Some hypothesized relationships in this model are not significant – those paths
has been excluded and are not shown on figure 3. Model 3 in form shown on figure 3 fits data
very well (table 2).
R2= ,31
VC_PURCH
PC_PURCH
VC_RISK
PC_ADV PC_RISK
E1
-,24*
,29***
R2= ,25
V2P
R2=,08
P2V
E2
E3
,22*
,36***
TAM_PEOU
,24**-,24*
-,25*
VC_EXP
,33***
,26*
,19*
-,34**
,24*
-,23*
-,29**
Notes: *** p<0,001 ** p<0,01 * p < 0,05 (n.s.) – not significant
Figure 3: Model 3 – explanation of channel change
Channel change from virtual to physical relies on physical channel perceived advantages and
this channel connected risk, and also on frequent physical channel choice for purchasing (in
all cases positively), as well as virtual channel risk (negatively) and perceived ease of use of
the virtual channel (positively). Those connections are rather surprising, only H4a is partially
supported in this case, and different signs of relationships described in H4b and H4d has been
found. Channel change from physical to virtual is significantly negatively connected with
752
perceived risk of using virtual channel, which supports hypothesis H5b. Other hypotheses
under H5 group were not supported.
As adding consumer decision-making style to channel choice was promising, to model 3 paths
suggested in model 2 were added creating model 4 to check hypothesis H7.
R2=,35
VC_PURCH
R2= ,10
PC_PURCH
R2= ,05
VC_IU
R2= ,15
PC_IU
VC_RISK
R2=,15 PC_RISK
,17*
E1 E3
E4E2
RSC
PVC
CO
HBL
COMP
-,19*
,39***
-,17*
,25**
,26**
,22*
E5
ECO
-,23*
,22*
,32**
-,32**
,28**
,24*
R2=,31
V2P
R2= ,09
P2V
E9
E10
,25**
-,22*
,36***
-,29**
TAM_PEOU
,25**
,24**
-,27**
VC_EXP
-,36***
,25*
,29**
Notes: *** p<0,001 ** p<0,01 * p < 0,05 (n.s.) – not significant
Figure 4: Model 4 - consumer decision-making styles extension to model 3
Direct positive influence of price-value consciousness on virtual to physical has been found,
suggesting that virtual channel has disadvantages for price-value oriented consumers. Indirect
influence of habitual brand-loyal orientation (through preference for purchasing at physical
channel), and mentioned confusion from over choice as well as compulsive buying tendency
(through virtual channel purchasing preference) is present in the model 4. Overall fit of model
4 is comparable do model 3, and good (table 2).
753
DISCUSSION
This study, finds a support for the use of channel perceived characteristics and consumer
decision-making styles in explaining consumer channel choice for information seeking and
purchasing, as well as for channel change. Channel characteristics are influencing channel
choice and eventual change more substantially than other considered factors. Consumer
decision-making styles profile are helping to improve modeling of those factors.
Our approach made possible to characterize constant factors influencing channel choice and
change. Unfortunately predicting channel choice is much easier for virtual channel, also
channel change from virtual to physical is easier to describe. So this study has important
shortcoming – estimated models do not perform very well for explaining physical channel
choice and change from physical to virtual channel.
Table 2 contains comparison of fit measures for all estimated models. Models 1b and 2
explained quite well virtual channel purchasing and physical channel information search. In
models 3 and 4 attempts to explain virtual to physical channel is rather successful. Model 4
linking models 2 and 3 shows better whole system of relationships. Consumer decision-
making style paths, added in models 2 and 4, allow to explain additional 5-7% of variance of
virtual channel purchasing, physical channel information search, and change from virtual
channel to physical one.
Table 2: Estimated models fit comparisons
Measures of fit Models
1a 1b 2 3 4
Variance Explained: Virtual channel
information search (VC_IU) 0.143 0.187 0.212 n/a 0.048
Variance Explained: Virtual channel
purchasing (VC_PURCH) 0.313 0.317 0.381 0.306 0.350
Variance Explained: Physical channel
information search (PC_IU) 0.220 0.213 0.258 n/a 0.152
Variance Explained: Physical channel
purchasing (PC_PURCH) 0.053 0.020 0.101 n/a 0.101
754
Variance Explained: Virtual to physical
channel change (V2P) n/a n/a n/a 0.251 0.313
Variance Explained: Physical to virtual
channel change (P2V) n/a n/a n/a 0.084 0.085
χ2/df (below 2 or 3 better)a 1.191 0.989 1.104 1.212 1.166
P (not significant better)b 0.226 0.475 0.470 0.224 0.127
GFI (above .9 is good fit) 0.938 0.952 0.895 0.940 0.867
AGFI (above .8 is good fit) 0.873 0.901 0.846 0.877 0.813
NFI (above .9 good fit) 0.800 0.856 0.708 0.783 0.592
RMSEA (.05 or less better) 0.046 0.000 0.006 0.048 0.043
PCLOSE (not significant better)c 0.510 0.727 0.888 0.478 0.624
Notes: a “rules of the thumb” suggested by Carmines and McIver (1981, p. 80) or Byrne (1989, p. 55). b for lager samples it is often unreasonable to have significant p value (Jöreskog, 1969, p. 200). c “p value” for testing the null hypothesis that the population RMSEA is no greater than .05 indicating close fit
(Browne and Cudeck, 1993).
CONCLUSION
The goal of this study was to explore whether perceived channel characteristics and consumer
decision-making styles can explain consumer channel choice for information search and
purchasing, as well as for channel change. This was quite successful – estimated models were
performing very well in terms of fit.
Unfortunately many findings from the models are rather surprising, not supporting some of
hypotheses. Other important limitation of this study is relatively small and not diversified
sample. To deal with this limitation the authors are suggesting replication of results on larger
and more diversified sample.
The effects of this research (despite mentioned limitations) have several practical implications
for physical and virtual retail management, including possibilities of better market
segmentation, marketing planning and positioning.
755
REFERENCES
Bagozzi R. P. 2007. The Legacy of the Technology Acceptance Model and a Proposal for a
Paradigm Shift, Journal of the Association for Information Systems, 8(4): 244-254.
Browne, M.W., and R. Cudeck. 1993. Alternative ways of assessing model fit, In Bollen.
K.A. and Long, J.S. [Eds.] Testing structural equation models, Newbury Park, CA:
Sage: 136-162.
Byrne, B.M. 1989. A primer of LISREL: Basic applications and programming for
confirmatory factor analytic models, New York: Springer-Verlag.
Carmines, E.G., and J.P. McIver. 1981. Analyzing models with unobserved variables. In
Bohrnstedt, G.W., and Borgatta, E.F. [Eds.] Social Measurement: Current Issues,
Beverly Hills: Sage.
Chen, L., M. Gillenson, and D. Sherrell. 2002. Enticing online consumers: An extended
technology acceptance perspective, Information and Management, 39(8): 705-719.
Childers, T., C. Carr, J. Peck, and S. Carson. 2001. Hedonic and utilitarian motivations for
online retail shopping behavior, Journal of Retailing, 77 (4): 511- 535.
Cronbach, L.J. 1951. Coefficient alpha and the internal structure of the tests, Psychometrika,
16: 297-334.
Davis F.D. 1989. Perceived usefulness, perceived ease of use, and user acceptance of
information technology, MIS Quarterly, 13(3): 319-339.
Davis, F.D., R.P. Bagozzi, and P.R. Warshaw. 1989. User acceptance of computer
technology: A comparison of two theoretical models, Management Science, 35(8): 982-
1003.
Dishaw, M.T., and D.M. Strong. 1999. Extending the technology acceptance model with task-
technology fit constructs, Information and Management, 36 (1): 9-21.
Faber, R.J., and T.C. O’Guinn. 1992. A clinical screener for compulsive buying, Journal of
Consumer Research, 19(3): 459-69.
Fishbein, M., and I. Azjen. 1975. Belief, attitude, intention, and behavior, Reading, MA:
Addison-Wesley.
Gefen, D., and M. Keil. 1998. The impact of developer responsiveness on perceptions of
usefulness and ease of use: An extension of the technology acceptance model, The
DATA BASE for Advances in Information Systems, 29 (2): 35-49.
Hausman, A. 2000. A multi-method investigation of consumer motivations in impulse buying
behavior, Journal of Consumer Marketing, 17 (5): 403-419.
756
Internet Standard. 2010. E-commerce 2010 [report in Polish], online:
http://lead.download.idg.pl/sbwps/41bc88a066fc06e14849f847a927dd9c/4d5b134c/is/R
aportE_Commerce2010.zip
Jöreskog, K.G. 1969. A general approach to confirmatory maximum likelihood factor
analysis, Psychometrika, 34: 183-202.
Klopping. I.M., and E. McKinney. 2004. Extending the Technology Acceptance Model and
the Task-Technology Fit Model to Consumer E-Commerce, Information Technology,
Learning, and Performance Journal, 22 (1): 35-47.
Lederer, A.L., D.J. Maupin, M.P Sena, and Y. Zhuang. 2000. The technology acceptance
model and the World Wide Web, Decision Support Systems, 29 (3): 269-282.
Lee, D., J. Park, and J. Ahn. 2001. On the explanation of factors affecting e-commerce
adoption, Proceedings of the Twenty-Second International Conference in Information
Systems: 109-120.
Lin, J.C.C., and H. Lu. 2000. Towards an understanding of the behavioural intention to use a
web site, International Journal of Information Management, 20(3): 197-208.
Lysonski, S., S. Durvasula, and Y. Zotos. 1995. Consumer decision-making styles: a multi-
country investigation, European Journal of Marketing, 30 (12): 10-21.
Mącik, R., and D. Mącik. 2005. Unplanned buying tendency – measurement scale proposal,
Prace Naukowe IZiM UMCS w Lublinie, 12 (1): 53-72 [in Polish].
Mącik, R., and D. Mącik. 2008. Consumer’s e-commerce adoption models – the case of
polish consumers, Intercultural Dialogue and Management, Proceedings of the 2008
Management International Conference, Barcelona, Spain, November 26th–29th, 2008:
1003-1022.
Mącik, R., and D. Mącik. 2009. Consumer’s Decision-Making Styles and e-Commerce
Adoption – Two Samples Comparison, Creativity, Innovation and Management,
Proceedings of the 2009 Management International Conference, Sousse, Tunisia,
November 25th–28th: 1269-1290.
Magal, S. R., and D.A. Mirchandani. 2001. Validation of the technology acceptance model for
Internet tools, Proceedings of the Americas Conference on Information Systems: 1-16.
Malhotra, Y., and D. F. Galetta. 1999. Extending the Technology Acceptance Model to
Account for Social Influence: Theoretical Bases and Empirical Validation, Proceedings
of the 32nd Hawaii International Conference on System Sciences – 1999: 1-14.
757
Mokhtarian P. L., and W. L. Tang. 2009. Accounting for Taste Heterogeneity in Purchase
Channel Intention Modeling: An Example from Northern California for Book
Purchases, Journal of Choice Modelling, 2(2): 148-172.
Mokhtarian P. L., and W. L. Tang. 2011. Trivariate probit models of pre-purchase/purchase
shopping channel choice: clothing purchases in northern California, paper presented at
International Choice Modelling Conference 2011, Oulton Hall (UK) 4 – 6 July 2011,
http://www.icmconference.org.uk/index.php/icmc/ICMC2011/paper/view/426/119
Moon, J.W., and Y.G. Kim. 2001. Extending the TAM for a world-wide-web context.
Information and Management, 38 (4): 217-230.
Pastuszak, Z. 2008. E-Business Reception in Service and Manufacturing Companies,
Intercultural Dialogue and Management, Proceedings of the 2008 Management
International Conference, Barcelona, Spain, November 26th–29th, 2008: 193-203.
Rook, D.W., and R.J. Fisher. 1995. Normative influences on impulsive buying behavior,
Journal of Consumer Research, 22 (December): 305-313.
Sproles, G. B., and E. L Kendall. 1986. Methodology for Profiling Consumers’ Decision-
Making Styles, The Journal of Consumer Affairs, 20 (2): 267-279.
Tai, S. H. C. 2005. Shopping Styles of Working Chinese Females, Journal of Retailing and
Consumer Services, 12 (May): 191-203.
Taylor, S., and P. A. Todd. 1995. Understanding information technology usage: A test of
competing models, Information Systems Research, 6 (2): 144-176.
Teering, M.L., and E.K.R.E. Huizingh. 2005. The complementarity between online and
offline consumer attitudes and behavior. Working Paper, University of Groningen, The
Netherlands
Teo, T.S.H., V.K.G. Lim, and R.Y.C. Lai. 1999. Intrinsic and extrinsic motivation in Internet
usage, Omega, 27 (1), 25-37.
Venkatesh V, and H. Balla. 2008. Technology Acceptance Model 3 and a Research Agenda
on Interventions, Decision Sciences, 39(2): 273-315.
Venkatesh, V., and F. D. Davis. 2000. A theoretical extension of the technology acceptance
model: Four longitudinal field studies, Management Science, 46 (2): 186-204.
Venkatesh V., M.G. Morris, G.B. Davis, F.D. Davis. 2003. User Acceptance Of Information
Technology: Toward A Unified View, MIS Quarterly, 27(3): 425-478.
Walsh, G., T., T. Hennig-Thurau, V.-W. Mitchell, K.P. Wiedmann. 2002. Segmenting
Markets Using Consumers’ Decision Making Style, Journal of Targeting. Measurement
and Analysis in Marketing, 10: 117-131.
758
Young-Hyuck, J., and M. Hyung-Jin Park. 2008. Information Search and Purchase Channel
Choice Across In-home Shopping Retail Formats. Academy of Marketing Studies
Journal, 12(2): 49-61.
Zigurs, I., B.K. Buckland, J.R. Connolly, and E.V. Wilson. 1999. A test of task-technology fit
theory for group support systems, ACM SIGMIS Database, 30 (3-4), 34-50.
759