T h e E m era ld R es ea rc h R e g ister fo r th is jo u rn a l is a v a ila b le a t T h e c u rren t is su e a n d fu ll tex t a rc h iv e o f th is jo u rn a l is a v a ila b le a t

http://www.emeraldinsight.com/researchregister http://www.emeraldinsight.com/0956-4233.htm

Understanding consumer Consumer

motivation and
motivation and behavior behavior

related to self-scanning in
59
retailing
Implications for strategy and research on
technology-based self-service
Pratibha A. Dabholkar
University of Tennessee, Knoxville, Tennessee, USA
L. Michelle Bobbitt
Bradley University, Peoria, Illinois, USA, and
Eun-Ju Lee
California State University, Los Angeles, California, USA
Keywords Self-service, Supermarkets, Retailing, Technological innovation, Consumer attitudes
Abstract Self-scanning technology is being tested by major supermarket chains as well as other
types of retailers across the world, but the success of the new technology from the consumer’s
perspective is not yet clear. This study investigates consumer reasons for both using and avoiding
self-scanning checkouts with a view to addressing these practitioner issues. In addition, the study
advances theory on consumer motivation and behavior related to technology-based self-service in
general. Factors driving preference or avoidance of self-scanning checkouts include attributes of self-
scanners, consumer differences, and situational influences. Reasons for preference of other types of
technology-based self-service over traditional service alternatives are also explored to determine
motivational and behavioral patterns across service contexts. A combination of research methods is
used to investigate these issues and offers richer findings than any one method used alone.
Implications are discussed for managerial strategy as well as for future research.

Recent advances in technology have created a surge in ``technology-based self-

service’’ delivery options ranging from in-room hotel checkout and automated
airline ticketing by telephone to Internet shopping (Dabholkar, 1994a). In
banking, for example, although automated teller machines (ATMs) were not
well received when first introduced more than 20 years ago, technology-based
self-service options now include telephone banking, Internet banking, smart-
card banking, and home banking via television, in addition to the ATM
(Prendergast and Marr, 1994). Such developments are changing the way service
firms and consumers interact, and are raising many new research issues for International Journal of Service
Industry Management
investigation (Dabholkar, 2000). Vol. 14 No. 1, 2003
pp. 59-95
# MCB UP Limited
The study was made possible by a grant from the College of Business Administration, 0956-4233
University of Tennessee, Knoxville. DOI 10.1108/09564230310465994
IJSIM Like the ATM in its early years, retail self-scanners first introduced in
14,1 grocery stores nearly two decades ago were not well received at the time.
However, unlike ATMs which were eventually accepted, self-scanning
checkouts were met with stubborn resistance. In fact, self-scanning in retail
stores represents a classic case of a technology-based self-service that failed on
its first inception. It is possible that this option was introduced at a time when
60 consumers were simply not ready to change their behaviors to adopt a new
way of shopping. Or, it may be that unlike the ATM, this particular technology-
based self-service was viewed as requiring too much effort on the part of the
consumer.
The modern consumer is much more technologically aware and comfortable,
and supermarket chains are beginning to experiment again with self-scanners
(Hennessy, 1998). According to the Food Marketing Institute, 18 supermarket
chains in North America had self-scanning lanes in 1998 (Discount Store News,
1998a), and by 2004, nearly half of the grocery retailers in the USA will offer
some form of self-scanning to consumers (Chain Store Age, 2002). That may be
an optimistic estimate perhaps; by the end of 2000, 14 percent of grocery
retailers in the USA with 11 or more stores offered self-scanning (Rohland,
2001), and currently about 20 percent offer self-scanning along with the
traditional checkouts (Forster, 2002).
The phenomenon is not restricted to the USA; Costco is testing self-checkout
systems in Canada (Discount Store News, 1998b), and NCR reports interest in
stationary self-scanners from supermarket chains in The Netherlands and
Germany (Hunt, 1998). Not that Europe has been behind the USA in this area.
Handheld self-scanners have been used by preferred customers in major
supermarket chains in several European countries since the mid-1990s – Albert
Heijn in The Netherlands, Safeway in the UK, SuperQuinn in Ireland, Metro in
Germany, and Monoprix in France (Ross, 1997).
Nor are supermarket chains the only retailers to try self-scanning checkouts.
Walmart is currently testing four different self-checkout systems and estimates
that even the smallest machines in each supercenter would add a cost of more
than $20 million (Bowden, 2002). Kmart, before filing for bankruptcy, had
reported satisfactory usage from all of the self-scanners it was testing (Chain
Store Age, 2002). Retailing consultants predict that if self-scanning checkouts
become widely accepted in supermarkets, the next service industries to
consider these systems will be drugstores and home improvement chains,
given their on-going problems with recruiting and the long lines at their
checkouts (Grant, 2001).
Given the expense and difficulty related to retaining a sufficient number
of reliable employees, retailers in general, and supermarkets in particular,
are viewing self-scanners as a smart alternative to their constant hiring and
training woes. Furthermore, if successful, self-scanners represent a huge
source of potential savings at a time of economic uncertainty. However, it is
not at all clear if these investments are paying off at present. Although the
number of US consumers who have tried self-scanning checkouts increased Consumer
from 6 percent in 1999 to 16 percent in 2001 (McDonald, 2002), this increase motivation and
is not a huge jump and is below industry expectations. Target, for example, behavior
is refusing to jump on the bandwagon of self-scanning (even though its
supercenters sell groceries), based on a firm belief that customers value the
human touch at checkouts (Grant, 2001).
In fact, early tests have shown mixed results as some shoppers seem willing 61
to try the self-scanners and others studiously ignore them (Discount Store
News, 1998a). Feedback from customers suggests that some shoppers like the
shorter lines and privacy (Chain Store Age, 2002), but others find the systems
difficult to use and feel uncomfortable with the machines screaming
instructions at them (Wisely, 2002). As stationary self-scanners can cost as
much as $90,000 (Grant, 2001), before investing money in this particular
technology-based self-service option on a large scale, retailers need to
determine its future potential through systematic, in-depth research on
consumer reasons for using or avoiding self-scanning checkouts.
It is our objective to conduct research that provides the needed answers for
this pressing practical issue not only in the supermarket industry, but for
retailing in general. In addition, the results should be useful for understanding
consumer evaluation and use of other technology-based self-service options,
and thus serve to advance theory in the services marketing field.
The only academic research on consumer perceptions of self-scanning is
a study by Anselmsson (2001). He examines determinants of perceived
service quality for grocery and library self-scanning in Sweden and
indicates what is important to consumers in evaluating this form of service
delivery. His research includes attributes of technology-based self-service
as well as consumer characteristics. We build on Anselmsson’s research in a
number of ways. The grocery store used as an empirical setting in his
study only offers self-scanning checkouts so that consumers do not have a
choice as to service delivery options. We broaden the investigation to
include a choice situation and to probe consumer reasons for both using and
avoiding the grocery self-scanning checkout. Anselmsson’s study includes
qualitative research which is used as a guide for his quantitative study.
We use a different approach in that we base our quantitative study on
past theory and simultaneously conduct a qualitative study to
allow comparison of the efficacy of different research methods. The data in
Anselmsson’s study were collected through mail surveys. Our interviews
are conducted in the field to draw on the immediacy of the setting to
invoke more relevant responses. Finally, we include possible situational
influences on the use of self-scanners, which were not considered in
Anselmsson’s study.
Past research on technology-based self-service (e.g. Dabholkar, 1996; Meuter
et al., 2000) has found that perceived attributes of the technology and consumer
preferences regarding interaction with the employee play a role in whether or
IJSIM not consumers will use such options. The methods used to study technology-
14,1 based self-service have included rigorous multivariate analysis of survey data
and critical incident techniques.
This study uses a combination of research methods to determine the factors
that influence consumers to use (or avoid) self-scanning checkouts. Survey
methodology and quantitative analysis are used to examine the relevance of
62 factors found to be important in past research on technology-based self-service.
In addition, tightly structured interviews and detailed content analysis are used
to extract factors that are important to consumers in using or avoiding the self-
scanning checkout. The comparison of results from different research methods
has interesting implications for future research methodology in services
marketing.
The factors investigated encompass perceived attributes of the self-
scanners, situational influences, and consumer differences in terms of related
behaviors and demographics. The study provides meaningful strategic
implications for retailers as well as advances theory in services marketing that
can be applied to a host of other service industries where technology-based
self-service options are offered or being considered.

Conceptual framework
Today, consumers can choose between a variety of technological options to
perform services for themselves; at the same time, companies can employ
technology at various stages in the service delivery process to improve the
quality and productivity of their service offering (Blumberg, 1994; Quinn,
1996). Providing these technological innovations for self-service is challenging
the notion that provider-client interaction is an essential feature of service
delivery (Prendergast and Marr, 1994) and is raising a host of significant
research issues that need to be investigated (Dabholkar, 2000; Lovelock, 1995;
Meuter and Bitner, 1998).
In implementing technology-based self-service, many service firms hope to
offer better service to consumers. But what do consumers see as the
constituents of better service? Dabholkar (1996) proposed that speed, control,
reliability, ease of use, and enjoyment are all important attributes to consumers
in evaluating and using technology-based self-service. She found ease of use,
control, and enjoyment to be strong determinants of perceived service quality
in her study on touch screen ordering in fast food restaurants. Although speed
was not found to be significant, its effect may have been masked by the
inclusion of waiting time in her study. Similarly, the effect of reliability may
have been masked by its high correlation to control. In fact, Dabholkar’s
(1994b) earlier research found performance, encompassing reliability and
accuracy, to be an important determinant of evaluation and use of technology-
based self-service.
Other researchers also support these five attributes. Studies have found
speed to be an important determinant of preference for self-service in general
(Bateson, 1985) and self-scanning in particular (Anselmsson, 2001). Similarly, Consumer
research on self-service (Bateson, 1985) and on-line shopping (Hoffman and motivation and
Novak, 1996) shows that consumers perceive increased control in using such behavior
options and that it positively affects their evaluation. In a discussion on
automated self-service, Evans and Brown (1988) suggest that reliability of the
technology plays a critical part in consumer acceptance of such service options.
Finally, studies on the adoption of computer technology (Davis et al., 1989; 63
1992), preference for self-scanning (Anselmsson, 2001), and evaluation of
on-line shopping (Childers et al., 2001) show that ease of use and enjoyment are
important aspects for using such options.
Consumers who regularly use self-scanning are likely to think all of these
attributes important. They would view self-scanners as performing well on
these attributes and this would guide their preference and use of the option.
Based on the literature, we therefore propose the following hypothesis related
to attributes of technology-based self-service.
H1. Compared to those who do not plan to use it regularly, consumers who
plan to use self-scanning regularly will:
(a) perceive it as faster;
(b) perceive it as offering greater control;
(c) perceive it as more reliable;
(d) perceive it as easier to use;
(e) perceive it as more enjoyable;
(f ) prefer it to the traditional checkout.
But what are the attributes important to those who prefer the traditional
checkout? Are they concerned about speed, control, and so on, and believe the
traditional checkout performs better on these attributes, or do they have other
reasons to avoid the self-scanning checkout?
Anselmsson (2001) found that only 25 percent of the respondents thought
self-scanning was faster than employees scanning the purchases. It is possible
that only those who prefer self-scanning will view it as faster, whereas those
who prefer the traditional checkout will perceive self-scanning as slower. In
addition, if the technology is cumbersome or complex, or if the consumer is not
technologically proficient, the self-scanning checkout could actually increase
the service delivery time. On the other hand, both groups may view the self-
scanning checkout as faster, but the second group may have other reasons for
choosing the traditional checkout, such as the human interaction involved.
The same type of possibilities exist for the other attributes. For example,
consumers may see self-scanning as reliable but still prefer the traditional
checkout in order to interact with an employee. Or, they may think the
traditional checkout is actually more reliable. Consumers who prefer the
traditional checkout may feel greater control in using that option, or control
IJSIM may not even be an important factor in their evaluation. Given the lack of
14,1 theory on consumer avoidance of technology-based self-service based on
specific attributes of such options, we plan to investigate possible reasons for
avoidance through content analysis.
Although speed is closely associated with service quality for many services
(Sellers, 1990), Ledingham (1984) suggests that efficiency and speed are more
64
important to consumers who use technology to serve themselves. Other
consumers, however, value human interaction above anything else in service
delivery (Cowles and Crosby, 1990; Dabholkar, 1996; Prendergast and Marr,
1994). In contrast, consumers who prefer technology-based self-service may
actually wish to avoid interaction with a service employee (Anselmsson, 2001;
Dabholkar, 1996; Meuter et al., 2000). Thus, consumer attitudes toward
interaction with service employees are likely to influence their use of
technology-based self-service. We hope to discover this relationship through
our content analysis and propose the following:
H2a. Consumers who like self-scanning (and use or plan to use it) will wish
to avoid interaction with service employees.
H2b. Consumers who dislike self-scanning (and have not used it or plan not
to use it) will value interaction with service employees.
Greater familiarity with technology results in more favorable attitudes toward
using technology-based self-service options in general (Dabholkar, 1992; 1996).
Further, once consumers become used to a particular technology, they more
readily adopt other technologies (Dickerson and Gentry, 1983; Korgaonkar and
Moschis, 1987). Thus, attitudes toward using technology in general (which are
linked to consumer familiarity with technology in general) have a direct
bearing on consumer attitudes and behavior toward a specific technology-
based self-service. As in the previous case, we hope to discover this relationship
through our content analysis and propose the following:
H3a. Consumers who like self-scanning (and use or plan to use it) will have
favorable attitudes toward using technology in general.
H3b. Consumers who dislike self-scanning (and have not used it or plan not
to use it) will have unfavorable attitudes toward using technology in
general.
We also expect a carryover effect of consumer familiarity and preference for
other technology-based self-service options to use of self-scanning. Similarly,
we expect a carryover effect of consumer avoidance for interacting with
employees in other shopping options to use of self-scanning. Considering the
two effects together, we propose the following hypothesis:
H4. Consumer who use self-scanning in grocery stores will prefer:
(a) shopping from home to shopping at the store;
(b) Internet shopping to telephone shopping;
 (c) using touch-tone dialing to speaking to a person when telephone Consumer
shopping; motivation and
(d) using a computer touch screen in the store to ordering verbally to an behavior
employee in the store;
(e) using an ATM to using a bank teller.
65
In addition, from the behavioral and motivational patterns that emerge for
these different forms of technology-based self-service options, we propose four
hypotheses parallel to hypotheses H2a-b and H3a-b. Based on the theory
discussed earlier for hypotheses H2 and H3, we expect similar relationships for
using and avoiding these other technology-based self-service options as we do
for using and avoiding self-scanning. Thus:
H5a. Consumers who prefer a particular technology-based self-service to its
alternative traditional service option will wish to avoid interaction
with service employees.
H5b. Consumers who prefer the alternative traditional service option to a
particular technology-based self-service, will value interaction with
service employees.
H6a. Consumers who prefer a particular technology-based self-service to its
alternative traditional service option will have favorable attitudes
toward using technology in general.
H6b. Consumers who prefer the alternative traditional service option to a
particular technology-based self-service, will have unfavorable
attitudes toward using technology in general.
Past research (e.g. Dickerson and Gentry, 1983; Prendergast and Marr, 1994)
found that younger, better educated, and affluent males were more likely to use
technology-based self-service. Some retailers have observed that older
consumers do seem somewhat reluctant to use self-scanning (Discount Store
News, 1998a; Grant, 2001), possibly due to being accustomed to service by
employees in this context. Yet, Anselmsson (2001) found the opposite to be the
case. It is unlikely today that demographic factors play a major role in the
evaluation and use of in-store technology-based self-service. Therefore, we
measure demographics to investigate these issues only in an exploratory sense.
The one hypothesis we propose on demographics that is supported by theory is
related to Internet access, which may be viewed as a more relevant surrogate
for education and income in this context. We have noted that as consumers
become comfortable with technology in one service industry, they are more
willing to try technologies in other service industries (Dickerson and Gentry,
1983; Korgaonkar and Moschis, 1987). This suggests that consumers more
likely to use technology-based self-service options are the ones more familiar
with technology in general, such as indicated by greater Internet access. Thus:
H7. Consumers who use self-scanning will have greater access to the
Internet than consumers who avoid self-scanning.
IJSIM As before, we propose a parallel hypothesis for the other forms of technology-
14,1 based self-service options explored in this study. Thus:
H8. Consumers who prefer a particular technology-based self-service will
have greater access to the Internet than consumers who prefer the
alternative traditional service option.
Researchers suggest that situational factors can influence the use of
66 technology-based self-service, including Internet shopping (Bobbitt and
Dabholkar, 2001; McMellon et al., 1997). Empirical studies (Dabholkar, 1996;
Dabholkar and Bagozzi, 2002) have substantiated the influence of situational
factors, such as waiting time and crowding, on the use of technology-based self-
service. Based on the literature as well as on observation, it is likely that a
variety of situational factors, such as the length of lines for alternative checkout
options, time of day, day of the week, whether the store is crowded, and
whether the consumer is in a hurry, will influence evaluations of the self-
scanning checkout. We do not propose hypotheses for situational factors
because a rigorous test of these would necessitate an experimental research
design. Instead, we study them in an exploratory sense, both by observing
conditions at the time of the interview and also by specifically asking
consumers to name situations under which they would use self-scanning
checkouts.
To sum, the principal research objective for this study is to determine the
reasons consumers use (or avoid) self-scanning checkouts. Based on classic
adoption literature (e.g. Rogers, 1983; Gatignon and Robertson, 1985) as well as
the literature on technology-based self-service (e.g. Dabholkar 1992; 1994b;
1996; Meuter et al., 2000; Prendergast and Marr, 1994), we expect these reasons
to include ``innovation characteristics’’ (i.e. attributes of self-scanners) as well as
``personal characteristics’’ (i.e. consumer attitudes toward interacting with
employees and toward using technology). Another research objective is to
compare use of self-scanning checkouts with consumer shopping preferences
for other technology-based self-service options to uncover possible behavioral
and motivational patterns. We also plan to explore the influence of
demographic factors on the use and avoidance of self-scanning checkouts and
other technology-based self-service options. Yet another research objective is to
investigate the effect of situational factors on usage of self-scanning checkouts.
Finally, we plan to compare results from different research approaches to offer
insights on the approaches themselves. All of our findings should have
implications for managerial strategy as well as for future research on services.

Methodology
Data collection and sample selection
A large regional supermarket chain in the southeastern USA was selected for
the study. A representative store that offered both self-scanning and traditional
checkout was chosen for data collection. Using tightly structured interviews,
data were collected from two groups of shoppers in the selected store. One
group consisted of consumers shopping at various locations throughout the Consumer
store. The other group consisted of shoppers at the self-scanners. motivation and
Undergraduate honor students, rigorously trained with practice interviews, behavior
collected the data. They were monitored on-site, and coached as needed, by
doctoral students.
Data collectors took up strategic spots in the store (including some at the
checkouts) at various times of the day and on various days of the week. Every 67
fifth person encountered in the store by each data collector was approached and
asked if they would be willing to answer a few questions. In the second group,
respondents were actually in the process of checking out using the self-
scanners. The students identified themselves to potential respondents in both
groups and explained that this was a university research project. This resulted
in an unusually high response rate of 83.33 percent. The students were
equipped with clipboards and made quick but comprehensive notes as they
conducted the interviews.

Interview questions
For respondents located throughout the store (i.e. the ``in-store’’ group), a set of
closed-ended questions measured awareness of the self-scanning checkout in
the store, as well as past usage, attitude, and intentions for future use of the
``self-scan option.’’ For respondents at the self-scanners, awareness was evident.
Hence, similar closed-ended questions were limited to past usage of self-
scanning and intentions to use the self-scan option in the future.
For the in-store group, open-ended questions were included to capture why
respondents liked or disliked the self-scanning checkout. For respondents at the
self-scanners, preference for this option over the traditional checkout was
measured quantitatively, as described later in this section. For both groups,
open-ended questions were included to capture why the respondents planned to
use or not use the self-scan option in the future.
Both groups were also questioned about their shopping preferences for
technology-based self-service versus the alternative, traditional service
provided by an employee. Specifically, respondent preferences were measured
for shopping from home vs shopping at the store, Internet shopping vs
telephone shopping, using touch-tone dialing vs talking to a service employee
by telephone, using a computer touch screen in the store vs ordering verbally to
an employee in the store, and using an ATM vs using a bank teller.
Demographic questions on age, gender, education, and measures of Internet
access were included for both groups. Situational factors, including time of day,
day of the week, crowded conditions, relative length of lines at alternative
checkouts, and whether the consumer was in a hurry, were noted by the
interviewer. In addition, respondents were asked under what situations they
would use the self-scan option.
For the group at the self-scanners, additional questions measured
perceptions of speed, control, reliability, ease of use, and enjoyment related to
IJSIM the self-scanning checkout, as well as their overall preference for the self-scan
14,1 option over the traditional checkout. Each construct was measured using two
seven-point Likert items; the phrasing was adapted from Dabholkar’s (1996)
study for the self-scanning context (see Appendix).

Quantitative analysis
68 Confirmatory factor analysis and reliability tests were performed on the items
used to measure perceptions and preference related to the self-scanning
checkout. T-tests (and ANOVAs) were conducted to determine differences in
these perceptions and preferences, for situational and demographic differences
as well as for different groups of respondents. Frequencies were computed and
nonparametric statistical tests conducted to determine differences between the
two major respondent groups (i.e. in-store and at the self-scanners) in terms of
demographic factors. Similar tests were conducted to determine differences in
shopping preferences between these two groups. The tests were repeated for
certain relevant sub-groups of respondents to look for possible differences.

Content analysis
The qualitative data collected from both groups were recorded in detail. Two
researchers independently identified categories for all the responses recorded,
then discussed these categories to determine agreement on labeling. Inter-judge
reliability can be ascertained by a number of possible measures. Initial
agreement between the two researchers was 92.6 percent. Differences in
opinion regarding the categories were discussed so that agreement on labels
rose to 97 percent. A third researcher examined the agreed-upon categories and
after further discussion, 7.4 percent of these were relabeled. The third
researcher also reconciled the differences for the categories (3 percent) where
agreement had not been reached. Final agreement on labeling was 100 percent.

Results
Sample breakdown by research design
The sample of consumers shopping throughout the store included 101
respondents, and the sample of consumers at the self-scanners included 49
respondents. A breakdown for the 101 in-store respondents, in relation to
awareness, past use, and attitudes related to the self-scan option, as well as
their intentions to use this option in the future, is shown in Figure 1. A
breakdown for the 49 respondents at the self-scanners, in relation to past use
and future intentions, is shown in Figure 2.

Results of quantitative analysis

Respondents at the self-scanners had answered a survey to measure their
perceptions and preference for the self-scan option (see Appendix).
Confirmatory factor analysis (Jöreskog and Sörbom, 1993) was run on all the
items capturing perceptions (speed, control, reliability, ease of use, and
 Consumer
motivation and
behavior

69

Figure 1.
Distribution of in-store
survey respondents

Figure 2.
Distribution of survey
respondents at
self-scanners

enjoyment) of the self-scan option and preference for the same over the
traditional checkout. The results strongly supported the six factor structure
with a chi-squared value of 71.58, with df = 39, RMSR = 0.03, NNFI = 0.92, and
CFI = 0.95. Cronbach’s alpha values for these constructs were 0.97 for speed,
0.92 for control, 0.87 for reliability, 0.84 for ease of use, 0.86 for enjoyment, and
0.86 for preference for the self-scan option.
The sample (n = 49) was too small to run structural equations, and
regression analysis did not discriminate sufficiently among the constructs.
Separate simple regressions showed all factors (perceptions) to be significant
determinants of preference, but multiple regression showed only ease of use to
be significant (b = 0.85, p < 0.001), masking the effect of other factors.
IJSIM In any case, the research question of interest was whether consumers who
14,1 planned to use self-scanning regularly had different perceptions of it from those
who did not plan to use it regularly, and whether these consumers indeed
preferred the option to the traditional checkout (H1). It was expected that the
sample would be somewhat equally divided between these two groups. Instead,
39 respondents planned to use self-scanning regularly, four did not plan to use
70 it, and six respondents indicated they might use it depending on the situation
(see Figure 2). T-tests were conducted to compare perceptions of the 39 who
planned to use self-scanning regularly versus the ten who did not plan to use it
or would only use it under certain situations.
Despite the one small group (n = 10), the t-tests worked well. Consumers
who planned to use self-scanning regularly viewed it as offering greater
control (t = 2.12, p < 0.05), more reliable (t = 2.05, p < 0.05), easier to use
(t = 2.45, p < 0.05), and offering greater enjoyment (t = 3.41, p < 0.01) than
those who did not plan to use this option regularly. Thus, hypotheses H1b,
H1c, H1d, and H1e were supported. Only speed was not significantly
different for the two groups, thus failing to support hypothesis H1a.
This result does not however indicate that speed was not important to
these groups. The mean value for speed was 5.67 (on a scale of 1-7),
higher than the means for all the other perceptions. This suggests
that irrespective of whether consumers planned to use self-scanning
regularly, they saw it as a fast option. Finally, a t-test confirmed that
consumers who planned to use self-scanning regularly showed greater
overall preference for the option over the traditional checkout (t = 3.12,
p < 0.05) than the group that did not plan to use this option regularly, thus
supporting hypothesis H1f.
Although hypothesis H2 was to be tested with content analysis, a
nonparametric test statistic (Mann-Whitney) also showed some support for
H2 as follows. Shopping preferences of consumers within the in-store group
were compared between those who had used the self-scan and those who
had not used the self-scan or were not aware of it. The only significant
difference was that the first group preferred using touch screen ordering in
a store to ordering verbally to an employee in a store (z = 1.93, p < 0.054). It
appears that consumers who had used the self-scan do want to avoid contact
with employees, thus offering support for hypothesis H2a. The finding
also suggests simultaneously that consumers who had not used the self-
scan prefer interacting with an employee, thus offering support for
hypothesis H2b.
Table I shows the shopping preferences for in-store respondents (n = 101)
and for respondents at the self-scanners (n = 49). A series of Mann-Whitney
tests showed no differences between the two groups for preferences related to:
shopping from home vs shopping at the store; using touch-tone dialing vs
speaking to a person when telephone shopping; and using a computer touch
 In-store At self-scanners Consumer
n = 101 n = 49 motivation and
Comparison of shopping methods n Percent n Percent behavior
1. Shopping from home 15 21 7 19
Shopping at the store 56 79 30 81
Total 71 100 37 100 71
2. Internet shopping from home 18 25 19 51
Telephone shopping from home 52 73 16 43
Total 70 98 35 94
3. Using touchtone dialing when telephone shopping 7 10 6 16
Speaking to a person when telephone shopping 61 86 29 78
Total 68 96 35 94
4. Using a computer touch screen in a retail store 17 17 14 29
Ordering verbally to an employee in a retail store 81 80 34 69
Total 98 97 48 98
5. Using an ATM for banking transactions 49 49 35 71
Using a bank teller for banking transactions 51 50 12 24
Total 100 99 47 95
Notes: 1. A screening question was used to determine the people who had shopped from
home before. Only these people (71 and 37 in the two groups respectively) answered the first Table I.
three questions. (The entire samples answered questions 4 and 5.) 2. Percentages may not Shopping preferences
add up to 100 per cent due to some non-response on questions of respondents

screen vs ordering verbally to an employee in a store (see categories 1, 3, and 4

in Table I).
However, compared to in-store respondents, respondents at the self-
scanners preferred Internet shopping to telephone shopping (z = 2.88,
p < 0.01) and (2) using ATMs to using bank tellers (z = 2.75, p < 0.01) (see
categories 2 and 5 in Table I). Thus, hypotheses H4a, H4c, and H4d were not
supported, but hypotheses H4b and H4e were supported. Given that the
in-store respondents included those who had used and liked the self-scan,
this difference across the two groups for hypotheses H4b and H4e is even
more striking.
Table II shows the demographic profiles of the two major groups of
respondents (i.e. in-store and at the self-scanners). Within the second group,
i.e. respondents at the self-scanners, t-tests were conducted for gender and
overall Internet access (yes/no), and ANOVAs were run for age, education,
and specific Internet access (home/work/both). No differences were found in
perceptions of attributes or in overall preference for the self-scanning
checkout across any of the basic demographic categories or for Internet
access.
To compare differences across the two groups of respondents without
reference to attributes or preference, a nonparametric test statistic was used
IJSIM In-store At self-scanners
14,1 Demographics n Percent n Percent

Age
18-24 29 28.7 16 32.7
25-34 18 17.8 17 34.7
35-44 22 21.8 4 8.2
72 45-54 15 14.9 5 10.2
55-64 10 9.9 5 10.2
65 and over 7 6.9 2 4.1
Total 101 100.0 49 100.0
Gender
Male 37 36.6 21 42.9
Female 64 63.4 28 57.1
Total 101 100.0 49 100.0
Education
Grade school 1 1.0 2 4.1
High school 13 12.9 2 4.1
Some college 28 27.7 11 22.4
Undergraduate degree 31 30.7 13 26.5
Graduate degree 20 19.8 15 30.6
More than one graduate degree 8 7.9 5 10.2
Total 101 100.0 48a 97.9
Overall Internet access
Yes 77 76.2 44 89.8
No 24 23.8 5 10.2
Total 101 100.0 49 100.0
Specific Internet accessb
Home 23 29.9 20 45.5
Work 18 23.4 20 45.5
Both 29 37.6 4 9.1
Total 70c 90.9 44 100.0
Notes: a One person did not provide this information; b this question was asked only to
Table II. those who had Internet access (i.e. 77 in the in-store group and 44 in the self-scan group);
c
Demographic profiles seven people did not provide this information

given the independence of the samples. The Mann-Whitney test showed no

significant differences for age, education, and gender across these two
groups. This is a good finding in that it verifies that the demographic
profiles of the randomly selected respondents in the two major groups are
similar.
Other sub-groups within the in-store group were also compared for possible
demographic differences using the same procedure, i.e. the Mann-Whitney test.
No difference in demographic profiles was found between respondents who had
used the self-scan and those who had not used it or who were not aware of it.
Nor were any demographic differences found between respondents who had
liked the self-scan and those who had disliked it. These findings show
that demographic factors do not influence use, preference, or avoidance of Consumer
self-scanning. motivation and
Similarly, when demographic profiles were compared for shopping behavior
preferences, no differences were found for shopping from home vs shopping at
the store, for using touch-tone dialing vs speaking to a person when telephone
shopping, or for using a touch screen vs ordering verbally to an employee in the
store. However, those who preferred using ATMs to using bank tellers tended 73
to be younger (z = 3.41, p < 0.001) and those who preferred Internet shopping to
telephone shopping also tended to be younger (z = 1.87, p < 0.05) and were more
likely to be male (z = 2.64, p < 0.01), supporting earlier research on these specific
contexts.
As predicted, Internet access across the two major groups of respondents
was significantly different. Respondents at the self-scanners were more likely
to have Internet access than those who were interviewed in other areas of the
store. The Mann-Whitney statistic was significant for overall Internet access
(z = 1.97, p < 0.05) and for specific Internet access (z = 2.97, p < 0.01). Thus,
hypothesis H7 is supported at two levels.
There was no difference in Internet access across the same three earlier sets
of shopping preferences that had shown no demographic differences in
consumer profiles. However, Internet access was higher for those who preferred
using ATMs to using tellers (z = 3.29, p < 0.001) and understandably for those
who preferred Internet shopping to telephone shopping (z = 3.40, p < 0.001),
thus partially supporting hypothesis H8.
Finally, t-tests were conducted for a variety of situational factors:
weekday vs weekend, morning vs evening, longer vs shorter wait lines at
the self-scanning checkout, whether the consumer was in a hurry or not,
and whether the store was crowded or not. None of the first four situational
factors changed perceptions of attributes or preference for the self-scan.
The only situational factor that showed a difference was whether the
store was crowded. A test for crowding showed that respondents using
the self-scanning checkout under crowded conditions thought it was
faster than those who were using it under normal conditions (t = 2.13,
p < 0.05).

Results of content analysis

Preference, avoidance and situational use of self-scanning option. Table III
compares reasons consumers like or plan to use the self-scan option among
three groups: respondents at the self-scanners (n = 39), in-store respondents
who had used and liked this option (n = 29), and in-store respondents who had
not used this option but were thinking of trying it (n = 18). It is seen that the
most important reason that all these consumers would want to use the self-scan
option is that they perceive it as ``fast.’’ This result offers additional support to
explain why the relevance of speed could not be differentiated across groups in
the earlier quantitative analysis (see H1a). The content analysis results suggest
 74
14,1
IJSIM

Table III.

self-scan option
Reasons for using
What was liked about self-scan Why self-scan option may be used in
Why self-scan option will be used option (In-store respondents who the future (In-store respondents who
regularly (Respondents at self-scanners) n = 39 had used self-scan option) n = 29 had not used self-scan option) n = 18

Fast 30 Fast 22 Fast 8

No waiting 11 Easy to use 8 Enjoyable 3
Easy to use 11 Convenient 7 Easy to use 3
Convenient 7 No waiting 6 Convenient 2
Avoid interaction with employee 4 Can use for few products 5 No waiting 2
Can use for few products 3 Enjoyable 4 Control 1
Control 2 Novelty 4 Accurate 1
Enjoyable 2 Accurate 2
Can use for certain products 1 Assistance 2
Employees are rude, unhelpful, etc. 1 Avoid interaction with employee 1
Advanced technology 1 Control 1
Loves the option 1 Familiarity with technology 1
Ability to see prices 1 Likes self-service 1
Novelty 1 Time pressure 1
Using self-service to fill time 1 Self-scan is challenging 1
(as expected, based on the comparison of attribute means) that, irrespective of Consumer
where they were interviewed, consumers who had used or planned to use self- motivation and
scanning viewed this option as fast. behavior
Other frequently cited reasons for using self-scanning were that there was
no waiting (related to speed), and it was easy to use, convenient, and enjoyable.
Control and accuracy were also mentioned but less frequently. In addition to
reasons related to attributes of self-scanning, respondents mentioned ``avoid 75
interaction with employee,’’ and ``employees are rude, unhelpful, etc.,’’ thus
offering some support for hypothesis H2a. Favorable attitudes toward using
technology were mentioned in a variety of ways – ``advanced technology,’’
``familiarity with technology,’’ ``self-scan is challenging,’’ ``loves the option,’’ and
``novelty’’ – thus offering support for hypothesis H3a.
Table IV compares reasons consumers do not like or do not plan to use the
self-scan option among three groups: respondents at the self-scanners (n = 4),
in-store respondents who had used but disliked this option (n = 17), and in-store
respondents who had not used this option and were not planning to try it
(n = 18). The most important reason these consumers would want to avoid the
self-scan option is that they like to interact with employees, thus supporting
hypothesis H2b. As additional support for H2b, some respondents said that
self-scanning was impersonal or they liked the social experience and the
relationship with employees.
Other important reasons for avoiding self-scanning include perceptions that it is
difficult to use or the customer is not familiar with it. There is
also a sense that the customer has a right to expect service, the
self-scan involves too much effort, and the price should be lower for
self-service. These along with other reasons such as ``dislikes automation,’’ ``lack of
familiarity,’’ and ``uncomfortable with using self-scan’’ suggest unfavorable
attitudes toward using technology, thus supporting hypothesis H3b. Perceptions
of the self-scan as slow, inaccurate, difficult to use, having process problems, and
inconvenient add indirect but further support for hypothesis H3b.
Table V compares situations where consumers would want to use the self-
scan option among three groups: respondents at the self-scanners (n = 6);
in-store respondents who had used this option (and either liked or disliked it)
(n = 16); and in-store respondents who had not used this option (n = 16). The
most frequently mentioned situation relates to number of items. In other words,
if the store did not have a policy restricting the number of items a customer
could have in order to use the self-scan, more consumers would be willing to
use this option.
Other situations cited frequently were ``if (there is a) line at regular checkout’’
and ``if (customer is) in a hurry.’’ These situations along with ``if line at self-scan is
short’’ reveal the importance of speed and/or the perception of the self-scan as fast.
Those relatively unfamiliar with the self-scan however said they would use the
option if they had the time to learn how, if someone helped them,and if they gained
 76
14,1
IJSIM

Table IV.

self-scan option
Reasons for not using
What was not liked about Why self-scan option may not be used
Why self-scan option will not be used self-scan option (In-store respondents in the future (In-store respondents who
regularly (Respondents at self-scanners) n=4 who had used self-scan option) n = 17 had not used self-scan option) n = 18

Likes interaction with employee 2 Difficult to use 6 Lack of familiarity 7

Habit (using regular checkout) 1 Likes interaction with employee 4 Likes interaction with employee 4
Need incentive for self-service 1 Customer has a right to service 3 Customer has a right to service 6
Time pressure 1 Too much effort 3 Price should be lower for self-service 3
Dislikes automation 1 Process problems 3 Habit (using regular checkout) 2
Lack of accessibility 1 Computer voice annoying 2 Difficult to use 2
Self-scan is impersonal 1 Price should be lower for 2 Not accurate 1
self-service Lack of accessibility 1
Slow 2 Relationship with employees 1
Dislikes automation 1 Social experience 1
Not accurate 1
Inconvenient 1
Lack of familiarity 1
Long lines at self-scan 1
Not enough time saved 1
Limit on number of products to 1
be scanned
Prefers traditional check-out 1
method
Type of products 1
Uncomfortable with using 1
 Under what situations would
Under what situations would self-scan self-scan option be used (In-store Under what situations would self-scan
option be used (Respondents at respondents who had used option be used (In-store respondents
self-scanners) n=6 self-scan option) n = 16 who had not used self-scan option) n = 16

If few products 4 If few products 16 If few products 8

If line at regular checkout 2 If line at regular checkout 7 If line at regular checkout 4
If child wants to use 1 If more experience 2 If in a hurry 3
If time available to use 2 If more experience 3
If in a hurry 2 If time available to use 2
If motivated 2 If purchasing certain types of products 2
If purchasing certain types of 1 If option is easy to use 1
products If faster 1
If line at self-scan is short 1 If assistance provided in learning to 1
To avoid interaction with 1 use
employee
If assistance provided in 1
learning to use
If bank did not charge for cash 1
withdrawals
If correct change 1
If food stamps can be used 1

option
Situations influencing
Table V.

the use of the self-scan

behavior
motivation and
Consumer

77
IJSIM more experience in using this option. Other concerns related to the type of
14,1 products they were buying or the payment options related to the self-scan.
Shopping preferences for other technology-based self-service options. For both
groups of respondents (in-store and at the self-scanners), shopping preferences
were determined for other technology-based self-service options versus
traditional alternatives. As mentioned, these included:
78 shopping from home vs shopping at the store;
Internet shopping vs telephone shopping;
using touch-tone dialing vs talking to a person when telephone
shopping;
using a computer touch screen in the store vs ordering verbally in the
store; and
using an ATM vs using a teller.
Content analysis determined the reasons for these preferences.
Irrespective of the group, the majority of consumers preferred shopping
at the store to shopping from home (see Tables I and VI). Table VI compares
reasons consumers gave for shopping from home vs shopping at the store.
The reasons supporting an option are marked as positive (+) and those
against the alternative option are marked as negative (–). A reason related
to a particular situation is marked as ``depends’’ with a (D). The most
important reason for shopping from home is convenience, followed by ease
of use and ease of shopping. In contrast, the most important reason for
shopping at the store is the ability to see products, followed by verification
of items, ability to touch products, and social experience. Other reasons
mentioned frequently for shopping at the store are avoiding returns, ease of
return, and convenience, which are parallel to the reasons for shopping from
home. Thus, consumers who prefer a particular option think it is faster,
offers more control, is more reliable (accurate), easier to use, and more
enjoyable than the alternative option. Some reasons against shopping from
home are similar to reasons for avoiding self-scanning (e.g. too much effort
or discomfort).
In-store respondents clearly preferred telephone shopping, whereas
respondents at the self-scanners preferred Internet shopping (see Tables I and
VII). Table VII compares reasons consumers gave for Internet shopping vs
telephone shopping. Ironically, the most important reason for preferring
Internet shopping is ``the ability to see products,’’ also the most important
reason for shopping at the store. This is followed by ease of use, control, and
security. The most important reason against telephone shopping is to avoid
interaction with employees, thus supporting hypothesis H5a. In contrast, the
most important reasons for preferring telephone shopping are ``likes to interact
with employees,’’ and ``employees are friendly, helpful, etc.’’ thus supporting
hypothesis H5b. This is followed by security, familiarity, ease of use,
 At self- At self-
In-store scanners In-store scanners
Category Prefer home shopping (n = 15) (n = 7) Category Prefer store shopping (n = 56) (n = 30)

+ Convenient 12 4 + Ability to see products 32 19

+ Easy to use 3 + Verification of items 15 9
+ Ease of shopping 3 + Ability to touch products 9
+ Ability to see products 1 1 + Social experience 5 2
(catalogue/pictures) + Avoid returning purchases 3
+ Accurate 1 + Ease of return 2
+ Control 1 + Convenient 3 1
+ Fast 1 1 + Accurate 2
+ Novelty 1 + Familiarity 2 1
+ Enjoyable 1 + Ability to see prices 1
+ Ability to try product 1
+ Easy to use 1
+ Enjoy shopping 1
+ Freshness of products 1
+ Habit 1 1
+ Likes to interact with employee 1
+ Employees are friendly, helpful, etc. 1
+ Method of payment– cash 1
+ More product variety 1
+ Control 1
+ No waiting on delivery 1
+ Wider selection 1 1
+ Security 1
+ Cheaper 1
(continued)

Table VI.

home vs store
Shopping
preference:
behavior
motivation and
Consumer

79
 80
14,1
IJSIM

Table VI.
At self- At self-
In-store scanners In-store scanners
Category Prefer home shopping (n = 15) (n = 7) Category Prefer store shopping (n = 56) (n = 30)

– Dislikes store shopping 3 2 – Cost of shipping with Internet 1

– Avoid crowds 2 – Lack of capability with Internet (no credit 1
card)
– Avoid interaction with 1 – Lack of product visibility (Internet) 1
employees Too much effort (home shopping) 1
– Time pressure 1 – Unfavorable experience (Internet) 1
– Dislikes shopping from home or work 1
– Lack of security (Internet) 1
– Not comfortable (with home shopping) 1
– Doesn’t think about online shopping 1
D Type of product 2 1 D Time available 1
D Only known providers 1 D Type of product 1 1
with high quality
Total 28 15 Total 90 45
 At self- At self-
In-store scanners In-store scanners
Category Prefer Internet shopping (n = 18) (n = 19) Category Prefer telephone shopping (n = 52) (n = 16)

+ Ability to see products 4 11 + Likes to interact with employee 14 1

+ Easy to use 3 3 + Employees are friendly, helpful, etc. 2
+ Control 2 1 + Security 6 1
+ Security 2 2 + Familiarity 5 2
+ Fast 1 2 + Assistance – questions 4
+ Accurate 1 + Easy to use 3 3
+ Confirmation 1 + Fast 3 3
+ Convenient 1 1 + Ability to see products (catalogue) 3
+ No waiting 1 + Habit 3
+ Familiarity with Internet 2 + Convenient 2
+ Wider selection 1 + Assistance – information 1
+ Enjoyable 1 + No waiting 1
+ Like it 1 + Accurate 1
+ Easier to compare prices 1
– Avoid interaction with 6 3 – Lack of accessibility (Internet/computer) 11 5
employee
– Employees are rude, 1 – Lack of familiarity (Internet) 11 4
unhelpful, etc.
– Dislikes using phone 1 – Lack of trust (Internet) 1 2
– Dislikes recordings 1 – Difficult to use (computer) 1 1
– Physical limitation (eyesight) 1
D If everything were 1
available on-line
Total 25 30 Total 71 25
behavior

Internet shopping vs
Table VII.

telephone shopping
motivation and
Consumer

81

Shopping preference:
IJSIM convenience, etc. showing that preference for an option makes consumers think
14,1 it does better on the same attributes. Finally, lack of familiarity and lack of
accessibility are major reasons against shopping on the Internet.
Irrespective of the group, the majority of consumers preferred speaking
to a person when telephone shopping to using touch-tone dialing (see Tables
I and VIII). Table VIII compares reasons consumers gave for using touch-
82 tone dialing vs speaking to a person when telephone shopping. The most
important reasons for using touch-tone dialing are that it is easy to use and
fast, offering indirect sypport for hypothesis H6a. The only reason given
against speaking to a person when telephone shopping is to avoid
interaction with employees, thus supporting hypothesis H5a. In contrast,
the most important reason for talking to a person when telephone shopping
is ``likes to interact with employee.’’ This together with ``employees are
friendly, helpful, etc.’’ and several reasons related to assistance by
employees on the telephone strongly support hypothesis H5b. Again,
interestingly, consumers who prefer this option see it as fast, easy to use,
and offering control, and also see the touch-tone option as difficult to use,
inconvenient, slow, annoying, impersonal, and not enjoyable. Their other
reasons against using touch-tone dialing point to unfavorable attitudes
toward technology (e.g. dislikes automation, dislikes using machines, do not
trust technology, not comfortable with technology, etc.), thus supporting
hypothesis H6b.
Irrespective of the group, the majority of consumers preferred ordering
verbally to an employee to using a touch screen in a store, although a
higher percentage of respondents using the self-scan preferred the touch
screen option (see Tables I and IX). Table IX compares reasons consumers
gave for using a touch screen in the store vs ordering verbally to an
employee in the store. The most important reasons for using a touch screen
in the store are that it is easy to use, fast, convenient, and accurate. These
reasons along with reasons such as superiority, novelty, familiarity,
and enjoyable are indicative of a favorable attitude toward using
technology, thus supporting hypothesis H6a. The main reason given
against ordering verbally is to avoid interaction with employees, thus
supporting hypothesis H5a. In contrast, the most important reasons
for ordering verbally in a store are to interact with employees, to
get assistance, and employees are friendly, helpful, etc., thus
supporting hypothesis H5b. Again, consumers who prefer this option see it
as fast, easy to use, and offering control, and also see the touch screen
option as difficult to use, inflexible, slow, inconvenient, and involving
too much effort. These reasons along with other reasons against using
a touch screen (e.g. dislikes automation, dislikes using machines, do
not trust technology, not comfortable with technology, etc.)
point to unfavorable attitudes toward technology, thus supporting
hypothesis H6b.
 At self- At self-
In-store scanners In-store scanners
Category Prefer touch-tone dialing (n = 7) (n = 6) Category Prefer speaking to a person (n = 61) (n = 29)

+ Easy to use 5 2 + Likes to interact with employee 21 1

+ Fast 3 2 + Assistance – answer questions 21 3
+ Convenient 1 + Assistance - information 9 1
+ Likes touch-tone 1 + Employees are friendly, helpful, etc. 8
+ Accurate 1 + Accurate 6 6
+ Assistance – product 1 + Assistance – handle problems 6 3
knowledge + Fast 5 2
+ Easy to use 5 3
+ Control 3
+ No waiting 1 2
+ Familiarity 1
+ People perform better 1
+ Trust 1
– Avoid interaction with 2 1 – Dislikes automation 3 1
employee – Dislikes using machines 2
– Lack of familiarity with technology 2 2
– Too much effort 2
– Difficult to use (touch-tone) 2
– Lack of control (touch-tone) 2
– Annoying (touch-tone) 1
– Inconvenient to use (touch-tone) 1 1
(continued)

dialing vs speaking to
behavior

a person when
using touch-tone
Table VIII.

telephone shopping
motivation and
Consumer

83

Shopping preference:
 84
14,1
IJSIM

Table VIII.
At self- At self-
In-store scanners In-store scanners
Category Prefer touch-tone dialing (n = 7) (n = 6) Category Prefer speaking to a person (n = 61) (n = 29)

– Slow (touch-tone) 1 1
– Impersonal (touch-tone) 1
– Process problems (touch-tone) 1
– Don’t trust technology 1
– Not enjoyable (touch-tone) 1 1
– Not comfortable with technology 1
– Physical limitation 1
D If order is simple 1
D Type of product 1
Total 13 8 Total 93 44
 At self- At self-
Prefer using touch In-store scanners In-store scanners
Category screen (n = 17) (n = 14) Category Prefer ordering verbally (n = 81) (n = 34)

+ Easy to use 7 6 + Likes to interact with employee 29 4

+ Fast 5 7 + Assistance – questions 25 9
+ Convenient 4 2 + Employees are friendly, helpful, etc. 2 10
+ Accurate 3 2 + Accurate 7 3
+ No waiting 2 1 + Easy to use 6 1
+ Familiarity 1 2 + Assistance – information 4 3
+ Less effort 1 + Assistance – product knowledge 3
+ Novelty 1 + Familiarity 3 4
+ Superiority 1 + Fast 2 1
+ Enjoyable 1 + Assistance – handle problems 2 2
+ Social experience 2 1
+ Control 1
+ Personalized service 1
+ Prefers personal assistance 1
+ Relationship with personnel 1
+ Trust 1
+ Verification of order 1
+ Flexibility with employees 1
+ Likes to be served 1
+ Customer has right to service 1
– Avoid interaction with 3 4 – Dislikes automation 5 2
employees – Possible service failure 4
– Avoid bothering 1 – Dislikes using machines 3
employees – Too much effort 2
– Concern about accuracy 1
– Difficult to use (touch screen) 1 1
(continued)

using touch screen in

behavior

Table IX.

verbally in store
motivation and
Consumer

store vs ordering
85

Shopping preference:
 86
14,1
IJSIM

Table IX.
At self- At self-
Prefer using touch In-store scanners In-store scanners
Category screen (n = 17) (n = 14) Category Prefer ordering verbally (n = 81) (n = 34)

– Inflexibility of computers 1
– Lack of experience (touch screen) 1
– Lack of familiarity (touch screen) 1
– Lack of familiarity with technology 1
– Technophobic 1
– Slow (touch screen) 1
– Inconvenient to use (touch screen) 1
D Type of product 1 D Type of product 1
D Type of store (grocery) 1 D Easy to use – if large number of products 1
D Easy to use – if no lines 1
D Type of store (non-grocery) 1
Total 30 26 Total 110 52
In-store respondents were equally divided as to preference for using ATMs Consumer
and bank tellers, whereas respondents at the self-scanners clearly preferred motivation and
using ATMs (see Tables I and X). Table X compares reasons consumers behavior
gave for using an ATM vs using a teller. The most important reasons for
using an ATM are that it is fast, convenient, accessible, and easy to use,
again offering indirect support for hypothesis H6a. The main reason given
against using tellers is to avoid interaction with employees, which together 87
with the reason that employees were rude, unhelpful, etc., offers support for
hypothesis H5a. In contrast, the most important reasons for using a teller
are liking to interact with employees (in-store respondents) and employees
are friendly, helpful, etc. (respondents at self-scanners), thus supporting
hypothesis H5b. Other reasons such as assistance and social experience also
offer support for H5b. The main reasons against using ATMs (e.g. dislikes
automation, dislikes using machines, unfavorable experiences, etc.) indicate
an unfavorable attitude toward using technology, thus supporting
hypothesis H6b.

Discussion
A main research issue in the study was to determine consumer reasons for
using or avoiding self-scanning checkouts in retail stores. Our quantitative
analysis showed that control, reliability, ease of use, and enjoyment were
indeed important to consumers in using the self-scanning option. Although
speed was not differentiated among consumers who planned to use this option
regularly and those who did not, mean values of attribute perceptions showed
clearly that self-scanning was very much viewed as a fast option by consumers
who had tried it.
Our content analysis supported these findings but in addition showed the
predominance of speed as a reason for liking the self-scan option. The other
reasons tested in the quantitative analysis (i.e. control, reliability, ease of use,
and enjoyment) were also mentioned, but less frequently. One factor, not
included in the theroretical framework, but frequently mentioned by
respondents, was convenience. Future studies will need to determine whether
convenience is a separate construct, or whether it overlaps with speed and/or
ease of use.
In addition to reasons related to attributes, consumers planned to use this
option to avoid interaction with employees and/or because they had favorable
attitudes toward using technology in general. So far, this is good news for
supermarket chains, as well as for other retailers considering this option;
consumers who prefer self-scanning see it as offering many benefits and they
seem inclined to use it whenever possible.
With regard to reasons for avoidance of self-scanning, we found that many
consumers truly like to interact with employees and the self-scanning checkout
cannot fulfill this need. These consumers did have unfavorable attitudes
toward using technology in general and the stores would have little control
 88

teller
14,1
IJSIM

Table X.
Banking preference:
using ATM vs using
At self- At self-
In-store scanners In-store scanners
Category Prefer using ATM (n = 49) (n = 35) Category Prefer using teller (n = 51) (n = 12)

+ Fast 24 26 + Likes to interact with employees 20 1

+ Convenient 17 13 + Accurate 9
+ Accessibility – after hours 11 4 + Assistance – questions 6 2
+ Easy to use 6 8 + Security 5
+ Familiarity 3 1 + Easy to use 4 1
+ Accessibility – multiple locations 3 1 + Familiarity 3
+ No waiting 2 3 + Habit 3
+ Confirmation 2 + Social experience 2 2
+ Less effort 1 1 + Assistance-handle problems 2
+ Does not need assistance 1 + Confirmation of deposit 2
+ Accurate 1 + Fast 2
+ Enjoyable 1 + Cost 1
+ Habit 1 + Greater satisfaction 1
+ Security 1 + Prefers traditional method 1
+ Trust 1
+ Employees are friendly, helpful, etc. 3
+ Flexibility with employees 1
+ Versatility – other services 1
– Avoid interaction with employee 2 4 – Unfavorable experience (ATM) 3 1
– Employees are rude, unhelpful, etc. 1 2 – Dislikes automation 3 1
– Time pressure 1 – Dislikes using machines 2
– Lack of experience with teller 1 – Dislikes ATMs 1
– Cost associated with ATM 1
– Lack of accessibility (ATM down) 1 1
– Lack of familiarity (ATM) 1 1
– Lack of trust (ATM) 1
– Inconvenient to use (ATM) 1
D For withdrawals 1 1 D Purpose of transaction 2
Total 76 68 Total 75 18
over changing such attitudes, especially in the short term. Also, earlier we had Consumer
raised the issue of effort and wondered whether this might be why self- motivation and
scanning checkouts seem to face some resistance, in contrast to ATMs which behavior
are so widely accepted. Our study did show that consumers who disliked self-
scanning expressed a sense of having a right to be served and that self-
scanning involved too much effort. Given these findings, it would not be
prudent to switch completely to self-scanning as some grocery stores in 89
Sweden have done. Retailers that do this would stand to lose a large part of
their clientele, especially in regions with healthy competition in the particular
service industry. The best scenario is to offer both options and gradually win
over more and more consumers to the self-scanning checkout. The good news is
that there is a sizeable segment that does prefer this option so as to make the
investment in it economically viable for grocery chains as well as for other
retailers.
Another research issue was to explore the possible influence of demographic
factors. Our quantitative analysis showed that demographic factors such as
age, gender, and education, had no influence on the use of self-scanning. The
only demographic factor that was different was that those who used self-
scanning had greater access to the Internet. The implication for grocery stores
or other retailers planning to offer self-scanning is that as Internet access is
widened – as it will undoubtedly – consumer acceptance and use of the
technology-based self-service options within stores should increase as well.
With regard to the effect of demographic influences on other technology-
based self-service options, our study did find that younger people were more
likely to prefer using ATMs to using tellers and younger males more likely to
prefer Internet shopping to telephone shopping. Greater Internet access was
also related to both these preferences. Practitioners, especially those who hope
to increase Internet sales significantly, should ensure that younger males know
and approve of their Websites. A related implication for Internet marketers is
to increase access to and familiarity with the Internet for a wider audience,
especially as inaccessibility and unfamiliarity were two major reasons cited by
consumers against Internet shopping.
Yet another research objective was to investigate the influence of situational
factors on the evaluation and use of self-scanning checkouts. Our quantitative
analysis found only one situational factor to be relevant. Under crowded
conditions, consumers viewed the self-scan as faster than under normal
conditions. This is an important finding for practitioners; at crowded times, the
presence of a self-scanning checkout will assure its good utilization and help
pay toward the investment in this technology.
Our content analysis revealed many possible situations where consumers
would use self-scanning, all of which have managerial implications. The most
important situation cited was the number of products purchased. Currently,
many stores set a maximum number of products that can be bought through
the self-scanning checkout. Although NCR reports that 60 percent of US
shoppers have fewer than 12 items at checkout (Solomon, 1997), the restriction
IJSIM keeps consumers from using the self-scanning checkout when they have many
14,1 items to purchase. Respondents indicated that they would be likely to use self-
scanning when they have fewer products than the maximum allowed. By
removing this constraint, or raising the number of items allowed, grocery
stores should be able to increase the use of self-scanning. This is an implication
that retailers considering offering self-scanning checkouts should bear in mind
90 when considering alternative systems with or without such constraints.
Other reasons cited showed a willingness to try self-scanning if there is
assistance in showing the customer how self-scanners work and if payment
options preferred by the customer apply. Clearly, managers have an
opportunity here to increase utilization of the self-scanning checkout by having
an employee stand by and offer to actively help consumers learn how to use it
(as banks did years ago when the ATM was first introduced). Utilization may
also be increased by adopting systems that are flexible in allowing consumers
to pay for their purchases using a variety of methods, as is possible through the
traditional checkout. These implications apply to grocery chains as well as to
retailers in general.
A fourth research objective was to compare the use of self-scanning
checkouts with shopping preferences for other types of technology-based self-
service. We found that consumers who use self-scanning prefer Internet
shopping to telephone shopping and also prefer using ATMs to using tellers.
This is understandable because the reasons driving preference for these
different technology-based self-service options are similar: fast, convenient,
accessible, reliable, avoiding interaction with an employee, and so on. The
broad implication for practitioners is that for technologies that work well,
consumers with favorable attitudes toward using technology in general will
transfer those attitudes to a variety of technology-based self-service options.
Other implications relate to the set of attributes that consumers found
important in each case. Practitioners should ensure that their particular
technology-based self-service option offers the specific attributes consumers
seek from that service.
In contrast, most consumers irrespective of their use of self-scanning prefer
shopping at the store vs shopping at home, speaking to a person when
telephone shopping vs using touch-tone dialing, and ordering verbally with an
employee vs using a touch screen in a store. It is interesting that whereas some
consumers prefer the Internet to telephone shopping, they still prefer to shop at
the store to shopping from home. The ability to see and touch products and to
verify items is extremely important to a majority of consumers, and ease of
return is also a consideration. Until Websites make it easy for consumers to
view products and Internet marketers simplify returns, this preference is
unlikely to change.
It is not surprising that preference for talking to a person when telephone
shopping to using touch-tone dialing is almost universal. Given the frustration
brought about by interminable directions on automated telephone systems,
most consumers are rightly reluctant to use these systems. To change attitudes
toward this particular technology-based self-service, automated touch-tone Consumer
systems need enormous improvement in terms of speed, information as to motivation and
waiting time, and easy options to connect with a person or to leave a voice behavior
message.
It is not clear why most consumers prefer ordering verbally in a store to
using a touch screen. We had expected that those who use self-scanning would
prefer using a touch screen as well. Although the percentage was higher in this 91
group as expected, the difference across groups was not statistically
significant. Perhaps, this option is so new that respondents were not assured
that it would be faster than the verbal option. Unlike the ATM, touch screens in
retail stores vary widely in terms of user-friendliness, and respondents chose to
be conservative in answering a question where they could not be sure of
attributes as they could with the widely familiar ATM. The implication for
practitioners is to design better touch screens in terms of flexibility and user-
friendliness so that consumers will eventually be as comfortable with using
them as they are with using ATMs.
Having discussed managerial implications along with the detailed
discussion of our findings, it remains to acknowledge limitations of the study,
compare efficacies of different research methodologies, and suggest directions
for future research. In terms of limitations, our sample was relatively small and
we collected information from only one store; to that extent, the results may not
be widely generalizable. The small sample also precluded the use of structural
equations, but this was relevant for only a small part of our overall research
plan. The sample size was more than sufficient for thorough content analysis
as well as for conducting t-tests, ANOVAs, and nonparametric statistical tests.
A second limitation is that consumer time constraints (especially while grocery
shopping) prevented us from including many more questions of interest in our
surveys. Still, we were able to capture much useful information in relatively
short, but carefully structured interviews.
Having used a variety of research and analytical methods in this study, our
overall conclusion not surprisingly is that where possible, a combination of
methods yields the most information. For example, our quantitative analysis
supported past theory with respect to attributes important for using self-
scanning. Our content analysis corroborated these results but the frequencies
for the reasons cited gave a clearer indication of the most important reasons
motivating consumers to use or avoid the self-scanning checkout. In addition,
where theory was lacking, we were able to determine through content analysis
that consumers who avoided self-scanning perceived the traditional checkout
as performing better on the same attributes that were important for using self-
scanning checkouts. We had conjectured that this might be one of two
alternatives. The other possibility was that consumers may think the self-scan
does better on some of these attributes, but they choose the traditional checkout
in spite of this in order to interact with employees. The findings showed that
consumers who preferred the traditional checkout viewed it as performing
better on all the same attributes and also liked to interact with employees. The
IJSIM two sets of reasons together form a strong basis for their choice of the
14,1 traditional checkout, suggesting to managers that they need to offer both
options for the foreseeable future.
Our content analysis also revealed that consumers who used self-scanning
had favorable attitudes toward using technology in general and wanted to
avoid interaction with employees. This was also true for consumers who
92 preferred Internet shopping, using touch-tone dialing, touch screens, or ATMs.
In contrast, our content analysis revealed that consumers who avoided self-
scanning had unfavorable attitudes toward using technology in general, and as
mentioned earlier, wanted to interact with employees. This was also true for
consumers who preferred telephone shopping, speaking to a person, ordering
verbally in a store, or using a teller. Certainly, these findings could have been
verified through quantitative analysis as in Dabholkar’s (1996) study on touch
screens; however, it would have considerably lengthened the questionnaire to
capture items measuring all of these constructs for the various contexts. In
addition, support from a different research approach for these hypotheses, on
reasons for using and avoiding self-scanning checkouts as well as several other
technology-based self-service options, advances services marketing theory
even further.
In addition, our content analysis revealed that attributes similar to those
cited for using or avoiding self-scanning (e.g. speed, control, enjoyment)
motivated the use or avoidance of various technology-based self-service
options. Again, to verify this through quantitative analysis would have been
cumbersome and close to impossible in a field setting. Our research approach in
this case allowed us to garner huge amounts of relevant information in an
efficient way. The content analysis also revealed attributes not included in
previous models, and it is up to future research to determine rigorously if these
attributes are unique constructs or if there is overlap.
But content analysis could not tell us much about the influence of
demographic factors. In contrast, our quantitative findings showed that
demographic factors (other than Internet access) were not relevant for using or
avoiding self-scanning. In most cases, however, findings from content analysis
and quantitative methods supported each other as discussed earlier. Another
case in point is that eye-balling frequencies for shopping preferences (Table I)
revealed which ones were different for the two main respondent groups, but
quantitative analysis offered statistical support for this assumption. As for
situational factors influencing the use of self-scanning, quantitative analysis
supported only one factor (crowding) to be significant whereas content analysis
extracted a number of new ideas that managers could work on to possibly
improve utilization of self-scanning. Again, the two methods together allowed
us to confirm hypotheses based on theory as well as to uncover motivational
and behavioral patterns and raise new issues for managers to consider and for
future researchers to investigate further.
Based on our results, we recommend a combination of research methods,
including content analysis and different types of quantitative testing, for future
research on technology-based self-service. In addition, we offer the following Consumer
suggestions for future studies. Studies similar to ours but conducted for other motivation and
contexts where new technology-based self-service options are being proposed behavior
or tested would be very useful to practitioners in that industry. The attributes
extracted from our content analysis could be measured through surveys in
future research to allow statistical testing of their relative significance for a
variety of contexts offering technology-based self-service. Situational factors 93
uncovered in our content analysis could be controlled and tested in future
studies with lab or field settings. Finally, future research could test a
combination of technology-based ``self-service’’ with varying degrees of
interaction with service employees in a variety of contexts. The idea would be
to gauge the viability of such combinations in an attempt to win over those
consumers who like to interact with service employees as well as those who
have somewhat unfavorable attitudes toward using technology entirely on
their own.

References
Anselmsson, J. (2001), ```Customer-perceived service quality and technology-based self-service’’,
doctoral dissertation, Lund Business Press, Lund University, Lund.
Bateson, J.E.G. (1985), ``Self-service consumer: an exploratory study’’, Journal of Retailing, Vol. 61
No. 3, pp. 49-76.
Blumberg, D.F. (1994), ``Strategies for improving field service operations productivity and
quality’’, The Service Industries Journal, Vol. 14 No. 2, pp. 262-77.
Bobbitt, L.M. and Dabholkar, P.A. (2001), ``Integrating attitudinal theories to understand and
predict use of technology-based self-service: the Internet as an illustration’’, International
Journal of Service Industry Management, Vol. 12 No. 5, pp. 423-50.
Bowden, B. (2002), ``Customers help check out new technology’’, Arkansas Business, Vol. 19 No. 5,
pp. 15-8.
Chain Store Age (2002), ``Time flies when you’re scanning for fun’’, Chain Store Age, Vol. 76 No. 2,
pp. 2C-3C.
Childers, T.L., Christopher, C.L., Peck, J. and Carson, S. (2001), ``Hedonic and utilitarian
motivations for online retail shopping behavior’’, Journal of Retailing, Vol. 77, pp. 511-35.
Cowles, D. and Crosby, L.A. (1990), ``Consumer acceptance of interactive media’’, The Service
Industries Journal, Vol. 10 No. 3, pp. 521-40.
Dabholkar, P.A. (1992), ``The role of prior behavior and category-based affect in on-site service
encounters’’, in Sherry, J.F. and Sternthal, B. (Eds), Diversity in Consumer Behavior,
Vol. XIX, Association for Consumer Research, Provo, UT, pp. 563-9.
Dabholkar, P.A. (1994a), ``Technology-based service delivery: a classification scheme for
developing marketing strategies’’, in Swartz, T.A., Bowen, D.E. and Brown, S.W. (Eds),
Advances in Services Marketing and Management, Vol. 3, JAI Press, Greenwich, CT,
pp. 241-71.
Dabholkar, P.A. (1994b), ``Incorporating choice into an attitudinal framework: analyzing models
of mental comparison processes’’, Journal of Consumer Research, Vol. 21, pp. 100-18.
Dabholkar, P.A. (1996), ``Consumer evaluations of new technology-based self-service options: an
investigation of alternative models of service quality’’, International Journal of Research in
Marketing, Vol. 13 No. 1, pp. 29-51.
IJSIM Dabholkar, P.A. (2000), ``Technology in service delivery: implications for self-service and service
support’’, in Swartz, T.A. and Iacobucci, D. (Eds), Handbook of Services Marketing and
14,1 Management, Sage Publications, New York, NY, pp. 103-10.
Dabholkar, P.A. and Bagozzi, R.P. (2002), ``An attitudinal model of technology-based self-service:
moderating effects of consumer traits and situational factors’’, Journal of the Academy of
Marketing Science, Vol. 30 No. 3, pp. 184-201.
94 Davis, F.D., Bagozzi, R.P. and Warshaw, P.R. (1989), ``User acceptance of cmputer technology: a
comparison of two theoretical models’’, Management Science, Vol. 35 No. 8, pp. 982-1003.
Davis, F.D., Bagozzi, R.P. and Warshaw, P.R. (1992), ``Extrinsic and intrinsic motivation to use
computers in the workplace’’, Journal of Applied Social Psychology, Vol. 22 No. 14,
pp. 1109-30.
Dickerson, M.D. and Gentry, J.W. (1983), ``Characteristics of adopters and non-adopters of home
computers’’, Journal of Consumer Research, Vol. 10, pp. 225-35.
Discount Store News (1998a), ``Meijer to adopt self-checkout’’, Discount Store News, Vol. 37 No. 18,
pp. 5-6.
Discount Store News (1998b), ``Self-checkout systems add `on-line’ efficiency’’, Discount Store
News, Vol. 37 No. 11, pp. 70-1.
Evans, K. and Brown, S.W. (1988), ``Strategic options for service delivery systems’’, in
Ingene, C.A. and Frazier, G.L. (Eds), Proceedings of the AMA Summer Educators’
Conference, American Marketing Association, Chicago, IL, pp. 207-12.
Forster, J. (2002), ``Hungry for convenience’’, Business Week, Industry/Technology ed., No. 3765,
pp. 120-3.
Gatignon, H. and Robertson, T.S. (1985), ``A propositional inventory for new diffusion research’’,
Journal of Consumer Research, Vol. 11, March, pp. 849-67.
Grant, L. (2001), ``Scan-it-yourself catching on in supermarkets’’, USA Today, 7 June, pp. 1-6,
available at: www.usatoday.com
Hennessy, T. (1998), ``Taking control’’, Progressive Grocer, December, pp. 83-6.
Hoffman, D.L. and Novak, T.P. (1996), ``Marketing in hypermedia computer-mediated
environments: conceptual foundations’’, Journal of Marketing, Vol. 60, pp. 50-68.
Hunt, J. (1998), ``NCR ready for global rollout’’, Grocer, Vol. 221 No. 7361, p. 19.
Jöreskog, K.G. and Sörbom, D. (1993), LISREL8 User’s Reference Guide, Scientific Software,
Chicago, IL.
Korgaonkar, P. and Moschis, G.P. (1987), ``Consumer adoption of videotex services’’, Journal of
Direct Marketing, Vol. 1 No. 4, pp. 63-71.
Ledingham, J.A. (1984), ``Are consumers ready for the information age?’’, Journal of Advertising
Research, Vol. 24 No. 4, pp. 31-7.
Lovelock, C.H. (1995), ``Technology: servant or master in the delivery of services?’’ in Swartz,
T.A., Bowen, D.E. and Brown, S.W. (Eds), Advances in Services Marketing and
Management, Vol. 4, JAI Press Inc., Greenwich, CT, pp. 63-90.
McDonald, S.B. (2002), ``Retailers get to checkout PSC’s new scanner’’, Knight Ridder Tribune
Business News, January 15, p. 1.
McMellon, C.A., Schiffman, L.G. and Sherman, E. (1997), ``Consuming cyberseniors: some
personal and situational characteristics that influence their on-line behavior’’, Advances in
Consumer Research, Vol. 24, pp. 517-21.
Meuter, M. and Bitner, M.J. (1998), ``Self-service technologies: extending service frameworks and Consumer
identifying issues for research’’, in Grewal, D. and Pechman, C. (Eds), Marketing Theory
and Applications, American Marketing Association, Chicago, IL, pp. 12-9. motivation and
Meuter, M., Ostrom, A.L., Roundtree, R.I. and Bitner, M.J. (2000), ``Self-service technologies: behavior
understanding consumer satisfaction with technology-based service encounters’’, Journal
of Marketing, Vol. 64 No. 3, pp. 50-64.
Prendergast, G.P. and Marr, N.E. (1994), ``Disenchantment discontinuance in the diffusion of 95
technologies in the service industry: a case study in retail banking’’, Journal of
International Consumer Marketing, Vol. 7 No. 2, pp. 25-40.
Quinn, J.B. (1996), ``The productivity paradox is false: information technology improves service
performance’’, in Swartz, T.A., Bowen, D.E. and Brown, S.W. (Eds), Advances in Services
Marketing and Management, Vol. 5, JAI Press, Greenwich, CT, pp. 71-84.
Rogers, E.M. (1983), Diffusion of Innovations, The Free Press, New York, NY.
Rohland, P. (2001), ``New service lets supermarket shopper check themselves out’’, Eastern
Pennsylvania Business Journal, Vol. 12 No. 2, pp. 4-5.
Ross, J.R. (1997), ``Scanning to please’’, January, p. 1, available at: www.idsystems.com/reader
Sellers, P. (1990), ``What consumers really want’’, Fortune, 4 June, pp. 58-68.
Solomon, H. (1997), ``Can NCR succeed where others have failed?’’, Computing Canada, Vol. 23
No. 25, pp. 18-9.
Wisely, R. (2002), ``Self-serve checkout lanes on groceries’ leading edge’’, The Detroit News
(Technology), 27 February, pp. 1-2.

Appendix. Measures for attributes and preference (for self-scan)

Speed
The self-scan saves me time.
The self-scan lets me check out quickly.

Control
The self-scan gives me control.
The self-scan lets the customer be in charge.

Reliability
The self-scan is accurate.
The self-scan is reliable.

Ease of use
The self-scan is easy to use.
The self-scan does not take much effort.

Enjoyment
I enjoy using the self-scan.
It is fun to scan the items yourself.

Preference
The self-scan is better than the regular checkout.
I prefer using the self-scan to using the regular checkout.

Source: Adapted from Dabholkar (1996); all items used seven-point Likert scales.