An Updated and Streamlined Technology Parasuraman2014

Journal of Service Research http://jsr.sagepub.
com/
An Updated and Streamlined Technology Readiness Index: TRI 2.0

A. Parasuraman and Charles L. Colby
Journal of Service Research published online 22 June 2014
DOI: 10.1177/1094670514539730
The online version of this article can be found at:

http://jsr.sagepub.com/content/early/2014/06/19/1094670514539730
Published by:
http://www.sagepublications.com
On behalf of:
Center for Excellence in Service, University of Maryland
Additional services and information for Journal of Service Research can be found at:
Email Alerts: http://jsr.sagepub.com/cgi/alerts
Subscriptions: http://jsr.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
>> OnlineFirst Version of Record - Jun 22, 2014
What is This?
Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Article
Journal of Service Research
1-16
An Updated and Streamlined Technology ª The Author(s) 2014
Reprints and permission:
sagepub.com/journalsPermissions.nav
Readiness Index: TRI 2.0 DOI: 10.1177/1094670514539730
jsr.sagepub.com
A. Parasuraman1 and Charles L. Colby2
Abstract
The Technology Readiness Index (TRI), a 36-item scale to measure people’s propensity to embrace and use cutting-edge
technologies, was published in the Journal of Service Research over a decade ago. Researchers have since used it in a variety of
contexts in over two dozen countries. Meanwhile, several revolutionary technologies (mobile commerce, social media, and cloud
computing) that were in their infancy just a decade ago are now pervasive and significantly impacting people’s lives. Based on
insights from extensive experience with the TRI and given the significant changes in the technology landscape, the authors under-
took a two-phase research project to update and streamline the TRI. After providing a brief overview of technology readiness and
the original TRI, this article (a) describes the multiple research stages and analyses that produced TRI 2.0, a 16-item scale; (b)
compares TRI 2.0 with the original TRI in terms of content, structure, and psychometric properties; and (c) demonstrates TRI
2.0’s reliability, validity, and usefulness as a customer segmentation tool. The article concludes with potential applications of TRI
2.0 and directions for future research.
Keywords
technology readiness, technology adoption, technology use, TRI, TR-based segments
Introduction Technology-triggered transformation in services is likely to

accelerate in the future because current technologies are
The Technology Readiness Index (TRI), a 36-item scale to
increasing rapidly in speed, capacity, connectivity, functional-
measure ‘‘technology readiness’’—defined as ‘‘people’s pro- ity, and ease of use, while potentially groundbreaking innova-
pensity to embrace and use new technologies for accomplish-
tions are still nascent. The Consumer Electronics Association
ing goals in home life and at work’’—was published over a
identified five important future technologies likely to signifi-
decade ago (Parasuraman 2000, p. 308). Since then, technology
cantly affect service delivery and consumption: (1) increased
has revolutionized service delivery in virtually every service
linking of physical objects across billions of nodes, (2) driver-
category. The magnitude of change is evident in the growth and
less vehicles, (3) digital health care, (4) robotic technology, and
penetration statistics of foundational technologies. As of 2013,
(5) empowerment of consumers as curators of digital content
2.7 billion people worldwide had Internet access, with global
(Chisholm et al. 2013). These technologies will have major
penetration growing from 7% in 2000 to 39% in 2013 (Brahima implications for service providers, customers, and employees.
2013). Growth in mobile technology is even more prolific,
For instance, the billions of interconnected nodes will enable
with mobile cellular subscriptions worldwide growing from
ultimate service personalization through continually respond-
2.3 billion in 2005 to 6.8 billion in 2013 (Brahima 2013).
ing to real-time information about customers and their environ-
Another example of increasing connectivity is social
ments. Driverless vehicles will free up a substantial block of
media—Facebook had 1 million subscribers worldwide in
time and create a new service delivery channel (the vehicle).
2004, 350 million in 2009, and over 1 billion in 2012
Digital health care will present opportunities wrought by wear-
(Yahoo! Finance 2012).
able devices, robotic aids, telemedicine, and so on. Robots will
The impact of these technologies in the service domain is open a revolutionary frontier that could upset traditional
evident in statistics tracked in the United States since 1999
by the National Technology Readiness Survey (NTRS).1 In
the financial services category, for instance, while just 30%
1
of consumers with Internet access checked bank account University of Miami, Coral Gables, FL, USA
2
information online in 1999, 51% did so in 2004, and 76% Rockbridge Associates, Inc., Great Falls, VA, USA
in 2012. Similar technology-induced growth trends are
Corresponding Author:
evident in other categories such as investing, travel, online A. Parasuraman, University of Miami, PO Box 248027, Coral Gables, FL 33124,
education, e-government, health care and "Customer to Cus- USA.
tomer (C2C)"commerce. Email: parsu@miami.edu

2 Journal of Service Research
customer-employee relationships. And, media service provi- Contemporaneous with and subsequent to the TRI’s develop-
ders must adopt new business models to survive as their audi- ment, other scholars have examined the advantages and draw-
ence takes control of content. backs of new technology-based systems and their implications
For service providers, customers, and employees, the technol- for fostering consumer acceptance. For instance, Hoffman,
ogy revolution has caused tension between the positive aspects Novak, and Peralta (1999) discussed the need and strategies for
of increased value and the negative aspects of having to learn bolstering consumer trust in e-commerce, which was
and develop trust in new methods of doing business. Previously, still nascent and hence cutting edge at that time. Alluding to the
managers in service firms may have been preoccupied with the technology paradoxes uncovered by Mick and Fournier (1998),
challenge of converting a bricks-and-mortar distribution system Bitner (2001) elaborated on the challenges vis-à-vis consumer
into one with electronic, self-service interfaces. But going for- and employee acceptance of technology-based service systems.
ward, as technology revolutionizes services, managers must Meuter et al. (2003, 2005) posited and empirically examined asso-
cope with more complex challenges associated with delivering ciations between consumers’ technology-related characteristics
innovative service experiences, while ensuring that customers (e.g., technology anxiety) and their usage of SSTs.
are receptive to those experiences, and potential adverse effects Since the TRI’s publication, the pace of technological
on employees are minimal. change has accelerated, with the advent of advances such as
Customers also face trade-offs associated with trying to get high-speed Internet access, mobile commerce, social media,
maximum value from technology-based service options without and cloud computing. Against this backdrop, and informed
encountering frustration or failure. To illustrate, a music con- by over 12 years of experience using the TRI, the authors ini-
sumer who subscribes to a service like iTunesTM will have con- tiated the development of an updated and streamlined TRI 2.0.
venient access to a vast array of music and videos, but will To distinguish the original TRI from TRI 2.0, hereafter the for-
have to master the skills of downloading and managing content; mer is referred to as TRI 1.0.
or, a patient needing a lifesaving procedure could forgo travel to The remainder of this article first offers a brief overview of
a hospital, but must learn to trust a remotely controlled robot. the technology readiness construct and the 36-item TRI 1.0. It
Likewise, employees, especially customer-facing employ- then discusses (a) the motivation for developing TRI 2.0 and
ees, must feel confident about and comfortable with new the multiphase process employed in doing so, (b) TRI 2.0’s
technology-based service options; otherwise, their morale and psychometric properties and summary statistics vis-à-vis those
productivity may decline. Understanding employees’ reactions of TRI 1.0, and (c) a TR-based segmentation of consumers
to cutting-edge technologies is as critical as understanding cus- using latent class analysis (Magidson and Vermunt 2003). The
tomers’ reactions. Thus, the importance and practical relevance article concludes with TRI 2.0’s potential applications and
of the Technology Readiness (TR) construct will continue to some future research directions.
grow, commensurate with rapidly evolving technologies.
The Structure of Technology Readiness and

Conceptual Underpinnings and Domain of Technology TRI 1.0
Readiness
Technology readiness represents a gestalt of mental motivators
The original TRI was anchored in literature on adoption of new and inhibitors that collectively determine a person’s predispo-
technologies and people-technology interactions. A brief recap sition to use new technologies (Parasuraman 2000). The con-
of that literature would be helpful. As summarized in Parasura- struct is multifaceted, comprising four dimensions:
man (2000), Mick and Fournier’s (1998) seminal work, based
on extensive qualitative research on peoples’ reactions to tech- Optimism—a positive view of technology and a belief
nology, identified eight technology paradoxes (e.g., freedom/ that it offers people increased control, flexibility, and
enslavement, assimilation/isolation, efficiency/inefficiency), efficiency in their lives.
implying that technology may trigger both positive and nega- Innovativeness—a tendency to be a technology pioneer
tive feelings. Based on Mick and Fournier’s findings, and and thought leader.
insights from previous studies in the context of interactive Discomfort—a perceived lack of control over technol-
media (Cowles and Crosby 1990), teleshopping (Eastlick ogy and a feeling of being overwhelmed by it.
1996), and self-service technologies (SSTs; Dabholkar 1996), Insecurity—distrust of technology, stemming from
Parasuraman argued that the relative dominance of positive and skepticism about its ability to work properly and con-
negative feelings about technology would vary across people cerns about its potential harmful consequences.
and cause corresponding variations in people’s propensity to
embrace and employ new technologies. Other studies (e.g., Of the four dimensions, optimism and innovativeness are
Davis, Bagozzi, and Warshaw 1989) also have identified spe- ‘‘motivators,’’ contributing to TR, whereas discomfort and
cific consumer beliefs and motivations that may enhance insecurity ‘‘inhibitors,’’ detracting from it. Moreover, the four
(e.g., perceived ease of use, fun) or curtail (e.g., perceived risk) dimensions are relatively distinct, meaning that an individual
new technology adoption. can possess different combinations of technology-related traits,

Parasuraman and Colby 3
sometimes leading to a paradoxical state that consists of strong to (a) reassess scale statements referencing contexts that were
motivations tempered by strong inhibitions. no longer innovative, (b) examine and incorporate relevant
TRI 1.0 consists of 36 belief statements, each with a fully implications of a changing technology environment, and
anchored 5-point scale (strongly disagree ¼ 1 to strongly (c) make the instrument more parsimonious.
agree ¼ 5). Of the 36 statements, 10 measure Optimism, 7 An inherent challenge with a scale that measures technology
measure Innovativeness, 10 measure Discomfort, and 9 mea- attitudes is that the technologies themselves change over time, the
sure Insecurity. Thus, TRI 1.0 provides dimension-specific as pace being particularly fast in the realm of service delivery tech-
well as overall measures of TR. nologies. Scale items that refer to specific technologies to make
TR is an individual-level characteristic that does not vary in the statements meaningful and clear to respondents risk losing
the short term nor does it change suddenly in response to a sti- their relevance if the referenced technologies become obsolete
mulus. Higher TR levels are correlated with higher adoption or commonplace. Perhaps because of this, the overall TRI as mea-
rates of cutting-edge technology, more intense usage of tech- sured by the NTRS tended to shift gradually upward over time.
nology, and greater perceived ease in doing so (e.g., see Kuo For example, an abbreviated 10-item version of TRI 1.0 that
2011; Lin and Chang 2011; Massey, Khatri, and Montoya- Rockbridge tracked averaged 100 in 1999, but reached 103 in
Weiss 2007). 2009. Though this change was relatively minor, the trend signaled
that some scale items might be losing their relevance. For exam-
ple, for the item ‘‘I do not consider it safe to do any kind of finan-
The Technology Readiness Research
cial business online,’’ the percentage of NTRS respondents that
Program strongly or somewhat agreed was 58% in 1999, 50% in 2004, and
TRI 1.0’s development was a collaborative effort between the 38% in 2009. Because online financial services were becoming
authors and Rockbridge Associates (a Virginia-based market increasingly common in the United States, the item became less
research company specializing in technology and services relevant and contributed to the TR index’s upward drift. Other
issues). As part of an ongoing research program on TR, the TRI 1.0 items contained words such as ‘‘machine,’’ ‘‘computer,’’
authors have conducted multiyear surveys of the U.S. adult and ‘‘computer program’’ that were becoming outdated.
population (these are the NTRS mentioned previously), derived Another issue was that many formative technologies of today
normative measures of TR from the survey data, and monitored were in their infancy in 1999. Examples include smartphones,
technology behavior trends. Data from the 1999 NTRS were wireless Internet services, social media, home videoconferen-
used in developing, fine-tuning, and validating TRI 1.0 (Para- cing, and cloud applications. In addition, new issues were emer-
suraman 2000). Subsequent NTRS studies included TRI 1.0 as ging in societal conversations about technology, including
well as measures of technology-related behaviors, intentions, concerns about social dependency and distraction. There was a
and preferences. need to ensure that the TR index captured contemporary
To date, the authors have provided academic licenses for technology-related themes to ensure its continued relevance.
TRI 1.0’s use to 127 researchers in 30 countries, where it has often Finally, a commonly expressed concern was that the 36-item index
been translated into local languages. Almost a third (29%) of the was too long. The benefit of having an extensive item battery was that
applications have been in the United States, but other countries it offered psychometrically sound measures of the four distinct TR
with many users include Germany (9%), Malaysia (6%), Turkey dimensions. However, many researchers seeking permission to use the
(6%), United Kingdom (6%), China (5%), India (5%), Brazil scale were only interested in measuring overall TR, as just one con-
(4%), Canada (4%), Philippines (4%), and South Africa (4%). struct in comprehensive multiconstruct frameworks. They therefore
Most of the studies involved business-to-consumer contexts needed survey space to measure other constructs and opted to use the
(41%), but a good number were in business-to-business (30%) concise 10-item version mentioned previously. Consequently, the
and educational (29%) contexts. TRI 1.0’s applications span a question arose as to whether a shorter version of the TR index—one
variety of services, attesting to the proliferating influence of tech- that could measure overall TR as well as its four components without
nology in the service domain. A longitudinal analysis of the aca- sacrificing reliability and validity—could be developed through a rig-
demic license requests suggests that TRI 1.0’s early applications orous scale development process. Recently, Lin and Hsieh (2012)
were in sectors that were early adopters of technology-based ser- derived a 16-item TR scale from the original 36-item scale and empiri-
vice delivery models, including financial services and retailing. cally verified its reliability and validity in contexts involving the use of
The applications then spread to government and nonprofit ser- SSTs. However, their research effort was primarily empirical and
vices, and more recently to health care. focused solely on condensing the original scale by eliminating items.
It did not involve updating the scale by adding new items or modifying
current items to address the changing technology environment and to
Motivation for Developing TRI 2.0 capture emerging technology themes.
The collective experience from over a decade of TRI 1.0’s use
in the NTRS and other studies surfaced some key issues that
provided the impetus for updating the scale. The feedback from
Development of TRI 2.0
other researchers came through a LinkedIn TR forum as well as The cumulative experience of the authors’ and other research-
through personal communications. The issues included a need ers’ use of TRI 1.0 provided preliminary ideas about the types

of changes the instrument needed. To augment and flesh out What is it about you in particular that influences the
those ideas, the authors undertook additional exploratory extent of your technology use? (To identify personal
research. This qualitative phase was followed by an extensive characteristics, such as demographics and lifestyle, that
quantitative research phase to refine TRI 1.0 and assess the could affect technology use)
refined scale’s psychometric soundness.
Forum participants were also given an opportunity to pose
questions about pertinent issues not covered in the discussion.
This research phase produced a rich data set of 317 com-
Qualitative Research Phase
ments (about 5 per respondent). To assimilate the results and
The exploratory phase of TRI 2.0’s development consisted of an identify technology themes, the research team convened a
interactive discussion with consumers using a virtual data collec- brainstorming session that included four analysts who helped
tion forum, OpinionPondTM (Woodall and Colby 2011). The moderate the forum and one of the authors (who monitored the
OpinionPond forum mirrors a social media discussion, allow- forum but did not moderate). Before the brainstorming, the
ing consumers to post comments and respond to others’ com- team reviewed the comments (maintained in a spreadsheet) and
ments in a user-friendly, graphically attractive environment. a memorandum summary prepared by the lead moderator. Dur-
A representative sample of U.S. adults (age 18 and older) was ing the brainstorming, the team identified important themes
recruited from an online panel to participate in the discussion. related to technology drivers, technology inhibitors, technol-
The sample was prescreened and balanced to include a mix of ogy behaviors, and technology correlates.
consumers by gender (half male), age, education (at least a Insights from the exploratory research reaffirmed broad
fourth had no college), employment status (at least half were themes from previous technology readiness research suggesting
employed), and technology readiness (at least one quarter scored that a combination of positive and negative beliefs influences
low based on a short list of 6 items). A total of 61 respondents technology behavior. Themes emerging from the forum discus-
participated in a 1-weeklong discussion. The research team mon- sions mapped on to the four TRI 1.0 dimensions fairly well. For
itored the discussion and posted follow-up questions to seek clar- instance, technology motivators that forum participants dis-
ification and stimulate further discussion when necessary. cussed (e.g., improved quality of life through greater freedom,
A principal objective of the OpinionPond discussion was to control, and mobility; technology’s ability to connect people to
obtain information about motivators and inhibitors underlying their social networks and to the world at large) corresponded
the adoption and use of cutting-edge technologies. Another to themes in the optimism dimension. Technology inhibitors
objective was to identify unique traits that might help differ- emerging from the forum corresponded to themes in both the
entiate consumers who adopt at different stages of a technol- discomfort and insecurity dimensions (e.g., a lack of confidence
ogy’s life cycle. A final objective was to identify what in using technology, risks and perceived costs, concerns with
consumers consider to be ‘‘cutting edge’’ in terms of products, security and privacy). However, one negative theme not appar-
services, and behaviors. Consistent with TR’s definition, the ent in the original research was a concern that technology had
discussion topics addressed both the personal and occupational a potentially dehumanizing effect. This effect manifested itself
spheres of technology adoption and use. in forum participants’ comments about overdependence on tech-
During the weeklong discussion period, respondents were nology, a diminished quality of personal relationships, and a ten-
asked to comment at least twice on several technology-related dency to become distracted (to a point that is potentially
topics to generate relevant information for modifying TRI 1.0, dangerous). Table 1 summarizes common themes derived from
as well as for ensuring that the survey instrument contained an the forum discussions and provides illustrative quotes.
adequate number of questions about (a) personal, demographic, The forum discussions also offered insights about correlates
and occupational characteristics that might be potential corre- and consequences of technology adoption. Participants identi-
lates of technology adoption and use and (b) contemporary fied what they believed were factors that correlated with recep-
tech-oriented behaviors (in order to validate TRI 2.0). Specifi- tiveness to new technologies. Their comments confirmed
cally, the following open-ended questions were posted on the several characteristics already identified in past TR research,
forum and participants were asked to comment on them: including age, occupation, and education. More important to
the next research phase was identifying technologies consu-
What’s cutting-edge technology at home? (To generate mers consider to be ‘‘cutting edge,’’ an area that by definition
examples of technologies for personal use) continually changes. In the personal sphere, forum participants
What’s cutting-edge technology at work? (To generate mentioned new digital technologies, such as smart devices,
examples of technologies for work use) eBooks, wireless connectivity, and video chats. In the occupa-
What motivates you to try new technology? (To uncover tional sphere, they described technologies that potentially
underlying beliefs that contribute to technology change the space and time structure of the workplace, such
adoption and use) as becoming an e-mail-dominant workplace, using portable
What makes you hesitant to try new technology? (To devices, and videoconferencing.
uncover underlying beliefs that inhibit technology adop- After completing the exploratory research, the authors
tion and use) reviewed the 36 items in TRI 1.0 and identified potential

Table 1. Illustrative Comments by Respondents in Qualitative Collectively, these items captured new issues underlying tech-
Research Phase. nology adoption (including freedom to choose locations, distrac-
Theme Quote
tion, impacts on relationships, dependency, and social pressures)
as well as ways consumers viewed technology in their lives, with
Technology motivators more nuance on personal and work influences. Examples of the
Improved quality Saving time is usually what motivates me to newly added items include the following:
of life upgrade or try new technology
Social influence My friends and coworkers have a huge influence Technology adoption motivators
on my decisions to try new technologies
Staying We’re all over the world and we now have the
connected closest thing to living next door . . .
Technology makes me more productive in my personal
Feeling . . . having an iPhone increases my confidence life.
empowered and decreases my stress when I am traveling Communications technology and the Internet help peo-
... ple build stronger relationships.
Being Love the fact that I can take my phone or iTouch
entertained/ and entertain my daughter at a restaurant . . . Technology adoption inhibitors
amused
Technology inhibitors People are too dependent on technology to do things for
Lack of I was afraid to try different things because I felt if them.
confidence I screwed it up, I could not fix it . . . Too much technology distracts people to a point that is
Dependency Why do we feel the need to sit on Facebook and
harmful.
Myspace for hours?
Security and . . . I sometimes am alarmed at spam mail I
Table 2 presents the full battery of 45 TR items included in
safety receive after merely visiting a website
Risk of early . . . if it’s a new product out sure there will be the survey employed in the quantitative phase. The table iden-
adoption bugs and there will be updates for the product tifies 16 items that subsequently comprise TRI 2.0 (these items
and that’s why I’ll wait a bit . . . have new reference labels in the first column; items without
Cost barriers The high cost of acquiring these [technologies] is reference labels in this column were eliminated during the scale
actually very discouraging refinement process). The table’s last column shows the original
reference labels (from Parasuraman 2000) for the 36 TRI 1.0
refinements in the form of (a) wording changes to current scale items. The 9 newly added items are identified as such in the last
items and (b) additional items to capture contemporary themes. column.
The wording changes were intended to update questions to
reflect how consumers describe technology currently and/or
Quantitative Research Phase
to make items less dependent on a specific type of technology The quantitative phase consisted of a mail and an online survey
(that has or will become obsolete). Of the original 36 items, 11 of a representative cross-section of U.S. adults (ages 18 or
were updated and 25 were deemed appropriate and left older). The survey’s core consisted of the 45 TR statements
unchanged. The following are examples of item rewordings: with the same fully anchored 5-point agreement scale used in
TRI 1.0 (strongly/somewhat disagree, neutral, somewhat/
Optimism Statement: strongly agree). The survey also included:
Original: ‘‘I like computer programs that allow me to tai-
lor things to fit my own needs.’’ 33 behavioral items concerning ownership and use of, or
Reworded: ‘‘I like technologies that allow me to tailor intent to purchase, a variety of innovative technology-
things to fit my own needs.’’ based products and services; these items included 13
Discomfort Statement: consumer behaviors, 5 occupational behaviors, 4 beha-
Original: ‘‘There should be caution in replacing important viors that could be personal or work related, and 11
people tasks with technology because new technology social media behaviors;
can break down or get disconnected.’’ 31 items concerning Internet-based activities such as
Reworded: ‘‘There should be caution in replacing impor- e-commerce transactions and/or the use of e-services
tant people tasks with technology because new tech- (e.g., online banking);
nology is not dependable.’’ 9 demographic questions concerning type of commu-
Insecurity Statement: nity, homeownership, children in the household, gender,
Original: ‘‘I worry that information I send over the Inter- ethnicity, age, education, and income;
net will be seen by other people.’’ 7 work-related questions concerning occupation,
Reworded: ‘‘I worry that information I make available employment status, and whether the employment was
over the Internet may be misused by others.’’ technology related.
Apart from wording changes, 9 new items reflecting contem- In the mail survey, the 45 technology readiness state-
porary themes emerging from the exploratory phase were added. ments were randomized twice (i.e., two questionnaire

Table 2. 45 TR Items in the 2012 National Technology Readiness Survey (NTRS).
TRI 2.0 (NTRS TRI 1.0 (NTRS

2012)a Scale Item (2012 Wording) 1999)b
Optimism
OPT1 New technologies contribute to a better quality of life New item
OPT2 Technology gives me more freedom of mobility OPT8
OPT3 Technology gives people more control over their daily lives OPT1
OPT4 Technology makes me more productive in my personal life New item
Technology gives people more freedom to live and work where they please New item
I like technologies that allow me to tailor things to fit my own needs OPT5
Technology makes me more efficient in my occupation OPT6
I like the idea of doing business online because I am not limited to regular business hours OPT3
I feel confident that technology-based systems will follow through with what I instruct them to do OPT10
Products and services that use the newest technologies are much more convenient to use OPT2
I rely on technology to keep up to date on topics I care about New item
Communications technology and the Internet help people build stronger relationships New item
Innovativeness
INN1 Other people come to me for advice on new technologies INN1
INN2 In general, I am among the first in my circle of friends to acquire new technology when it appears INN3
INN3 I can usually figure out new high-tech products and services without help from others INN4
INN4 I keep up with the latest technological developments in my areas of interest INN5
I enjoy the challenge of figuring out high-tech gadgets INN6
I find I have fewer problems than other people in making technology work for me INN7
I prefer to use the most advanced technology available OPT4
I find new technologies to be mentally stimulating OPT7
Learning about technology can be as rewarding as the technology itself OPT9
Discomfort
DIS1 When I get technical support from a provider of a high-tech product or service, I sometimes feel as if I am DIS4
being taken advantage of by someone who knows more than I do
DIS2 Technical support lines are not helpful because they don’t explain things in terms I understand DIS1
DIS3 Sometimes, I think that technology systems are not designed for use by ordinary people DIS2
DIS4 There is no such thing as a manual for a high-tech product or service that’s written in plain language DIS3
It is embarrassing when I have trouble with a high-tech gadget while people are watching DIS6
If you provide information to a technology-based system, you can never be sure it really gets to the right INS9
place
It seems my friends are learning more about the newest technologies than I am INN2
There should be caution in replacing important people tasks with technology because new technology is DIS7
not dependable
I do not consider it safe to do business online INS2
Technology always seems to fail at the worst possible time DIS10
Many new technologies have health or safety risks that are not discovered until after people have used DIS8
them
If I buy a high-tech product or service, I prefer to have the basic model over one with a lot of extra DIS5
features
In my circle of friends, people are admired more if they own the latest gadgets New item
Insecurity
INS1 People are too dependent on technology to do things for them New item
INS2 Too much technology distracts people to a point that is harmful New item
INS3 Technology lowers the quality of relationships by reducing personal interaction New item
INS4 I do not feel confident doing business with a place that can only be reached online INS4
I worry that information I make available over the Internet may be misused by others INS3
The human touch is very important when doing business with a company INS7
When I call a business, I prefer talking to a person rather than interacting with an automated system INS8
Whenever something gets automated, you need to check carefully that the system is not making mistakes INS6
Any business transaction you do electronically should be confirmed later with a separate communication INS5
New technology makes it too easy for governments and companies to spy on people DIS9
I do not consider it safe to provide personal information over the Internet INS1
Note. Technology Readiness Index (TRI) 1.0 and TRI 2.0 are copyrighted by Rockbridge Associates and the authors; their use requires written permission from the
authors.
a
Labels in the left column are shown only for items retained in the final (revised) TRI 2.0; bulleted items without labels were included in the NTRS but eliminated
during the scale refinement process. bItems with labels in right column were included in the original scale; the labels correspond to those in Parasuraman (2000).

versions were created) and the sample was split evenly components analysis with Varimax Rotation of the factor load-
between the two to minimize order effects. The mail survey ings. A scree plot of Eigenvalues for different numbers of com-
was administered by Readex Research, a professional mar- ponents affirmed a four-factor solution, with the incremental
keting research firm, to a representative sample of 2,500 variance explained by additional factors being relatively small.
U.S. residential postal addresses. To encourage response, a The four-factor solution, shown in Table 3, explained 44% of
prenotification postcard was mailed to each household. The the variance across the items. The 45 items by and large
subsequent mail kit included a cover letter, a postage-paid grouped around the same four dimensions as in TRI 1.0, with
return envelope, and a US$1 bill. Questionnaire recipients the reliability coefficients (Cronbach’s as) for the 4 item clus-
were offered the choice of responding online, but nearly all ters ranging from .68 to .90. Thus, even though some new tech-
who responded returned their completed questionnaires by nology belief themes emerged that were not prevalent over a
mail. A total of 354 usable questionnaires were returned for decade ago, they still seemed to fit well within TRI 1.0’s
a response rate of 14%. dimensional structure and were likely reflective of changes in
The online survey was administered to a census-balanced emerging technologies rather than any fundamental shift in
sample of U.S. adults recruited from an online panel main- people’s mind-set underlying their adoption and use of new
tained by Survey Sampling International (www.surveysam- technology-based offerings.
pling.com). In this version, the 45 TR statements were fully A second factor analysis (using the same procedure as in the
randomized. To account for inattentive survey respondents, a first step) was conducted on just the 36 statements from TRI
control question was included that required carefully reading 1.0. Again, a scree plot confirmed a four-factor structure, mir-
the question to answer correctly. Respondents who failed on roring the four dimensions in the original scale. The resulting
this question or who completed the survey within an unrealis- solution explained 46% of the variance in the items. Table 4
tically brief period were removed from the sample. The online shows the factor loadings for the items (in the columns labeled
survey yielded a total of 524 usable questionnaires. ‘‘2012’’). The table also shows the loadings for the same items
The number of questions in the online survey was greater from the 1999 NTRS reported in Parasuraman (2000). With
than in the mail survey, where space was limited; for example, few exceptions the factor-loading patterns across the 2012 and
only 23 online behaviors were captured in the mail survey, and 1999 surveys are consistent, attesting to the temporal stability
the mail survey did not include social media questions. The of the TR index’s dimensional structure. The reliability coeffi-
subsequent analyses were based on items contained in both ver- cients for the four dimensions are also consistently good, rang-
sions. The combined sample of 878 respondents from both sur- ing from .77 to .86 in 2012 and.74 to .81 in 1999.
veys included 51% females and 49% males. The median age Because a third of the items were reworded, it is perhaps not
was 51 years, with 21% being 18–34 years, 16% 35–44 years, surprising that 7 of the items in the new factor analysis did not
22% 45–54 years, 18% 55–64 years, and 23% 65þ years. correspond to the same dimensions as in the original TR index,
The reliance on two sampling frames helped balance the while the remaining 29 matched their original dimensions. Spe-
demographic mix in the sample. The mail survey respondents cifically, of the 7 innovativeness items in the 1999 survey, 1
tended to be older (median age 56) and more male (57% male), shifted to discomfort; of the 10 optimism items, 3 shifted to
while the online sample tended to be younger (median age 47) innovativeness; of the 9 insecurity items, 2 shifted to discom-
and more female (44% male). Even after combining the mail and fort; of the 10 discomfort items, 1 shifted to insecurity. It is
online samples, the survey sample differed in some respects noteworthy that, with the exception of the 1 innovativeness
from data reported by the U.S. Census Bureau’s Current Popula- item (INN2) in the 1999 survey that shifted to discomfort in
tion Survey (CPS). The sample was very close to the U.S. adult 2012, the remaining six shifts occurred either within motivator
population on gender mix (49% male vs. 48% in the CPS), but dimensions (three from optimism to innovativeness) or within
skewed toward being older (42% age 55þ vs. 34% in the CPS), inhibitor dimensions (two from insecurity to discomfort and
and toward being more educated (39% had a 4-year college one from discomfort to insecurity). Thus, even the small num-
degree or above vs. 29% in the CPS). To check for potential non- ber of shifts are consistent with the fundamental motivator-
response bias, the data were subsequently weighted by gender, inhibitor conceptual structure of TRI 1.0. However, the fact
age, and education, and the results were compared with and that certain items shifted between positive and negative dimen-
without weighting for the critical measures obtained from the sions, even taking into account changes in wording, suggests an
survey (TRI 2.0 and a TR-based segmentation, described subse- opportunity for improving the index by making it more parsi-
quently). The differences were negligible, confirming that non- monious and enhancing its discriminant validity.
response bias did not impact the findings. Additional analyses were conducted to derive a more parsi-
monious scale by eliminating items from the augmented list of
45 existing and new items. The primary criteria guiding these
Assessment of Factor Structure, Reliability, and
analyses focused on ensuring sufficient reliability for the four
Discriminant Validity TR dimensions, while simultaneously limiting each dimension
The analyses employed in developing TRI 2.0 mirrored those to as few items as possible, and preserving the index’s dimen-
used in developing TRI 1.0. The first step was to assess the gen- sional structure. The authors began by dropping items that from
eral data structure across the 45 TR items using principal the outset had ambiguous loadings and low communality,

Table 3. Preliminary 45-Item TR Scale. selectively removed items from each of the four dimensions
based on their item-to-total correlations, communalities, and
TR Dimensions
impact on reliability. After identifying a shorter list of potential
Item Optimism Innovativeness Discomfort Insecurity scale attributes, the authors assessed the convergent and discri-
minant validity of attribute lists through confirmatory factor
OPT1 69 — — — analysis using AMOS. After several rounds of item deletions
OPT2 73 — — —
OPT3 68 — — —
and analyses to verify reliability and factor structure, the itera-
OPT4 68 — — — tive procedure converged on a 16-item solution.
63 — — — Table 5 shows the factor structure of the final 16-item TRI
64 — — — 2.0, with 4 items for each dimension. Of the 16 items, 11 were
70 — — — in TRI 1.0, while 5 are new (2 in the optimism dimension and 3
63 — — — in the insecurity dimension). The four-factor solution explains
57 — — — 61% of the variance across the 16 items. All dimensions meet the
54 37 — —
minimum reliability threshold: The lowest reliability (Cronbach’s
65 — — —
49 — — — a) is .70 for discomfort and the highest is .83 for innovativeness.
INN1 — 76 — — TRI 2.0’s factor structure is also distinct: The items load cleanly
INN2 — 73 — — on their respective dimensions (with just one exception all cross-
INN3 — 72 — — loadings are .30 or less) and all loadings are strong (.59 or higher).
INN4 48 59 — — To further verify the dimensional structure in Table 5, a con-
35 71 — — firmatory factor analysis of a measurement model (with four
36 64 — —
latent constructs representing the four subscales, and the corre-
48 59 — —
53 49 — — sponding items specified as manifest variables) was conducted
38 37 — — using AMOS. Although the model produced a significant w2 value
DIS1 — — 63 — (p < .01), which could be an artifact of the large sample size
DIS2 — — 54 31 (Bagozzi and Yi 1988), other goodness-of-fit statistics indicated
DIS3 — — 53 — satisfactory fit (goodness-of-fit index ¼ .95; nonnormed fit index
DIS4 — — 40 33 ¼ .92; comparative fit index ¼ .94; root mean square residual ¼
— — 59 —
.06). The scale was also assessed for common method bias by
— — 44 48
— — 49 -45 introducing a common latent factor (CLF) to capture overlapping
— — 32 53 variance among the manifest variables (see Williams, Hartman,
31 — 38 48 and Cavazotte 2010). The scale performed well with none of the
— — — 46 standardized regression weights changing substantially between
— — — 47 the models with and without the CLF.
— 33 — — An important benefit of TRI 1.0 and the revised TRI 2.0 is
33 33 41 —
INS1 — — — 60
the ability to measure TR’s four dimensions. A decade of expe-
INS2 — — — 54 rience with TRI 1.0, and the consistency of factor structures in
INS3 — — — 61 the various stages of analysis for TRI 2.0, strongly supports
INS4 — — — 56 TR’s four-dimensional structure, making it important for the
— — — 60 scale to provide measures of each dimension. To this end, CFA
— — — 58 was used to assess discriminant validity by comparing each
— — — 50 latent dimension’s average variance extracted (AVE) with the
— — — 54
— — — 48
correlations among dimensions to determine if items within
— — — 54 each dimension correlate more highly with one another than
— — — 52 with items outside their parent factors (see Hair et al. 2010).
Cronbach’s a .90 .88 .68 .86 As shown in Table 5, which reports the AVE and maximum
shared variance, optimism and innovativeness show a high
Note. Factor loadings after Varimax rotation. Item labels correspond to those
in Table 2. For each dimension, items with just bullet labels correspond to the level of discrimination, while discomfort and insecurity meet
bulleted items under that dimension in Table 2 (e.g., the third bulleted item the minimum threshold for acceptable discriminant validity.
under ‘‘Optimism’’ is ‘‘Technology makes me more efficient in my occupation.’’). Table 5 also provides results from the CFA to assess conver-
All numbers (except in the last row) are factor loadings multiplied by 100. gent validity—the ability of each subscale to be explained by its
Loadings with absolute value .30 or less are not shown.
component items. As Table 5 shows, the composite reliability for
each subscale exceeds its AVE. Another commonly accepted cri-
suggesting they did not contribute much to the overall index. terion for acceptable convergent validity is for the AVE to be
An example of such a problematic item, identified in the qua- equal to or greater than .5. Optimism and innovativeness meet this
litative research phase, was ‘‘In my circle of friends, people are threshold (.51 and .56, respectively), while discomfort and inse-
admired more if they own the latest gadgets.’’ The authors then curity fall short (.38 and .40, respectively).

Table 4. Factor Loadings After Varimax Rotation for Original 36 TR Items: 2012 Versus 1999.
TR Dimensions
TRI 2.0 TRI 1.0 Optimism Innovativeness Discomfort Insecurity
Item Labels Item Labels
(2012 NTRS) (1999 NTRS) 2012 1999 2012 1999 2012 1999 2012 1999
OPT2 OPT8 69 66 — — — — — —
OPT5 56 50 — — — — — —
OPT6 62 54 — — — — — —
OPT3 OPT1 66 62 — — — — — —
OPT3 65 54 — — — — — —
OPT10 55 53 — — — — — —
OPT2 51 60 — — — — — —
INN1 INN1 — — 79 70 — — — —
INN3 INN4 — — 76 65 — — — —
INN6 — 39 76 56 — — — —
INN2 INN3 — — 74 63 — — — —
INN7 — — 69 52 — — — —
INN4 INN5 — — 68 61 — — — —
OPT4 — 54 65 — — — — —
OPT7 — 57 56 — — — — —
OPT9 — 56 43 — — — — —
DIS1 DIS4 — — — — 69 62 — —
DIS2 DIS1 — — — — 62 66 — —
DIS3 DIS2 — — — — 62 47 — —
DIS6 — — — — 59 52 — —
DIS4 DIS3 — — — — 49 45 — —
INS9 — — — — 58 52 31 37
INN2 — — — 55 49 35 — —
DIS7 — — — — 49 45 35 —
INS2 — — — — 47 — 37 56
DIS10 — — — — 43 48 31 —
DIS8 — — — — 42 52 34 —
DIS5 — — — — 31 40 — —
INS4 INS4 — — — — — — 52 59
INS3 — — — — — 45 57 37
INS7 — — — — — — 63 46
INS8 — — — — — — 62 47
INS6 — — — — — 39 54 35
INS5 — — — — — — 51 58
DIS9 — — — — — 52 51 —
INS1 — — — — 36 — 47 58
Cronbach’s a .86 .81 .77 .80 .77 .75 .77 .74
Note. TRI ¼ Technology Readiness Index. Item labels in the first two columns correspond to those in the first and last columns, respectively, of Table 2. Items with
just bullets in the first column were eliminated during the development of TRI 2.0. For example, the fifth bulleted item under ‘‘Optimism’’ corresponds to OPT2 in
TRI 1.0 (‘‘Products and services that use the newest technologies are . . . ’’), but is not part of TRI 2.0. All numbers (except in the last row) are factor loadings
multiplied by 100. Loadings .30 or less are not shown.
Innovativeness and optimism, TR’s motivating dimensions, are all a little different (hence, an AVE less than .5), but tend to
are innate traits that can be easily measured with a set of attributes, be interrelated around a theme (hence, a high Cronbach’s a and
but the two inhibiting dimensions, discomfort and insecurity, are composite reliability). Thus, the low AVE values may not be due
complex and more challenging to measure. The various factor to issues such as ambiguous item wording, but due to the slightly
analyses conducted for this study do suggest the existence of two different themes each item captures that are not part of the core
distinct inhibiting dimensions. Yet, the creation of any meaning- construct they share.
ful list of items to measure them will necessarily include different
facets that are only moderately correlated because they represent
real-life themes covering multiple aspects of the dimension. For Assessment of Construct Validity
example, insecurity is a combination of safety concerns, concerns The high reliabilities and consistent factor structure of TRI
about other negative consequences of technology, and a need for 2.0’s four dimensions provide support for the scale’s trait valid-
assurance. Consequently, the items in the inhibiting dimensions ity (Peter 1981). However, trait validity is necessary, but not

Table 5. Final 16-Item Technology Readiness Index (TRI) 2.0.
TR Dimensions
Item Optimism Innovativeness Discomfort Insecurity
OPT1 80 — — —
OPT2 76 — — —
OPT3 75 — — —
OPT4 71 — — —
INN1 — 84 — —
INN2 — 78 — —
INN3 — 76 — —
INN4 42 68 — —
DIS1 — — 75 —
DIS2 — — 73 —
DIS3 — — 69 —
DIS4 — — 61 —
INS1 — — — 76
INS2 — — — 74
INS3 — — — 72
INS4 — — — 59
Cronbach’s a .80 .83 .70 .71
Composite reliability .81 .84 .71 .72
Average variance extracted .51 .56 .38 .40
Maximum shared variance .42 .42 .38 .38
Average shared variance .20 .20 .19 .22
Note. Item labels correspond to those shown on the left-hand side of Table 2. Factor loadings after Varimax rotation. Confirmatory factor analysis: goodness-of-fit
index ¼ .953; nonnormed fit index ¼ .920; comparative fit index ¼ .942; root mean square residual ¼ .065. All factor loadings have been multiplied by 100. Load-
ings .30 or less are not shown.
sufficient, for establishing construct validity (Churchill 1979). behavioral measures in each of the regression analyses. The
Construct validity—the extent to which a scale fully and unam- adjusted R2 values for models in which the predictors were the
biguously captures the underlying unobservable construct it is overall 16-item and 36-item TR scores (respectively) are as fol-
intended to measure—must be evaluated on additional criteria. lows: 12.9% and 14.0% for number of gadgets, 22.8% and
The principal conceptual yardstick for assessing construct 26.1% for number of online behaviors, and 25.7% and 27.7%
validity is face validity: Does the scale appear to measure what for number of technology-oriented behaviors. All models were
it is supposed to? Do the items capture key facets of the unob- significant at the .001 level. The analyses also suggest that TR
servable construct being measured? As discussed previously, is an important predictor of technology-related behaviors, par-
TRI 2.0’s content and dimensional structure are by and large ticularly in the e-services domain.
consistent with those of TRI 1.0, which itself emerged from As further evidence of TRI 2.0’s predictive validity, Table 6
an extensive multiphase research program. Moreover, the cur- shows statistically significant differences in mean TRI 2.0
rent study’s exploratory phase identified contemporary scores (on a 1–5 scale) for owners/users, intenders (who plan
technology-related themes and terminology that were incorpo- to acquire in the next 2 years), and nonintenders (all F values
rated into TRI 2.0. As such, the two aforementioned questions in the last column are significant at the .001 level). With just
can be answered in the affirmative. one exception (‘‘cell phone [without Internet capabilities]’’),
Empirical assessment of the16-item TRI 2.0’s construct the means for owners/users and intenders are higher than the
validity began with a comparison of its predictive performance mean for nonintenders. Table 6 lists the technologies in declin-
vis-à-vis that of the 36-item TRI 1.0. The authors ran three sim- ing order of F values to highlight where the biggest differences
ple linear regression analyses for both TRI 1.0 and TRI 2.0, are. Consistent with our expectation that TR is likely to most
with the total TR score as the independent variable and each influence behaviors pertaining to technologies that are truly
of the three broad measures of technology behavior, in turn, cutting edge, the list is headed by the smartphone, a technology
as the dependent variable: (1) number of technology gadgets in early stages of its life cycle at the time of data collection.
owned (0–13 items, listed in Table 6), (2) number of online Likewise, the aforementioned sole exception to the pattern of
behaviors engaged in during the past year (0–23, listed in Table declining mean TR scores across the three groups is also con-
7), and (3) number of technology-oriented behaviors (0–4 sistent with expectation since cell phones are no longer novel,
behaviors: connecting devices to a TV, using cloud applica- being replaced by smartphones.
tions, digitizing documents, using a Wi-Fi service). Both scale The authors also examined TRI 2.0’s association with each
versions explain a comparable amount of variance in the of the 23 online behaviors captured in the survey. In this

Table 6. Mean Value of Technology Readiness Index (TRI) 2.0 by Ownership of Different Technologies.
Mean TR Score (1 ¼ lowest, 5 ¼ highest) and Sample Size
Currently Own or Intend to Acquire Do Not Intend to F Value

Use Technology at Technology in Next 24 Acquire in Next 24 (Degrees of
Home Months Months Freedom)
Smart phone (like an iPhone or Blackberry) 3.29 (372) 3.01 (157) 2.74 (319) 81.4** (2, 845)
Tablet computer (like an iPad) 3.27 (205) 3.16 (283) 2.79 (361) 55.5** (2, 846)
Portable music player (e.g., MP3 player) 3.20 (470) 3.09 (61) 2.77 (319) 51.9** (2, 847)
Portable media player (plays video) 3.23 (322) 3.15 (80) 2.86 (443) 37.4** (2, 842)
Television that be connected to the Internet 3.11 (337) 3.19 (212) 2.82 (302) 30.1** (2, 848)
eBook reader (e.g., Kindle) 3.23 (219) 3.13 (194) 2.88 (433) 28.7** (2, 843)
Digital single-lens reflex (SLR) camera (has complete 3.25 (149) 3.23 (128) 2.92 (562) 26.4** (2, 836)
manual control, advanced exposure control, and
detachable lenses)
Digital video camera 3.17 (393) 3.03 (127) 2.86 (333) 25.3** (2, 850)
Digital video recorder for your television (e.g., Tivo) 3.16 (358) 3.13 (129) 2.87 (363) 22.9** (2, 847)
Cell phone (without Internet capabilities) 2.93 (526) 2.87 (44) 3.22 (274) 21.9** (2, 841)
Digital point and shoot camera 3.09 (575) 3.05 (81) 2.83 (199) 13.4** (2, 852)
Portable navigation device (GPS) 3.12 (425) 3.08 (121) 2.88 (302) 13.1** (2, 845)
Stationary computer or laptop 3.05 (797) 2.78 (28) 2.60 (33) 11.1** (2, 855)
**p < .001.
Table 7. Incidence of Consumers’ Online Activities in Past 12 Months by TR Tier.
Technology Readiness Tier (Tercile, based on TRI 2.0)
Pearson
Lower One Middle One Upper One Chi-Square
Third (2.82 or Third (2.83– Third (3.25 or (Two Degrees of
Lower) 3.24) Higher) Freedom)
Sample size (273–281) (279–284) (278–289)

Booked travel arrangements online 37% 52% 66% 46.8**
Purchased an item costing less than US$10 online 49% 66% 81% 60.5**
Purchased an item costing between US$10 and US$100 online 73% 84% 93% 36.6**
Purchased an item costing more than US$100 online 44% 58% 77% 66.2**
Bought or sold stock or securities online 6% 12% 22% 31.9**
Checked information on my bank account online 61% 76% 89% 63.2**
Moved money between bank accounts, made deposits, or made 44% 57% 77% 64.3**
withdrawals online
Paid a bill online using a bill paying service, such as my bank’s 43% 59% 77% 67.9**
online bill pay service
Managed a credit card account online 37% 55% 73% 71.5**
Checked account information with a utility such as a phone, 40% 62% 77% 77.7**
gas, or electric company online
Taken a course taught online 13% 21% 27% 16.1**
Researched health information online 65% 77% 83% 25.9**
Read the news or a magazine online 58% 71% 85% 50.8**
Received medical test results online 10% 14% 21% 13.4*
Communicated with medical professionals online 11% 17% 28% 28.4**
Used a streaming music service 22% 45% 61% 84.9**
Listened to live radio programming 30% 45% 63% 64.1**
Watched a video online 49% 72% 86% 95.0**
Watched live TV programming 28% 40% 55% 42.2**
Downloaded songs online 29% 47% 66% 77.1**
Downloaded books online 18% 27% 45% 48.9**
Downloaded movies online 10% 26% 42% 76.9**
Made a phone call with a video connection (e.g., using Skype) 24% 33% 42% 20.6**
Note. TRI ¼ Technology Readiness Index.
*p < .01, **p < .001.

Table 8. Summary Statistics for Technology Readiness Index (TRI) 2.0 and Its Components.
M Correlation Coefficients*
TR Components 1999 NTRS (TRI 1.0) 2012 NTRS (TRI 2.0) SD Skewness Kurtosis OPT INN DIS INS
Optimism (OPT) 3.84 3.75 .80 .55 .10 1.00 .52 .32 .28
Innovativeness (INN) 3.18 3.02 1.02 .02 .76 .54 1.00 .40 .30
Discomfort (DIS) 3.46 3.09 .84 .11 .35 .14 .20 1.00 .56
Insecurity (INS) 4.03 3.58 .83 .44 .20 .29 .26 .44 1.00
Overall TRI 2.88 3.02 .61 .19 .29 .70 .75 .62 .70
Note. All mean values are on a 5-point scale. The overall TRI score for each respondent was the average score on the four dimensions (after reverse coding the
scores on discomfort and insecurity).
*All coefficients are significant at p < .01; coefficients in the upper triangle are from the 1999 National Technology Readiness Survey (NTRS) study.
analysis, survey respondents were divided into three approxi- scores for discomfort and insecurity are lower, but the magni-
mately equal-sized groups based on their TRI 2.0 scores—low tude of change is .40 or less. The distribution of TRI 2.0 scores
TR tier, middle TR tier, and high TR tier. Table 7 reports the is near normal, with skewness (.19) and kurtosis (.29) close to
percentage of respondents in each tier engaging in each online zero.
behavior. As the c2 test results summarized in the table’s last The 4 4 correlation matrix in Table 8 indicates that the
column show, TR is significantly associated with each of the pattern of pairwise correlations in the lower triangle is quite
23 behaviors, with increasing incidence of engagement from similar to that in the upper triangle. For instance, the correla-
low to high tiers. Even with widespread Internet access (92% tion between the two motivating TR dimensions is .54 for TRI
of the mail survey respondents reported having Internet 2.0 vis-à-vis .52 for TRI 1.0; likewise, the correlation between
access), there is a major difference between low- and high- the two inhibiting TR dimensions is .44 for TRI 2.0 vis-à-vis
TR consumers in terms of their engaging in a range of online .56 for TRI 1.0. Moreover, consistent with expectation, corre-
behaviors. TRI 2.0’s ability to consistently differentiate across lations for the various motivator-inhibitor combinations are all
a variety of technology-related consumer behaviors further negative, and smaller in magnitude, for both TRI 2.0 and TRI
supports its construct validity. 1.0. Thus, TRI 2.0 seems to preserve TRI 1.0’s overall distinct
TRI 2.0 also correlates with social media behavior. The dimensional structure, and patterns of association among its
online survey asked respondents if they have a ‘‘social media components.
page such as Facebook, LinkedIn, or Twitter.’’ The difference
in mean TR scores for those answering ‘‘yes’’ and those A TR-Based Segmentation Analysis
answering ‘‘no’’—3.20 versus 2.93—was statistically signifi-
cant (t ¼ 4.16, p < .0001). The survey also measured the fre- The four TR dimensions’ distinctiveness implied by the rela-
quency of 11 social media activities. For every activity, the tively low pairwise correlations (and hence even lower overlap-
TR score was significantly higher for those who engaged in ping variance) among combinations of dimensions suggest that
it more frequently. segmenting customers based on their TR scores can be insight-
ful. For instance, Parasuraman and Colby (2001) developed
and discussed a segmentation scheme (using K-means cluster
Summary Statistics for TRI 2.0 and Its Components analysis of TRI 1.0 scores) that consisted of five segments:
Table 8 reports means, standard deviations, and distributional Explorers (high motivation, low inhibition), pioneers (high
characteristics for the overall TRI 2.0 and its four components, motivation, high inhibition), skeptics (low motivation, low
as well as pairwise correlations among them. For comparison inhibition), paranoids (moderate motivation, high inhibition),
purposes, the table also reproduces the mean values and pair- and laggards (low motivation, high inhibition). In this study,
wise correlations from the 1999 NTRS study reported in Para- the authors conducted a latent class analysis (Magidson and
suraman (2000). Vermunt 2003) of scores on the 16 TRI 2.0 items. LCA is often
The TR scale ranges from 1.0 (strongly disagree) to 5.0 preferred as a segmentation method over traditional methods
(strongly agree), with 3.0 representing the scale’s midpoint such as K-means. It is more robust in handling different types
(neutral). The mean TRI 2.0 score is 3.02, close to the middle of measurements, including the Likert-type scale TRI 2.0 rat-
of the scale and not far from TRI 1.0’s mean value of 2.88. ings. LCA also does not require assumptions about the number
Consumers are generally optimistic about technology (M ¼ of classes/clusters, and provides greater confidence because it
3.75) and closer to the scale’s midpoint on innovativeness is based on the probability of cases belonging to latent classes
(3.02). The values on both of these motivating dimensions are rather than a more subjective algorithmic approach underlying
similar to those for TRI 1.0. Consumers are close to the scale’s traditional clustering methods.
midpoint on discomfort (3.09) and slightly above the midpoint Three separate LCA solutions—corresponding to four, five,
on insecurity (3.58). Compared to TRI 1.0, mean TRI 2.0 and six latent classes (i.e., clusters)—were derived and

Table 9. Latent Class Segmentation Using Technology Readiness Index (TRI) 2.0 and TRI 1.0 Data.
Means (Ranks) and R2 Values
Segments (n) % Optimism Innovativeness Discomfort Insecurity Total TR
TRI 2.0 (16 items)

1. Skeptics (293) 38% 3.47 (4) 3.03 (3) 2.81 (4) 3.46 (4) 3.06 (2)
2. Explorers (141) 18% 4.63 (1) 4.09 (1) 2.36 (5) 2.67 (5) 3.92 (1)
3. Avoiders (123) 16% 2.62 (5) 1.80 (5) 3.62 (2) 4.27 (1) 2.13 (5)
4. Pioneers (121) 16% 4.24 (2) 3.93 (2) 3.86 (1) 4.12 (2) 3.05 (3)
5. Hesitators (104) 13% 4.06 (3) 1.91 (4) 3.32 (3) 3.69 (3) 2.74 (4)
R2 (Adjusted) 65%* 71%* 39%* 40%* 76%*
TRI 1.0 (36 items)
1. Skeptics (302) 35% 3.56 (3) 3.11(3) 2.96 (4) 3.34 (4) 3.19 (3)
2. Explorers (107) 13% 4.55 (1) 4.24 (1) 2.43 (5) 2.98 (5) 4.00 (1)
3. Avoiders (70) 8% 2.33 (5) 1.61(5) 3.73 (1) 4.24 (1) 2.06 (5)
4. Pioneers (162) 19% 4.23 (2) 3.87 (2) 3.69 (2) 4.10 (2) 3.22 (2)
5. Hesitators (217) 25% 3.35 (4) 2.38 (4) 3.55 (3) 4.09 (3) 2.61 (4)
R2 (Adjusted) 67%* 69%* 47%* 50%* 76%*
*Significant at p < .001.
compared on three conventional criteria for evaluating LCA- The overall TR scores range from 2.13 for the least techno-
generated clusters: Akaike Information Criterion, Bayesian ready segment, the ‘‘avoiders,’’ to 3.92 for the most techno-
Information Criterion, and Consistent Akaike’s Information ready segment, the ‘‘explorers.’’ The segmentation scheme’s abil-
Criterion. The three solutions differed only slightly on these ity to explain overall TR and its components was assessed by con-
criteria. However, the five-cluster solution yielded a more even ducting five separate general linear model analyses, wherein
distribution of cases across the clusters than did the four- and segmentation was a categorical independent variable and TR
six-cluster solutions. Moreover, the five-cluster solution, sum- scores, overall and for each component, in turn, was the depen-
marized in Table 9’s top half, is similar to the TR-based dent variable. As the R2 values in Table 9 show, the segmentation
segmentation described by Parasuraman and Colby (2001), variable is significantly associated with each TR component and
reinforcing its superiority over the other two solutions. overall TR, explaining 76% of the variance in the latter.
A K-means solution was also generated and compared with A five-group LCA was also conducted using data on the 36
the LCA solution. They were very similar, with 84% of original TR items as input. The resulting segments, summar-
cases belonging to the same cluster in both solutions. ized in the bottom half of Table 9, are similar to those in the
The five segments were labeled as follows based on the dis- top half in terms of relative sizes and TR profiles. Moreover,
tinct combinations of technology-related beliefs associated the share of respondents classified into a given segment in the
with each: 36-item analysis who fall into the same segment in the 16-item
analysis was high for most segments (76% for skeptics, 94% for
‘‘Skeptics’’ (38% of consumers)—tend to have a explorers, 92% for avoiders, 65% for pioneers, and 37% for
detached view of technology, with less extreme positive hesitators).
and negative beliefs; The five TR segments have unique demographic character-
‘‘Explorers’’ (18%)—tend to have a high degree of moti- istics (Table 10). For example, the segment with the highest TR
vation and low degree of resistance; score, the explorers, is younger, higher educated, more likely to
‘‘Avoiders’’ (16%)—tend to have a high degree of resis- work in a technology profession, and owns the largest number
tance and low degree of motivation; of technology gadgets. The avoider segment is a mirror oppo-
‘‘Pioneers’’ (16%)—tend to hold both strong positive site of the explorers, but also stands out as being low on ethnic
and negative views about technology; and diversity. The pioneer segment is the most ethnically diverse,
‘‘Hesitators’’ (13%)—stand out due to their low degree while the hesitator segment has the lowest employment in tech-
of innovativeness. nology professions. Thus, the segments differ on demographics
and behavior, despite being derived entirely from scores on
TRI 2.0’s 16 items.
The Explorers are similar to the same early adopter
explorers identified in previous research, while the avoiders are
similar to the late adopter laggards. The other segments do not
fall into a tidy spectrum from early to late adopters, but instead
Limitations of TRI 2.0
reflect more complex combinations of beliefs that suggest dif- TRI 2.0 has some limitations that should be acknowledged.
ferent challenges in marketing technology-based services. Specifically, TRI 2.0’s subscales for the inhibitor dimensions

Table 10. Demographic Characteristics of Technology Readiness Index (TRI) 2.0-Based Segments.
Femalea % Aged % With Minimum 4-Year % Non- % Employed in Technology Number of High-Tech
Segment (n) (%) 50þb (%) College Degreec (%) Whited (%) Professionb (%) Gadgets Owned (M)e
1. Skeptics (293) 52 50 42 23 8 6.0

2. Explorers (141) 44 34 46 20 18 7.4
3. Avoiders (123) 59 79 29 14 7 4.4
4. Pioneers (121) 53 41 42 30 13 6.7
5. Hesitators (104) 48 69 39 12 2 5.5
a
Chi-square not significant.
b
Chi-square significant at p < .001.
c
d
e
F value significant at p < .001.
of discomfort and insecurity are somewhat weak on some psy- Second, TR can be an important moderating variable in stud-
chometric criteria, especially AVE. While these dimensions ies involving multivariate frameworks. Scholars can use TRI 2.0
emerged as clean factors in the exploratory factor analysis, to explain the dynamics between variables in a technology-
their conceptual core is challenging to represent as a set of influenced context and as a diagnostic or control variable in
homogeneous attributes. The TRI 2.0 items capturing the inhi- experiments. In past research involving multiple constructs,
bitor dimensions tap into diverse beliefs/experiences about (a) researchers have frequently opted to use a small subset of items
perceived control of technology and (b) distrust of technology. from TRI 1.0 to reduce the burden on respondents. The more
While this item diversity could be viewed as a strength, in that parsimonious, yet rigorously developed, TRI 2.0 should be more
it makes TRI 2.0 more robust, future research should investi- practical to use in its entirety in future studies. Moreover, the fact
gate if the AVE for the inhibitor dimensions could be improved that TRI 2.0 has the same number of items (4) for each TR
through refined wording, adding more attributes, or using alter- dimension may be a desirable feature for some research studies.
nate measurement scales. In the meantime, researchers should Technology readiness measured by TRI 2.0 can be used as a
exercise caution when using TRI 2.0 primarily to study a single potentially valuable psychographic variable in applied,
facet of beliefs that inhibit technology usage. Other than this decision-oriented research in contexts where technology-
issue, TRI 2.0 and its subscales are psychometrically sound. based innovation plays an important role. The TR of a given
market or customer base can be scored and compared to base-
lines, providing implications for marketing. High-TR custom-
Applications of TRI 2.0 ers (the ‘‘explorers’’) will be interested in advanced
A primary reason for TRI 1.0’s widespread use for over a decade functionality and capable of mastering new high-tech offerings
is the exponential growth in technology’s influence in the ser- with minimal help. Low-TR customers (the ‘‘avoiders’’ and
vice domain. While scores of studies involving technology- ‘‘hesitators’’) will be more satisfied with basic functionality but
related consumer behavior have employed the scale, a signifi- will need more support and reassurance. TRI 2.0 will also
cant shortcoming of TRI 1.0 is its length. TRI 2.0 has wider identify the ‘‘skeptics’’—persuasive messages that pro-
applicability because it is more concise, resulting in less burden vide concrete reasons for adoption must be developed to
on surveys measuring multiple constructs besides just TR, and attract them—and the ‘‘pioneers’’—they need little con-
various other refinements to the index make it more robust for vincing to adopt technology but require more support to
use across different contexts and over time. be satisfied. Finally, there is often a special interest in the
Parasuraman (2000) suggested some potential applications highest TR consumers because they could serve as ‘‘evan-
of the then newly developed TR index. Building on those sug- gelists’’ in motivating others to try a new technology-
gestions and based on insights from research proposals and per- based offering.
mission requests received during the past dozen years, TRI 2.0 Researchers can also use TRI 2.0 to tier consumers into
has two general applications. First, it can be used to assess TR groups with varying TR levels, providing a unique lens
levels within a given population, which can consist of a coun- for understanding the role of technology beliefs in the market-
try, a particular demographic group of special interest (e.g., place. A strong correlation between TR and interest level in
African Americans in the United States, teens, seniors), a pro- (or adoption rate of) a new offering signals that the offering
fession (e.g., teachers, nurses), or a market segment (e.g., pur- is inherently cutting edge and, therefore, requires specialized
chasers of high-tech products). TRI 2.0 facilitates marketing strategies that differ from conventional strategies
understanding the dynamics behind adoption of various tech- for marketing lower tech offerings. Moreover, in usability
nologies by providing measures of the four TR dimensions as research pertaining to a new technology-based offering, it is
well as overall TR. As demonstrated earlier, TRI 2.0 is a robust prudent to set quotas of users with varying TR levels to ensure
predictor of technology-related behavioral intentions as well as that the offering is intuitive to all consumer types, not just to
actual behaviors. self-learners.

Directions for Future Research studies. However, most of them have included just a subset
of items from the 36-item TRI 1.0, and none has involved a
Apart from the general applications outlined in the preceding
rigorous cross-country comparison of the scale’s psycho-
section, there are specific TR-related issues worthy of scholarly
metric properties. Conducting multicountry, across-context
research. One potentially fruitful research avenue is to investi-
studies incorporating the full TR scale should now be more
gate the causes and correlates of TR. Research findings to date,
feasible with the availability of the much shorter TRI 2.0.
including those reported herein (Table 10), suggest that demo-
graphic characteristics such as age, education, and occupation
correlate with TR. More in-depth research is needed to validate Concluding Remarks
these correlates, to identify other correlates, and to understand The primary objective of the research discussed in this article
why they matter. One interesting issue is whether education was to update the original TR index (TRI 1.0) and produce a
and occupation cause changes in TR, or if more techno-ready more concise and contemporary scale. Starting with the 36-
people seek out higher education and gravitate to technology- item TRI 1.0, as well as user feedback from a variety of applica-
related professions. Another research-worthy issue is whether tions over the years, the authors augmented and updated it based
TR declines naturally after a certain age as extant data seem on insights from an extensive exploratory research phase. This
to suggest or if the age effect might be due to differences across was followed by a quantitative research phase that (1) iteratively
generational cohorts in terms of exposure to and experience condensed the augmented TR scale to produce the 16-item TRI
with cutting-edge technologies. Longitudinal research designs 2.0 and (2) verified its reliability and validity on a variety of cri-
are needed to shed light on these causes. teria. Though similar to TRI 1.0 in overall structure and content,
Another research-worthy issue is if people’s values, emo- TRI 2.0 is less than half as long, and its items are more technol-
tions, and other inherent traits are related to their TR and ogy neutral in that they do not refer to technologies or contain
whether any uncovered associations are causal or merely corre- terms that are outdated. The availability of TRI 2.0 will hope-
lational. In-depth qualitative research might be an appropriate fully accelerate (1) practical applications of the TR construct
starting point for addressing these questions. Another intri- (e.g., examining a TR-based segmentation of the target market
guing, even if peculiar, issue to investigate is whether there are before full-scale introduction of a cutting-edge technology for
genetic factors related to TR. As the science of using DNA data serving customers) and (2) inclusion of the TR construct in scho-
advances, it may be possible to ascertain if TR, as measured by larly investigations (e.g., as a moderator of the link between per-
TRI 2.0, is linked to genetic factors. ceived benefits of a new technology and its actual adoption/use).
The interaction between TR, an inherent individual-level trait,
and the characteristics of technology-based offerings is also worth Declaration of Conflicting Interests
exploring. Frameworks such as the technology assessment The author(s) declared no potential conflicts of interest with respect to
model—TAM (Davis 1989; Venkatesh and Davis 2000) have the research, authorship, and/or publication of this article.
been employed to study how perceived characteristics of a
technology-based product influence its adoption and usage; and, Funding
scales such as E-S-QUAL (Parasuraman, Zeithaml, and Malhotra The author(s) received no financial support for the research, authorship,
2005) have been used for assessing the service quality of elec- and/or publication of this article.
tronic interfaces and its downstream effects. There have also been
some attempts to study the joint effects of individual and Note
technology-specific characteristics. In the realm of adopting 1. The National Technology Readiness Survey (NTRS), administered
SSTs, Meuter et al. (2005) proposed and empirically demon- by Rockbridge Associates and A. Parasuraman, has been con-
strated the role of ‘‘consumer readiness,’’ a construct reflecting ducted in 1999, 2000, 2001, 2002, 2004, 2006, 2007, 2009, and
consumers’ role clarity, motivation, and ability vis-à-vis SSTs. 2012.
Lin, Shih, and Sher (2007) integrated TR and TAM to propose and
test the technology readiness acceptance model; but they also References
called for more comprehensive research to investigate additional
Bagozzi, Richard P. and Youjae Yi (1988), ‘‘On Evaluation of Struc-
variables and their interactions that could influence adoption.
tural Equation Models,’’ Journal of the Academy of Marketing
Likewise, Zeithaml, Parasuraman, and Malhotra (2002) have
Science, 16 (Spring), 74-94.
called for research on whether TR moderates the link between
Bitner, Mary Jo (2001), ‘‘Service & Technology: Opportunities and
perceived quality of electronic interfaces and their usage by
Paradoxes,’’ Managing Service Quality, 11 (6), 375-379.
consumers.
Brahima, Sanou (2013), ‘‘The World in 2013: ICT Facts and Figures,’’
Because TR is an individual-level construct, examining
ICT Data and Statistics Division, Geneva Switzerland: Telecommu-
TRI 2.0’s structural stability and invariance across different
nication Development Bureau, International Telecommunications
environmental contexts (e.g., countries with different cul-
Union, (February), 1-2.
tures, technology infrastructures, etc.) is also a promising
Chisholm, Mark, Jack Cutts, Shawn DuBravac, Rachel Horn, and Rick
research avenue. The authors have received permission
Kowalski (2013), 5 Technology Trends to Watch. Arlington, VA:
requests from around the world to use TRI 1.0 in research
Consumer Electronics Association.

Churchill, Gilbert A. Jr. (1979), ‘‘A Paradigm for Developing Better Parasuraman, A., Valarie A. Zeithaml, and Arvind Malhotra (2005),
Measures of Marketing Constructs,’’ Journal of Marketing ‘‘E-S-QUAL: A Multiple-item Scale for Assessing Electronic Ser-
Research, 16 (February), 64-73. vice Quality,’’ Journal of Service Research, 7 (3), 213-233.
Cowles, Deborah and Lawrence A. Crosby (1990), ‘‘Consumer Parasuraman, A. and Charles L. Colby (2001), Techno-ready Market-
Acceptance of Interactive Media in Service Marketing Encoun- ing: How and Why Your Customers Adopt Technology. New York,
ters,’’ Service Industries Journal, 10 (July), 521-540. NY: The Free Press, Division of Simon & Schuster.
Dabholkar, Pratibha A. (1996), ‘‘Consumer Evaluations of New Parasuraman, A. (2000), ‘‘Technology Readiness Index (TRI): A
Technology-based Self-service Options: An Investigation of Alter- Multiple-item Scale to Measure Readiness to Embrace New Tech-
native Models of Service Quality,’’ International Journal of nologies,’’ Journal of Service Research, 2 (May), 307-320.
Research in Marketing, 13 (1), 29-51. Peter, J. Paul (1981), ‘‘Construct Validity: A Review of Basic Issues
Davis, Fred D. (1989), ‘‘Perceived Usefulness, Perceived Ease of Use, and Marketing Practices,’’ Journal of Marketing Research, 18
and User Acceptance of Information Technology,’’ MIS Quarterly, (May), 133-145.
13 (3), 319-340. Venkatesh, Viswanath and Fred D. Davis (2000), ‘‘A Theoretical
Davis, Fred D., Richard P. Bagozzi, and Paul R. Warshaw (1989), Extension of the Technology Acceptance Model: Four Longitudi-
‘‘User Acceptance of Computer Technology: A Comparison of Two nal Field Studies,’’ Management Science, 46 (2), 186-204.
Theoretical Models,’’ Management Science, 35 (8), 982-1003. Williams, Larry J., Nathan Hartman, and Flavia Cavazotte (2010),
Eastlick, Mary Ann (1996), ‘‘Consumer Intention to Adopt Interactive ‘‘Method Variance and Marker Variables: A Review and Compre-
Teleshopping,’’ Marketing Science Institute working paper, Report hensive CFA Marker Technique,’’ Organizational Research Meth-
No. 96-113 (August), Cambridge, MA. ods, 13 (3), 477-514.
Hair, Joseph F. Jr., William C. Black, Barry J. Babin, and Rolph E. Woodall, Gina and Charles L. Colby (2011), ‘‘The Results Are In:
Anderson (2010), Multivariate data analysis. 7th ed. Upper Saddle Social Media Techniques vs. Focus Groups for Qualitative
River, NJ: Prentice Hall. Research,’’ MRA’s Alert! Magazine, March, 23-27.
Hoffman, Donna L., Thomas P. Novak, and Marcos Peralta (1999), Yahoo! Finance (2012), ‘‘Number of Active Users at Facebook over the
‘‘Building Consumer Trust Online,’’ Communications of the ACM, Years,’’ (accessed October 23, 2012), [available at finance.yahoo.
42 (4), 80-85. com/news/number-active-users-facebook-over-years-214600186–fi
Kuo, Tsung-Hsien (2011), ‘‘The Antecedents of Customer Relation- nance.html].
ship in E-banking Industry,’’ The Journal of Computer Information Zeithaml, Valarie A., A. Parasuraman, and Arvind Malhotra (2002),
Systems, 51 (3), 57-66. ‘‘Service Quality Delivery Through Web Sites: A Critical Review
Lin, Jiun-Sheng Chris and Pei-Ling Hsieh (2012), ‘‘Refinement of the of Extant Knowledge,’’ Journal of the Academy of Marketing Sci-
Technology Readiness Index Scale: A Replication and Cross- ence, 30 (4), 362-375.
validation in the Self-service Technology Context,’’ Journal of
Service Management, 23 (1), 34-53.
Lin, Jiun-Sheng Chris and Hsing-Chi Chang (2011), ‘‘The Role of
Author Biographies
Technology Readiness in Self-service Technology Acceptance,’’
Managing Service Quality, 21 (4), 424-444. A. Parasuraman (‘‘Parsu’’) is a professor and holder of the James W.
Lin, Chien-Hsin, Hsin-Yu Shih, and Peter J. Sher (2007), ‘‘Integrating McLamore Chair in marketing at the School of Business, Univer-
Technology Readiness into Technology Acceptance: The TRAM sity of Miami. His research focuses on service quality assessment
Model,’’ Psychology & Marketing, 24 (7), 641-657. and improvement, service innovation, service productivity, and
Magidson, Jay and Jeroen K. Vermunt (2003), ‘‘Latent Class Models,’’ technology’s role in marketing to and serving customers. He has
in Handbook of Quantitative Methods in Social Science Research, published extensively in scholarly journals, written a textbook on
D. Kaplan, 175-198, ed. Newbury Park, CA: Sage. marketing research, and coauthored a number of research mono-
Massey, Anne P., Vijay Khatri, and Mitzi M. Montoya-Weiss (2007), graphs and three business books. He is affiliated with several
‘‘Usability of Online Services: The Role of Technology Readiness service research centers around the world and is a frequent speaker
and Context,’’ Decision Sciences, 38 (2), 277-308. at international conferences.
Meuter, Matthew L., Mary Jo Bitner, Amy L. Ostrom, and Stephen W.
Brown (2005), ‘‘Choosing Among Alternative Service Delivery Charles L. Colby is the founder and chief methodologist of Rockbridge
Modes: An Investigation of Customer Trial of Self-service Tech- Associates, Inc., a market research firm specializing in the services and
nologies,’’ Journal of Marketing, 69 (April), 61-83. technology sectors. He has over 30 years of experience in market
Meuter, Matthew L., Amy L. Ostrom, Mary Jo Bitner, and Robert research and strategy formulation. Prior to founding his company, he
Roundtree (2003), ‘‘The Influence of Technology Anxiety on Con- directed a research function for Citigroup and worked for leading
sumer Use and Experiences with Self-service Technologies,’’ research organizations. He has presented widely at conferences and
Journal of Business Research, 56 (11), 899-906. coauthored the book Techno-ready Marketing with A. Parasuraman.
Mick, David Glenn and Susan Fournier (1998), ‘‘Paradoxes of Tech- He is a senior fellow at the Center for Excellence in Service at the
nology: Consumer Cognizance, Emotions, and Coping Strategies,’’ Robert H. Smith School of Business, University of Maryland, and has
Journal of Consumer Research, 25 (September), 123-147. an MBA from the same school.

An Updated and Streamlined Technology Parasuraman2014

Uploaded by

Copyright:

Available Formats

An Updated and Streamlined Technology Parasuraman2014

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

An Updated and Streamlined Technology Parasuraman2014

Uploaded by

Copyright:

Available Formats

Journal of Service Research http://jsr.sagepub.

An Updated and Streamlined Technology Readiness Index: TRI 2.0

The online version of this article can be found at:

Center for Excellence in Service, University of Maryland

Email Alerts: http://jsr.sagepub.com/cgi/alerts

>> OnlineFirst Version of Record - Jun 22, 2014

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

A. Parasuraman1 and Charles L. Colby2

Introduction Technology-triggered transformation in services is likely to

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

The Structure of Technology Readiness and

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Table 2. 45 TR Items in the 2012 National Technology Readiness Survey (NTRS).

TRI 2.0 (NTRS TRI 1.0 (NTRS

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Table 5. Final 16-Item Technology Readiness Index (TRI) 2.0.

Item Optimism Innovativeness Discomfort Insecurity

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Mean TR Score (1 ¼ lowest, 5 ¼ highest) and Sample Size

Currently Own or Intend to Acquire Do Not Intend to F Value

Table 7. Incidence of Consumers’ Online Activities in Past 12 Months by TR Tier.

Technology Readiness Tier (Tercile, based on TRI 2.0)

Sample size (273–281) (279–284) (278–289)

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Means (Ranks) and R2 Values

Segments (n) % Optimism Innovativeness Discomfort Insecurity Total TR

TRI 2.0 (16 items)

*Significant at p < .001.

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

1. Skeptics (293) 52 50 42 23 8 6.0

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

Downloaded from jsr.sagepub.com at University of Waikato Library on June 24, 2014

You might also like