Impact of Point-of-Sale-Systems (POS) in Traditional and Fast Food Restaurants

Users’ intentions to employ a Point-Of-Sale system

Yun Ji Moon a , Woody Kim

b and Sunny Ham

c∗

a Department of Management Information Systems, Catholic University of Pusan, Pusan, Republic of

Korea; b Dedman School of Hospitality, College of Business, Florida State University, Tallahassee,

FL, USA; c Department of Tourism Management, Gachon University, Seongnam-si, Republic of

Korea

(Received 24 January 2012; accepted 21 April 2013)

This study proposes that task, technology, and individual characteristics affect the Point-Of-Sale (POS) utilization of employees in service industry, specifically in restaurants. The integrated technology acceptance model and task – technology fit (TTF) model is appropriate for explaining service employees’ behavioral intentions to use POS. Data were obtained from 167 service employees. The hypothesized model resulted in a good fit, supporting all eight proposed hypotheses. The TTF construct was confirmed to be a mediator of task, technology, and individual characteristics affecting intention to use. Our integrated model is expected to help researchers and practitioners better understand why service employees choose POS for their tasks and, further, how the technology characteristics of POS and its fit- with-task characteristics in a service sector lead to service employee choices.

Keywords: Point-Of-Sale; task – technology fit; technology acceptance model

Introduction

Recently, a key concern of service and marketing research has been to identify the ade-

quate technology for improving customer contact personnel’s productivity (Honeycutt,

Thelen, Thelen, & Hodge, 2005; Rangarajan, Jones, & Chin, 2005; Sundaram, Schwarz,

Jones, & Chin, 2007). To generate a positive impact on individual performance, the tech-

nology must be utilized, and it must also be a good fit with the task it supports (Goodhue &

Thompson, 1995). The information technology literature reveals that when examining

information technology utilization or adoption, the two most employed theories are tech-

nology acceptance model (TAM) and task – technology fit (TTF) model. TAM assumes

users’ beliefs and attitudes toward a particular information technology largely determine

whether they use the information technology, while TTF proposes that users choose infor-

mation technology that provides benefits, such as improved job performance.

However, Dishaw and Strong (1999) and Klopping and Mckinney (2004) insisted that

the integration of the two models offers better explanatory power than either one does by

itself. The introduction of the TTF variables into TAM as an antecedent of perceived ease

of use and perceived usefulness is worth emphasizing because it explores the fit between

task and technology as a determinant of users’ perceptions (Lee, Park, Chung, & Blakeney,

2012). This argument concurs with the work of Mathieson (1991), who claimed that

# 2014 Taylor & Francis

∗Corresponding author. Email: [email protected]

The Service Industries Journal, 2014

Vol. 34, No. 11, 901 – 921, http://dx.doi.org/10.1080/02642069.2014.915947

perceived ease of use can be a function of TTF as can perceived usefulness. Indeed, the

level of TTF is quite likely to affect users’ perceptions since user beliefs about usefulness

and ease of use are likely to be developed from rational assessments of the information

technology’s characteristics and the tasks for which it could be used (Dishaw & Strong,

1999). Therefore, the integrated model in this study is expected to explain more in the var-

iance of the relationships among TTF, perceived ease of use, perceived usefulness, and

intention to use.

Academics and practitioners in the service industry have paid attention to research on

technology adoption, such as self-service, sales automation, online grocers, or retail scan-

ning (Hui & Wan, 2009; Jones, 1985; Liu, Huang, & Chiou, 2012). The present study

intends to determine the factors affecting service employees’ intentions to use the

Point-Of-Sale (POS) technology in restaurants. In the restaurant industry, the POS

system is defined as ‘a network of cashier and server terminals that typically handles

food and beverage orders, transmission of orders to the kitchen and bar, check settlement,

credit card authorization, labor scheduling, and timekeeping’ (Collins, Cobanoglu, &

Malik, 2003, p. 233). The POS system equipped with powerful functions has revolutio-

nized the restaurant industry, achieving greater food, labor, and beverage cost-savings,

better employee management, increased revenue management, and enhanced ability to

understand customer preferences.

For instance, ‘a Pizza Hut franchise was able to strengthen its operations control by

using the POS computer systems in the area of sales, cost of sales, payroll, labor forecast-

ing, promotions, and cash payouts’ (Tricon Global Restaurant, 2012). The POS system has

aided the goal of creating happier customers through accurate order information to the

kitchen, thus speeding up the delivery of the correct order, and by accurate billing and

a faster cash-out for each customer. Since enhancing the speed of customer service

and improving the quality of transactions is still important in restaurants, POS usages

and new technologies are important investments that cannot be overlooked in the

service industry (Hui & Wan, 2009; Jones, 1985). Before opening, new restaurants

spend 3 – 6% of their total initial investment in POS technology. Information technology

investment for restaurant operators becomes a necessary tool like many other required

investments in kitchen equipment. Regardless of the importance of this topic, however,

little research has been conducted to explore factors influencing effective POS utilization

in the restaurant industry. To reduce this dearth in the literature, therefore, this paper inves-

tigates the interrelationships among factors affecting POS adoption intention.

An integrated TAM and TTF model, which was developed by Dishaw and Strong

(1999), was used as the basis of the proposed research model. Specifically, the proposed

exogenous constructs influencing foodservice employees’ POS system adoption intentions

are task, technology, and individual characteristics (Karahanna, Straub, & Chervany,

1999; Thompson, Compeau, & Higgins, 2006; Venkatesh, Morris, Davis, & Davis,

2003). Task characteristic and technology characteristic constructs are core concepts of

TTF. It is also suggested that individual characteristics affect technology utilization in a

variety of work settings. An information technology is more likely to be adopted if the

functions reflecting task, technology, and individual characteristics available to users

support the activities of the users. In other words, a system that does not reflect those

characteristics and does not offer sufficient support will not be used.

While TAM and TTF are the key theories in explaining user’s intention or behavior,

task, technology, and individual characteristics are evidenced in the academic literature

affecting technology utilization. In the sense, adding the three variables, task, technology,

and individual characteristics, into the integrated TAM and TTF strengthen a research

902 Y.J. Moon et al.

model to explain technology user behavior. Therefore, this study postulates that the fitness

is determined by the task, technology, and individual characteristics and that the degree of

fitness influences users’ perceptions, followed by the level of intention to use.

In summary, the objectives of the proposed model are threefold: (1) to determine if

influencing components – task, technology, and individual characteristics – have a sig-

nificant (either positive or negative) relationship with TTF; (2) to examine if a positive

link exists between TTF and perceived usefulness as well as between TTF and perceived

usefulness; and (3) to examine if there are positive relationships among perceived ease of

use, perceived usefulness, and the intention to adopt a restaurant POS system. At the con-

ceptual level, the study provides an empirical validation of TTF and TAM factor linkage.

Also, since there has been little known about the meanings of task, technology, and indi-

vidual characteristics tailored to the restaurant context, the lack of specificity of each

characteristic influencing TTF is likely to be supplemented. At the managerial level,

this research provides insights into critical drivers encouraging POS utilization by satisfy-

ing service employee’s needs in restaurants. Food service employees performing certain

tasks, such as food and beverage cost control, labor scheduling, fast order taking, and

prompt food delivery, will choose a sophisticated POS system that is well-suited to

their tasks. The findings of this study would also serve as valuable and supplementary

information for the developers of POS hardware and software programs.

TAM and TTF

TAM and TTF model

TAM has been widely used as one of the most effective tools to assess an information tech-

nology’s acceptance/adoption in a variety of working environments (Hui & Wan, 2009;

Klopping & McKinney, 2004). Although TAM proves satisfactorily adequate in predicting

usage and intention, its one weakness, in understanding information technology accep-

tance and usage behavior, is its lack of task focus. Information technology will be

adopted if it is a good fit for the task it supports (Goodhue & Thompson, 1995). While

the perceived usefulness construct in TAM includes an implicitly task-related piece that

merely considers that usefulness means useful for something, the items in perceived

ease of use do not reflect specifically or explicitly any certain characteristics of tasks. In

other words, one drawback of TAM in the understanding of an information technology’s

acceptance and usage behavior is its lack of explicitly included task characteristic

(Goodhue & Thompson, 1995).

The TTF model, on the other hand, extends the TAM by considering how the task

affects the use. More specifically, the TTF model suggests that technology adoption

depends on how well the technology fits the requirements of a particular task. For that

reason, researchers capture the task concept in utilization of technologies, suggesting

that ‘more explicit inclusion of task characteristics may provide a better model of infor-

mation technology utilization’ (Dishaw & Strong, 1999, p. 11). The TTF model considers

the importance of TTF constructs (fit) on how a task influences information technology

use. The core of the TTF model is based on its ability to match an information technology

to the demands of a task (Goodhue & Thompson, 1995). The TTF model includes four key

constructs (i.e. task characteristics, technology characteristics, perceived fit, and perform-

ance/utilization) where task characteristics and technology characteristics, respectively,

influence the TTF construct, which will eventually affect the available outcome of

either the performance or the utilization. Later, an individual abilities construct was

added to the TTF model as an exogenous variable, assuming that TTF results from the

The Service Industries Journal 903

correspondence among task requirements, individual abilities, and the functionality of the

technology. Individuals may use technologies to assist them in their performance of their

tasks. Individual characteristics such as different training level and computer experiences

could affect how easily and well he/she will utilize the technology (Goodhue & Thomp-

son, 1995). Thus, TTF is determined by the interactions among task, technology, and indi-

vidual characteristics.

Combination of TAM and TTF

Although TAM and TTF, respectively, serve well in predicting information technology

utilization or adoption, Dishaw and Strong (1999) suggest an integration of the two

models. Testing the newly combined model, they demonstrated that the integrated

model offers more explanatory power than either model alone, and they implied that

the application of the integrated model should lead to a better understanding on the adop-

tion/choice of information technology systems. The rationale behind combining the two

models is that each model explains different aspects of users’ choices. TAM, an atti-

tude/behavior model, argues that users’ behaviors are primarily determined by their

beliefs or attitudes toward a particular technology. While perceived usefulness is a

measure of the individual’s subjective perception of the utility offered by the technology

in a specific task-oriented context, perceived ease of use is an indicator of the individual’s

cognitive effort needed to learn and to utilize the technology. Both indicators are assessed

by individual-level beliefs and attitudes toward information technology.

The TTF model demonstrates that users, with a rational assessment, choose a particular

information technology that provides benefits, such as improved job performance, regard-

less of their attitude toward it, specifically, or information technology, more generally.

Organizational structural contingency theories and the TTF model share some similarity

because both theories are interested in fit-related technology (Khazanchi, 2005). Contin-

gency theories ask how organizations can match various technologies to the design of

units to accomplish a higher unit performance at the organizational level, such as the

organizational context which is used (Cohen & Levinthal, 1990). On the other hand, the

TTF model is more concerned with individual work related to various contexts.

From the perspectives of the effects of beliefs toward information technology, as well as

the rationally assessed expectations from using information technology on users’ infor-

mation technology adoption, Goodhue and Thompson (1995) tested the integrated model

with working programmer-analysts completing maintenance projects in three Fortune

500 firms. The integrated model explained much more of the variance in the dependent vari-

able, utilization, than did either TAM or TTF alone. Utilization is the behavior of employing

the technology to complete tasks such as the frequency of use or the diversity of applications

applied (Davis, Bagozzi, & Warshaw, 1989; Thompson, Higgins, & Howell, 1994). Adding

TTF constructs to TAM results in a significant improvement in explanatory power.

Considering this improved explanatory power, this paper incorporates TAM and TTF

into a combined model. Originally, TTF was designed to evaluate workplace technology

adoption and the impact of adoption on performance. According to Klopping and

McKinney (2004), this is the reason why the combined model of TTF and TAM has

always been applied to predicting management information systems (MIS). For instance,

as the service industry has also seen a rise in customer use of technology applications to

perform tasks such as self-services (Hui & Wan, 2009; Liu et al., 2012; Schrier, Erdem, &

Brewer, 2010), the technology has been given some attention in service literature in recent

years. With regard to self-service technology (SST), customers can use SSTs either

904 Y.J. Moon et al.

‘on-site’ or ‘off-site’. Examples of on-site applications include physical devices such as

department store touch-screen displays, hotel information kiosks, and self-check-out

counters at grocery stores (Chandler, 1995). Common examples of off-site applications

are automated telephone systems and online transaction web sites (Dabholkar, 1994).

The increasing presence of SST has changed the role of the customer to that of an

active participant in the service delivery process.

With this changed role of the customer, most service research focuses on the utilization

of a model that examines intentions to use an online or offline SST, making the integrated

TAM and TTF model a good fit. For example, Schrier et al. (2010) examined a fit of cus-

tomers’ in-room entertainment tasks and guest empowerment technologies (for example,

on-demand services, video gaming, mp3 player docking stations, etc.) by an integrated

TAM and TTF model. Accordingly, recent service studies have investigated a variety

of forms of online technologies and purchasing tasks via an integrated TAM and TTF

model (Chen, Gillenson, & Sherrell, 2002; Klopping & McKinney, 2004; Lederer,

Maupin, Sena, & Zhuang, 2000). For example, in an e-commerce study by Lederer

et al. (2000), self-reports of time online and frequency of use are related to how well

the consumer feels web technology fits the task. Chen et al. (2002) found the integrated

TAM and TTF model is effective in evaluating online shopping and information

seeking at a particular virtual online store.

Thus, recent combined models of TAM and TTF in service industries focus on custo-

mers’ SSTs. However, TTF was originally designed for organizational performance of

workplace technologies. Though an integrated TAM and TTF model needs to be based

on employee behaviors and attitudes in technology utilization, limitations due to customer

intervention exist in service industries. In SSTs, customers play the roles of users in utiliz-

ing technologies, and then it is not easy to discuss customers’ tasks from a perspective of

non-routineness and interdependence. Compared with previous service literature, this

study is the first attempt to investigate a fit of task and technology in a pure offline

service workplace environment, which is its academic contribution. Based on the signifi-

cant results of previous studies, as well as the increased explanatory power of the inte-

grated model, this study adopts the integrated model to attempt a better understanding

of restaurant employees’ POS usage.

Research hypotheses

Drawing on previous literature (Dishaw & Strong, 1999; Goodhue & Thompson, 1995;

Klopping & McKinney, 2004; Strong, Dishaw, & Bandy, 2006), we adopt an integrated

TAM and TTF model. Figure 1 demonstrates the proposed research model with the

eight hypotheses. The proposed study examines whether the current POS provides a

good fit to the tasks food service employees perform, and how the fit affects their POS

utilization.

Task, technology, and individual characteristics in TTF

The TTF theory states that the correspondence between information systems’ (IS) func-

tionality and task requirements leads to positive user evaluations, higher utilization, and

positive performance impacts (Goodhue, 1995; Goodhue & Thompson, 1995). According

to Goodhue (1995), in the context of IS research, tasks are defined as the actions of indi-

viduals who turn inputs into outputs. IS give value by being instrumental in some tasks,

and users will reflect this in their evaluations of the systems.

The Service Industries Journal 905

Following Goodhue’s (1995) two-dimensional construct of a general characteriz-

ation of tasks, task characteristics are defined as non-routineness (i.e. equivocality

and lack of analyzable search behavior) and interdependence (with other organizational

units). An employee’s perception of equivocality and task analyzability provides infor-

mation on the perceived uncertainty, re-utilization, and/or challenge experienced with a

task. The greatest uncertainty would be experienced when tasks are low in analyzability

and high in equivocality (Dunegan, Duchon, & Uhl-Bien, 1992). In other words, the

perceived task challenge will be highest when unexpected events occur with some regu-

larity and when the employee is unsure about the correct procedure to follow when

resolving task problems. Also, the more tasks are interdependent across other organiz-

ational units that have various task characteristics, the more employees would encoun-

ter task uncertainty and challenge. If the task is highly structured and the work is

relatively routine, the employee would be more certain about the correct procedure

to follow.

Rational users more frequently will use the information technology that enables them

to reduce the possibility of two task characteristics (i.e. non-routineness and interdepen-

dence) and consequently to complete their tasks with the greatest net benefit. For instance,

in a pizza restaurant, a server would often encounter non-routine events because of a

special order based on promotional menu. A server may make a mistake in placing

orders or notifying customers about sales offers. In this case, if the POS interface can

provide easily understandable sequences for placing orders by enabling simple and

error free order placement for combos and other sales offers (e.g. buy one pizza at X

get a second at Y), users of the POS technology in restaurants will think that the non-

routine aspects of their work are diminished by TTF. Also, when a customer orders a

pizza from the Internet, his/her one-click event is interrelated with kitchen preparations,

delivery services, credit card processings, and managerial accountings. If this event is

not processed at the same time by all of these work stations, then the workflow of all

Figure 1. Research model.

906 Y.J. Moon et al.

employees involved will be severely delayed. Thus, POS can reduce the need for the

detailed product and task training of individual servers.

Feratt and Vlahos (1998) investigated from the perspective of TTF the managerial

decision making support of computer-based IS, and they found that managers value IS

highly for their non-routine and interdependent tasks, such as allocating resources, evalu-

ating alternatives, identifying problems, and making short-term decisions. These managers

will complete their tasks best when the problem representation and any tools or aids all

support the processes required to perform that task. Therefore, the level of non-routineness

and the interdependence of the task would negatively affect users’ perceptions of TTF.

Thus, it is hypothesized that:

Hypothesis 1: The degree of task characteristics in non-routineness and interdependence will be negatively associated with TTF.

In the TTF literature, technology, also referred to as tools, includes a wide range of

information technology including hardware, software, data, user support services, or

any combination of these (Goodhue & Thompson, 1995). Systems implementation

research notes the need for a fit between tasks and technology (Palvia & Chervany,

1995). However, since the definition of technology includes a wide range of information

technology, different technology characteristics have been adapted according to different

information technology types used in the various contexts.

Dishaw and Strong (2003) use software maintenance tool functionality to better under-

stand technology characteristics. The software maintenance tool functionality variable was

designed to elicit a priori functionality anticipated by the programmer to be available in the

tool to complete the maintenance project (e.g. the functionality of the construct represen-

tations of entities, relationships, or processes in a diagram or model). Zigurs and Buckland

(1998) chose group support systems for testing the TTF model. In their study, the charac-

teristics of group support systems are classified into three groups: communication support,

process structuring, and information processing. Thus, although the technology can be

characterized from many different perspectives, regardless of the type of information tech-

nology, technology characteristics should reflect the generic functions needed for perform-

ing the specific tasks. In this manner, the technology characteristics of POS in restaurants

relate to credit/check cashing authorizations, cooking instructions, sales analyses, trans-

actions, and price look-ups (Weber & Kantamneni, 2002).

Previous studies have tested the theory of TTF and discovered that users’ perceptions are

significantly more positive when the technology reflects the functionality needed for their

specific tasks. For example, Murthy and Kerr (2004) showed that group members perceived

the TTF level more positively when communicating face to face for problem-solving tasks

and when communicating via group support systems for idea-generation tasks. Lim and

Benbasat (2000) tested their task-representation fit model and discovered that multimedia

representations are perceived as more fit than text-based representations in the context of

individual decision makers utilizing organizational data. Hence, it is hypothesized that:

Hypothesis 2: The degree of technology characteristics (credit/check cashing authorizations, cooking instructions delivered to kitchen, etc.) of restaurant POS systems will be positively associated with TTF.

The inclusion of individual experience is supported by Work Adjustment Theory,

which describes the relationship of the individual to his or her work environment

(Dawis & Lofquist, 1984). Based on Work Adjustment Theory, TTF posits that users’

experiences with particular information technology are generally associated with higher

utilization of that information technology (Thompson et al., 1994). Individual

The Service Industries Journal 907

characteristics such as users’ experiences with information technology have been reported

to play an important role in adopting technologies (Meuter, Bitner, Ostrom, & Brown,

2005), because the acceptance of the computer technology depends on the technology

itself, as well as the level of skill or expertise of the individual using the technology.

In testing the TTF model, individual abilities were operationalized as computer lit-

eracy (Goodhue, 1995) or as experience with the particular technology (Thompson

et al., 1994), experience with technology and tasks (Dishaw & Strong, 2003), and personal

innovativeness and computer playfulness (Agarwal & Prasad, 1998). Goodhue (1995)

tested the proposed TTF model in relation to user evaluations, while industry employees’

and staffs’ computer literacy was found to have a direct and significant effect on the TTF

construct. We can assume accordingly that users’ individual characteristics, such as com-

puter experience, are the key determinants of fit between task and technology. It is there-

fore hypothesized that:

Hypothesis 3: The degree of individual characteristics (or abilities) in using a restaurant POS system will be positively associated with TTF.

TTF, perceived usefulness, and perceived ease of use

TTF is defined as the extent to which information technology satisfies a user’s job needs

(Goodhue, 1995). Research has confirmed that TTF has a strong influence on a user’s per-

ception and behavior. A person engages in a behavior because he/she has evaluated the

benefits of engaging in that behavior and expects a certain positive result. Specifically,

we believe that TTF determines, in part, perceived usefulness. This path follows directly

from the definitions of both TTF and perceived usefulness. In TAM, perceived usefulness

refers to ‘the degree to which an employee believes that using a particular technology

would enhance his or her job performance’, while perceived ease of use means ‘the

degree to which an employee believes that using or learning a particular system would

be free of effort’ (Davis et al., 1989, p. 985).

If the technology provides a good fit for the task, users perceive that the technology is

useful for that task. Similarly, increased experience with the technology may lead to an

increased perception of usefulness as the user develops an understanding of how the func-

tionality of the information technology can be used to accomplish tasks (Lee et al., 2012).

Dishaw and Strong (1999) showed that TTF has a significant effect on perceived ease of

use, but its impact on perceived usefulness was not confirmed. According to Cooper and

Zmud (1990), a high compatibility between task and technology would stimulate infor-

mation technology acceptance by increasing users’ positive perceptions of the technology.

Venkatesh and Davis (2000) corroborated that job relevance, a similar concept to TTF,

defined as the extent to which technology is applicable to an individual job, is strongly

associated with perceived usefulness and perceived ease of use. Accordingly, we hypoth-

esize as follows:

Hypothesis 4: TTF has a positive relationship with perceived ease of use.

Hypothesis 5: TTF has a positive relationship with perceived usefulness.

Perceived usefulness, perceived ease of use, and intention to use

TAM explains that people’s behaviors are determined by intentions to perform the beha-

viors. Intention is influenced by attitude toward the behavior using technologies. TAM

identifies two beliefs that influence attitudes toward information technology use, perceived

usefulness and perceived ease of use, which serve as key independent variables in TAM.

908 Y.J. Moon et al.

The rationale for hypotheses and relationships among perceived usefulness, perceived ease

of use, and intention to use has been well established by previous research (Lee, Kozar, &

Larsen, 2003; Venkatesh et al., 2003). This study also suggested a relationship between

ease of use and usefulness. People create opinions concerning perceived usefulness, for

the most part, by judging a system’s capabilities for helping them accomplish their

jobs’ tasks (Venkatesh & Davis, 2000). Therefore, when users perceive that the

system offers to make their work complete with less effort and within a shorter time,

users’ perceptions about usefulness may increase. Moreover, all else being equal, if the

system is easier to use, then the more it is used, the more it can increase job performance.

Concurring with the TAM theory, it is predicted that the perception of the POS’s ease of

use will show a positive relationship with perceived usefulness, hence the following

hypothesis:

Hypothesis 6: Perceived ease of use has a positive relationship with perceived usefulness.

Furthermore, judgments about a system’s usefulness affected by an individual’s cog-

nitive matching of his/her job goals with the consequences of system use are expected to

influence intention to use. In addition, perceived ease of use could be a potential catalyst to

increasing the likelihood of a user’s usage intention (Venkatesh & Morris, 2000). Since the

POS system in restaurants is a required part of the job, if users perceive a system matches

with their job goals and has user-friendly interface such as a touch screen, a user’s inten-

tion to use is likely to increase. Thus, based on the TAM model theories, we hypothesize

that perceived ease of use and perceived usefulness will positively influence the subjects’

willingness to reuse the POS:

Hypothesis 7: Perceived ease of use has a positive relationship with intention to use.

Hypothesis 8: Perceived usefulness has a positive relationship with intention to use.

Research methodology

Data collection

The target population of this study is restaurant managers who have worked in the restau-

rants for at least three years. Data were collected via mailed surveys. A Directory of the

Restaurant Association in Midwestern state was used to collect the contact information

for restaurants in the state. The initial survey was mailed to 689 restaurants in 2007. In

the first round of surveys, a total of 30 surveys were returned to survey administrators:

the businesses were either closed or inaccurate addresses were used (15); or blank

surveys were returned because the facilities did not use POS (15). A week later, a

follow-up letter was sent to all the restaurants. Out of the 659 eligible restaurants, 178

of them returned the questionnaire, and of these, 167 were used for the final data analysis;

11 of the forms were not used because they lacked sufficient information. Thus, the ques-

tionnaire resulted in an effective response rate of 25.34%.

While Armstrong and Overton (1977) have suggested interviewing a non-respondent

sample to uncover any effects of non-response bias, such a task was not undertaken due to

the authors’ desires to preserve the anonymity of the survey respondents. An alternate

approach is one based on the belief that the vast majority of non-respondents’ profiles

mirror those of the average late respondents. A demographic characteristic comparison

was performed between the early respondents who were within the category of the first

10% of respondents to those who were within the last 10%. The chi-square tests confirmed

that no significant differences were detected, demographically, between the two groups of

The Service Industries Journal 909

respondents. Thus, we can conclude that non-response bias may not have been a serious

concern in this study.

Table 1 summarizes the profile of the respondents’ restaurant operations. The respon-

dent restaurants’ demographic information is also shown in Table 1. In terms of location,

over a quarter (28.1%) of the restaurants are located in small towns, while 71.9% are in

cities or suburbs. About 80% of the respondent restaurants belong to either casual/

family (59.3%) or quick service eating places (19.2%), and majorities (72.5%) of the res-

taurants serve American cuisine. About 60% of the restaurants (58.7%) have more than 30

employees. All of the restaurants included in the study use a POS system. Among the most

popular POSs are Micros-HIS (29.3%) and Aloha (10.2%). The others include Digital

Dining, Posi-Touch, Restaurant Manager, Panasonic, Wand, Dinerware, Jonas, Squirrel,

and Radiant. As explained in the data collection, the sample is confined to employees

who were experienced with POS.

Measurement

The design of the questionnaire was based on multiple-item measurement scales that have

been validated in previous studies. A multi-item approach was chosen to measure all con-

structs. Thirty-eight measurement items were used to capture the latent constructs. Based

on the review of the literature, 26 items were developed to measure the 4 elements of the

TTF model (i.e. individual characteristics, technology characteristics, task characteristics,

and perceived fit), and 12 items were drawn to measure the 3 TAM-related constructs (i.e.

perceived ease of use, perceived usefulness, and intention to use). The items in all scales

except for technology characteristics were measured on a seven-point, Likert-type scale

anchored from 1 (strongly disagree) to 5 (strongly agree).

Table 1. Demographic characteristics of samples.

Measure Frequency Percentage Measure Frequency Percentage

Location Number of employees Small town 47 28.1 5 or less 5 3.0 City/suburb 104 62.3 6 – 10 8 4.8 Rural 16 9.6 11 – 15 15 9.0

Restaurant type 16 – 20 14 8.4 Casual/family dining 99 59.3 21 – 30 27 16.1 Quick service 32 19.2 More than 30 98 58.7 Take-out/delivery 3 1.8 Name of POS Upscale/fine dining 9 5.4 Micros-HIS 49 29.3 Private country club 8 4.8 Aloha 17 10.2 Mixed type 8 4.8 Panasonic 14 8.4 Pub-tavern 3 1.8 Digital Dining 10 6.0 Others 5 2.9 Radiant 9 5.4

Food type Dinerware 8 4.8 American 121 72.5 Posi-Touch 7 4.2 Mexican 4 2.4 Jonas 6 3.6 Chinese 4 2.4 Restaurant Manager 5 3.0 Italian 17 10.2 Squirrel 5 3.0 Others 21 12.6 Wand 4 2.4

Others 33 19.7 Total 167 100 Total 167 100

910 Y.J. Moon et al.

Task characteristics

The task characteristics were adapted from measures developed by Goodhue (1995). Fol-

lowing Fry and Slocum’s (1984) suggestion of a general characterization of tasks,

Goodhue combined Perrow’s (1967) and Thompson’s (1967) dimensions and successfully

measured a two-dimensional construct of task characteristics: non-routineness (lack of

analyzable search behavior) and interdependence (with other organizational units). Five

measures of task characteristics were adopted from Goodhue and Thompson’s (1995)

study. Three items (i.e.‘I frequently deal with ill-defined business problems’, ‘I frequently

deal with ad hoc, non-routine business problems’, and ‘frequently the business problems I

work on involve answering questions that have never been asked in quite that form

before’) were used to measure the non-routine task characteristics. Two items (i.e.‘the

business problems I deal with frequently involve more than one e-business function’

and ‘the problems I deal with frequently involve more than one business function’)

were employed to measure the interdependence of task characteristics.

Technology characteristics

Measures of technology characteristics in restaurant operations were modified from the

POS support activities of retail businesses developed by Weber and Kantamneni (2002).

The following five items relevant to technology characteristics of restaurant operation

were developed as follows: (1) credit/check cashing authorizations using POS; (2)

cooking instructions delivered to the kitchen staff; (3) POS data used for sales analysis

and inventory control; (4) touch-screen monitor available at the POS terminal to increase

the speed and accuracy of transactions; and (5) price look-ups performed with the POS.

Individual characteristics

To measure the individual characteristics, we referred to Igbaria and Chakrabarti’s (1990)

scales of end users’ computer literacies. The individual characteristics measure employ-

ees’ experiences with using four different types of software in their routine work, and

they are (1) Microsoft Windows applications (e.g. Word, Excel, and PowerPoint); (2)

Internet, email, or web sites; (3) restaurant/business-specific applications (e.g. inventory

and purchase order); and (4) portable tools used for the restaurant operations (e.g. hand-

held devices, fixed touch-screen terminals, and wireless headsets for servers).

Task – technology fit

The perception of TTF is measured as users’ evaluation of TTF, and the constructs of per-

ceived TTF are defined as perception of individuals about TTF in previous studies

(Goodhue & Thompson, 1995; Gu & Wang, 2009). According to Gu and Wang’s

(2009) suggestion, we asked users to evaluate the systems and services based on fit

with their personal task needs. More specifically, we measured TTF by six items including

adequacy, usefulness, compatibility with the task, helpfulness, sufficiency, and fit the task.

Technology acceptance model

The measurement items of perceived ease of use, perceived usefulness, and intention to

use were taken from Davis et al. (1989) and Venkatesh and Davis (2000), and were slightly

modified to make them relevant to computers/POS in restaurant operations. Perceived

The Service Industries Journal 911

usefulness and perceived ease of use are the two measurements of beliefs in TAM used in

predicting technology acceptance. TAM measures the beliefs that using the computing

system would enhance the respondent’s job performance (perceived usefulness) and learn-

ing/using the computing system would be free of effort. For our study, PU and ease of use

constructs were measured with five and four items, respectively. The five items used to

measure perceived usefulness were as follows: to enhance performance, increase pro-

ductivity, enhance accuracy, enable me to accomplish tasks quicker, and find POS

useful in my job. The four items used to measure ease of use were as follows: ‘Learning

to operate POS is easy for me’, ‘I find it easy to get POS to do what I want it to do’, ‘It is

easy for me to become skillful at using POS’, and ‘The more I use the POS the easier it

becomes’. The intention to use measure asked respondents to rate their agreement on state-

ments about the use of POS in their working environment. The three items were intend to

use POS for the restaurant operation management, to continue using POS in the future, and

to expect the manager to support the use of POS in the future.

Data analysis

Unidimensionality assessment of the data included estimating reliability for those vari-

ables based on multiple items. A multi-step approach was used to check the convergent

and discriminant validity of the measures and to test the hypothesized relationships.

Seven constructs (i.e. individual/task/technology characteristics, TTF, perceived ease of

use, perceived usefulness, and intention to use) were evaluated using confirmatory

factor analysis (CFA). We employed structural equation modeling, using AMOS 7.0

based on a correlation matrix with maximum-likelihood estimation, to determine

whether the endogenous constructs (i.e. TTF, perceived usefulness, perceived ease of

use, and intention to use) were affected by the exogenous constructs (i.e. individual/

task/technology characteristics). Using AMOS, multivariate normality is determined by

using skewness and kurtosis test. In the current paper, skewness and kurtosis do not

appear to be significant problems in the data set. Using the benchmark +2.0, no items exhibited significant skewness and kurtosis (Arbuckle, 1997).

Results

Unidimensionality assessment

To assess the unidimensionality of each scale, internal consistency and CFAs were per-

formed. First, a reliability test was used to purify the measurement scale for each construct.

All coefficient alphas of the seven constructs surpassed Nunnally’s (1978) 0.70 criteria for

reliability acceptability. Items with weak contributions to coefficient alpha and low item-

to-total correlations (,0.40) were dropped. One item of individual characteristics and two

items of TTF were dropped. To examine an acceptable fit of the proposed measurement

model, each of the constructs was evaluated by examining the statistical significance of

each estimated loading, and the overall model fit indices were evaluated. All loadings

exceeded cutoff of 0.5, which was suggested by Hair, Black, Babin, Anderson, and

Tatham (2006) and each indicator t-value exceeded 10.55 (p , .001) (Table 2).

The x 2

fit statistics showed 643.62 with 523 degrees of freedom (p , .01). The root

mean squared error of approximation (RMSEA) was 0.04, less than the recommended

0.08 threshold (Newcomb, 1994) and the recommended 0.05 cutoff (Marsh & Hau,

1996). The comparative fit index (CFI ¼ 0.98) and the Tucker – Lewis coefficient (TLI

¼ 0.98) values exceeded the recommended 0.90 (Newcomb, 1994). Only the goodness-

912 Y.J. Moon et al.

Table 2. Description of items used to measure the independent and dependent constructs∗.

Standardized loading (t-value) AVE CCR

Item-to-total correlation

Cronbach’s a

(1) Task characteristics 0.67 0.91 0.91 Task 1 0.81 (fixed) 0.76 Task 2 0.77 (11.52) 0.76 Task 3 0.78 (11.58) 0.75 Task 4 0.88 (13.68) 0.82 Task 5 0.85 (13.06) 0.78

(2) Technology characteristics

0.76 0.94 0.94

Technology 1 0.88 (fixed) 0.84 Technology 2 0.89 (17.55) 0.86 Technology 3 0.95 (20.21) 0.91 Technology 4 0.90 (17.71) 0.87 Technology 5 0.72 (11.66) 0.69

(3) Individual characteristics

0.71 0.88 0.86

Individual 1 0.86 (fixed) 0.76 Individual 2 0.94 (14.44) 0.82 Individual 3 0.71 (10.85) 0.66 Individual 4

† 0.38

(4) TTF 0.79 0.96 0.95 TTF 1 0.89 (fixed) 0.83 TTF 2 0.93 (17.81) 0.81 TTF 3 0.91 (16.61) 0.84 TTF 4 0.92 (16.35) 0.82 TTF 5 0.86 (14.81) 0.84 TTF 6 0.82 (11.24) 0.79

x 2 ¼ 248.00, df ¼ 129 (p , .01); CFI ¼ 0.98; RMSEA ¼ 0.06; TLI ¼ 0.97; and GFI ¼ 0.87.

(5) Perceived ease of use

0.71 0.91 0.89

PEU 1 0.83 (fixed) 0.77 PEU 2 0.85 (12.59) 0.76 PEU 3 0.89 (13.27) 0.81 PEU 4 0.80 (10.78) 0.72

(6) Perceived usefulness

0.81 0.96 0.97

PU 1 0.89 (fixed) 0.92 PU 2 0.91 (19.56) 0.93 PU 3 0.93 (15.88) 0.92 PU 4 0.88 (14.78) 0.90 PU 5 0.90 (15.29) 0.91

(7) Intention to use 0.86 0.95 0.94 Intention 1 0.97 (fixed) 0.92 Intention 2 0.92 (24.07) 0.89 Intention 3 0.89 (21.38) 0.86

x 2 ¼ 643.62, df ¼ 523 (p , .01); CFI ¼ 0.98; RMSEA ¼ 0.04; TLI ¼ 0.98; and GFI ¼ 0.87.

Note: PU, perceived usefulness; PEU, perceived ease of use; CCR, composite construct reliability; AVE, average variance extracted. ∗Hypothesized model with standardized parameter estimates for the full sample (N ¼ 178). † Items were deleted after a reliability test.

The Service Industries Journal 913

of-fit index (GFI ¼ 0.87) was slightly smaller than the recommended 0.90. Thus, all stat-

istics supported the overall, satisfactory measurement quality, given the number of

indicators.

Common method variance refers to the amount of spurious covariance shared among

variables because of the common method used in collecting data (Buckley, Cote, & Com-

stock, 1990). Self-report surveys are the most common form of data collection in the social

sciences, including marketing (Malhotra, Kim, & Patil, 2006). In typical survey studies in

which the same rater responds to the items in a single questionnaire at the same point in

time, data are likely to be susceptible to common method variance (Lindell & Whitney,

2001). Harman’s (1976) single-factor test is arguably the most widely known approach

for assessing common method variance in a single-method research design. Typically,

in this single-factor test, all of the items in a study are subject to exploratory factor analy-

sis. Then, common method variance is assumed to exist if (1) a single factor emerges from

unrotated factor solutions or (2) a first factor explains the majority of the variance in the

variables (Podsakoff & Organ, 1986). In the result in our paper, seven factors with eigen

value above 1 were elicited and a first factor explains only 16.35% of total variance. There-

fore, Harman’s single-factor test shows that our model does not have any problems by

common method bias.

Next, to assess discriminant validity, Fornell and Larcker (1981) suggest using the

average variance extracted (AVE) shared between a construct and its measures. Evidence

of discriminant validity exists when the proportion of variance extracted in each construct

exceeds the square of the zero-order correlation coefficients, representing its correlation

with other factors. One pair of scales with the highest correlation was TTF and perceived

usefulness (F ¼ 0.61, F 2 ¼ 0.37) (Table 3). The AVE estimates for TTF and perceived

usefulness were 0.65 and 0.86, respectively, exceeding the square of the correlation coef-

ficient of 0.37, which indicates adequate discriminant validity. Therefore, according to this

assessment, the measures had satisfactory levels of validity.

Structural equation models and hypothesis testing

The hypotheses of the research model were tested with a structural equation path

model using AMOS version 7.0. The proposed model provided an adequate fit to the

data, x 2

(536) ¼ 739.12, p , .001; GFI ¼ 0.86; CFI ¼ 0.97; TLI ¼ 0.96; RMSEA ¼

0.05. Overall, the proposed model explained 45% of the variance in TTF (squared multiple

correlation (SMC) ¼ 0.45), 20% of the variance in perceived ease of use (SMC ¼ 0.20),

Table 3. Correlation estimates and construct means.

1 2 3 4 5 6 7 M SD

Individual characteristics 1.00 4.38 0.89 Technology

characteristics 0.29∗∗ 1.00 3.82 0.96

Task characteristics 0.46∗∗ 0.52∗∗ 1.00 3.54 0.89 TTF 0.37∗∗ 0.53∗∗ 0.49∗∗ 1.00 3.88 0.78 Perceived usefulness 0.32∗∗ 0.58∗∗ 0.52∗∗ 0.61∗∗ 1.00 4.15 0.86 Perceived ease of use 0.25∗∗ 0.56∗∗ 0.43∗∗ 0.50∗∗ 0.51∗∗ 1.00 3.66 0.84 Intention to use 0.26∗∗ 0.48∗∗ 0.38∗∗ 0.57∗∗ 0.51∗∗ 0.46∗∗ 1.00 4.5 0.85

∗p , .05 (two-tailed). ∗∗p , .01 (two-tailed).

914 Y.J. Moon et al.

57% of the variance in perceived usefulness (SMC ¼ .57), and 40% of the variance in

intention to use (SMC ¼ 0.40).

Within the model, the estimates of the structural coefficients provided the basic tests of

the hypothesized relationships. The effects of the three antecedents on TTF were first

addressed (Hypotheses 1 – 3). The expected relationship between individual characteristics

and TTF (Hypothesis 1) is supported by the positive path coefficient (g1 ¼ 0.13), statisti-

cally significant at the p , .05 level. Technology characteristics affected TTF (g2 ¼ 0.47,

p , .001), thus supporting Hypothesis 2. Hypothesis 3, concerning the negative relation-

ship between task and TTF, is also supported (g3 ¼ 20.19, p , .05). Fit decreases as task

requirements increase; that is, tasks can become too large and complex for information

technology to provide adequate support (Dishaw & Strong, 1999).

Hypotheses 4 and 5 suggest that TTF has a positive effect on perceived usefulness and

ease of use, respectively. The empirical results suggest that TTF does increase perceived

usefulness (b1 ¼ 0.71, p , .001) and ease of use (b2 ¼ 0.93, p , .001). Thus, Hypoth-

eses 4 and 5 are supported. Next, the typical relationships among perceived ease of use,

perceived usefulness, and intention to use, Hypotheses 6 – 8, are all supported. Perceived

ease of use has a significant impact on perceived usefulness (b3 ¼ 0.42, p , .001). Both

perceived ease of use (b4 ¼ 0.27, p , .01) and perceived usefulness (b5 ¼ 0.43, p ,

.001) increase intention to use; thus Hypotheses 7 and 8 are supported.

In addition, indirect effects were examined and tested for significance using the Boot-

strap estimation procedure in AMOS (Benetti & Kambouropoulos, 2006). It displays the

indirect effects and their associated 95% confidence intervals. As shown in Table 4,

Table 4. Structural model results.

Hypothesized relationship

Proposed model

Standardized path coefficient t-Value

Hypotheses testing results

Direct effect H1 Task characteristics � TTF (g3) 20.19 22.52∗ ∗∗ Supported H2 Technology characteristics �

TTF (g2) 0.47 6.34∗∗∗ Supported

H3 Individual characteristics � TTF (g1)

0.13 2.33∗∗ Supported

H4 TTF � PEU (b2) 0.93 7.11∗∗∗ Supported H5 TTF � PU (b1) 0.71 9.37∗∗∗ Supported H6 PEU � PU (b3) 0.42 7.15∗∗∗ Supported H7 PEU � intention to use (b4) 0.27 4.68∗∗∗ Supported H8 PU � intention to use (b5) 0.43 4.82∗∗∗ Supported Indirect effect Individual characteristics � intention

to use 0.09 2.84∗∗∗

Technology characteristics � intention to use

0.13 1.75 ns

Task characteristics � intention to use

20.13 1.64 ns

TTF �intention to use 0.72 9.20∗∗∗

Note: x 2

(536) ¼ 739.13, p , .001; CFI ¼ 0.97; TLI ¼ 0.96; GFI ¼ 0.86; and RMSEA ¼ 0.05. ∗p , .05 (1.96). ∗∗p , .01 (2.32). ∗ ∗ ∗p , .001 (.2.57).

The Service Industries Journal 915

individual characteristics exerted a significant indirect effect on intention to use via TTF,

perceived usefulness, and perceived ease of use (g ¼ 0.09, p , .001). The indirect effect

of TTF on intention to use (b ¼ 0.72, p , .001) via perceived usefulness and perceived

ease of use was also significant. However, technology characteristics and task character-

istics showed no significant indirect effects.

Discussions

During the past few decades, two significant models of information technology utilization

behavior have emerged in the MIS literature, namely the TAM and the TTF model.

Although both models explore the factors that explain software utilization and its link

with user performance, these two models offer different perspectives on utilization behav-

ior. According to Dishaw and Strong (1999), TAM focuses on attitudes toward using a

particular information technology, which users develop based on perceived usefulness

and ease of use. On the other hand, TTF focuses on the match between user task needs

and the available functionality of the information technology. In the view of Dishaw

and Strong (1999), an integrated model of TAM and TTF should lead to a better under-

standing of choices regarding information technology usage. Under this proposition,

this study adopts an integrated TAM and TTF construct in order to provide more useful

results than either TAM or TTF alone, regarding POS usage in restaurant operations.

In the TTF part of the integrated model, an explicit conceptualization of task is its non-

routineness and its interdependence (Fry & Slocum, 1984; Goodhue, 1995; Perrow, 1967;

Thompson, 1967). This conceptualization reflects a fundamentally important aspect of

task that is particularly relevant in a restaurant setting. Because guest-contact restaurant

employees frequently deal with special orders based on a promotion, they face non-

routine service encounters, and therefore, the chances are high that the service provider

will likely face many unexpected service failures due to his/her lack of understanding

about the specific promotional information. In addition, all restaurant functions are

mutually interrelated. For example, the point of receiving customers’ orders, the functions

of forwarding those orders to the kitchen, the deliverance of food to customers, and the

check settlement are all closely interdependent, and a failure in one of these functional

areas will cause a delay in another function; these delays may, in turn, result in disgruntled

customers, as well as inefficient operations. Furthermore, the characterization of technol-

ogy is based on key dimensions from prior studies and is specifically linked to task

demands. The technology characteristics reflect task demands requiring POS applications

in restaurant operations: delivering cooking instructions to kitchen, cashing, ordering, con-

ducting sales analyses, and controlling inventory.

In sum, the lower the level of non-routineness and interdependence of task character-

istics, the higher the level of TTF. In other words, if task characteristics in restaurant oper-

ations deal less frequently with ad hoc, non-routine business problems and lower levels of

interdependence, then a POS system will result in much better TTF by providing higher

system reliability, convenience, and ease of use. By decreasing the task characteristics

of non-routineness and interdependence, restaurant employees who use a system, which

is equipped with more flexibility to accommodate non-routine business problems, will

eventually be able to streamline their operations and increase customer satisfaction,

which in turn will attract more POS users.

The findings also suggest that as users gain more experience with information technol-

ogy applications, such as Microsoft Windows applications, Internet, or restaurant/

business-specific applications, the fit of the POS tool and restaurant tasks is enhanced.

916 Y.J. Moon et al.

As shown in Table 4, the general empirical results support our propositions about three

determinants of user evaluation of TTF, which was found to be influenced directly by

task characteristics, technology characteristics, and individual characteristics. Technology

characteristics provide the strongest total effect, followed by task characteristics and indi-

vidual characteristics (total effects of 0.47, 20.19, and 0.13, respectively).

Next, as stated earlier, TAM and TTF are combined in this study to capture two differ-

ent aspects of users’ choices to utilize information technology. TAM assumes users’

beliefs and attitudes toward a particular information technology largely determine

whether they utilize the information technology. TTF proposes that users choose to use

information technology that provides benefits, such as improved job performances. By

combining both models, we assumed that users’ perceptions such as perceived ease of

use and perceived usefulness were determined by users’ rationales for expected conse-

quences from using the information technology, namely TTF. Although Dishaw and

Strong’s (1999) analysis indicated that there is a non-significant relationship between

TTF and perceived usefulness and also that TTF affects perceived ease of use and intention

to use, which is not consistent with Chen’s findings (Chen et al., 2002), the empirical

results indicate that TAM’s two independent variables, perceived ease of use and per-

ceived usefulness, are dependent on the level of TTF (total effects of 0.93 and 0.71,

respectively) and are important contributors to explaining intention to use (total effects

of 0.27 and 0.43, respectively). Perceived ease of use of an information technology

should be mainly determined by the functionality available in the tool used by individuals

in carrying out their tasks. If POS provides a good fit for restaurant employees’ tasks, users

should perceive that the POS technology is easy to use as well as useful for their tasks.

Furthermore, although the TAM literature presents external variables, such as system

usability, having an impact on perceptions (Venkatesh & Davis, 2000), little has been

empirically shown to represent the antecedents of perceived ease of use and perceived use-

fulness. We denote that TTF can represent an objective construct for these exogenous vari-

ables (i.e. perceived usefulness and perceived ease of use), causing a variance of

perception in technology acceptance. Finally, we also found that there was a typically sig-

nificant relationship between ease of use and usefulness (total effects of 0.42).

Moreover, we tested the indirect effects of task/technology/individual characteristics

on intention to use. The findings show that individual characteristics (i.e. employees’

experiences with using computers) affect TTF directly, and it also has an indirect effect

on intention to use mediated by TTF, perceived usefulness, and perceived ease of use.

These findings are supported by considerable research showing that computer experience

influences intentions, use, and/or performance (Karahanna et al., 1999; Thompson et al.,

2006; Venkatesh et al., 2003). Apart from TTF, we can propose that as an individual

gains experience with the technology, he/she has intentions (positive and/or negative)

that are internalized, and in turn, these lead to actual computer usage. Also, with the

test of TTF’s indirect effects on intention to use strongly supported by the data, it has

become obvious that TTF adequately reflects the extent of intention to use. These findings

suggest that as much as TTF plays an important role in shaping and changing users’ per-

ceptions about POS in restaurants, TTF is equally influential when it comes to the

decisions of these users regarding their intentions to actually use POS.

Theoretical and managerial implications

For practitioners, this study suggests that the decision to adopt the actual system depends

on task fit, ease of use, and usefulness. Clearly, this study suggests restaurant managers

The Service Industries Journal 917

carefully evaluate the methods and the reasons for employees using the POS system. More

specifically, considerations include the nature of the design of technological functions that

support restaurant managers and employees, and consequently, the degree to which the

POS technology fits a specific restaurant’s tasks. For example, employees use the POS

for usual restaurant operations such as credit/cashing authorization, delivery of cooking

instructions, transactions, or pricelist reference. For each of these potential tasks that a

POS may serve, the developer should assess the efficiency of the POS system’s fit for

these needs. Assessment requires the developer’s awareness of sufficiently detailed infor-

mation, accuracy of the information, and ease of locating that information.

Although restaurant tasks can be clearly classified into pre-process, in-process, and

post-process from the perspective of restaurants’ functions (Kimes, 2004), little consider-

ation focuses on the nature of restaurant tasks’ characteristics, such as aberrations to rou-

tines and interdependent aspects. Thus, the current investigation attempts to adapt the

combination of TAM and TTF to the restaurant industry.

Limitations and future research directions

While this study is careful to assure conducting a valid study, limitations do exist. One

significant advantage of this study is that the collection of data involves restaurants’

employees who had experience with restaurants’ operations and a POS system. The down-

side of the restricted requirements for participation is a relatively modest sample. While

the data are realistic and the model’s fit to the statistics is acceptable, the relatively

small sample size necessitates further testing of the proposed integrated model in other

situations.

Second, future study needs to consider additional measurement items related to indi-

vidual characteristics constructs, such as position or tenure profile. The current study

measures individual characteristics with users’ computer experience because the computer

literacy of an end user reflects the user’s knowledge of software-related tasks so that com-

puter literacy of individual characteristics can discern the fit between restaurant task and

POS. However, future study needs to broaden the effect of individual characteristics

through including other possibly important variables.

Some concerns are attributable to the specific situation of the restaurant industry. One

limitation of the study arises from the incomprehensive measurement of restaurants’ tasks.

The conceptualization of task characteristics in this study has its basis in Goodhue’s

general operationalization. In other words, the foundation of the task’s dimension rec-

ommended here is the underlying non-routineness and interdependence of the task as a

whole. However, most tasks involve various subtasks that constitute phases of an

overall task. For example, according to Kimes (2004), categorization of restaurants’ oper-

ations consists of four major subtasks: pre-process (e.g. seat to greet, greet to drinks, and

drinks to order), production (e.g. order to entrée served), in-process (entrée served to check

presented), and post-process (final settlement, departure to bussed, and bussed to be

reseated). Last, the constructs measured for the study are not specific for using POS,

but rather are general in terms of use of technology. One reason may be a scarcity of pre-

vious research on the use of technology in restaurants. The measurement of some con-

structs (e.g. task characteristics and TTF) needs refinement to reflect industry-specific

characteristics.

Needed future research, therefore, would provide more detailed typology of tasks. A

usable coding scheme for types of task types would ease identification of tasks’ topology

(McGrath, Arrow, Gruenfeld, Hollingshead, & O’Connor, 1993). In addition, future work

918 Y.J. Moon et al.

needs to investigate whether or not the intention to use, and the actual utilization, depends

not only on perceived usefulness and ease of use, but also on the degree to which the tech-

nology’s characteristics, such as tool functionality, match the needs of the task. Despite

these limitations, the expectation is that this study’s combined TAM and TTF model is

an aid to researchers’ and practitioners’ better understanding the reasons for restaurant

employees’ choosing to use POS for their tasks and the degree to which the POS technol-

ogy’s characteristics fit with tasks’ characteristics in restaurant operations leading to users’

choices. Such understanding is also important for POS software developers to consider

when creating new products or upgrading systems.

References

Agarwal, R., & Prasad, J. (1998). A conceptual and operational definition of personal innovativeness in the domain of information technology. Information Systems Research, 9(2), 204 – 215.

Arbuckle, J. L. (1997). Amos users’ guide version 3.6. Chicago, IL: SmallWaters. Armstrong, J. S., & Overton, T. S. (1977). Estimating nonresponse bias in mail surveys. Journal of

Marketing Research, 14(3), 396 – 402. Benetti, C., & Kambouropoulos, N. (2006). Affect-regulated indirect effects of trait anxiety and trait

resilience on self-esteem. Personality and Individual Differences, 41, 341 – 352. Buckley, M. R., Cote, J. A., & Comstock, S. M. (1990). Measurement errors in behavioral sciences:

The case of personality/attitude research. Educational and Psychological Measurement, 50(3), 447 – 474.

Chandler, S. (1995, September 11). The grocery cart in your PC. Business Week, pp. 63 – 64. Chen, L., Gillenson, M., & Sherrell, D. (2002). Enticing online consumers: An extended technology

acceptance perspective. Information and Management, 39(8), 705 – 719. Cohen, W. M., & Levinthal, D. A. (1990). Absorptive capacity: A new perspective on learning and

innovation. Administrative Science Quarterly, 35, 128 – 152. Collins, G. R., Cobanoglu, C., & Malik, T. (2003). Hospitality information technology: Learning

how to use it. Dubuque, IA: Kendall-Hunt. Cooper, R. B., & Zmud, R. W. (1990). Information technology implementation research: A techno-

logical diffusion approach. Management Science, 36(2), 123 – 139. Dabholkar, P. (1994). Technology based self service delivery: A classification scheme for develop-

ing marketing strategies. Advances in Services Marketing and Management, 3, 241 – 271. Davis, F. D., Bagozzi, R., & Warshaw, P. R. (1989). User acceptance of computer technology: A

comparison of two theoretical models. Management Science, 35(8), 982 – 1003. Dawis, R. V., & Lofquist, L. H. (1984). Psychological theory of work adjustment: An individual-

differences model and its applications. Minneapolis: University of Minnesota Press. Dishaw, M. T., & Strong, D. M. (1999). Extending the technology acceptance model with task-tech-

nology fit constructs. Information & Management, 36(1), 9 – 21. Dishaw, M. T., & Strong, D. M. (2003). The effect of task and tool experience on maintenance case

tool usage. Information Resources Management Journal, 16(3), 1 – 16. Dunegan, K. J., Duchon, D., & Uhl-Bien, M. (1992). Examining the link between leader – member

exchange and subordinate performance: The role of task analyzability and variety as modera- tors. Journal of Management, 18, 59 – 76.

Feratt, T. W., & Vlahos, G. E. (1998). An investigation of task-technology fit for managers in Greece and the US. European Journal of Information Systems, 7(2), 123 – 135.

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable vari- ables and measurement error. Journal of Marketing Research, 18(1), 39 – 40.

Fry, L. W., & Slocum, J. W. (1984). Technology, structure, and workgroup effectiveness: A test of a contingency model. Academy of Management Journal, 27(2), 221 – 246.

Goodhue, D. L. (1995). Understanding user evaluations of information systems. Management Science, 41(2), 1827 – 1844.

Goodhue, D. L., & Thompson, R. L. (1995). Task-technology fit and individual performance. MIS Quarterly, 19(2), 213 – 236.

Gu, L., & Wang, J. (2009). A study of exploring the ‘Big Five’ and task technology fit in web-based decision support systems. Issues in Information Systems, 10(2), 210 – 217.

The Service Industries Journal 919

Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate data analysis (6th ed.). Englewood Cliffs, NJ: Prentice Hall.

Harman, H. H. (1976). Modern factor analysis (3rd ed.). Chicago, LA: The University of Chicago Press.

Honeycutt, E. D. Jr., Thelen, T., Thelen, S. T., & Hodge, S. K. (2005). Impediments to sales force automation. Industrial Marketing Management, 34(4), 345 – 354.

Hui, T.-K., & Wan, D. (2009). Who are the online grocers? The Service Industries Journal, 29(11), 1479 – 1489.

Igbaria, M., & Chakrabarti, A. (1990). Computer anxiety and attitudes towards microcomputer use. Behaviour & Information Technology, 9(3), 229 – 241.

Jones, P. (1985). The spread of article numbering and retail scanning in Europe. The Service Industries Journal, 5(3), 273 – 279.

Karahanna, E., Straub, D. W., & Chervany, N. L. (1999). Information technology adoption across time: A cross-sectional comparison of pre-adoption beliefs. MIS Quarterly, 23(2), 183 – 213.

Khazanchi, D. (2005). Information technology (IT) appropriateness: The contingency theory of ‘fit’ and IT implementation in small and medium enterprises. The Journal of Computer Information Systems, 45(3), 88 – 95.

Kimes, S. (2004). Restaurant revenue management at Chevys Arrowhead. Cornell Hotel and Restaurant Quarterly, 45(1), 52 – 67.

Klopping, I. M., & McKinney, E. (2004). Extending the technology acceptance model and the task- technology fit model to consumer e-commerce. Information Technology, Learning, and Performance Journal, 22(1), 35 – 48.

Lederer, A. L., Maupin, D. J., Sena, M. P., & Zhuang, Y. (2000). The technology acceptance model and the World Wide Web. Decision Support Systems, 29(3), 269 – 282.

Lee, Y., Kozar, K. A., & Larsen, K. R. T. (2003). The technology acceptance model: Past, present, and future. Communications of the Association for Information Systems, 12, 752 – 780.

Lee, Y.-K., Park, J.-H., Chung, J., & Blakeney, A. (2012). A unified perspective on the factors influ- encing usage intention toward mobile financial services. Journal of Business Research, 65(11), 1590 – 1599.

Lim, K. H., & Benbasat, I. (2000). The effect of multimedia on perceived equivocality and perceived usefulness of information systems. MIS Quarterly, 24(3), 449 – 271.

Lindell, M. K., & Whitney, D. J. (2001). Accounting for common method variance in cross-sectional research designs. Journal of Applied Psychology, 86(1), 114 – 121.

Liu, S.-F., Huang, L.-S., & Chiou, Y.-H. (2012). An integrated attitude model of self-service tech- nologies: Evidence from online stock trading systems brokers. The Service Industries Journal, 32(11), 1823 – 1835.

Malhotra, M. R., Kim, S. S., & Patil, A. (2006). Common method variance in IS research: A com- parison of alternative approaches and a reanalysis of past research. Management Science, 52(12), 1865 – 1883.

Marsh, H. W., & Hau, K.-T. (1996). Assessing goodness of fit: Is patrimony always desirable? Journal of Experimental Education, 64, 364 – 390.

Mathieson, K. (1991). Predicting user intentions: Comparing the technology acceptance model with the theory of planned behavior. Information Systems Research, 2(3), 173 – 191.

McGrath, J. E., Arrow, H., Gruenfeld, D. H., Hollingshead, A. B., & O’Connor, K. M. (1993). Group tasks and technology: The effects of experience and change. Small Group Research, 24, 406 – 420.

Meuter, M. L., Bitner, M. J., Ostrom, A. L., & Brown, S. W. (2005). Choosing among alternative service delivery modes: An investigation of customer trial of self-service technologies. Journal of Marketing, 69(2), 61 – 83.

Murthy, U. S., & Kerr, D. S. (2004). Comparing audit team effectiveness via alternative modes of computer-mediated communication. Auditing: A Journal of Practice & Theory, 23(1), 141 – 152.

Newcomb, M. D. (1994). Drug use and intimate relationships among women and men: Separating specific from general effects in prospective data using structural equation models. Journal of Consulting and Clinical Psychology, 62(3), 463 – 476.

Nunnally, J. C. (1978). Psychometric theory (2nd ed.). New York, NY: McGraw-Hill. Palvia, S. C., & Chervany, N. L. (1995). An experimental investigation of factors influencing pre-

dicted success in DSS implementation. Information and Management, 29(1), 43 – 53.

920 Y.J. Moon et al.

Perrow, C. (1967). A framework for the comparative analysis of organizations. American Sociological Review, 32(2), 194 – 208.

Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organizational research: Problems and pro- spects. Journal of Management, 12(4), 531 – 544.

Rangarajan, D., Jones, E., & Chin, W. (2005). Impact of sales force automation on technology- related stress, effort, and technology usage among salespeople. Industrial Marketing Management, 34(4), 345 – 354.

Schrier, T., Erdem, M., & Brewer, P. (2010). Merging task-technology fit and technology acceptance models to assess guest empowerment technology usage in hotels. Journal of Hospitality and Tourism Technology, 1(30), 201 – 217.

Strong, D. M., Dishaw, M. T., & Bandy, D. B. (2006). Extending task technology fit with computer self-efficacy. DATA BASE, 37(2/3), 96 – 107.

Sundaram, S., Schwarz, A., Jones, E., & Chin, W. W. (2007). Technology use on the front line: How information technology enhances individual performance. Journal of the Academy of Marketing Science, 35(1), 101 – 112.

Thompson, J. D. (1967). Organizations in action. New York, NY: McGraw-Hill. Thompson, R., Compeau, D., & Higgins, C. (2006). Intention to use information technologies: An

integrative model. Journal of Organizational and End User Computing, 18(3), 25 – 46. Thompson, R. L., Higgins, C. A., & Howell, J. M. (1994). Influence of experience on personal com-

puter utilization: Testing a conceptual model. Journal of Management Information System, 11(1), 167 – 187.

Tricon Global Restaurant. (2012). Retrieved from http://www.yum.com Venkatesh, V., & Davis, F. (2000). A theoretical extension of the technology acceptance model: Four

longitudinal field studies. Management Science, 46(2), 186 – 204. Venkatesh, V., Morris, M., Davis, G., & Davis, F. (2003). User acceptance of information technol-

ogy: Toward a unified view. MIS Quarterly, 27(3), 425 – 478. Venkatesh, V., & Morris, M. G. (2000). Why don’t men ever stop to ask for directions? Gender,

social influence, and their role in technology acceptance and user behavior. MIS Quarterly, 24, 115 – 139.

Weber, M. M., & Kantamneni, S. P. (2002). POS and EDI in retailing: An examination of underlying benefits and barriers. Supply Chain Management: An International Journal, 7(5), 311 – 317.

Zigurs, I., & Buckland, B. K. (1998). A theory of task/technology fit and group support systems effectiveness. MIS Quarterly, 22(3), 313 – 334.

The Service Industries Journal 921

Copyright of Service Industries Journal is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.