Information technology adoption and use has been extensively studied in the industrialized world, yet there is dearth of knowledge regarding less developed countries (LDCs). The need for understanding how and why technology has or has not adopted by these potential knowledge workers is important for managers and providers alike. This study extends Davis’ diffusion model ¾ the Technology Acceptance Model (TAM) ¾ and builds toward an understanding of the problem and possible solutions.

Toward that end, this study tested TAM successfully in five LDC Arab nations in the Middle East. In addition to demonstrating the robustness of TAM in the developing world, the study extends TAM methodology. This was accomplished by expanding TAM’s artifact of study from specific technologies to generic personal computing. This is critical for understanding information technology adoption and use in the developing world where there may be formidable barriers to computer technology in general.
 

Keywords: Diffusion of Innovation, Technology Acceptance Model, less developed countries, Arab culture and IT, IT cross-cultural research

Introduction

Research into information technology (IT) adoption and use has been motivated by the desire to predict the factors which lead to IT use (Thompson and Rose, 1994). Over the years, this area of study has received a great deal of attention and includes a wealth of research. A literature review by Prescott and Conger (1995), for instance, included seventy IT adoption and use articles based on the diffusion of innovation (DOI) paradigm alone.

Study of diffusion and adoption of new technologies recently gained new attendance after being very popular during the 1980s. During this period, organizations throughout the Western developed countries started to use computer technology, especially personal computers, on a large scale. The new wave of attention was at least partly initiated by the increasing diffusion of networking technologies, such as the Internet, and the decreasing importance of geographical distances. Yet, while research in this area has been extensive, it has taken place almost exclusively in developed economies.

The overwhelming majority of information technology adoption and use research has been carried out in the technologically developed world. In fact, of the one hundred IT adoption and use studies covered in two recent literature reviews (Prescott and Conger, 1995; Thompson and Rose, 1994), no studies took place in lesser developed countries (LDCs). Perhaps this is understandable in that the majority of academic institutions and IT users are located in the industrialized world. The consequence, however, is that study of these phenomenon in the less developed world, where IT has thus far failed to transfer effectively (Goodman and Press, 1995; Knight, 1993; Mahmood, Gemoets, & Gosler 1995; Odedra, Lawrie, & Goodman, 1993), has been severely limited.

Business relations are intensifying and becoming more complex, not only within individual countries, but also across national borders. The way business is done is increasingly being shaped by the use of computer technology, especially in developed countries. The lack of similar developments in other parts of the world could widen existing cultural gaps.

Mutual understanding between business players from different parts of the world and cultural backgrounds is essential to ensure smooth interaction between the players and mutually beneficial relationships. Currently less developed countries not only open markets for organizations operating in highly competitive environments in other parts of the world, they are also becoming serious competitors themselves exporting their goods and services to the rest of the world. Over the past years many companies from developed nations have been globalizing their business, while specializing at the same time. As this trend involves more and more LDCs as equal business partners interaction across national borders, is intensifying.

The need for understanding how and why technology has or has not been adopted by these hundreds of millions of potential computer literate knowledge workers is important for both managers and providers of technology alike (Danowitz, Nassef, & Goodman, 1995; Mahmood, et al., 1995; Mesher, Goodman, Snyder, Briggs, & Press, 1992; Nidumolu and Goodman, 1993). As stated, most of the major educational facilities and research institutions are located in Northern America and Western Europe. Research being done in these universities is generally localized to the regions containing the institutions. As a result, most of the research being done is carried out in the developed world. Additionally, distribution of knowledge out of universities is likely inhibiting the distribution of knowledge of information technology adoption and use within the developing world to the developed world.

Many students from LDCs attend Western universities and go back to their home countries. Students from LDCs who study abroad do so not only to apply technical and business knowledge but also to bring back some understanding about Western culture they experienced while they were living abroad. A transfer of cultural knowledge in the opposite direction is not necessarily happening.

The relatively few research institutions in the developing world traditionally do not attract large numbers of students from the developed world. As a result, members of the developed world need to rely on themselves to begin a more concerted effort to acquire a greater level of understanding of LDC cultures. A mindful study of information technology adoption and use in developing nations should be pursued to advance our understanding of why information technology has consistently failed.

Therefore, the research question being investigated in this study is: Do explanations for IT diffusion (specifically, the Technology Acceptance Model) originating in the developed world also apply to developing nations and cultures? The culture being studied in this research is the Arab world.

Literature Review

IT Adoption and Use in Developing Countries

Even in the technologically developed West, IT adoption and use is impeded by such barriers as a lack of "top management support," poor quality IS design, or inadequately "motivated and capable" users (Kwon and Zmud, 1987, p. 228). In the developing world, equivalent barriers appear to be often insurmountable (Danowitz, et al., 1995; Knight, 1993; Mahmood, et al., 1995; Nidumolu and Goodman, 1993). While dysfunctional behavior such as non-use of equipment has been documented in both environments, it seems most troublesome in regions like sub-Saharan Africa which have little or no functioning IT (Odedra, et al., 1993).

Some of these problems are attributable to a lack of national infrastructure (Odedra, et al., 1993), capital resources (Goodman and Press, 1995), or government policies set in place to prevent technology transfer (Goodman and Green, 1992). However, there are reports of countries where sufficient resources and government support exist yet technology has still failed to transfer effectively (Atiyyah, 1989; Goodman and Green, 1992; Ibrahim, 1985). While the problem has been documented, little more than anecdotal evidence as to why systems fail to transfer or what motivates use of computer technology is available. Explanations supported by empirical evidence are needed to address these issues in order to both develop business opportunities for IT producers and to provide value to these nations which often have the scarcest resources.

Diffusion of Innovations and LDCs

While information technology-specific adoption and use has not been evaluated across cultures of varying technological development levels, diffusion of non-IT innovations has been tested successfully (Rogers, 1995). In Roger’s seminal work Diffusion of Innovations (1995), five antecedents: relative advantage, compatibility, complexity, observability, and trialability affecting diffusion (adoption and use) of a technology were identified. Included among hundreds of studies were a variety of nations representing both developed and developing economies of the world. Tests of these models in the developing world have primarily centered around diffusion of "farm innovations" (Rogers, 1995, p. 59), but also include diffusion of "nutrition, health, and family planning innovations to villagers" (Rogers, 1995, p.59).

While these studies do suggest that information technology adoption and use models tested in developed nations may be applicable to LDCs, no hard evidence presently exists. Of the seventy IT-based studies which either confirmed or extended the Rogers’ diffusion of innovation (DOI) model evaluated by Prescott and Conger (1995), none were conducted within developing nations. Without empirical verification, therefore, it is unclear whether diffusion of IT models validated in developed countries will apply equally well to LDCs.

A suitable first model for testing in the developing world would be one which has shown robustness across the spectrum of IT applications. This robust model should have the highest probability of success in future transfers across economic and cultural boundaries. Additionally, this model would have been well researched and, as a result, give easily interpretable results. Davis’ Technology Acceptance Model (TAM) is a model closely related to Rogers’ DOI model which has demonstrated this robustness. For this reason, TAM was selected as an appropriate model for studying IT diffusion in LDCs.

Constructs in the Technology Acceptance Model (TAM)

Davis’ TAM is a well respected model of IT adoption and use. Based on the more general Theory of Reasoned Action (Fishbein and Ajzen, 1975), TAM has been tailored to explain computer usage. The two antecedents to computer technology use are:

• Perceived Ease of Use (PEU), defined as "the degree to which a person believes that using a particular system would be free of effort" (Davis, 1989, p. 320), and

• Perceived Usefulness (PU), defined as "the degree to which a person believes that using a particular system would enhance his or her performance" (Davis, 1989, p. 320).

Robustness of TAM

As is true with most IT adoption and use models, TAM has not been studied outside of the industrialized world. In fact, it has only been studied in two countries outside of North America (Straub, 1994; Straub, Keil, & Brenner, 1997). However, within the industrialized world, it has proven to be applicable across a wide variety of IT applications (see Table 1 for examples).

 
 

	Study	Computer Technology Examined
	(Adams, et al., 1992)	E-Mail, Voice-Mail, graphics, spreadsheet, word processing software
	(Davis, 1986; Davis, 1989)	E-Mail and graphics
	(Karahanna, 1993)	E-Mail and Voice-Mail
	(Goette, 1995)	Adaptive technology for the disabled
	(Hendrickson, et al., 1993)	Spreadsheet, DBMS
	(Mathieson, 1991)	Spreadsheet software

Although TAM has not been studied in LDCs, it has been a consistently good predictor of IT use in technologically advanced countries. As a result, it is a sensible choice for testing in the developing world as well.

IT Use in the Arab World

Arab society provides a robust test of the TAM model in one of the most complex cultural and social systems in the world. Furthermore, according to one descriptive analysis of the region, there is a tremendous variation in the uses of IT (Goodman and Green, 1992). Egypt, for example, was noted as having the largest and most internationally oriented computer system in the area, and uses IT in most, if not all, governmental agencies and non-governmental organizations. In contrast, Jordan, uses computers in public and private domains, also uses them to maintain extensive cultural and archaeological archives. Although the institutional leaders in Saudi Arabia view computers as signs of modernization, many of their systems are incompatible with each other (Atiyyah, 1989). While the uses of IT are varied, the common bond of computer use in the region is one of limited diffusion (Goodman and Green, 1992).

While data about IT use in the region is sparse, indications are that this trend of limited diffusion continues for various reasons--in some cases by decree. Al-Sulimani (1994) and Bukhari and Meadows 1992), attribute difficulties in Saudi IT transfer (ITT) to technical, organizational, and human problems (statistics regarding ITT in Saudi Arabia compared to the developed West and the world are shown in Table 2). As of 1997, problems with ITT in Saudi Arabia could still be seen in such areas as the Saudi construction industry, showing "under-utilization" and in "limited" use in this industry (Sash and Al-Amir, 1997, p. 195). Additionally, Saudi use of the Internet is tightly controlled in an effort to limit access to pornography and unacceptable discussion of such topics as atheism (Ambah, 1995). Consistent with Saudi Arabia, Internet access is also carefully limited by state owned or controlled provider services in other countries including Egypt, Bahrain, Kuwait, Jordan and the U.A.E. (Albrecht, 1996).
 

	Size--Square Miles (#K)	Population --1991 Est. (#M)	GNP--1990 Est. ($B)	GNP/ Capita --1990 Est. ($)	Computers In Use--1992 (#M)	Computers in Use--Share (%)	Computers/ Person--1992 (#)	Computer Power--MIPS (#K)	Computer Power--Share (%)	MIPS/ 1000 People--1992 (#)
USA	3539	252.7	5465	21,626	67.04	45.33	0.265	130,373	51.39	515.9
W. Europe	1330	369	6985	18,939	34.8	23.5	0.0942	56,927	22.44	154.35
Worldwide	50,524	5,423	25,395	4,683	147.9	100	0.027	253,676	100	46.78
Saudi Arabia	830	17.9	125	6,983	0.39	0.26	0.022	611	0.24	34.1

In spite of existing limitations to ITT in the Arab region, there are indications of growing demand for IT Transfer to the region and opportunities for vendors in the West. As a result of the war with Iraq, demand for defense systems in Kuwait offers billions of dollars in potential IT sales (Morrocco, 1997). Further, demand in the U.A.E. for computer equipment is predicted to grow by 20% per year through the year 2002 (Anonymous, 1997). Therefore, understanding what factors contribute to successful ITT is crucial for those wishing to tap into these billions of dollars in potential sales. It is with this potential in mind that the Arab region was chosen as the first LDC region for testing of TAM.

Hypotheses

Lacking an a priori reason for why TAM would not apply to LDCs, such as those in the Arab region, we posit that TAM will predict use of IT in the developing world. As a result, the following hypotheses, implicitly framed in the setting of lesser developed countries, are expected to be true:

A cross-sectional survey of knowledge workers in the Arab world was conducted to test these hypotheses. This developing region is economically diverse, yet culturally unified in many ways (Barakat, 1993). As such, it makes a fitting region to study how knowledge workers in LDCs respond to the introduction of computer technology.

Instrument Development and Implementation

The sample was drawn from knowledge workers in Jordan, Egypt, Saudi Arabia, Lebanon, and the Sudan across the spectrum of IT penetration. The sampling approach in Jordan provides an exemplar for how this can be accomplished. In this case, a top Jordanian government official contacted managers and professionals in the airline, public, and health care sectors to gain commitment for interviews and surveys. In that these organizations represented IT diffusion at the relatively high and low ends of the spectrum, the research team used other contacts to fill out the sample with private firms who had a middle range level of computer usage.

Two versions of the instrument ¾ one in Arabic and one in English ¾ were prepared so that respondents who were not fluent in English would still be able to participate in the study. Arabic versions incorporated dialectical differences for each country in which the data were to be collected and were then back-translated for greater accuracy.

Generic IT Use by Knowledge Workers in LDCs

If greater understanding of IT adoption and use is to be achieved for developing regions, it is desirable that the focus of the study be computer technology in the broadest sense. In many of these nations, knowledge of even the most basic IT applications such as word processors or graphics packages is not widespread and thus the key issue is to determine why computers in general would or would not be received well. Most of the 500 million citizens of sub-Saharan Africa, for instance, have access to neither reliable telephone service or computers (Odedra, et al., 1993) so that in LDCs where there may be formidable barriers to computer use, it is most appropriate to test potential explanatory models with dependent variables measuring IT use at the most generic level. Thus, the reaction of knowledge workers to the prospect of computerization is the central issue for planning units in the Arab world and elsewhere.

TAM has not been tested with general IT in any environment. There is precedent, however, for studying adoption and use of general IT in the developed world. Moore and Benbasat (1991) measured DOI antecedent effects on use of personal workstations (PCs). Consistent with Moore and Benbasat (1991), we have extended TAM’s artifact of study beyond specific IT applications to general computing in LDCs.

Instrument Validation

Survey methodologies permit gathering relatively large samples, but instrument validity depends on pretesting content validity, reliability, and construct validity (Straub, 1989). Construct validity was assessed via a principal components factor analysis. With loadings greater than .40, the independent variables perceived usefulness and perceived ease-of-use loaded separately from each other (Varimax rotation), demonstrating convergent and discriminant validity (Campbell and Fiske, 1959). System use, the dependent variable, was compared to other dependent variables not being examined in the current study. These variables, which were also IT diffusion outcome variables, were respondent ratings of likely implementation success of different IT scenarios. Again, the variable of interest, system use, showed convergent and discriminant validity by loading only on one factor and being clearly discriminated from the other dependent variables, all of which loaded onto a second factor. Relevant factor loadings, along with reliabilities for the TAM endogenous and exogenous variables, are shown in Table 3 below. All exceed Nunnally’s (1967) rule of thumb standard for Cronbach a s in exploratory research. All items were measured via Likert scales except for V13, which asked respondents to estimate the amount of time averaged per day on computers. Items used in the instrument are listed in Table 3 as well.

 
 

Construct	Item Code	Items	Factor 1 Loadings	Factor 2 Loadings	Factor 3  Loadings	a
Perceived Usefulness	V17A V17E V17H	· Using computers in my job enables me to accomplish tasks more quickly.  · Using computer systems increases my job performance.  · Using computer systems in my job increases my productivity.	.89423  .88642 .69826			.819
Perceived Ease-of-Use	V17C  V17F V17I	· Learning to operate computers is easy for me.  · I find computers easy to use.  · I find it easy to get computers to do what I want them to do.		.89661 .88310 .54141		.758
System Use	V14C      V14D  V17B  V14E V13  V14A	· I use computer technology or computers (for example, PCs, mainframes, FAX machines, and Electronic-Mail) extensively in my work.   · I use computers extensively at home.  · While I am at work, I am an enthusiastic user of computers.  · My children use computers extensively at home.  · Time in an average work day spent working with a computer  · Computers are used extensively in my organization.			.73171     .66178 .62122 .58871 .54955 .53582	.693

Sampling Procedure

The total sample size was 274 across five Arab countries. The sample base was composed of 121 from Jordan; 45 from Egypt; 28 from Saudi Arabia; 35 from Lebanon; and 45 from the Sudan. IT penetration varied by nations with Egypt and Saudi Arabia at the high end and the Sudan at the low end. Through contacts in each of these countries, questionnaires were distributed and personally collected from the sampled groups. Under such circumstances, it is not possible to know the exact response rate, but there is every reason to believe that it was very high, likely exceeding 80%.

In an attempt to maximize variance in both endogenous and exogenous variables, samples were taken from individuals, countries, and organizations and across a wide variety of IT penetration. The process of data collection in Jordan is representative of how goals of the research program were met. Initially, contact with a top Jordanian government official was made. That individual arranged for the interview and survey of managers and professionals in the public and health care sectors. Additional arrangements were made to include a sizable group of private sector employees. IT use was widespread in the public organization and in one of the private firms. Low rates of diffusion were seen in one of the health care organizations. Additionally, a private firm and a second health care organization had moderate IT use.

The population sought was knowledge workers capable of using IT in the Arab world. A relatively homogeneous sample of organizational knowledge workers in urban areas across these countries was collected. All samples were drawn from this population although, since characteristics of the overall population are not generally known, it was not possible to test how representative our sampling was. Computer system users are typically well educated in many Arab countries (Danowitz, 1995). As a result, it appears that this was a representative sample of the population of interest, as shown in Table 4.
 

	Degrees   (type of degree, percent of sample population, and number of individuals)			Location Degree Conferred  (region degree conferred, percent of sample population, and number of individuals)
	High School Diploma	10.4% ( 28)		Arab Country	81.0% (218)
	Undergraduate	69.3% (187)		Developed Country	19.6% ( 51)
	Masters	13.3% ( 36)
	Doctorate	7.0% ( 19)

 

DATA ANALYSIS

The hypotheses were tested through Partial Least Squares (PLS). PLS permits researchers to assess the strength of structural relationships in a model while retaining multiple indicators of the latent constructs. The technique enables testing hypotheses of coefficient differences from zero to be tested via T-statistics. These statistics are generated through a jackknife analysis of the raw data. Interpretation of coefficients in PLS is similar to interpreting regression coefficients and, under many circumstances, the results can be very similar to those of other structural modeling techniques like LISREL (Fornell and Bookstein, 1982).

Figure 1 summarizes the major findings. All hypotheses are supported by the data. PEOU is strongly related to PU and both PU and PEOU impact systems use. PEOU affects PU at the .196 level of explained variance, indicating that PU does mediate the relationship between PEOU and system use. Overall explained variance of the dependent variable system use is slightly over 40%, comparing favorably to prior TAM studies. For example, Davis (1989) found explained variance for XEDIT to be 46%, for Chartmaster, 51%, and for Pendraw 71%. In Davis (1989), the explained variances for two studies were 45% and 49%.

Details of the PLS loadings are shown in Table 6. As can be seen from the generally large T-statistics, all items loadings were significant at the .05 level, indicating that the items were related to the constructs specified in the model.

 
 

	Construct	Indicators/ Item Codes	Entire Sample Estimate	Jackknife Estimate	T-Statistic
	System	V13	0.5461	0.5380	8.3235
	Use	V14A	0.4800	0.4953	5.6055
		V14C	0.7719	0.7646	21.5228
		V14D	0.5581	0.5676	8.7525
		V14E	0.4449	0.4419	5.0440
		V17B	0.7935	0.7973	27.1589
	PU	V17A	0.7326	0.7370	10.7943
		V17E	0.9001	0.9118	40.0806
		V17H	0.9172	0.9196	61.1979
	PEOU	V17C	0.8388	0.8381	31.5961
		V17F	0.8304	0.8319	23.6052
		V17I	0.7746	0.7809	22.5287

Discussion and implications

This study provides insight into information technology adoption and use outside of the technologically advanced world. As predicted, TAM transferred successfully to the Arab world. The effectiveness of TAM in predicting general computer technology adoption and use raises the question of whether other adoption and use models might not be applicable as well.

For practitioners, successful transfer of TAM to less developed countries suggests certain implementation strategies. Approaches which are suitable for introducing IT in technologically advanced cultures may be applicable to a certain extent in less developed cultures. Stressing rational factors like usefulness and ease-of-use of a new system in user training could lead to better implementation, for example. Caution must be used when interpreting these findings, however. We know that social and cultural norms are also good predictors of technology use (Karahanna, 1993; Mathieson, 1991; Straub, et al., 1997) and, thus, it is possible that over emphasis of rational factors could lead to cultural backlash rather than cultural acceptance.

Social and cultural beliefs may be very specific to certain cultures and need to be addressed in training and implementation. Managers should attempt to work with, rather than against the dominant culture. Yavas, Luqmani, & Quraeshi, (1992) has found that top management buy-in and championship must be ensured before attempting to introduce new IT in a highly patriarchal, tribal, and communal society like the Arab culture, for example. In that face-to-face meetings are an essential part of Arab society, managers should probably not stress the efficiency of face-to-face replacement systems, such as an E-Mail or groupware.

Assuming, however, that a balanced emphasis can be placed on the value of computers in making the work of knowledge workers easier and more productive, certain training and implementation tactics suggest themselves. Knowledge that ease of use and perceived usefulness predicts usage might aid in the development of implementation and training strategies in the Arab world such as the following:

• EOU: Stress features of the computer that make it easy to understand and operate. Newer, GUI operating systems should be intuitively appealing to Arabs, as they are to knowledge workers in developed countries. Trainers and developers who are attempting to gain acceptance of the technology should reinforce users’ natural responses to the embedded meaning of the icons in the interface.

• PU: Compare the use of computer applications to manual tools for accomplishing the same tasks. This should help users to see the usefulness of computers. Clearly, for example, the use of computers as point-of-sales terminals gives far more information than a comparable manual system. Ordering systems that are computerized should also strike knowledge workers as superior to paper-and-pencil systems.

Inevitably, as firms expand their global reach, they will be confronted with the need to encourage computer use in countries within which they hope to do business. Knowledge workers in cultures that differ in critical ways from the industrialized West and East may resist learning and using computers. For this reason, there is a growing need to understand how multinationals can develop and nurture a computer literate work force.

By the same token, government planning agencies in developing countries can improve their forecasting and policy-making with deeper insights into computer penetration of the business community and the population in general. Studies in IT diffusion in LDCs can assist in providing this kind of useful knowledge.

FUTURE RESEARCH OPPORTUNITIES AND LIMITATIONS

While this study supports the belief that TAM transfers to LDCs, much work still needs to be done. For example, future research focusing on instrument development appears desirable. Although validated, the present instrument may have systematic measurement error leading from its construction for, or its use in the Middle East. For example, system usage seems to be a difficult variable to measure in the Arab world (a = .693). Such problems suggest that cultural biases may be in play. For instance, Arabs have a greater sense of collectivism than North Americans, according to Hofstede’s indices (1984). Compared to subjects in other cultures, Arab respondents may be more likely to give answers to interview questions which reflect their group leader’s opinions of instead of their own. Future IT studies in LDCs, therefore, could be greatly enhanced by instrument construction which did everything possible to reduce culturally-dependent biases.

As the first study of technology acceptance in LDCs, these findings need to be replicated through additional studies in the developing world. Moreover, in spite of the fact that Arab countries typically demonstrate strong cultural cohesiveness, individual Arab LDCs are unique and results from one nation or group of nations may not apply to all others. That is to say, overgeneralization of these results is possible. Moreover, other developing nations may differ, in whether people respond to rational factors like perceived usefulness, for example. As a result, models which seem to explain technology transfer in specific LDCs, such as the present study, should be confirmed through diverse samples in other LDC nations and regions.

One other factor suggests that tests in other LDCs are desirable. The applicability of the TAM model suggests that Arab culture does not strongly mitigate against the cultural beliefs of perceived usefulness and perceived ease of use. Based on Straub (1994), it would be premature to suggest that all cultures are responsive to these beliefs. It may be necessary to empirically test TAM in multiple LDC cultures before any such generalization could be made.

REFERENCES

Adams, D.A., Nelson, R.R. & Todd, P.A. (1992). Perceived Usefulness, Ease of Use and Usage of Information Technology: A Replication. MIS Quarterly, 16(2), 227-247.

Al-Sulimani, T. & Sharad, D. (1994). Advent of TQM in Saudi Arabia. American Association of Cost Engineers Transactions: INT9.1-INT9.4.

Albrecht, K. (1996, June 3). Cybersurfers of Arabia. Business Week,3478, 108.

Ambah, F. S. (1995, August 21). An intruder in the kingdom. Business Week, 3438, 40.

Anonymous (1997). U.A.E.'s dynamic market for computers/peripherals to grow at least 20% a year. Middle East Executive Reports, 20(8), 8, 18+.

Atiyyah, H.S. (1989). Determinants of Computer System Effectiveness in Saudi Arabian Public Organizations. International Studies of Management & Organization, 19(2), 85-103.

Barakat, H. (1993). The Arab World. Berkeley, CA: University of California Press.

Bukhari, A. A. & Meadows, A. J. (1992). The Use of Information Technology by Scientists in British and Saudi Arabian Universities: A Comparative Study. Journal of Information Science Principles & Practice 18(5), 409-415.

Campbell, D.T. & Fiske, D.W. (1959). Convergent and Discriminant Validation by the Multitrait-Multimethod Matrix. Psychological Bulletin, 56(2), 81-105.

Computer Industry Almanac Inc.(1992). The 5th Annual Computer Industry Almanac. Incline Village, NV.

Danowitz, A.K., Nassef, Y. & Goodman, S.E. (1995). Cyberspace Across the Sahara: Computing in North Africa. Communications of the ACM, 38(12), 23-28.

Davis, F. (1986). Technology Acceptance Model for Empirically Testing New End-User Information Systems: Theory and Results. Unpublished Doctoral Dissertation, Massachussetts Institute of Technology.

Davis, F. (1989). Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology. MIS Quarterly, 13, 319-339.

Davis, F.D., Bagozzi, R.P. & Warshaw, P.R. (1989). User Acceptance of Computer Technology: A Comparison of Two Theoretical Models. Management Science, 35(8), 982-1003.

Fishbein, M. & Ajzen, I. (1975). Belief, Attitude, Intentions and Behavior: An Introduction to Theory and Research. Boston: Addison-Wesley.

Fornell, C. & Bookstein, F.L. (1982). Two Structural Equation Models: LISREL and PLS Applied to Consumer Exit-Voice Theory. Journal of Marketing Research, 19(4), 440-452.

Goette, T. (1995). Determining Factors in the Successful Use of Adaptive Technology by Individuals with Disabilities: A Field Study. unpublished Dissertation, Georgia State University.

Goodman, S.E. & Green, J.D. (1992). Computing in the Middle East. Communications of the ACM,35(8), 21-25.

Goodman, S.E. & Press, L.I. (1995). Computing in Vietnam: An Asian Tiger in the Rough, Communications of the ACM,38(1), 11-16.

Hendrickson, A.R., Massey, P.D. & Cronan, T.P. (1993). On the Test-Retest Reliability of Perceived Usefulness and Perceived Ease of Use Scales, MIS Quarterly,17(2), 227-230.

Hill, C. E., Loch, K. D., Straub, D. W., El-Sheshai, K. (1998). A Qualitative Assessment of Arab Culture and Information Technology Transfer. Journal of Global Information Management forthcoming.

Hofstede, G. (1984). Culture's Consequences: International Differences in Work-Related Values, Newbury Park, CA: Sage.

Ibrahim, R.L.R. (1985). Computer Usage in Developing Countries: Case Study Kuwait. Information & Management, 8(2), 103-112.

Karahanna, E. (1993). Evaluative Criteria and User Acceptance of End-User Information Technology: A Study of End User Cognitive and Affective Processes. unpublished Dissertation, University of Minnesota,.

Knight, J. (1993). Contumacious Computer, In W. W. Cotterman & M. B. Malik (Ed.), Informational Technology in Support of Economic Development. Atlanta, GA, USA: Georgia State University Business Press.

Kwon, T.H. & Zmud, R.W. (1987). Unifying the Fragmented Models of Information Systems Implementation. In R. J. Boland Jr. & R. A. Hirscheim (Ed.), Critical Issues in Information Systems Research, (pp. 227-251). New York: John Wiley & Sons,.

Mahmood, M.A., Gemoets, L.A. & Gosler, M.D. (1995). Information Technology Transfer and Diffusion to Mexico: A Preliminary Analysis. Journal of Global Information Management, 3(4), 5-15.

Mathieson, K. (1991). Predicting User Intentions: Comparing the Technology Acceptance Model with the Theory of Planned Behavior. Information Systems Research,2(3), 173-191.

Mesher, G.M., Goodman, S.E., Snyder, J.M., Briggs, R.O. & Press, L.I. (1992). Cuba, Communism, and Computing. Communications of the ACM,35(11), 27-29, 112.

Moore, G.C. & Benbasat, I. (1991). Development of an Instrument to Measure the Perceptions of Adopting an Information Technology Innovation. Information Systems Research,2(3), 192-222.

Morrocco, J. D. (1997). Bidders Vie for Kuwaiti Integrated C4I Contract. Aviation Week & Space Technology, 147(21), 31-33.

Nidumolu, S.R. & Goodman, S.E. (1993). Computing in India: An Asian Elephant Learning to Dance. Communications of the ACM,36(6), 15-22.

Nunnally, J.C. (1967). Psychometric Theory. New York: McGraw-Hill.

Odedra, M., Lawrie, M., Bennett, M. & Goodman, S. (1993). Sub-Saharan Africa: A Technological Desert. Communications of the ACM,36(2), 25-29.

Prescott, M. & Conger, S. (1995). Diffusion of Innovation Theory: Borrowing, Extensions and Modifications from IT Researchers. DATABASE for Advances in Information Systems,26(2&3), 20-41.

Rogers, E.M. (1995). Diffusion of Innovations. New York: The Free Press.

Sash, A. A. & Al-Amir, M. (1997). Information Technology in Contractors' Firms in Saudi Arabia. Construction Management and Economics,15, 187-200.

Straub, D.W. (1989). Validating Instruments in MIS Research. MIS Quarterly,13(2), 146-169.

Straub, D.W. (1994). The Effect of Culture on IT Diffusion: E-mail and FAX in Japan and the U.S. Information Systems Research,5(1), 23-47.

Straub, D.W., Keil, M. & Brenner, W. (1997). Testing the Technology Acceptance Model across Cultures: A Three Country Study. forthcoming in Information & Management,.

Straub, D., Loch, K., Hill, C., & El-Sheshai, K. (1998). Transfer of Information Technology to Developing Countries: A Test of Cultural Influence Modeling in the Arab World. Working paper, 1-50.

Thompson, R. & Rose, G. (1994). Information Technology Adoption and Use. Proceedings of the Administrative Sciences Association of Canada (ASAC) Annual Conference, 16-26.

Yavas, U., Luqmani, M., & Quraeshi, Z. (1992). Facilitating the Adoption of Information Technology in a Developing Country. Information & Management, 23(2), 75-82.