Predicting students usage of Internet in two emerging economies using an extended technology acceptance model (TAM).

Alshare, Khaled A. ; Alkhateeb, Faisal B.

ABSTRACT

This study employed an extended technology acceptance model (TAM) to predict Internet usage in two developing countries (Chile and United Arab Emirates (UAE)). In addition to investigating the impacts of perceived ease of use (PEOU), perceived usefulness (PU), and perceived Internet content (PIC) on students' usage of the Internet, it analyzed the direct impacts of external variables such as gender, educational background, income level, self-reported measure of computer knowledge, Internet cost, and Internet availability on Internet usage and their moderating role in the relationship between PEOU, PU, and PIC and Internet usage. To validate the research model, data was collected from 169 students from Chile and 194 students from United Arab Emirates (UAE). The results showed that only PU was a significant predictor of Internet usage for both Emirates and Chilean samples. Additionally, while gender significantly impacted Emirates students' usage of Internet, self-reported knowledge about computers significantly impacted Chilean students' usage of Internet. Income level was the only significant moderator for both countries. PU affected usage of the Internet more positively for students with high income level than it did for those students with low income. Discussion of practical implications of the results was included.

INTRODUCTION

Most studies of the Internet have focused primarily on adoption, e-commerce, and web design (e.g., Kim et al., 2005; Park, et al., 2004; Stanfield and Grant 2003; Ranganathan and Grandon 2002; Tan and Teo 1998; Teo and Pian 2004). Little research has been done on student usage of the Internet (Alshare et al, 2005a). Additionally, the majority of studies on Internet usage in the last decade have been carried out in developed countries. There is a need to understand not only why technology has or has not been adopted but also to comprehend the impacts of its adoption by developing countries. The Internet has major impacts upon the ability of developing countries and citizens to be more effective participants in the emerging global business environment.

The Technology Acceptance Model (TAM), introduced by Davis (1989), is the most popular model used in Information Systems (IS) literature to predict the intention or the usage of Information Technology (IT). According to the citation index of the Institute for Scientific Information (ISI 2005), by September 2005, there were 631 journal citations for the original Davis (1989) article. In his model, Davis introduced perceived ease of use and perceived usefulness as the two main factors that influence computer usage (e-mail). TAM has been used in predicting intention or usage of different computer applications primarily in the region of North America (Lapczynski 2004; Pijpers 2001). TAM was then extended by incorporating additional factors; see for examples, (Alshare et. al, 2004; Davis et al., 1989; Gefen and Straub 1997; Venkatesh et al. 2003; Venkatesh and Davis 2000; and Lucas and Spitler 2000). It is worth mentioning that there were few studies that tested TAM or extended versions outside the region of North America, primarily in developed countries (e.g., Al-Gahtani 2001, Huang et al., 2003; Lai and Wong 2003; Straub et al., 1997). Moreover, fewer studies applied TAM or extended versions to developing countries (e.g., Akour et al. 2006; Elbeltagi et al., 2005; Loch et al., 2003; McCoy et al. 2005; Parboteeah et al., 2005; Rose and Straub, 1998; Zakour 2004).

In this study, we extended the research that was conducted by Alshare et. al (2005a) by including more external variables such as Internet cost and Internet availability. Additionally, we tested a modified TAM model outside the region of North American in two developing countries, Chile and the United Arab Emirates (UAE). These two countries represent two emerging economies (The World Competitiveness Yearbook, 2002).

LITERATURE REVIEW OF CONSTRUCTS AND HYPOTHESES

Perceived Ease of Use (PEOU) and Perceived Usefulness (PU)

According to TAM, perceived ease of use refers to the extent to which a person feels that using a particular technology would be free of effort. On the other hand, perceived usefulness refers to the extent to which a person believes that using a particular technology would enhance his/her productivity and effectiveness (Davis 1989). TAM has been utilized in many studies and found that PEOU and PU were significantly related to computer usage (Adams 2002; Igbaria et al., 1997, Mccloskey (2003-2004); Seyal et. al. 2002; Venkatesh and Davis 2000). Seyal et al. (2002) developed a model to test whether PU, PEOU, and other variables determine Internet usage among college academics. They found that PU and PEOU were significant predictors of technology usage. In this study, we focused on PU and PEOU to explore their direct impact on Internet use. Thus, we developed the following hypotheses for testing:

H1: Perceived ease of use (PEOU) has a significant impact on Internet usage.

H2: Perceived usefulness (PU) has a significant impact on Internet usage.

Perceptions of Internet Content (PIC)

Many studies have focused on factors that make web sites more attractive to users (Liu and Arnett 2000; Park, et al., 2004; Ranganathan and Ganaphaty 2002; Ranganathan and Grandon 2002). Internet content was found to be a major factor that influenced user usage of the Internet (Huizingh 2000; Torkzadeh and Dhillon 2002; Palmer 2002). However, Alshare et al (2005a) investigated students' usage of the Internet in the USA, and they found that PEOU and PU, but not PIC had significant influence on Internet usage. Since this study focused on students in Chile and UAE who represent two different cultural settings (Hofstede 1997), it was natural to think that some people perceived Internet content as a threat to their values and culture and avoided its usage, while others had a positive perception toward it and would continue or start to use it (Alshare 2005b). Thus, we proposed the following research hypothesis:

H3: Perception of Internet content (PIC) has a significant impact on Internet usage.

The Impact External Variables

Many studies have explored the effect of external variables on the relationship between PEOU, PU, and technology usage (Alshare et al., 2005a; Alshare et al., 2004; Venkatesh et al., 2003; Venkatesh and Morris 2000). For example, Alshare et al. (2005a) explored the direct effect of external variables (gender, income level, educational background, computer users' classification, and self-reported measures of computer knowledge) on Internet usage and their moderating effect on the relationship between PEOU, PU, and PIC and Internet usage. They found that gender was the only significant moderator (PEOU affected usage of the Internet more strongly for female students than it did for male students). However, they found that classification of computer users and self-reported knowledge about computers were significant predictors of Internet usage. Venkatesh and Morris (2000) explored the moderation effect of gender on the relationship between PEOU, PU, and subjective norm with the intention to use a system for data and information retrieval. They found that male technology usage decisions were more strongly influenced by their perceptions of usefulness, while females were more strongly influenced by perceptions of ease of use. It was also reasonable to assume that the cost and the availability of the Internet would influence students' usage of the Internet. For example, Alshare et al., (2003) found that there was a significant relationship between Internet cost and its usage. Based on the above discussion we proposed the following hypotheses:

H4: Gender has a significant impact on Internet usage.

H5: Educational background has a significant impact on Internet usage.

H6: Income level has a significant impact on Internet usage.

H7: Self-reported knowledge about computers has a significant impact on Internet usage.

H8: Internet cost has a significant impact on Internet usage.

H9: Internet availability has a significant impact on Internet usage.

In this study, we followed the aforementioned research and proposed that gender, educational background, income level, self-reported measure of computer knowledge, Internet cost, and availability can be considered as moderating variables in the relationship between PEOU, PU, and PIC and Internet usage. Thus, we proposed the following set of hypotheses:

H10a: The impact of perceived ease of use on Internet usage depends on gender.

H10b The impact of perceived ease of use on Internet usage depends on educational background.

H10c: The impact of perceived ease of use on Internet usage depends on income level.

H10d: The impact of perceived ease of use on Internet usage depends on self-reported knowledge about computers.

H10e: The impact of perceived ease of use on Internet usage depends on Internet cost.

H10f: The impact of perceived ease of use on Internet usage depends on Internet availability.

H11a: The impact of perceived usefulness on Internet usage depends on gender.

H11b: The impact of perceived usefulness on Internet usage depends on educational background.

H11c: The impact of perceived usefulness on Internet usage depends on income level.

H11d: The impact of perceived usefulness on Internet usage depends on self-reported knowledge about computers.

H11e: The impact of perceived usefulness on Internet usage depends on Internet cost.

H11f: The impact of perceived usefulness on Internet usage depends on Internet availability.

H12a: The impact of perceived Internet content on Internet usage depends on gender.

H12b: The impact of perceived Internet content on Internet usage depends on educational background.

H12c: The impact of perceived Internet content on Internet usage depends on income level.

H12d: The impact of perceived Internet content on Internet usage depends on self-reported knowledge about computers.

H12e: The impact of perceived Internet content on Internet usage depends on Internet cost.

H12f: The impact of perceived Internet content on Internet usage depends on Internet availability.

RESEARCH METHOD

The Proposed Model

Based on the previous analysis, we proposed the following theoretical model.

2[FIGURE 1 OMITTED]

Survey Questionnaire

In addition to asking questions concerning demographic variables such as gender, age, educational background, and income level, the questionnaire solicited information about Internet usage, PEOU, PU, and PIC. Five items were used to measure each of the PEOU and PU constructs that were taken directly from Davis' (1989) scale and modified to measure Internet usage. The PIC construct was adopted from Alshare et al., (2005a). Four items were used to measure the construct PIC. The survey instrument was developed, reviewed for content as well as readability, and pilot tested; then, the survey was modified accordingly. Survey participants responded to statements using a 5-point Likert scale ranging from strongly disagree to strongly agree. The Statistical Packages for the Social Services (SPSS) was used to compute frequencies, means, percentage, factor analysis, and reliability (Cronbach alpha coefficient). The regression procedure was utilized to test the hypotheses.

Samples and Data Collection

The survey questionnaire was administered to convenient samples of college students in the UAE and Chile during Fall 2003-Spring 2004. In Chile and UAE, colleagues of the authors were approached and asked to distribute the survey to students in their schools. Students completed the survey during class time; then the surveys were collected by the instructors and sent back to the USA via postage mail. The questionnaire was distributed to 300 college students in each country. Since English is the second spoken language in the UAE and students and instructors were familiar with it (AMIDEAST 2005), the questionnaire was administrated in English to the UAE sample. In Chile, however, the questionnaire was administered in Spanish, since only 2% of Chileans older than 15 years were fluent in English (Miranda, 2004). Back translation procedure (Brislin, 1986) was used to ensure that the meaning of the questions was not lost during the translation process.

Measures of Variables and Constructs

The dependent variable Internet usage (IU) was measured with a single item that represented the number of hours devoted to the usage of Internet per day. The independent variables included three constructs: perceived ease of use (PEOU) measured using 5 items, perceived usefulness (PU) measured using 5 items, and perceived of internet content (PIC) measured using 4 items.

The external variables considered in this study were gender (GEN), educational background (EDBACK), family-monthly income (INC), self-reported knowledge about computers (KNOWL), Internet cost (ICO), and Internet availability (IAV). The above external variables were operationalized as follows:

Gender is a dummy variable that takes on a value of 0 for males and a value of 1 for females.

EDBACK is a dummy variable that takes on a value of 0 for business majors and 1 for other majors.

INC is a dummy variable that takes on a value of 0 for low family-monthly income and 1 for high family-monthly income.

KNOWL is a dummy variable that takes on a value of 0 for very good-to-excellent computer knowledge and 1 for poor-to-good computer knowledge.

ICO is a dummy variable that takes on a value of 0 for expensive-very expensive and 1 for very cheap-fair.

IAV is a dummy variable that takes on a value of 0 for very good-excellent and 1 for fairgood.

Regression Models

Three regression models were used to test hypotheses (H1-H12f). The average of the items for each construct was used in the regression analysis. The external variables were included in the multiple regression equations as dummy variables. We evaluated their effects on the relationship between PEOU, PU, PIC and IU by adding the interaction term between the external variable and each of the independent variables.

Model 1: IU = [[alpha].sub.1] + [[beta].sub.1] (PEOU) + [[beta].sub.2](PU) + [[beta].sub.3](PIC) + [e.sub.1]

Model 2: IU = [[alpha].sub.2] + [[beta].sub.1] (GEN) + [[beta].sub.2](EDBACK)+ [[beta].sub.3](INC) + [[beta].sub.4](KNOWL) + [[beta].sub.5] (ICO) + [[beta].sub.6](IAV) + [e.sub.2]

Model 3: [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Where:

IU: Internet usage

PEOU: Perceived ease of use

PU: Perceived usefulness

PIC: Perceived Internet content

GEN: Gender

EDBACK: Educational background

INC: Family-monthly income

KNOWL: Self-reported knowledge about computers

ICO: Internet cost

IAV: Internet availability

e: error term

DATA ANALYSIS

Characteristics of the Samples

One-hundred sixty nine Chilean students and 194 Emirates students returned completed surveys. This represented response rates of 56 and 65 percent respectively. A summary of frequency distributions by country for relevant variables is presented in Table 1.

Seventy-eight percent of Chilean students were males, compared with 49 percent in the Emirates sample. In both samples, students were undergraduate and younger than 30 years old. Forty-six percent of Chilean and 69 percent of Emirates students had business majors. Forty-seven percent of Chilean students and 25 percent of Emirates students had low family incomes. Ninety-five percent of students in Chile and 96 percent of students in UAE reported having a computer at home. Seventy-nine percent (127/160) of Chilean students who had computers at home also had access to the Internet from home, as did 72 percent (139/186) of the Emirates students.

Thirty percent of Chilean, compared to 59 percent of Emirates students, stated that their knowledge about computers was very good to excellent. Six percent of Chilean compared to 4 percent of Emirates students reported that their knowledge was poor to fair. Some 64 percent of Chilean students compared to 37 percent of Emirates students indicated that their knowledge about computers was good. This should be of no surprise, since more than one-half of the students, in both countries, used computers over two hours per day. Thirty-seven percent of Chilean, compared to 56 percent of Emirates students, used the Internet for more than two hours per day, mostly for class-related activities and communication (e-mail). Shopping on line was reported to be the least-used activity on the Internet in both countries. However, approximately 50 percent of students in both countries used the Internet for entertainment activities. While the majority of students in Chile reported that the cost of the Internet was "expensive-very expensive", the majority of students in UAE indicated that the cost of the Internet was "fair". Additionally, two-thirds of Chilean felt that the availability of the Internet in their country was "poor-good", while three-quarters of Emirates students felt that the availability of the Internet in their country was "very good-excellent".

Validation of the Measures

Exploratory factor analysis and Cornbach's alpha were used to assess the psychometric proprieties of the scales. Factor analysis (Principal component, with Varimax rotation) was performed to confirm that the items loaded according to the proposed model. According to Hair et al. (2006), the acceptable value for factor loading for a sample size of 150 is 0.45. Thus, items with loading less than 50 percent were dropped from further analysis. As a result, two items were dropped from each of PEOU, PU, and PIC. Appendix A presents the results of factor analysis and Appendix B shows the items and their descriptions that were used in the computations. Scale reliability was measured using Cronbach's alpha coefficient. As shown in Table 2, the values of alpha for the two samples ranged from 0.62 to 0.83. These values are considered to be sufficient according to Hair et al., 2006.

THE RESULTS OF THE STUDY

The results of the study are divided into three sections. The first section discusses the relationships between the usage of the Internet and PEOU, PU, and PIC. For each country, the multiple regression procedure was employed to test the hypotheses ([H.sub.1]-[H.sub.3]). The second section analyzes the direct effect of the external variables of gender, educational background (business vs. non-business), income level (low vs. high), a self-reported measure of computer knowledge (very good-excellent vs. poor-good), Internet cost (very cheap-fair vs. expensive-very expensive), and Internet availability (fair-good vs. very good-excellent) on Internet usage (hypotheses [H.sub.4]-[H.sub.9]). The third section reports the impact of the external variables on the relationship between PEOU, PU, PIC and usage of the Internet (hypotheses [H.sub.10a]-[H.sub.12f]).

Before testing the hypotheses, the assumptions of the multiple regression models were validated. Several tests such as multicollinearity, autocorrelation, plotted histogram, and the plots of the dependent variable against each of the independent variables were conducted. Multicollinearity was not a problem since the variance inflation factor (VIFs) were low (< 2.0) for both samples. Autocorrelation problem was not an issue since the D.W. values ranged from 1.80 to 1.96. The plotted histograms of the data depicted a normal distribution. Additionally, the plots of the dependent variable against each of the independent variables showed a linear relationship.

The Impact of PEOU, PU, and PIC on Internet Usage (IU)

Based on the regression results of Model 1 (Table 3), only the second hypothesis ([H.sub.2]) was supported by the data for both samples. Thus, PU was a significant predictor of Internet usage (t= 2.068, p= 0.04 for UAE, and t=2.104, p= 0.037 for Chile). On the other hand, PEOU (t= 1.292, p= 0.198 for UAE, and t= 0.151, p= 0.880 for Chile) and PIC (t= -0.636, p= 0.526 for UAE, and t= 0.663, p= 0.508 for Chile) were not significant in predicting Internet usage.

The Impact of External Variables on Internet Usage

In this section, hypotheses ([H.sub.4]-[H.sub.9]) were tested using multiple regression procedure as described earlier in Model 2. As shown in Table 3, two external variables had significant impacts on the Internet usage. Gender had significant impact on Internet usage for the case of the UAE sample (t = 2.312, p = 0.022). Female students in UAE would spend more time using the Internet compared to their male counterparts ([H.sub.4] was supported). For the case of the Chilean sample, self-reported knowledge about computers was a significant variable that impacted student's Internet usage (t = -2.605, p = 0.009). Chilean students who rated their knowledge about computers as "poor-good" would spend less time using the Internet compared to those who rated their knowledge "very-good-excellent" ([H.sub.7] was supported).

The Moderating Effect of the External Variables

In Model 3, the interaction terms between PEOU, PU, PIC and the external variables (gender, educational background, family-monthly income level, self-reported knowledge about computers, Internet cost, and Internet availability) were regressed to determine the role of the external variable in moderating the relationship between PEOU, PU, and PIC and Internet usage ([H.sub.10a]-[H.sub.12f])

As shown in Table 3 at the end of the text, only one interaction term was significant that is (PU *income) for both samples; UAE (t = 2.132, p = 0.035) and Chile (t = 2.211, p = 0.029). Therefore, all hypotheses (H10a-H12f) were not supported with the exception of H11c (The impact of perceived usefulness on Internet usage depends on income level). PU affected usage of the Internet more positively for Emirates and Chilean students with high-income level than it did for those students with low income.

DISCUSSION AND CONCLUSIONS

This study investigated the effect of PEOU, PU, and PIC on students' usage of the Internet in Chile and UAE. It also examined the direct impact of external variables such as gender, educational background, family-monthly income, self-reported knowledge about computers, Internet cost, and Internet availability on Internet usage. Additionally, the study evaluated the moderation role of the external variables in the relationship between PEOU, PU, and PIC and IU.

Table 4 shows a comparison between the results of this study and the results of Alshare et al. (2005a) study that was conducted in the U.S. The results showed that while PU was the only significant predictor of Internet usage for both Emirates and Chilean students, PEOU and PU were significant predictors of Internet usage for American students. It seems that students in the three countries regardless of their differences of cultural backgrounds felt that the use of Internet would be beneficial to them. On the other hand, American students, compared to Emirates and Chilean students, felt that the ease of use of Internet motivated them to use it more frequently. An explanation for this finding is that Emirates and Chilean students, compared to American students, would be more dependent on their teachers according to Hofstede's cultural dimensions (Alshare et. al. 2005b), and they expect support and help from their teachers. Thus, PEOU was not an important factor in predicting Internet usage. On the other hand, American students are more dependent on themselves in learning how to use the Internet; therefore, PEOU was a significant factor in predicting Internet usage. Another plausible suggestion might be that Emirates and Chilean students believed in the importance of using the Internet regardless of its learning difficulty. On the other hand, American students felt that the difficulty level of learning how to use the Internet was a major factor that influenced their usage of the Internet. According to the results of Alshare et al. (2005a) study, PEOU was the most influential factor that affected American students' usage of the Internet.

Even though PIC was found to be a valid and reliable construct, it did not significantly affect students' usage of the Internet. Once again this finding was consistent with Alshare et al (2005a) findings. This outcome should be of interest to instructors. It appeared that students considered Internet content as a trusted source for class-related activities. The role of instructors becomes more important to show students the correct way for obtaining quality information on the Internet. One explanation for not having PIC as a significant factor could be the fact that the Internet is considered by many students to be the most convenient way of finding information for class-related activities; therefore, students may not have much concern about the actual content.

With respect to the impact of external variables, educational background (business vs. non-business), family-monthly income level (low vs. high), Internet cost (very cheap-fair vs. expensive-very expensive), and Internet availability (fair-good vs. very good-excellent) did not influence students' usage of the Internet in the three countries as shown in Table 4. One can say that most of the teachers in many academic majors do request their students to utilize the Internet as a source for information; thus, educational background was not significant. As mentioned earlier and as shown in Table 1 the majority of students used the Internet for class related activities and communication (email). Since access to the Internet is available to all students in the three countries at their schools, income level, Internet cost, and Internet availability were not significant factors.

While gender significantly influenced Emirates students' usage of the Internet, self-reported knowledge about computers (very good-excellent vs. poor-good) significantly influenced American and Chilean students' usage of the Internet as reported in Table 4. As expected students with greater knowledge about computers would feel at ease in utilizing computer applications such as the Internet; and therefore, use it more frequently. While it might be easier to explain, based on the "gender gap" concept, why male students, compared to female students, would spend more time using the Internet, it is not quite easy to explain why Emirates female students would spend more time using the Internet. One explanation could be based on the UAE culture. According to Hofstede's cultural dimensions, the culture of the Arab countries, which UAE is one of them, is a conservative society (Alshare et al. 2005b). Therefore, Emirates female students, compared to their male counterparts, would use more frequently the Internet for communication (email); especially with their teachers. Additionally, Emirates female students who live on campus are limited in their social interactions outside the campus; therefore, the Internet would be their social outlet.

As shown in Table 4, family-monthly income level was the only significant moderator in the case of UAE and Chilean samples. It moderated the impact of PU on Internet usage. On the other hand, gender was the only significant moderator in the case of the American sample. It moderated the relationship between PEOU and Internet usage. PU influenced usage of the Internet more positively for Emirates and Chilean students with high income level than it did for those students with low income. It is reasonable to assume that people with high income level would have access to the Internet at home or Internet shops; and thus, they use the Internet more often for a variety of reasons, and they would appreciate its usefulness. As a matter of fact, the majority of students with high income level spend more than 4 hours per day using the Internet, while the majority of students with low income level spend lees than 3 hours. On the other hand, Income level was not a significant moderator for American sample because the Internet cost, compared to the cost in UAE and Chile, is considered cheap. Thus, both groups of income levels could afford to have internet access at home; and therefore, appreciate its usefulness. As reported by Alshare et al., (2005a) PEOU influenced usage of the Internet more positively for American female students than it did for males. Female students felt that the ease of use of the Internet, but not necessarily its usefulness or its content, motivated them to use the Internet more frequently. However, for Emirates and Chilean students this was not the case. The impact of gender on the relationship between PEOU and Internet usage was not significant.

Finally, the results revealed that an extended TAM model was partially valid in non-western cultures such as Chilean and Emirates cultures. Only PU impacted student's usage of the Internet regardless of their cultural backgrounds. Therefore, educators need to reinforce this concept (perceived usefulness) especially when deciding to teach online classes. Instructors might request students to use the Internet more frequently and demonstrate how easy it is for them to find the desired information.

The limitations of the study includes: first, the reliance on self-reported data on all constructs. Thus, relationships among the constructs might be inflated. Second, the use of students as the target population restricts the ability to generalize the results. Therefore, future research might use a more detailed questionnaire survey with a follow up with subjects. Another future research might be targeting all population segments; and thus, results of the study could be generalized. Finally, another plausible future research could be examining the impact of cultural dimensions on the proposed model by including constructs that represents cultural dimensions.

Appendix A
UAE and Chile

Rotated Component Matrix (Factor Analysis)

 Component

UAE 1 2 3

PEOU3 .880 .233 0.163
PEOU1 .803 .297 0.176
PEOU2 .723 .269 0.28
PU3 .131 .820 0.27
PU1 .276 .774 -0.009
PU4 .316 .694 0.172
PIC2 .105 .341 0.783
PIC1 .421 -.022 0.703

Total Variance Explained: 71.969%

 Component

Chile 1 2 3

PU1 .800 .211 -0.021
PU3 .794 .250 0.186
PU4 .717 .037 0.138
PEOU1 .339 .725 -0.068
PEOU2 .073 .719 -0.032
PEOU3 .115 .692 0.308
PIC1 -.033 .105 0.835
PIC2 .196 .228 0.702

Total Variance Explained: 62.434%.

Appendix B
Significant Items Considered in the Final Analysis

 Construct Item Description

Perceived Ease PEOU1 Learning to use the Internet would be
of Use (PEOU) easy for me

 PEOU2 I would find it is easy to get the
 Internet to do what I want it to do

 PEOU3 I would find the Internet easy to use

Perceived PU1 Using the Internet would increase my
Usefulness (PU) productivity

 PU3 I would find the Internet useful in
 my career

 PU4 Using the Internet would make my
 communication with others more efficient

Perceived Internet PIC1 The information provided by the
Content (PIC) Internet is reliable

 PIC2 I am satisfied with the quality of the
 information provided by the Internet

REFERENCES

Adams, N. (2002), Educational Computing Concerns of Postsecondary Faculty, Journal of Research on Technology in Education, 34(3): p. 285-303.

Akour, I., Alshare, K., Miller, D., and Dwairi, M. (2006). An Exploratory Analysis of Culture, Perceived Ease of Use, Perceived Usefulness, and Internet Acceptance: The Case of Jordan. The Journal of Internet Commerce, 5 (3) (forthcoming).

Al-Gahtani, S. (2001). The Applicability of TAM Outside North America: An Empirical Test in the United Kingdom, Information Resources Management Journal 14 (3), pp. 37-46.

Alshare, K., Grandon, E., and Miller, D., (2005a) Internet Usage in the Academic Environment: the Technology Acceptance Model Perspective. Academy of Educational Leadership Journal, 9(2), 81-97.

Alshare, K., Grandon, E., and Badri, M. (2005b). How Do College Students Perceive Arguments for and against the Impact of the Internet on a Country's Social and Political Affairs? Evidence from Three Countries. The International Journal of Business and Public Administration. 2 (1), 75-91.

Alshare, K., Grandon, E., and Miller, D. (2004). Antecedents of computer technology usage: Considerations of the technology acceptance model in the academic environment. The Journal of Computing Sciences in Colleges, April 2004, 164-180.

Alshare, K., Al-dwairi, M., and Akour, I. (2003) Student-Instructor Perception of Computer Technology in Developing Countries: the Case of Jordan. Journal of Computer Information Systems, 43(4), Summer 2003, p. 115-123.

AMIDEAST (2005), United Arab Emirates (UAE) Country Information, Retrieved December 11, 2005 from http://www.amideast.org/offices/uae/country_info.htm

Brislin R. (1986) The wording and translation of research instrument. In W. J. Lonner & J. W. Berry (Eds.). Field methods in cross-cultural research. Beverly Hills, CA: Sage. pp 137-164.

Davis, F. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 319-340.

Elbeltagi, I., McBride, N., and Hardaker, G. (2005). Evaluating thr factors affecting DSS usage by senior managers in local authorities in Egypt. Journal of Global Information Management, 13 (2), 42-65.

Gefen, D. and Straub, D. (1997). Gender differences in the perception and use of e-mail: An extension to the technology acceptance model. MIS Quarterly, 389-400.

Hair, J, Black, W., Babin, B., Anderson, R., and Tatham, R. (2006), Multivariate Data Analysis 6th, Prentice Hall Publishing Company.

Hofstede, G. (1997). Cultures and Organizations. Software of the Mind. Intercultural Cooperation and its Importance for Survival, New Jersey: McGraw-Hill.

Huang, L. Lu, M. and Wong, B. (2003). The Impact of Power Distance on Email Acceptance: Evidence from the PRC, Journal of Computer Information Systems, 44 (1), pp. 93-101.

Huizingh, E. K. (2000). The content and design of web sites: An empirical study. Information & Management. 37, 123-134.

Igbaria, M., Zinatelli, N., Cragg, P., and Cavaye, A. (1997, September). Personal computing acceptance factors in small firms: A structural equation model. MIS Quarterly, 279-302.

Institute for Scientific Information (2005). Requested September 5, 2005 from http://www.isinet.com/cit/

Kim, C., Galliers, R., and Hoon Yang, K., (2005), "Comparison of Web-based Shopping Systems in the UK and Korea," Journal of Global Information Technology Management, Vol. 8 (4), October 2005

Lai, V. and Wong, B. (2003). The moderating effect of local environment on a foreign affiliate's global IS strategy- effectiveness relationship, IEEE Transactions on Engineering Management 50(3), pp. 353-362.

Lapczynski, P. (2004), An Integrated Model of Technology Acceptance for Small Mobile Technology Devices, Unpublished Doctoral Dissertation, School of Computer Science and Information Systems, Pace University.

Liu, C., and Arnett, K. P. (2000). Exploring the factors associated with web site success in the content of electronic commerce. Information & Management, 38(1), 23-33.

Loch, K.D, Straub, D.W., Kamel, S. (2003). The role of social norms and technological culturation. IEEE Transactions on Engineering Management, 50 (1): 45-63.

Lucas, H. and Spitler, V. (2000). Implementation in a world of workstations and networks, Information and Management 38 (2), pp. 119-128.

McCloskey, D. (2003-2004). Evaluating electronic commerce acceptance with the technology acceptance model. Journal of Computer Information Systems. 49-57.

McCoy, S. Everard, A. and Jones, B. (2005). An Examination of the Technology Acceptance Model in Uruguay and the

US: A Focus on Culture, Journal of Global Information Technology Management 8 (2), pp. 27-45.

Miranda N (2004) Las formulas mas atractivas para aprender Ingles. La Tercera, online edition, 2004. Retrieved March 15, 2004, from www.tercera.cl/articulo/0,5819,3255_5726_63463513,00.html

Palmer, J. W. (2002). Web site usability, design, and performance metrics, Information Systems Research, 13(2), 151- 167.

Parboteeah, V., Parboteeah, P., Cullen, J., and Basu, C. (2005) "Perceived Usefulness of Information Technology: A Cross National Model" Journal of Global Information Technology Management, Vol. 8 (4), October 2005.

Park, J., Dongwon Lee, D., and Ahn, J., (2004), "Risk-Focused E-Commerce Adoption Model: A cross-Country Study," Journal of Global Information Technology Management, Vol. 7 (2), April 2004.

Pijpers, A., Senior Executives' Use of Information Technology: An examination of factors influencing managerial beliefs, attitude and use of Information Technology, Unpublished Doctoral Dissertation, Eindhoven University of Technology, Netherlands, 2001.

Ranganathan, C., and Ganapathy, S. (2002). Key dimensions of business-to-consumer web sites. Information & Management, 39, 457-465.

Ranganathan, C. and Grandon, E. (2002). An exploratory examination of factors affecting online sales. Journal of Computer Information Systems, 87-93.

Rose & Straub, D. (1998). Predicting General IT Use: A Study in Arab Developing Nations," Journal of Global Information Management, 6 (3), Summer, 1998, 39-46

Seyal, A. H., Rahman, M. N. A., and Rahim, M. M. (2002). Determinants of academic use of the Internet: A structural equation model. Behavior & Information Technology, 21(1), 71-86.

Stansfield, M., and Grant, K. (2003). An investigation into issues influencing the use of the Internet and e-commerce among small-medium sized enterprises. Journal of Electronic Commerce Research, 4(1), 15-33.

Straub, D. Keil, M. and Bernner, W. (1997). Testing the Technology Acceptance Model Across Cultures: A Three Country Study, Information & Management 33, pp. 1-11.

Tan, M., and Teo, T. S. H. (1998). Factors influencing the adoption of the Internet. International Journal of Electronic Commerce, 2(3), 5-18.

Teo, T. S. H., and Pian, Y. (2004). A model for Web adoption. Information & Management, 41, 457-468.

The World Competitiveness Yearbook, (2002). Retrieved February 10, 2004, from http://ecoluinfo. unige.ch/archives/ceeman02/0214.html

Torkzadeh, G., and Dhillon, G. (2002). Measuring factors that influence the success of Internet commerce. Information Systems Research, 13(2), 187-204.

Venkatesh, V. and Morris, M. (2000). Why don't men ever stop to ask for directions? Gender, social influence, and their role in technology acceptance and usage behavior. MIS Quarterly, 24(1), 115-139.

Venkatesh, V., Morris, M. G., Davis, G. B., and Davis, F. D. (2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 27(3),425-478.

Venkatesh, V., and Davis, F. (2000). A theoretical extension of the technology acceptance model: Four longitudinal field studies, Management Science 46(2), pp. 186-204.

Zakour, A. (2004). Cultural differences and information technology acceptance, Proceedings of the 7th Annual Conference of the Southern Association for Information Systems, pp. 156-161

Khaled A. Alshare, Emporia State University

Faisal B. Alkhateeb, United Arab Emirates University

Table 1: Frequency Distributions of Key Variables by Country

 Variable Chile (n2=169) UAE (n3=194)

 No. of No. of
 Responses (%) Responses (%)
Gender:
 Male 131 77.5 95 49
 Female 38 22.5 99 51

Educational background:
 Business 78 46.2 133 68.6
 Other 91 53.8 61 31.4

Family monthly income:
 Low income 79 46.7 48 24.7
 High income 90 53.3 146 75.3

Having computer at home:
 Yes 160 94.7 186 95.9
 No 9 5.3 8 4.1

Knowledge about computers:
 1. Excellent-Very good 51 30.0 115 59.3
 2. Good 108 64.0 72 37.1
 3. Fair--Poor 10 6.0 7 3.6

Using computer per day:
 Less than 2 hours 76 45.0 59 30.4
 More than 2 hours 93 55.1 135 69.6

Having access to the Internet
 at home:
 Yes 127 75.1 139 71.6
 No 42 24.9 55 28.4

Cost of Internet:
 1. Very Cheap-Cheap 11 6.5 43 22.2
 2. Fair 69 40.8 101 52.1
 3. Expensive-Very Expensive 89 52.7 50 25.7

Availability of Internet:
 1. Poor--Good 114 67.45 42 21.65
 2. Very good--Excellent 55 32.55 152 78.25

Using Internet per day:
 Less than 2 hours 107 63.4 86 44.3
 More than 2 hours 62 36.7 108 55.7

Internet applications usage:
 Class related activities 130 76.92 141 72.68
 Communication 132 78.11 128 65.97
 Entertainment 84 49.70 104 53.60
 Other activities 78 46.15 116 59.79
 Selling/buying 18 10.65 33 17.01

Table 2: Reliability Analysis (Cronbach Alpha Coefficient)

 Construct UAE Chile

Perceived Ease of Use (PEOU) (3 items) 0.83 0.65
Perceived Usefulness (PU)--(3 items) 0.76 0.70
Perceived Internet Content (PIC)--(2 items) 0.62 0.69

Table 3: Results of Regression Analysis (Coefficient [beta], p-value)

Dependent Variable IU

 Reg.
 Model Independent Variables

Model 1 PEOU PU PIC
 ([R.sup.2])

UAE 0.151 0.261 -0.082
 (6.2) (0.198) (0.04) (b) (0.526)

Chile 0.024 0.323 0.113
 (4.1) (0.88) (0.037) (b) (0.508)

Model 2 Gen Edback Inc Knowl
 ([R.sup.2])

UAE 0.340 -0.12 0.145 -0.168
 (6.9) (0.022) (b) (0.455) (0.390) (0.249)

Chile 0.088 -0.034 0.133 -0.502
 (5.8) (0.681) (0.849) (0.475) (0.009) (c)

Model 3 (Peou * (Peou * (Peou * (Peou *
 ([R.sup.2]) Gen) Edback) Inc) Knowl)

UAE .127 .163 .020 .093
 (17) (0.676) .(565) (0.946) (0.325)

Chile -.009 -.088 -.274 -.100
 (16) (0.985) (0.810) (0.483) (0.807)

 Reg.
 Model Independent Variables

Model 1
 ([R.sup.2])

UAE
 (6.2)

Chile
 (4.1)

Model 2 Ico Iav
 ([R.sup.2])

UAE 0.113 -0.228
 (6.9) (0.447) (0.162)

Chile 0.107 -0.101
 (5.8) (0.554) (0.595)

Model 3 (Peou * (Peou * (Pu * (Pu * (Pu *
 ([R.sup.2]) Ico) Iav) Gen) Edback) Inc)

UAE -.053 .307 .014 .063 .679
 (17) (0.567) (0.291) (0.962) (0.841) (.035) (b)

Chile .452 .515 -.252 .438 .698
 (16) (0.258) (0.207) (0.573) (0.226) (.029) (b)

 Reg.
 Model Independent Variables

Model 1
 ([R.sup.2])

UAE
 (6.2)

Chile
 (4.1)

Model 2
 ([R.sup.2])

UAE
 (6.9)

Chile
 (5.8)

Model 3 (Pu * (Pu * (Pu * (Pic * (Pic *
 ([R.sup.2]) Knowl) Ico) Iav) Gen) Edback)

UAE .413 -.003 .129 .090 -.071
 (17) (0.143) (0.992) (0.733) (0.752) (0.805)

Chile -.210 -.574 -.407 -.127 .131
 (16) (0.593) (0.140) (0.308) (0.799) (0.770)

 Reg.
 Model Independent Variables

Model 1
 ([R.sup.2])

UAE
 (6.2)

Chile
 (4.1)

Model 2
 ([R.sup.2])

UAE
 (6.9)

Chile
 (5.8)

Model 3 (Pic * (Pic * (Pic * (Pic *
 ([R.sup.2]) Inc) Knowl) Ico) Iav)

UAE -.425 -.324 .151 -.007
 (17) (0.206) (0.255) (0.557) (0.985)

Chile -.449 -.164 -.033 .299
 (16) (0.311) (.243 (0.783) (0.520)

Values in parenthesis represent the P-value: a = P < 0.1,
b = P < 0.05, c = P < 0.01

Table 4: A Comparison of three-countries (USA, Chile, UAE) *

 Variable Alshare et
 al. (2005a) This Study

 USA CHILE UAE

PEOU Sig. Not Sig. Not Sig.
PU Sig. Sig. Sig.
PIC Not Sig. Not Sig. Not Sig.
Gender Not Sig. Not Sig. Sig.
Education background Not Sig. Not Sig. Not Sig.
Income Not Sig. Not Sig. Not Sig.
Self-reported knowledge Sig. Sig. Not Sig.
 about computers
Internet cost NA Not Sig. Not Sig.
Internet availability NA Not Sig. Not Sig.
Gender * PEOU Sig. Not Sig. Not Sig.
Income * PU Not Sig. Sig. Sig.

* The remaining of interaction terms were not significant.