USER ACCEPTANCE OF INFORMATION TECHNOLOGY ACROSS CULTURES

Transcription

1 USER ACCEPTANCE OF INFORMATION TECHNOLOGY ACROSS CULTURES Kakoli Bandyopadhyay Lamar University, P.O. Box 10033, Beaumont, TX Phone: (409) , Fax: (409) Soumava Bandyopadhyay Lamar University, P.O. Box 10025, Beaumont, TX Phone: (409) , Fax : (409) bandyopasu@my.lamar.edu ABSTRACT The user acceptance of prepayment metering systems, an innovative Information Technology based application, is examined with the help of a structural equation model in the two different cultures of India and the United States. The intention to use prepayment metering systems is measured along three dimensions: performance expectancy, effort expectance, and social influence. The same measures of the various aspects of user acceptance of prepayment metering systems are applied in the two countries. In both countries, performance expectancy, effort expectancy, and social influence, taken together, are found to have a significant positive impact on consumers intention to use the prepayment metering systems. Individually, different dimensions have significant impacts on behavioral intention in the two countries. INTRODUCTION User Acceptance of information technology (IT) has become a significant area of research in the discipline of management information systems in recent times. Most empirical studies on user acceptance of information technology have been conducted in North America. As the competition in the IT industry in the domestic U.S. market intensifies, more firms are now considering marketing their products and services to foreign countries. To succeed in overseas markets, firms need to gain a better understanding of the differences in consumers perception and adoption of information technology applications. Gaining access to any foreign market will require careful analysis of consumer usage behavior. We add a cross-cultural dimension to our study by comparing the user acceptance of an innovative technology, prepayment metering systems, in the United States with that in another country with a vastly different cultural setting India. The Indian market presents lucrative opportunities for multinational marketers. Even though the overall per-capita gross national income of the 1.1 billion population of India is relatively low at $820, there are about 180 million middle-class consumers in the country with growing purchasing power. In recent times, the declining value of the dollar has accelerated these opportunities. According to a NASSCOM- 553

2 McKinsey report, annual revenue projections for India s IT industry in 2008 are US $ 87 billion and market openings are emerging across four broad sectors: IT services, software products, IT enabled services, and e-businesses. India s GDP, currently growing at around 7 percent, makes it one of the fastest growing economies in the world. Import tariffs have been reduced progressively since the early 1990 s. India s emerging market holds great commercial opportunities for American IT firms, and many American firms such as Microsoft, Google, and others have entered the Indian market. With so many firms competing for the Indian consumer, a comparison study of user acceptance of information technology in both countries is very timely. The technology that we have chosen for this study is prepayment metering systems. The prepayment metering system is an IT -based innovation that involves the payment of electricity by consumers prior to its use (Ghosh, 2002). The consumer purchases credit and uses the electricity till the credit runs out. A prepayment metering system consists of three components: 1) an electricity dispenser; 2) a vending station; and 3) a system master station. An electricity dispenser is an intelligent meter with a built-in disconnecting device and a means of external inputs (smart card reader, keypad, etc.). The dispenser maintains the consumer s electricity credit account and disconnects the supply when the prepayment runs out. A vending station, managed by third party agents, receives customer payments in advance and issues a coded set of information to be entered into the dispenser. A system master station communicates with various vending stations via electronic data linkages. It maintains a common database for reporting information on consumers, tariff changes and detailed customer sales, and provides better administration and financial control. There has been an upsurge of interest in using prepayment systems in recent times and more than 40 countries are experimenting with such systems (Prepayment metering report, 2005). At present, there are about 14 million prepayment installations around the world, which clearly indicates that prepayment metering is already considered a viable alternative to credit purchase of electricity. Hence, the choice of prepayment metering systems is appropriate, since it allows us to study the acceptance of a technology which is new, innovative, and in its early stage of acceptance in both India and the U.S. THEORETICAL FOUNDATION Multiple models that provide justification for the variables under consideration have been used in the research of user acceptance of information technology. Each model will be examined briefly as to its relevance to the present study. The TAM model, based on the theory of reasoned action by Fishbein and Ajzen (1975), was developed by Davis (1989) and expanded in Davis et al. (1989). The model suggests that behavioral intentions to use technology can be predicted by two separate but interconnected variables perceived usefulness and perceived ease of use. The authors believe that external variables such as system design characteristics, user characteristics, task characteristics, nature of the device or implementation process, political influence, organizational structure, etc. are variables that influence perceived usefulness and perceived ease of use. 554

3 The Theory of Planned Behavior (TPB) is based on Fishbein and Ajzen s (1975) Theory of Reasoned Action (TRA). The TPB proposes that the perceived control the consumer has over the situation can also influence consumers intentions. This theory includes all of the elements of the TRA with an added component of perceived behavioral control. Another model considered is the decomposed theory of planned behavior (DTPB) by Taylor and Todd (1995). It is based on components of the TAM model, the TRA, and the TPB. Taylor and Todd (1995) found that all three components attitude, subjective norm, and perceived behavioral control contribute to behavioral intentions. The latest model to be developed from this body of research is a synthesis and unification of eight different models called the unified theory of acceptance and use of technology (UTAUT) by Venkatesh et al. (2003). This model examined the determinants of user acceptance and usage behavior performance expectancy, effort expectancy, social influence, and facilitating conditions and found that all contribute to the usage behavior either directly (facilitating conditions) or through behavioral intentions (performance expectancy, effort expectancy, and social influence). UTAUT does consider factors such as gender, age, experience, and whether or not use is voluntary. The UTAUT model was chosen for this study. The UTAUT model has been found to provide as much as 70 percent of the variance in intention to use technology. As such, it is very promising in terms of helping to determine what factors are important to consider when introducing a new technology to workers in different cultural settings. THE MODEL The research model employed in this study involves three individual issues (Venkatesh et al., 2003) which collectively can represent consumers intentions to use Prepayment Metering Systems. These dimensions are as follows: 1.Performance Expectancy: This refers to the extent to which a consumer perceives the Prepayment Metering Systems to be more useful in accomplishing the electricity account management tasks than using the conventional metering system. The Prepayment Metering Systems involve the payment of electricity by consumers prior to its use. The consumer purchases credit and uses the electricity until the credit runs out. The relationship between performance expectancy and behavioral intention is moderated by gender, and income. The consumer s use of Prepayment Metering Systems depends on the usefulness of the technology to allow consumers to budget, control, and monitor their electricity consumption. 2.Effort Expectancy: This represents the degree of ease that a consumer associates with using the Prepayment Metering Systems to accomplish the electricity account management tasks. Consumer perceptions about the clarity, understandability, flexibility, and ease of using the 555

4 system are taken into consideration. The relationship between effort expectancy and behavioral intention is moderated by gender, income, and experience. 3.Social Influence: This is defined as the social pressure felt by a consumer to use the Prepayment Metering Systems for electricity account management tasks. The social pressure generated from those individuals that the consumer perceives to be important influences the decision of a consumer to use the Prepayment Metering Systems. The effect of social influence on behavioral intention is moderated by gender, income, experience, and voluntariness to use the technology. The structural equation model relating the above three distinct but related issues with the behavioral intention to use the Prepayment Metering Systems is illustrated in Figure 1. The same model is empirically tested in the United States and in India. In the model, Performance Expectancy ( ξ 1), Effort Expectancy ( ξ 2), and Social Influence ( ξ 3) are three independent latent variables measured through the three observed variables X1, X2, and X3, respectively. Thus, each independent variable is measured by one observed variable which is the average score from a multi-item scale intended to measure that dimension. The moderating variables are gender ( ξ 4 ), income ( ξ 5), experience ( ξ 6), and voluntariness of use ( ξ 7 ). The dependent variable, Behavioral Intention ( η 1), is measured by one observed variable, Y1. Figure 1: The Research Model δ 1 X 1 λ x1 Performance Expectancy ξ 1 Y1 δ 2 δ 3 X 2 X3 λx2 λx3 Effort Expectancy ξ 2 Social Influence ξ 3 γ 15 γ 16 γ 18 γ 19 γ 20 γ 13 γ 11 γ 12 γ 21 λy1 Behavioral Intention γ 22 ς 1 η1 ε 1 γ 14 γ 17 Gender ξ 4 ξ 5 Income Experience ξ 6 Voluntariness of Use ξ 7 556

5 RESEARCH DESIGN OPERATIONALIZATION OF CONSTRUCTS The measures used to operationalize the constructs were taken from relevant prior studies. Multiitems scales were used in the questionnaire to measure the three issues described earlier: performance expectancy, effort expectancy, and social influence. The respondents were asked to determine on a seven-point Likert scale how much they agreed (1 = strongly disagree, and 7 = strongly agree) with each statement describing an issue. Three measures of behavioral intention were used in the study. The various items in the measures of Performance Expectancy (PE), Effort Expectancy (EE), Social Influence (SI), and Behavioral Intention (BI) were adapted from existing measures developed by Chau and Hu (2001), Davis (1989), Venkatesh et al. (2003), and Taylor and Todd (1995), and are listed in the Appendix. SAMPLE The purposive sampling method was used for this study. Purposive sampling searches for a specific profile based on target respondent definition for the concerned survey. The target respondents for this study were: Primary decision makers regarding the payment of electricity bills. Professionals/self employed belonging to the middle and upper levels of management. The same definition of target respondents was used in the United States and India for meaningful comparison of the results. The respondents were located in urban areas in both countries. The survey questionnaires in the two countries were identical. Altogether, 762 and 502 usable responses were obtained from the United States and India, respectively. RESULTS ASSESSMENT OF THE PSYCHOMETRIC PROPERTIES OF THE MEASURES Table 1 displays the estimated error variances, item reliabilities (Cronbach s alpha), and factor loadings for the scale items. The relatively low error variances, as well as the high item reliabilities and factor loadings indicate the rigor of the measurement constructs used. MODEL ESTIMATION The parameters for the structural equation model illustrated in Figure 1 were estimated by the maximum likelihood method using AMOS 5.0. The model fit indices for the structural equation model in India and in the United States are listed in Table 2. All measures of the causal model for both countries exceeded the acceptable levels thereby exhibiting that the Structural Equation Model presented a good fit with the data. Table 3 shows the detailed model test results. 557

6 Table 1: Psychometric Properties of the Measures Construct Error Variances Item Reliability Factor Loadings (Cronbach s Alpha) India U.S. India U.S. India U.S. Performance Expectancy (PE) PE PE PE PE Effort Expectancy (EE) EE EE EE Social Influence (SI) SI SI SI Behavioral Intention (BI) BI BI BI Table 2: Overall Model Parameters Parameter Estimate - India Estimate - Recommended United States Value Chisquare/Degree <= 3.0 of freedom Goodness-of-fit index >= 0.9 Adjusted goodness-of-fit index Normal fit index Comparative fit index Standardized root mean square residual >= >= >= <=

7 Table 3: Parameter Estimates for the Structural Model Parameter Estimate - India Estimate - U.S. R 2 η R 2 ξ R 2 ξ R 2 ξ γ **.270 ** γ **.025 γ **.766 ** γ γ γ γ γ γ γ γ γ ** p-value <.001 The explanatory power of the model is examined using the R 2 value for Behavioral Intention (R 2 η 1 ). The combination of Performance Expectancy (PE), Effort Expectancy (EE), and Social Influence (SI) accounted for 72% and 86% of the variances observed in consumers intention to use the Prepayment Metering Systems technology in India and the United States, respectively. The path coefficients from PE and SI for both countries are significant at p <.001 level. The path coefficient from EE is significant at p <.001 level in India. Even though PE, EE, and SI are all significant determinants of BI, EE exhibited the strongest direct and total effects on BI in India and SI exhibited the strongest direct and total effects on BI in the United States. The effects of PE, EE, and SI were found to be moderated by gender, experience, income, and voluntariness to use the technology in various degrees in the two countries, as shown by the values of the γ- variables in Table 3. DISCUSSION The good fit of the overall model in the two diverse cultures of the U.S. and India indicates a trend toward increased globalization that seems to blur traditional national cultural differences. This would be particularly applicable to the context of our study (technology adoption), since technology is a prime driver of globalization. An examination of the demographics of the study population in the two countries found the educational qualifications to be quite similar. This could be a common factor in defining customer segments in the two countries that exhibit similar technology adoption behavior, and would be in accordance with Levitt s (1983) idea that similar customer segments are likely to emerge in multiple national markets. 559

8 In terms of the impact of the individual dimensions on behavioral intention, the only major difference observed between the two cultures was that effort expectancy was found to influence behavioral intention significantly in India, while it did not do so in the United States. A possible explanation of this outcome may be derived from Strite (2006), who applied Hofstede s (2001) cultural dimension of masculinity to explain the different ways in which perceived ease of product use (similar to effort expectancy in our study) might impact behavioral intention in different cultures. From Hofstede s (2001) findings, the United States ranks higher than India on the masculinity scale. Strite (2006) argued that cultures that are less masculine might be more concerned with the ease of use of a technology, since such cultures place less emphasis on instrumental goals and more on the quality of life. When instrumental goals are less emphasized, effort-free use of technology will be more important (Davis et al., 1989). CONCLUSION Our study compares the user adoption behavior of a new technology (prepayment metering systems) in two different cultural environments (the United States and India) by applying the UTAUT model (Venkatesh et al., 2003) in both situations. The good model fit in both countries provides evidence of technology globalization, while the difference between the two countries in terms of the impact of one factor (effort expectancy) indicates the continuing existence of some national cultural differences. DESCRIPTION OF SCALE ITEMS APPENDIX Performance expectancy (PE): PE1. Using prepayment metering systems can enhance my effectiveness in managing electricity consumption. PE2. Using prepayment metering systems can improve my EAM PE3. Using prepayment metering systems can increase my productivity in EAM PE4. Overall I will find prepayment metering systems useful for my EAM. EAM (Electricity Account Management) includes purchasing and budgeting for electricity, and monitoring electricity usage/consumption). Effort expectancy (EE): EE1. I would find it easy to use prepayment metering systems to accomplish my EAM tasks. EE2. My interaction with prepayment metering systems would be clear and understandable. EE3. I would find prepayment metering systems to be flexible to interact with. EE4. Overall I believe that prepayment metering systems would be easy to use. Social influence (SI): SI1. Those people who are important to me would support my using prepayment metering systems rather than conventional metering for EAM. SI2. I think that those people who are important to me would want me to use prepayment metering systems rather than conventional metering for EAM. SI3. People whose opinions I value would prefer me to use prepayment metering systems rather than conventional metering for EAM. 560

9 Behavioral intention (BI) to use the prepayment metering system: BI1. I would use prepayment metering systems rather than conventional metering for EAM when it becomes available to me. BI2. I intend to use prepayment metering systems rather than conventional metering for EAM when it becomes available to me. BI3. Given that I had access to prepayment metering systems, I predict that I would use prepayment metering systems rather than conventional metering for EAM. REFERENCES Chau, P., and Hu, P. (2001) Information technology acceptance by individual professionals: a model comparison approach. Decision Sciences, 32(4), pp Davis, F. D. (1989) Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), pp Davis, F.D., Baggozzi, R.P., and Warshaw, P.R. (1989) User acceptance of computer technology: a comparison of two theoretical models. Management Science, 35(8), pp Fishbein, M., and Ajzen, I. (1975) Belief, attitude, intention, and behavior: an introduction to theory and research. Reading, MA: Addison-Wesley. Ghosh, K. (2002) Prepayment metering: the Indian context, Proceedings of the International Seminar on Energy Conservation. Audit and Metering, Mumbai, India. Hofstede, G. (2001) Culture's Consequences: Comparing Values, Behaviors, Institutions, and Organizations across Nations. Sage Publications, Thousand Oaks, California. Levitt, T. (1983) The Globalization of Markets, Harvard Business Review, May-June, Prepayment metering report (2005), ABS Energy Research, London, U.K. Available at: Strite, M. (2006) Culture as an explanation of technology acceptance differences: An empirical investigation of Chinese and US users. Australasian Journal of Information Systems, 14 (1), pp Taylor, S., and Todd, P. (1995) Understanding information technology usage: a test of competing models. Information Systems Research, 6(2), pp 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 (4),