Information and Communications Technology ...

Information Technology for Development

ISSN: 0268-1102 (Print) 1554-0170 (Online) Journal homepage: http://www.tandfonline.com/loi/titd20

Information and Communications Technology Acceptance by Youth Entrepreneurs in Rural Malaysian Communities: The Mediating Effects of Attitude and Entrepreneurial Intention Zeinab Zaremohzzabieh, Bahaman Abu Samah, Mahazan Muhammad, Siti Zobidah Omar, Jusang Bolong, Salleh Bin Hj Hassan & Hayrol Azril Mohamed Shaffril To cite this article: Zeinab Zaremohzzabieh, Bahaman Abu Samah, Mahazan Muhammad, Siti Zobidah Omar, Jusang Bolong, Salleh Bin Hj Hassan & Hayrol Azril Mohamed Shaffril (2016): Information and Communications Technology Acceptance by Youth Entrepreneurs in Rural Malaysian Communities: The Mediating Effects of Attitude and Entrepreneurial Intention, Information Technology for Development To link to this article: http://dx.doi.org/10.1080/02681102.2015.1128384

Published online: 19 Jan 2016.

Submit your article to this journal

Article views: 49

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=titd20 Download by: [Universiti Putra Malaysia]

Date: 09 March 2016, At: 21:30

Information Technology for Development, 2016 http://dx.doi.org/10.1080/02681102.2015.1128384

Information and Communications Technology Acceptance by Youth Entrepreneurs in Rural Malaysian Communities: The Mediating Effects of Attitude and Entrepreneurial Intention ∗

Downloaded by [Universiti Putra Malaysia] at 21:30 09 March 2016

Zeinab Zaremohzzabieha , Bahaman Abu Samahb, Mahazan Muhammada, Siti Zobidah Omarc, Jusang Bolongc, Salleh Bin Hj Hassanc and Hayrol Azril Mohamed Shaffrila a Institute of Social Science Studies, Universiti Putra Malaysia Putra, Putra Infoport, 43400 Serdang, Selangor Darul Ehsan, Malaysia; bFaculty of Educational Studies, Universiti Putra Malaysia Putra, 43400 Serdang, Selangor Darul Ehsan, Malaysia; cFaculty of Modern Languages and Communication, Universiti Putra Malaysia, 43400 Serdang, Selangor Darul Ehsan, Malaysia

This study aims to examine information and communication technology (ICT) acceptance among youth entrepreneurs in rural Malaysian communities by employing two tailored technology acceptance models based on attitudes (AT) and entrepreneurial intention (EI) that influence actual use (AU). The study involved 400 rural youth entrepreneurs selected from four states in Malaysia, and both mediating effects were analyzed using bootstrapping procedures through structural equation modeling. The two models were tested, and vary in terms of different conceptualization of the pathways of mediating entrepreneurship latent factors. The models in the present paper are closely related to the model in a prior study that used the same data set [Zaremohzzabieh et al., 2015. A test of the technology acceptance model for understanding the ICT adoption behavior of rural young entrepreneurs. International Journal of Business and Management, 10(2), 158–169.]. The results revealed that the first model is better than the second model. The results also revealed that the confirmatory strength of the two models improved the initial TAM through some form of mediating effects (i.e. AT and EI). The outcomes of this study contribute to our theoretical understanding of variables that influence ICT acceptance, and inform practice by recognizing methods to improve ICT acceptance among rural youth entrepreneurs in the country. The results provide new insights for small rural businesses and help to explain ICT acceptance, which is relatively underresearched in these growing nations. Keywords: youth entrepreneurship; ICT acceptance; technology acceptance model; and rural community

Introduction Youth entrepreneurship is recognized as the foundation of rural community development. Over the past couple of years, many developing countries, including Malaysia, have come to acknowledge the fact that rural youth entrepreneurship is a possible source of economic development and business creation (Ahmad, Fauziah, Noor, & Ramin, 2011). According to Sidhu and Kaur (2006), entrepreneurship offers a simple answer to the problem of how to generate work for young people in rural communities, inside their own social arrangements. In Malaysia, the importance of enterprise development, mostly in the agricultural industry, is articulated within standard policies as a means of making a sustainable livelihood that will not only contribute to the mitigation of the rate of unemployment among young people in rural areas (Samah ∗

Corresponding author. Email: [email protected] James Pick is the accepting Editor for this article.

# 2016 Commonwealth Secretariat


2

Z. Zaremohzzabieh et al.

et al., 2009), but will also prompt them to obtain employment within the farming sector (Othman, Ghazali, & Cheng, 2005). Yet, despite this attention, existing statistics reveal that the majority of those currently working in the farming sector in the country are aged 45 years and over, while only a minority are aged in the range of 15– 40 years (Hassan, Shaffril, Abu Hassan, & D’Silva, 2009); furthermore, Norsida (2007) also claims that, among Malaysian youth, there is a low tendency to be involved in the agrarian industry. Thus, encouraging young people to become a part of the agrarian sector is a great challenge, and it is necessary to shift the paradigm of rural youth toward perceiving the agricultural sector as one of the opportunities for them to become agro entrepreneurs. Recently, most findings have revealed a substantial relationship between rural entrepreneurship and information and communication technology (ICT, Khurshid & Ghani, 2001). Batchelor, Scott, Woolnough, and Tambo (2005) specified that ICT is among the instruments required to eliminate poverty. Additionally, rural youth entrepreneurs perceive and use ICT not just as an opportunity to develop sustainable livelihood, but also as an opportunity to integrate themselves into their residential areas. Indeed, advancement in ICT has eliminated some of the limitations to such development, as it is a great generator of income, and extends tasks, boosts productivity, produces more jobs, and speeds up communication (Rahman, Mahfuz, Ahmed, & Rajatheva, 2005). It represents the hub of an economic development strategy where small businesses are able to effectively utilize information systems (ISs) to their advantage, and in turn can harvest the full benefit of their technology, and become more profitable. On the other hand, in spite of the numerous training programs convincing entrepreneurs to use ICT as a new way of performing a task, the overall acceptance of ICT by entrepreneurs is still quite low, as most entrepreneurs believe that IS adoption is very difficult (Ahmad, Abu Bakar, Faziharudean, & Mohamad Zaki, 2014). A previous study by Ramsey, Ibbotson, Bell, and Gray (2003) has shown that, for small and medium enterprises (SMEs), only 33% of self-employees owned a website, and that most did not accept e-entrepreneurship business cases. A number of possible explanations exist for this lack of adoption. Firstly, Malaysia, like many Asian countries, is less technologically developed than its Western counterparts, and as such, Malaysian entrepreneurs are still incapable of fully utilizing ICTs. Secondly, there is a lack of successful locally based ICT adoption models for these entrepreneurs to emulate. Therefore, further studies and investigation are required to seek more accurate and reliable ICT adoption models for ICT acceptance behavior of rural youth. Currently, a large body of literature focuses on and investigates ICT use behavior, and numerous models have been utilized to investigate IS acceptance among SMEs (Ramdani, Kawalek, & Lorenzo, 2009). Among several models to recognize the procedure of user acceptance of ISs, the technology acceptance model (TAM; Davis, 1989) has been utilized in most of the previous studies in the context of ISs. According to the TAM, user inspiration is elucidated by perceived ease of use (PEoU), perceived usefulness (PU) and attitude (AT) toward using an IS (Davis, 1989). The TAM postulates that a user’s acceptance of an information system is influenced by that user’s intention to use the system, which in turn is influenced by the user’s AT toward the system. Also, TAM2, an extension of the TAM, includes subjective norm (SN) as an additional key factor of the usage intention construct, which is required to elucidate the changes in IS acceptance over time, as users gain experience in using the system (Venkatesh & Davis, 2000). To date, numerous studies that have focused on the adoption of ICT by small businesses in Malaysia have duplicated, extended, and employed the TAM (e.g. Ndubisi & Jantan, 2003; Ramayah & Lo, 2007). Currently, a limited number of studies have utilized the TAM, modified with entrepreneurship latent factors, i.e. attitude toward entrepreneurship (AT) and entrepreneurship intention (EI), to answer the questions of how and why rural young entrepreneurs


3

adopt and use ICT, and also to trace actual ICT usage (AU) to classify any AT changes in outdated business behavior, and the intention of these entrepreneurs regarding marketplaces and commercial innovations. Thus, this study tries to examine the mediating effects through AT and EI. There is a basic research question guiding this study; how does AT (and/or the EI) mediate the association between each of PEoU, PU, and SN; and AU?


Literature review ICT and rural enterprises ICT has proven to be a potent instrument for the empowerment of rural residential areas. ICT is an electronic method of capturing, processing, storing and distributing data (Duncombe & Heeks, 2002). Grimes (2000) noted that ICT offers opportunities to rural populations to explore new things, and this will make a rural community more knowledgeable. If ICT can be utilized wisely by the rural community, it can sweep over the disadvantage of the distance from the city, isolation, and remoteness, and the underdevelopment of rural regions. Currently, the functions of ICT help businesses create channels, open opportunities, and connect people for the development of firms. The use of ICT can also formulate a singular perspective in the areas of management, marketing, processing and purchasing goods and services. Empowering small enterprises to manufacture, produce, and access new markets has become a key component regarding the consumption of ICTs in emerging economic systems. According to BarbaSańchez, Martıńez-Ruiz, and Jimeńez-Zarco (2007), it can stimulate small enterprises to embrace new technologies, make modern products, and be competitive. Early studies have indicated that the usage of ICT can take on an important part in the development of small enterprises. In a previous case study in Botswana, Duncombe and Heeks (2002) discuss the role of ISs in many rural microenterprises. The findings of the study proved that the information needs of rural micro-initiatives are likely to gain more exposure by localized informal ISs (social networks), rather than by the formal aspects of ICT-based systems. They claim that information delivered through social linkages was often restrained and limited because the social network of the entrepreneur himself was limited in both quality and size. At this stage, the impact of ICT on the development of information must be considered more cautiously. Qureshi’s (2005) model of information technology for development suggests that ICTs can aid these businesses to attain the greatest benefits in terms of administrative efficiencies, learning and labor productivity, competitiveness, and access to new markets, knowledge, and expertise (or more generally, stakeholders). The end result of these improvements leads to the alleviation of poverty. Eggleston, Jensen, and Zeckhauser (2002) claim that the role of ICT is mainly to enhance the market functioning that is important for the welfare of poor people. Furthermore, they state that ICTs allow manufacturers to plan their product in an inexpensive way, and producers to trade their products directly to the most lucrative markets. It regulates the best sale time, access to more credit opportunities and price information, and augmented information about the accessibility of jobs. Hence, it could potentially result in faster matching between available jobs and landless agricultural laborers. In essence, the use of ICTs can play an important role in strengthening ISs development practice, and offers incredible opportunities for the poverty reduction and employment generation, as well as bringing on higher achievement among those who control it. ICT acceptance and rural youth entrepreneurs Today’s youth is different from youth of the past. They are exposed to digital technology in almost all aspects of their daily existence (Tapscott, 1998). They are characterized as the


4


“Net Generation” (Tapscott, 1998), “Cyberkids” (Valentine & Holloway, 2002), as well as “techsavvy,” “wired,” and “specialists” in the field of utilizing ICT. Their habit of using ICTs is important for development results within the effects of youth changes – and directly through the great contribution to general ICT use. ICT has allowed young villagers to become urbanized without necessarily drifting to cities. Their access to the ICT of network services normally coordinates urban and rural development, increases knowledge, and changes any negative AT toward rural enterprises. For example, the increasing acceptance of ICT in the marketplace for digital products and services generates more opportunities for youth to find jobs in rural areas. According to Curtain (2003), some developing countries put effort into creating selfemployment for young people in rural areas via public access to ICT, such as telecenters, and these ICT facilities in themselves can offer small and micro enterprises for rural youth entrepreneurs. This has particularly been the case in Malaysia, wherein the agricultural sector is becoming more knowledge-intensive; and as this trend intensifies, opportunities in ICT-related jobs for rural youth become more abundant (Hassan, Shaffril, & Bahaman, 2012). In addition, ICT increased the incentive for young Malaysian farmers to get involved in farming, as well as their competence in relation to it, by providing up-to-date agricultural information at any time or place, thus enabling them to achieve a better quality of life. Moreover, ICT enables farmers to create networks with other farmers and development agencies; therefore, ICT can increase their ability to strengthen and develop their agriculture business. For instance, mobile telephones have been known as a successful instrument of market ISs in Malaysia’s agricultural development, because it enables access to market prices and other useful information. The results in an earlier study also indicate that mobile telephones supply agro-businessmen with needed information every time they demand it, hence improving networking to obtain expert information on farming and agriculture (Shaffril, Hassan, & Samah, 2009). Thus, encouraging the rural youth to run enterprises implies ICT use and acceptance which is the only way to secure information about market relationships that can be exploited to enhance ongoing competitive advantage in the rural community. This may, in turn, progressively be influenced by entrepreneurial factors of the rural youth in Malaysia. Theoretical background The TAM One of the common models related to technology acceptance and use is the TAM, which was used as the theoretical background for this study to investigate ICT acceptance by youth entrepreneurs in rural Malaysian communities. According to Venkatesh and Bala (2008), the prognostic power of the TAM model makes it easy to relate variables predicting ICT acceptance. It was originally developed by Davis in 1989, which itself was derived from the Theory of Reasoned Action model by Ajzen and Fishbein (1980). Figure 1 shows the primary TAM, which determines the influence of five internal variables upon the real use of a specific technology or system. The internal variables in the primary TAM are: two cognitive beliefs constructs, PEoU and PU, an affective construct, AT, an intention construct, BI, and a behavioral response construct, AU (Davis, Bagozzi, & Warshaw, 1989). The PEoU construct mentions the degree to which individuals agree that the purpose of the object system (Davis, 1989). Venkatesh and Davis (2000) confirmed that both PEoU and PU directly influence BI, as well as AU. Furthermore, PU is one of the earlier cognitive belief constructs improved by the TAM. It refers to the level a person agrees that the diligence of the target system helps them in furthering their work performance (Davis, 1989). Additionally, the TAM also helped to develop theoretical frameworks for gauging the link between PU and PEoU with AU, through cognitive AT and BI as mediating agents. “AT”



5

Figure 1. TAM (external variables ¼ EVs, perceived usefulness ¼ PU, perceived ease of use ¼ PEoU, attitude ¼ AT, behavioral intention ¼ BI, actual usage ¼ AU; Davis, 1989).

refers to a user’s insight of negative or positive affective approaches linked to the use of a target system. If rural youth entrepreneurs have a strong AT for starting a new venture, the relationship between AT and behavior is strong, and they have a strong inclination toward entrepreneurship. BI can also be seen in context of entrepreneurship development as entrepreneurial intention (EI). In this study, the intention used is conceptualized as the degree to which rural youth entrepreneurs have formulated conscious plans to use ICT to improve their businesses. According to the TAM, AU is influenced directly or indirectly by other TAM variables. As expressed by Davis (1989), AU is encouraged by intention to use ICT, which in turn is driven by the user’s AT and idea about the system. Besides, Davis et al. (1989) also discovered the presence of external variables that characterize distinctive alterations, situational restrictions and directorially manageable interferences to be significant causes of PU, PEoU, and even the behavior in using the system. With the intention of addressing the concerns by Davis (1989) regarding the external variables toward PEoU and PU, Venkatesh and Davis (2000) suggested a theoretical extension of TAM, which is referred to as the TAM2. The aim of the TAM2 is to test the factors of usefulness and usage, and to study how these affect modification over time, as users gained experience with the ISs. Likewise, the TAM2 eliminates AT toward use as an antecedent of behavioral intention (BI) (Venkatesh & Davis, 2000). The TAM2, a revised version of the original TAM, is exhibited in Figure 2. It suggests factors from two central categories: social influence processes (SN, voluntariness (V), and image (I)), and cognitive instrumental processes (job relevance (JR), output quality (OQ), result demonstrability (RD) and PEoU, that influence PU. Considering the shape of relationships in the model, one main concern is the traditionally weak role of SN among these antecedents. In this latter case, some studies have simply excluded SN (Veciana, Aponte, & Urbano, 2005), while others discovered them to be non-significant (Krueger, Reilly, & Carsrud, 2000). Likewise, in a meta-analysis of the theory of planned behavior (TPB), Armitage and Conner (2001) uncovered SN to exert the weakest impact on design. SN is defined as the approval (or disapproval) that important referent individuals (or groups) have in relation to the enactment of a given behavior. Other studies (e.g. Davidsson, 1995) have assumed that SN is embedded within perceived desirability. In Ajzen’s model, SN describes the perceived social pressure to perform a behavior. While the concept of SN refers to a role of the normative beliefs of significant others weighted by the individual’s motive to follow every normative belief (Schepers & Wetzels, 2007), SN is analytically verified to be the most effective factor of AT/BI, especially when users have little skill with the technology (Venkatesh & Davis, 2000). In this regard, the likelihood of indirect influences of SN on AU must be further investigated.


6


Figure 2. TAM2 (subjective norm ¼ SN, voluntariness ¼ V, and image ¼ I, job relevance ¼ JR, output quality ¼ OQ, result demonstrability ¼ RD, perceived usefulness ¼ PU, perceived ease of use ¼ PEoU, behavioral intention ¼ BI, actual usage ¼ AU; Venkatesh & Davis, 2000).

Empirical applications of TAM Enterprises research. As Davis (1989) proposed this model, numerous surveys have been conducted applying it in information technology usage settings, testing its suitability and changing it in many different contexts. Lately, in line with the growth of ICTs, applications of the TAM have been established in the fields of enterprise contexts, and many researchers use the TAM to recognize the conditions or factors that could help the consolidation of ISs into SMEs. In the small and medium firms (SMFs) setting, Ndubisi and Jantan (2003) utilized the TAM model to describe the influence of identity system features, technical backing, and computing skill on ISs usage by Malaysian SMFs. The findings revealed that there is a positive association between computing skill and technical backing, on one hand, and IS usage directly and indirectly through PU and ease of use, on the other hand. The researchers interchanged user intention with symbolic adoption, as in earlier works, and revealed that PU, PEoU, and AT have direct and indirect effects on symbolic adoption. Furthermore, the current study upon the same datasets of two previous papers have been conducted to investigate EI in Malaysia using the entrepreneurial potential model and TAM in the context of ICTs (Zaremohzzabieh et al., 2015; Zaremohzzabieh et al., in press). The results of the first study showed that AT mediated the link between PU and EI (Zaremohzzabieh et al., 2015). The results from second study confirmed that entrepreneurial knowledge and AT have a direct influence on EI (Zaremohzzabieh et al., 2015; Zaremohzzabieh et al., in press). The model in the first study (Zaremohzzabieh et al., 2015) is closely related to the present Models I and II. However, the present Model I differs from the earlier model (Zaremohzzabieh et al., 2015) in that the factor actual use (AU) replaces EI and the factor SN is added. The present Model II differs from the earlier model in adding the factors, AU and SN. Information technology contexts. Since 1989, the TAM model has been widely examined and supported as a valid model in predicting individual acceptance behavior across diverse



7

information technologies and their users in work and academics. In some prior studies in the IT literature, the TAM has been proven to be a robust model for understanding and work on ISs implementation, predicting about 40% of IS utilization (e.g. Nah, Tan, & Teh, 2004; Venkatesh & Davis, 2000). For example, many studies tried to apply the TAM in IT-related cases, in which the TAM has effectively described human behavior to accept a certain number of IT applications. These studies relate to previous researches about the adoption of IT hardware (Chen, Chen, & Yen, 2011), IT software (Gallego, Luna, & Bueno, 2008), web usage, and online applications (Li & Qiu, 2008). In Malaysia, some scholars used the TAM model in predicting Internet shopping, Internet banking, e-library usage, and personal computer use among scholars (Ramayah & Lo, 2007; Ramsey et al., 2003). A study of five Indian states by Pick, Gollakota, and Singh (2014) established that use of telecenters could improve the income and capabilities of 280 farmers who were selected for their study. The authors also attempt to realize factors that influence adoption and use of information technology by these farmers. In doing so, they used the dimensions of technology adoption, the diffusion theory of Rogers (2003) (i.e. relative advantage, compatibility, low complexity, and observability), and two dimensions of the TAM (i.e. usefulness and ease of use). The result of this study indicated that the dimensions of the diffusion theory and the TAM can significantly improve the use of telecenters by farmers. In summation, a small number of studies have applied the TAM2 in different settings, such as health information technology adoption among medical staff, adoption of Internet banking among undergraduate and graduate students, and consumers’ adoption of Internet TV (e.g. Chan & Lu, 2003). Thus, a review of these studies identified that the original TAM and the TAM2 are as robust to test perceptions of ICT adoption in the context of IT and rural enterprise development.

Research models and hypotheses Based on two questions outlined in this study, two models are proposed to determine the efficacy of entrepreneurship constructs (AT and EI) in explaining rural youth entrepreneurs’ use of ICT in Malaysia. These models integrate key constructs of the TAM and the TAM2. The constructs include PU, PEoU, SN, attitude toward entrepreneurship (AT), entrepreneurship intention (EI), and actual ICT usage (AU). In line with the first model, the findings of previous studies revealed that there exists a direct effect between selected independent variables and AT, as well as AU (e.g. Kleijnen, Wetzels, & de Ruyter, 2004; Venkatesh & Bala, 2008). For example, Kleijnen et al. (2004) found that the perceptions of PEoU, PU, and SN are positively associated with AT. In view of the above, it is expected that a higher intensity of entrepreneurial AT leads to a greater intention to use ICT by rural youth entrepreneurs, and the following 10 hypotheses were supported by these studies: H1a: There is a positive relationship between PU and AT. H2a: There is a positive relationship between PEoU and AT. H3a: There is a positive relationship between SN and AT. H4a: There is a positive relationship between AT and actual use of ICT (AU). H5a: There is a positive relationship between PU and actual use of ICT (AU). H6a: There is a positive relationship between PEoU and actual use of ICT (AU). H7a: There is a positive relationship between SN and actual use of ICT (AU). H8a: AT mediates the effects of PU on actual use of ICT (AU). H9a: AT mediates the effects of PEoU on actual use of ICT (AU). H10a: AT mediates the effects of SNs on actual use of ICT (AU).


8


The second model includes EI. It suggests that there are direct and indirect effects between PU, PEoU, and SN, and AU, EI, and AT, with shared AT and EI as potential mediating factors. The model posits that, even though the mediating effect of AT on the value-behavior link has been verified (Homer & Kahle, 1988), a number of theories, including the TPB, dominate EI research (e.g. Carr & Sequeira, 2007). Within the TPB, attitudes toward a given behavior are the closet cognitive antecedents to BI, and thus explain actual behavioral performance. Therefore, it is argued that the TAM may be improved by considering the mediating effect of these two constructs. Some researchers have already given proof for this is empirical studies. Lee, Fiore, and Kim (2006) use a simplified version of the TAM to explain the effects of image interactive technologies on AT and BI toward online retailers. They found that the TAM’s success in predicting behaviors from AT and BI is well documented in business and technology applications. Furthermore, Moon and Kim (2001) found considerable support for the mediation of AT and BI. In accordance with the related literature, this study tested the following hypotheses: H1b: There is a positive relationship between PU and EI. H2b: There is a positive relationship between PEoU and EI. H3b: There is a positive relationship between SN and EI. H4b: There is a positive relationship between EI and actual use of ICT (AU). H5b: There is a positive relationship between PU and actual use of ICT (AU). H6b: There is a positive relationship between PEoU and actual use of ICT (AU). H7b: There is a positive relationship between SN and actual use of ICT (AU). H8b: There is a positive relationship between PU and AT. H9b: There is a positive relationship between PEoU and AT. H10b: There is a positive relationship between SN and AT. H11b: There is a positive relationship between AT and actual use of ICT (AU). H12b: AT and EI mediate the effects of PU on actual use of ICT (AU). H13b: AT and EI mediate the effects of PEoU on actual use of ICT (AU). H14b: AT and EI mediate the effects of SNs on actual use of ICT (AU).

Material and methods In this study, a quantitative approach was employed, whereby data were collected by way of a self-administered survey. The primary units of analysis of the study were rural youth entrepreneurs aged between 17 and 40 in rural Malaysian communities. Grounded on a national pilot survey in 2006 in Malaysia, the total rural persons employed in enterprises were 265,100 and, out of this, a total of 400 respondents from the states of Perlis, Selangor, Negeri Sembilan, and Terengganu were randomly chosen through a simple random sampling technique to take part in the survey. Forty-nine items utilized to the constructs of all variables (i.e. PU (9 items), PEoU (7 items), PSN (11 items), AT (8 items), and EI (7 items)) in this study were mostly adapted from relevant earlier papers and are displayed in Table 1. The sample of 400 for this study met the requirement to use a structural equation modeling (SEM), as the ratio of sample size to estimated parameters should be at least 10:1 (Chou & Bentler, 1995; Kline, 2005). According to Hair, Black, Babin, Anderson, and Tatham (2010), the preferred minimum number of indicators is three; however, in most cases, five to seven indicators are commonly used to represent a construct. The indicators properly were successfully validated in a pilot study in the context of entrepreneurship and IT. We also rearrange the points to reduce monotony of statements gauging the same constructs. The questionnaires were prepared in Malay language to enable better understanding and conclusion. All items were rated on a 5-point Likert-type


9

Table 1. Question items used in this study. Constructs PU

Item PU1 PU2 PU3 PU4 PU5 PU6 PU7 PU8


PU9 PEoU

PEoU1 PEoU2 PEoU3 PEoU4 PEoU5

PEoU6 PEoU7 Attitude toward entrepreneurship (AT)

AT1 AT2 AT3 AT4 AT5 AT6 AT7 AT8 AT9

SN

SN1 SN2 SN3 SN14 SN15

Measure ICT knowledge is utilized to enhance my business activities ICT skill is utilized to enhance my business activities ICT helps me to dominate new market spaces ICT helps me to save time in my business ICT improves the quality of my business ICT doubles the quantity of my business ICT helps me to store large data capacity for business planning purposes ICT accelerates the process of two-way communication in my business Computer software to minimize grammatical errors in writing letters, stories, statistics, schedule, or schedule of my business activities Typing and performing editing tasks using a computer are easy Business information seeking via the Internet is easy Updating my business software is simple Utilizing the services of Internet search engines or “search engine” such as Yahoo, Google and MSN is easy Use of applications such as Yahoo Messenger, Gmail, Yahoo Mail, Skype, Facebook, Twitter and other online services for the purpose of communicating and disseminating/sharing information is easy Online service provided by the agency concerned is easy to apply It is easy for me to employ the latest ICT tools despite the accession of new functionality of the equipment The benefit of doing entrepreneurship activities for me is more than its consequences Career as an entrepreneur is interesting to me If I have the opportunity and the capital, I would like to start a business Becoming an entrepreneur gives more satisfaction and makes me feel good at all stages If I become an entrepreneur I believe I can compete with my entrepreneur friends Venture into entrepreneurship to secure the future of my kin In the Immediate family, I think entrepreneurial activity via ICT is worth, worse or more salutary than the activities and other careers Among my friends, I think entrepreneurial activity via ICT is worth, worse or more salutary than the activities and other careers Among your close friends, I think entrepreneurial activity via ICT is worth, worse or more salutary than the activities and other careers Entrepreneurship is a culture in my residential area Entrepreneurs play a vital role in the economy of a community Many people take for granted that it is so difficult to become entrepreneurs Activities viewed as too risky entrepreneurs to follow It is common when people say entrepreneurs to hold advantage of others (Continued)

10


Table 1. Continued. Constructs EI

Item EI1 EI2 EI3 EI4 EI5 EI6 EI7 EI8 EI9


E10 E11 EI12 Actual use of ICT (AU) in entrepreneurship

AU1 AU2 AU3 AU4 AU5 AU6 AU7

Measure I am serious about being a successful entrepreneur I am ready to do whatever it needs to be an entrepreneur Being an entrepreneur would be my ultimate goal I will try to run my own business I aim to create a future business I am seriously thinking of starting a business I hold the intention to initiate a business one day From the choices available, I choose to be an entrepreneur If I decide to form a business by ICT whether the immediate family agrees with the decision If I decide to form a business by ICT whether My friends agrees with the decision If I decide to form a business by ICT whether my close friends agree with the decision I induce a propensity to be an entrepreneur because of my Information on product market entrepreneurship Looking for info on the benefits realized in the area of entrepreneurship Looking for info about the problems and obstacles that will be met in the area of entrepreneurship Searching information about entrepreneurial opportunities offered by government agencies/individual Searching info on loan deals for entrepreneurship Looking for info about opportunities to study in the area of entrepreneurship Looking for info about the course/seminar held entrepreneurial Reading success stories of other entrepreneurs along the net

scale, with anchors (1) strongly disagree (0) and strongly agree (4). In order to address the research questions of the study, the bootstrap method was used through the SEM approach as the covariance-based SEM to test the significance of the indirect and direct effects. The SEM analysis was taken using a measurement model analysis and structural model analysis (Anderson & Gerbing, 1988). In this study, the statistical software package utilized for SEM was AMOS. Descriptive statistical analyses (i.e. Frequency, percentage, mean and standard deviation) also were employed in order to analyze the profile information provided by respondents. Results Profile of respondents The data shown in Table 2 will give a clearer impression of the demographic of the respondents, including data on gender, age, race, level of education, and monthly income. A total of 64% of the male respondents, versus 36% of females, believe that men have potential for entrepreneurship activities in the rural community. According to Haji Idris (2008), the majority of rural women in Malaysia are fearful of the risks involved in being entrepreneurs. From the perspective of history, we know better, Malaysian women have been involved in the creation of business for a long time, only in a lower portion compared to men. In addition, there are clear racial differences with local Malay having more potential entrepreneurs (97%) than any other racial groups. The respondents were sorted into three age categories. Those in the middle-youth stage (20 – 30) formed the majority, 40.5%.


11

Table 2. Basic characteristics of respondents (n ¼ 400).


Factors Gender Male Female Age Early youth (17 –19) Middle youth (20 –30) Late youth (31 –40) Race Malay Chinese Indian Others Education achievement Never been to school/primary school Secondary school Skill certificate Tertiary Income ,RM750 RM751-RM1500 .RM1500

Frequency

Percentage

256 144

64.0 36.0

111 162 127

27.8 40.5 31.7

389 4 6 1.3

97.3 1.0 1.5 0.3

13 254 29 104

3.3 63.5 7.3 26.0

54 144 53

21.5 57.4 21.1

Mean

26.2

1355.42

Standard deviation

7.30247

911.64

In terms of educational achievement background, most respondents (63.5%) have obtained education at secondary school, while about 3% (n ¼ 13) of the respondents did not have any formal education and also complete even primary education, and 26% (n ¼ 104) and held tertiary qualifications almost 7% (n ¼ 29) held skill certificates. That means, more than half of selected rural youth entrepreneurs in Malaysia have secondary education and are low-skilled. Likewise, an earlier study of low-medium entrepreneurs (SMEs) in Kelantan area, west Malaysia, showed that the vast majority of owners of SMEs had finished secondary school qualifications (Yusoff, Yaacob, & Ibrahim, 2010). Out of 251 respondents, a total of 144 of respondents earned RM751– RM1500 per month, constituting the biggest percentage (36%), and the average score for income per month was RM1355. The entrepreneurial activity of each of the respondents can vary considerably over time and often. The area cited most frequently as a primary source of Agra-based sector was propelled by crops, livestock, fishing sub-sector and also manufacturing sector backed by rubber and plastic wares, food, drinks and tobacco sub-sector, and other manufacturing. Under the service sector, the majority of rural small businesses are in wholesale and retail trade (e.g. Provision store owners, vendors involved in retail sale of vegetables, and mini market owners), accommodation and restaurant (e.g. Food stalls owners/hawkers, restaurant business owners, and coffee shop owners), and transport (e.g. Motor vehicle workshop owners and taxi service business owners) (see also Tuyon, Mohammad, Junaidi, & Ali, 2011). On economic contribution, enterprises formed about 80% of business establishment in Malaysian Census, 2005.1 These youth play a significant function in the informal and formal enterprise supply chain and the addedvalues are related to explore the market niches by themselves that larger firms are not capable of undertaking effectively.


12


Structural Equation Modeling This study utilized an SEM approach as the main statistical method. The main benefits of SEM are that it stipulates the scholar with a chance to implement a more holistic approach to model structure. Particularly, SEM is beneficial to scrutinize association directly and indirectly among independent variables and dependent variables, once a dependent variable in one equation converts an independent variable in additional equation (Hair et al., 2010). In addition, SEM has a common character which is to adapt the bias in the estimates because of the measurement error related to unsatisfactory measures in data through multiple indicators for all latent variables; thus, SEM can stipulate more precise parameter estimates and augmented statistical power. Furthermore, SEM assesses indirect effects among latent constructs which led to the estimation of the total effect; however, in multiple regressions, an indirect effect is generally ignored when a hypothesized direct effect is of value and then the variable or the relationship is totally discharged. For all these details, SEM is appropriate for examining mediation effects of EI and AT on actual ICT usage of Malaysian youth entrepreneurs, using the bootstrap method. The overall procedure of SEM consists of two functions: the measurement model and the structural model. Before running these two steps for structural equation model building, a test of normality by maximum likelihood estimation is required to operate for screening the data. Maximum likelihood estimation is a method employed to estimate models normally delivered. In this study, an assessment of normality exhibited that the Skewness ranged from 21.304 and .067 and kurtosis ranged from 21.67 and .93 in which the data were considered to be normal and there was no item being non-normal. In accordance with the suggestion of Byrne (2010), all items have acceptable values of Skewness (values between 22 and +2) and kurtosis (values between 27 and +7). Based on the outcomes of screening of the factor structure of indicators and constructs, the measurement model was investigated through confirmatory factor analysis (CFA). CFA is the first step in data preparation in SEM for the individual construct which is employed for testing model fit, convergent validity, and construct reliability (CR). For purposes of this stage, we declared the loading of the manifest indicators on their particular unobserved latent constructs and also examined the model fit. In this study, the required measurement model was enhanced regarding the entire standardized factor loadings that must be more than .05, not more than 1, and positive (Hayes, 2009). These results also confirmed the validity of a 33-item, 6-factor model that included SN, PU, perceived ease to use (PEoU), AT, EI, and actual ICT use (AU) which is indicated in Table 3. Furthermore, a group of descriptive goodness-of-fit indices were consulted to assess model fit in the confirmatory models. The fit statistics in Table 3 displays a reasonably good fit for the model with x2/df ¼ 2.736, AGFI ¼ .806, CFI ¼ .907, NFI ¼ .835, TLI ¼ .898, IFI ¼ .908, root mean square error of approximation (RMSEA) ¼ .066. The results of the CFA in relation to the measurement models for the constructs of the sixfactor model were evaluated for convergent validity. Precisely, convergent validity can be Table 3. Model-of-fit estimation of the research model. Fit indices Relative chi-square Adjusted Goodness-of-fit Index Comparative Fit Index The Tucker –Lewis index Fit Index Normal Fit Index RMSEA

Threshold

Authors

Results obtained

,5.0 ≥0.9 ≥0.9 ≥0.9 ≥0.9 ≥0.9 ,0.08

Hu and Bentler (1999) Chau (1997) Kleijnen et al. (2005) Hu and Bentler (1999) Bentler and Bonett (1980) Bentler and Bonett (1980) Byrne (2010)

2.736 0.806 0.907 0.898 0.908 0.835 0.066


13

Table 4. AVE and CR of study instruments. Construct

Initial items

Final no. of items

AVE

CR

9 7 5 9 12 7

7 6 4 4 7 5

0.574 0.579 0.647 0.581 0.653 0.650

0.904 0.891 0.803 0.752 0.908 0.902


PU PEoU SN AT EI AU

assessed by two analyses: CR, and average variance extracted (AVE). In the first analysis, construct composite reliability is assessed based on the criteria that the indicator’s assessed pattern coefficient is significant on its fundamental factor by all constructs’ composite reliability (CR . 0.7; Hayes, 2009). In the second analysis, the AVE for each construct to specify the quantity of variance in the item clarified through the construct and based on the criteria, measurement error should be greater than .05 (Fornell & Larcker, 1981) to indicate high convergent validity. Results of convergent validity testing showed that the model CR ranged between 0. 752 (AT) and 0.908 (EI) and the AVE values ranged between 0.574 (PU) and 0.653 (EI). From the results of CR, and AVE, it can be settled that all constructs measured in this study acceptably determine reliability. The summary of the findings is presented in Table 4. The second step in data preparation in SEM is to test for discriminant validity. Discriminant validity denotes the degree to which measures of different constructs are distinct from each other (Hayes, 2009). To reinforce discriminant validity, the AVE should be better than the squared correlations between the construct and all other variables (Fornell & Larcker, 1981). Table 5 displays the result of the discriminant analyses; it is evident that the six constructs are different from each other. Test of the models Two models were fit to the data to decide which model offered the most reasonable explanation of ICT acceptance among rural youth entrepreneurs. Additionally, model fit statistics are revealed in Table 7, findings suggesting that most of the fit indices meet the requirements for SEM analysis; even though the values for GFI and AGFI do not exceed 0.9 (the threshold value), they still meet the requirement proposed by Baumgartner and Homburg (1996): the value is satisfactory if above 0.8. The values of RMSEA for these models are under 0.08. The confidence interval (CI) at a 90% level does not include 0.08, meaning that it is not a coincidence that RMSEA was less than 0.08; so, the fit indices propose that all of the three models provided a good fit to the data. Table 5. Discriminant validity – comparing AVE and squared correlation coefficients? (R2) for study instrument. Construct AVE PU PEoU SN AT EI AU

PU

PEoU

SN

AT

EI

AU

0.574 1.00 0.346 0.138 0.194 0.171 0.095

0.579

0.647

0.581

0.653

0.650

1.00 0.053 0.094 0.076 0.056

1.00 0.249 0.437 0.138

1.00 0.539 0.203

1.00 0.116

1.00

14


Table 6. Model comparison. Model

x2

x2/df

GFI AGFI RMSEA TLI


Model x 720.301 2.492 0.875 0.849 (Figure 2) Model xx 1145.112 2.545 0.844 0.817 (Figure 3)

CFI

IFI

ECVI

AIC

BIC

P

0.061

0.921 0.930 0.930 2.116

844.301 1091.772 .000

0.062

0.909 0.917 0.918 3.261 1301.112 1612.447 .000

Model comparison For the purpose of comparing the two models, we used the fit indices developed for model comparison including expected cross validation index (ECVI), Akaike Information Criterion (AIC), and the Bayesian Information Criterion (BIC). Smaller values indicate better model data fit. Between these two models, the first model is superior to the other model on the fit indices stated above (see in Table 6). The AIC values from the first model in the second model are 844.301 and 1301.112. The BIC values from the first model in the second model are 1091.772 and 1612.447, respectively. The ECVI values from the first model to the second model are 2.116 and 3.261. The first model has the lowest values on AIC, BIC and ECVI. Thus, these three indices also confirm the result that the first model is better than the second model. Along with the amount of variance accounted for in the endogenous variables by the exogenous variables in this study, the results of Model I showed that the squared multiple correlations linked to the two endogenous variables, AT and AU, were 0.383 and 0.234, respectively. The squared multiple correlations for Model II showed a slightly higher proportion of the variance explained in the exogenous variables to AT (40%), EI (54%), and AU (25%). While the above indication proposes that Model I is in better fit through fit indices, higher variance explained relative to AU was present in Model II. Structural model analysis Two models were tested in order to examine the hypothesized associations among latent variables. The purpose of the two models were to test the mediating role of EI and AT in actual ICT use (AU) by youth entrepreneurs in Malaysian rural community. The hypothesized direct effects and mediated effects (direct effects) were tested following the SEM approach and also the bootstrap procedure. The latest technique of analyzing the mediation effect is referred to as “bootstrapping,” and presents an estimation of the importance of the indirect effect and its statistical significance, and thus establishes CIs for the point estimate (Mallinckrodt, Abraham, Wei, & Russell, 2006). Therefore, in this study, the process of bootstrapping was used to see both of the standardized indirect effects, together with direct and the standardized Table 7. Mediation of the effect of AT on independent variables through AU (actual ICT use). Independent variable (IV) PEoU SN PU

Total effects (SE)

Direct effect (SE)

Indirect effect (SE)

0.068 0.307∗∗ 0.148

0.053 0.174∗∗ 0.072

0.015 0.133∗∗ 0.076∗∗

95% CI Lower

Upper

20.022 0.063 0.070 0.219 0.029 0.150

R2 AT ¼ 0.383 AU ¼ 0.234

Note: Bootstrapping CIs; 5000 bootstrap samples; perceived usefulness (PU); perceived ease of use (PEoU); perceived social norm (PSN). ∗ p , .05. ∗∗ p , .01.



Figure 3. Path analysis for Model I. Note: For all estimates ∗ p , .05, ∗∗ p , .01,

∗∗∗

15

p , .001.

total effect. In an indirect effect, an exogenous variable affects an endogenous variable via the mediation of at least one other variable (Kaplan, 2000) if the bias fixed and accelerated bootstrapping CIs (set at 95% from 5000 bootstrap samples) did not comprise zero. Direct effect analysis of the models Figures 3 and 4 show the results of path coefficients and the hypothesis of the two models. The path coefficient is shown for each path, with the significance as asterisked, and the R2 is shown under each endogenous variable. After the final structural model was found to fit the observed data satisfactorily, the path coefficients were examined for their statistical significance. With respect to Model I, model fit to the data indicates good fit with; x2 (289) ¼ 720.301, p ¼ 0.000, x2/df ¼ 2.492, RMSEA ¼ .061, CFI ¼ 0.930, IFI ¼ 0.930, and TLI ¼ 0.921. Four standardized path coefficients were significant (see in Figure 3). According to Figure 3, the analytical result is consistent with Hypothesis 7a; the SN had a strong direct effect on AU (b ¼ 0.174, CR ¼ 2.719, p ¼ 0. 007). Furthermore, in accordance with Hypothesis 1a, the PU significantly and directly affected AT (b ¼ 0.246, CR ¼ 3.634, p ¼ .000). Hypothesis 3a on the relation between the SN and AT was supported (b ¼ 0.430, CR ¼ 7.285, p ¼ .000). Finally, consistent with H4a, AT directly affected AU (b ¼ 0. .310, CR ¼ 4.499, p ¼ .000). Figure 3 also demonstrates that 35.6% of the variance of AT was explained by all exogenous variables and 24.2% of the variance of AU was explained by the model.


16


Figure 4. Path analysis for Model II. Note: For all estimates ∗ p , .05, ∗∗ p , .01,

∗∗∗

p , .001.

Three endogenous variables (i.e. EI, AT, and AU) were tested in the model II. Similar to the model I, more than half the variance in EI (53.7%) is explained by PU, PEOU, and SN. This model also explains a substantial proportion of the variance in AT and AU (40% and 25.4%, respectively). The second model exhibited a good fit with the data; x2 (450) ¼ 1145.112, p ¼ .000, x2/df ¼ 2.545, RMSEA ¼ 0.062, CFI ¼ 0.917, IFI ¼ 0.918, and TLI ¼ 0. 909. As may be seen in Figure 4, the six hypotheses were supported in the model II. With respect to parameter estimates shown in Figure 4, there were six hypotheses (i.e. H1b, H3b, H7b, H8b, H10b, H11b) that reached significance. Specifically, the parameter between SN and AU was significant (b ¼ 0.219, CR ¼ 2.467, p ¼ .014). In addition, the parameter between PU and AT was significant (b ¼ 0.240, CR ¼ 3.635, p ¼ .000) as was the parameter between SN and AT (b ¼ 0.480, CR ¼ 8.145, p ¼ .000). The parameters between PU and EI (b ¼ 0.167, CR ¼ 2.951, p ¼ .000), between SN and EI (b ¼ 0.650, CR ¼ 11.154, p ¼ .000), and between AU and AT (b ¼ 0.329, CR ¼ 4.580, p ¼ .000) were significant. Mediating effect analysis of the models Model I showed that AT was the only mediating variable between exogenous variables and the endogenous variable, AU. Table 7 displays that AT did not mediate the effect of PEoU on AU (b ¼ .015, p ¼ .377; 95% bootstrap CI of 20. 022 to 0. 063, respectively), not supporting H9a. However, the result of the model I revealed that SN has direct and indirect effects on AU. Table 7 shows that the indirect effect estimates are smaller than the direct effects estimates.


17

Table 8. Mediation of the Effect of EI and AT on independent variables through AU. Independent variable (IV) PEoU SN PU

95% CI

Total effects (SE)

Direct effect (SE)

Indirect effect (SE)

Lower

Upper

R2

0.064 0.318∗ 0.146

0.052 0.219∗ 0.082

0.012 0.099 0.064∗

20.028 20.025 0.013

0.060 0.233 0.147

AT ¼ 0.40 EI ¼ 0.537 AU ¼ 0.254


Note: Bootstrapping CIs; 5000 bootstrap samples; perceived usefulness (PU); perceived ease of use (PEoU); perceived social norm (PSN). ∗ p , .05. ∗∗ p , .01.

Hence, H10a was not supported in each linkage. Finally, if the direct effect estimates are not significant, the indirect effect estimates are significant and bigger than the direct effects estimates, so mediation is present. The result shows that while the direct effects of PU on AU were insignificant (b ¼ 0.072; p ¼ .301), the indirect effects of PU on AU through AT were significant (b ¼ 0.076, p , .01; 95% bootstrap CI of 0.029 – 0.150, respectively). Thus, this model asserts that AT has a mediating effect on PU on AU and fully supports H8a. In connection with Model II, the mediational impact of AT and EI on an association between exogenous variables and AU was investigated. Results from the bootstrapping analysis, as illustrated in Table 8, showed that these mediating factors significantly mediated the relationship between PU and AU (b ¼ 0.064, p , .05; 95% bootstrap CI of 0.013 – 0.147, respectively; therefore, H12b was supported. In contrast, this result did not prove the indirect effect between PEoU and SN and AU so hypotheses H13b and H14b were not supported.

Discussion Previous studies in the literature greatly embraced the original TAM model to describe the PU and PEoU toward BI in using technology. While the model has confirmed its capability to anticipate technology acceptance among users, it might not apply in the ISs study area for a nation with an emerging economy. Young rural people as entrepreneurs have a different job context, autonomy, ethics and obligation compared to ordinary users. They may have a different technology acceptance choice. Thus, the model might not completely describe the factors for young entrepreneurs’ technology acceptance through entrepreneurship latent factors. Our objective in this analysis was to contribute to the sustained improvement of models of technology acceptance. The current study uses the TAM, and its modification, TAM2 (i.e. SN), so as to capture the dissimilar view of IT adoption behavior by youth entrepreneurs in rural Malaysian communities. Therefore, this study advances the TAM model by using AT and EI as mediators, either distinctly or together, in two different models, based on the initial research questions that are able to clarify indirect and direct effects of SN, PU, and PEoU on AU. The findings of the first model demonstrated that AT was directly influenced by PU, and this is consistent with previous TAM research (Teo, Lee, & Chai, 2008). This means that when the use of ICT is perceived to be an enhancement to one’s productivity and its usefulness (Davis et al., 1989), these entrepreneurs are likely to develop a positive attitude toward its use. In particular, our findings demonstrated that SN is the most predictive of AU. Even though SN is included in the TAM2, recent studies have suggested positive relationships between them (Pan, Sivo, & Brophy, 2003). This finding suggests that SN may be important, primarily among these entrepreneurs. The results of the second model also showed the role of AT as a


18


mediator in explaining ICT acceptance behavior. Most notably, AT mediates the effects of PU on AU. That is to say, the effect of PU is no longer significant in directly explaining ICT use, which is contrary to the findings of many technology adoption studies (e.g. Venkatesh & Davis, 2000). In the second model, the results support previous research that showed that PU and SN were the most influential predictors to EI (Davis & Venkatesh, 2004). The remarkable result of the second model is the significant direct effect of SN on AU. The finding was consistent with the first model. In addition, our findings demonstrated the direct effects of PU and SN on AT. It means that PU and SN influence rural youth entrepreneurs toward using ICT, because these entrepreneurs form positive attitudes to ICT, which they believe will positively influence their activities in a rural setting. When SN is considered, people can also assist them to inquire about solutions, locate resources, and assemble a team of willing people to address their problems. Therefore, AT is a protective factor that is influenced by PU and SN (Venkatesh & Bala, 2008). In particular, the empirical findings showed that there is a direct relationship between AU and AT, and also AT and EI, mediated between PU and AU. It means that when the level of effort required for the behavior is high, the mediating role of AT and EI increases, thus, no direct relationship could be found between PU and AU. However, Adams, Nelson, and Todd (1992) replicated a simplified version of the TAM without both the AT and BI constructs; the second model allows us to take into account the mediating effects of AT and EI by increasing the intellect of the indirect effect from PU to AU via AT and EI. In comparing all the two models to each other from AIC, BIC and ECVI, it shows that the first model is better than the second model. In this model, we call attention to the role of attitude as a mediator in explaining ICT acceptance behavior among rural youth entrepreneurs in Malaysia. According to Othman and Ishak (2009), attitude is an important determinant of an individual’s success in entrepreneurship. Thus, a true and positive attitude is needed to assist youth in choosing and participating in entrepreneurship. Despite the importance of attitude in predicting an individual’s behavior, a number of TAM studies on IT adoption have discounted the role of attitude in explaining technology acceptance behavior. Additionally, the TAM was a more parsimonious model, which excluded attitude, as it did not fully mediate the effect of PU on intention to use ISs and actual usage behavior. Nevertheless, our findings indicated that attitude toward entrepreneurship mediates the indirect effect of PU on AU. Thus, the first model has included attitude as a construct within the TAM, because a positive attitude is commonly a major part of technology acceptance, which will strengthen users’ belief that ICT will help their entrepreneurial activities. This finding alerts researchers to be cautious in removing attitude from their models that examine technology acceptance of individuals. Besides, the description of all two models demonstrates that SN had a positive effect on endogenous variables. This could suggest that those relatives, acquaintances, and other social surroundings could have a certain amount of influence on the actual use of ICT by rural youth entrepreneurs. Taylor and Todd (1995) discovered SN as a better predictor of system behavior with inexpert subjects. According to Venkatesh and Davis (2000), SN significantly affected intention under required situations, and that it declined over time. On the other hand, this issue is still in the initial stages of investigation, requiring more research and it will no longer be one research to find the causal linkage between social influences, and IT adoption and the incorporation of new socially important factors. Furthermore, the indirect effect from the PU to actual ICT use, as hypothesized by the first and second models, was found to be significant. The sturdiness of PU as a cause of usage is supported by the entrepreneurs’ theory of production (Harper, 1996). The theory proposes that entrepreneurs will recognize and use technologies that will result in improved procedures, yields, and facilities. It is observed among entrepreneurs that IT usage will be sustained, or



19

even increased, as long as the system brings benefits to the organization, irrespective of the system’s complexity or usability. This is one of the reasons why technology usage decisions have been typically characterized by a strong productivity adjustment. Finally, all two models demonstrate that PEoU has not had an indirect and direct relationship with the AU. A reasonable explanation for these findings shows that rural youth entrepreneurs in Malaysia are influenced by outcomes (usefulness) than by process ease (usability) in their ICT usage choices. In fact, ease of use is vital only where easy-to-use systems are considered beneficial systems. The drive to succeed, restricted choice of applications and vendors characterizing many businesses in developing economies add to the hard work or perseverance of entrepreneurs, especially in developing countries, where nothing comes as easy explanations. Thus, PEoU may not be important in technology acceptance among rural youth entrepreneurs in Malaysia, and, by extension to other developing nations, especially when the system is considered useful. For them, finding a reasonable, usefulness system may suffice, hence, the insignificance of PEoU.

Research limitations and future research Although the findings of this survey can be perceived as statistically significant in the majority of its portions, there are some limitations of the present study. First, the survey approach is not biasfree from respondents’ subjective responses to survey statements. The questionnaire was a portrait of a dynamic social context. Second, the main variables like AU have been too infrequently evaluated, preventing tests of numerous significant relationships, like the one between intention and AU. When the target IT is not yet implemented at the time of the survey, assessing variables such as AU may need longitudinal investigation, which is something that is prominently lacking in the previous studies. Third, the task of assessing the inclusion or exclusion of being a rural youth entrepreneur was the third limitation of this study. Future studies must try to contain all who qualify, whether or not they are members of a specific association. Fourth, this study was based on Malaysian rural micro businesses. It would be worthy to investigate findings if the sample included other urban micro businesses in the country as well. Therefore, the next replication of this study will incorporate other areas as well.

Strength of the current research The strength of this study is highlighted in the following points. First, the study intentionally concentrated on only young rural entrepreneurs, due to little research in this part, to study whether there are any distinctions in their ICT usage and derives. Next, the model is based on theories grounded in current management ISs research. Furthermore, actual ICT usage (AU) was employed rather than intentions (as the dependent variable), which has been questioned by Straub, Limayemm, and Karahanna (1995). According to Ajzen and Fishbein (1980), intentions should be gauged eventually close to the behavioral observation to ensure an accurate prediction. BI may lack practical value in predicting long-term future implication in AU, as numerous TAM studies in the past more frequently measured EI or BI rather than AU. Thus, our interest here is to anticipate AU rather than EI as the endogenous variable. In addition, since the TAM has mostly focused greatly on PU and PEoU, and ignored the indispensable social influences involved in IT evolution, this type of criticism fails TAM from its predictability of the social influence variables. Considerably SN has been validated by some previous surveys in the context of IT and enterprise; we implement SN in the two models through TAM2. Finally, while numerous empirical studies have consistently acknowledged direct effects of the TAM

20



variables, including PEoU and PU on AT/BI (Davis et al., 1989), potential mediating effects of these two constructs (EI and AT) properties appear to be ignored in many TAM studies. Implications Malaysia is on the correct track in promoting rural youth entrepreneurship as one of the transformation agendas. To expedite the process, more effort is needed. In this regard, several theoretical and practical implications for ISs managers can be drawn from the results. From a theoretical viewpoint, the models may contribute to the existing body of related literature, and have signified a new view to our theoretical perception of ICT acceptance in IS areas and the emerging local market in many ways. First, this study assessed models with and without AT and EI constructs. Our findings indicated that the models with AT toward AU had a significantly better model fit. Second, the findings provide preliminary evidence which suggests that ICT acceptance among Malaysian youth entrepreneurs in rural communities is determined by the TAM, even though the findings did not present support for the existing theoretical links of the TAM, nor for some links newly hypothesized in this study. This also has practical policy implications for governmental and non-governmental agencies supporting IT implementations in rural enterprises. For the pathway in Model I of PU-to-AT-toAU, a practical implication for the Malaysian government should emphasize usefulness of ICT to rural youths through providing financial support and other resources for ICT training programs at all levels in schools and universities, encouraging the engagement of rural youths in entrepreneurial activities and in implementing ICT in their enterprise activities. Another important practical implication for the pathway in Model II of PU-to-AT and EI -to-AU is that Malaysia needs to integrate its rural youth entrepreneurial attitude and intention into all its entrepreneurial planning activities, especially when it targets development and use of ICTs in SME development and planning. We believe that a joint effort from educational institutions, entrepreneurial development government offices and the private sector will promote more entrepreneurial attitude and intentions among Malaysia’s rural youth entrepreneurs and encourage them to use ICT in their enterprise activities. This will in turn inspire other entrepreneurs to look up to these adopters and follow them, thus increasing adoption and usage of ICT in rural communities. The findings of the study also highlight the powerful roles of social norms in the involvement of rural youths in entrepreneurial activities via ICTs. The activities that further enable such experiences may include establishing direct contact between rural youth and entrepreneurs who may act as good role models. Role models are known to positively influence their attitudes and entrepreneurial intention to use ICT. More importantly, our results from the second model suggest that rural youths’ attitudes indeed influence ICT acceptance indirectly. Also, IS managers can indirectly improve their acceptance of ICT by having an impact on their attitude. While the importance of their beliefs underlines the value of IT system design and training, the significance of their attitudes underlines the importance of communication with other users about such systems. Conclusion In conclusion, the use of the widely accepted TAM can also be employed in the study of the factors that contribute to ICT adoption by rural youth entrepreneurs in Malaysia. The results of hypotheses tested suggest that PU, SN, and ICT usage have strong direct effects on AT. Moreover, statistically significant evidence was found to indicate that SN has a direct effect on ICT usage in Model I. The first model also asserts that AT has a mediating effect on PU on ICT usage.


21


In addition to direct effects, the result of Model II supported the relationship between PU and SN, and entrepreneurship intention; the relationship between SN and ICT usage; and the relationship between PU, SN and ICT usage, and attitude. Furthermore, the second model emphasizes that the effects of PU on ICT usage are fully mediated by these two mediating factors (attitude and entrepreneurship intention). The findings of this study not only help advance the understanding of ICT adoption by young entrepreneurs in a rural community, but also provide a theoretical foundation to investigate the effects of entrepreneurial factors as mediators on the ultimate outcome variable of AU, and also uncover new knowledge perspectives. Overall, the findings of this study provide useful tools for policy-makers and government managers in developing countries which can assist them to better understand young entrepreneurs’ ICT adoption in rural communities, and to further emphasize the vital role of this context in the TAM.

Disclosure statement No potential conflict of interest was reported by the authors.

Note 1. Source: Department of Statistics Malaysia (2007), Census of Establishments and Enterprises 2005.

Notes on contributors Zeinab Zaremohzzabieh is the corresponding author of the paper and currently doing PhD program in the field of youth studies at the Institute for Social Science Studies, Universiti Putra Malaysia. She obtained her Master of Science in sociology from the Science and Research Branch of the Islamic Azad University. To date, she has completed a number of research studies related to the community development and ICT. Bahaman Abu Samah started his career as a lecturer at Universiti Putra Malaysia. He served as director at the Institute for Social Science Studies. His teaching focuses on statistics and computer application in Human Resource Development. As a researcher, he is actively involved in different research projects. Mahazan Muhammad is currently working as a Social researcher at the Institute for Social Science Studies, Universiti Putra Malaysia. He is currently doing his Master in the rural development field of study. To date, he has completed a number of research studies related to Entrepreneurship, corporate Social Responsibility, and Social Entrepreneurship. Siti Zobidah Omar is an Associate Professor and currently working as the Head of Laboratory of Cyber Generation, Institute for Social Science Studies, Universiti Putra Malaysia and as a senior lecturer at the Faculty of Modern Language and Communication, Universiti Putra Malaysia. Siti Zobidah Omar gained her PhD in Communication Technology and Culture from The University of Birmingham and most of her research focuses on technology usage among the community. Jusang Bolong is an associate professor, a senior lecturer and currently acts as the Deputy Dean of Research and Innovation at the Faculty of Modern Language and Communication, Universiti Putra Malaysia. He gained his Ph.D. in Human Communication from Universiti Putra Malaysia and most of his research focuses on technology usage among the community. Md Salleh Bin Hj Hassan is a professor and a senior lecturer at the Faculty of Modern Language and Communication, Universiti Putra Malaysia. He gained his Ph.D. in Mass Communication, Developmental from Ohio University and most of his research focuses on mass communication for development. Hayrol Azril Mohamed Shaffril is currently working as a Social Research Officer at the Institute for Social Science Studies, Universiti Putra Malaysia. He obtained his Master of Science in Rural Advancement from Universiti Putra Malaysia and is currently doing his PhD in the same field of study. To date, he has completed a number of research studies related to the community development.

22



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. Ahmad, A. R. W., Fauziah, W. Y., Noor, H. M., & Ramin, A. K. (2011). Preliminary study of rural entrepreneurship development program in Malaysia. Journal of Global Entrepreneurship, 2(1), 537–546. Ahmad, S. Z., Abu Bakar, A. R., Faziharudean, T. M., & Mohamad Zaki, K. A. (2014). An empirical study of factors affecting e-commerce adoption among small-and medium-sized enterprises in a developing country: Evidence from Malaysia. Information Technology for Development, 21(4), 1–18. Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice-Hall. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103(3), 411–423. Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40(4), 471–499. Barba-Sańchez, V., Martıńez-Ruiz, M. D. P., & Jimeńez-Zarco, A. I. (2007). Drivers, benefits and challenges of ICT adoption by small and medium sized enterprises (SMEs): A literature review. Problems and Perspectives in Management, 5(1), 104–115. Batchelor, M. S., Scott, N., Woolnough, D., & Tambo, I. (2005). Good Practice. OECD-DAC Members. Baumgartner, H., & Homburg, C. (1996). Applications of structural equation modeling in marketing and consumer research: A review. International Journal of Research in Marketing, 13(2), 139–161. Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88(3), 588 –606. Byrne, B. M. (2010). Structural equation modeling with AMOS; basic concepts, applications, and programming (2nd ed.). New York: Taylor & Francis. Carr, J. C., & Sequeira, J. M. (2007). Prior family business exposure as intergenerational influence and entrepreneurial intent: A theory of planned behavior approach. Journal of Business Research, 60(10), 1090–1098. Chan, S.-C., & Lu, M. T. (2003). Understanding internet banking adoption and use behavior: A Hong Kong perspective. Journal of Global Information Management, 12(3), 21–43. Chau, P. Y. K. (1997). Reexamining a model for evaluating information center success using a structural equation modeling approach. Decision Sciences, 28(2), 309–334. Chen, K., Chen, J. V., & Yen, D. C. (2011). Dimensions of self-efficacy in the study of smart phone acceptance. Computer Standards & Interfaces, 33(4), 422–431. Chou, C.-P., & Bentler, P. M. (1995). Estimates and tests in structural equation modeling. In R. H. Hoyle (Ed.), Structural equation modeling: Concepts, issues, and applications (pp. 37 –55). Thousand Oaks, CA: Sage. Curtain, R. (2003). Creating more opportunities for young people using information and communications technology. The World Summit on Information Societies, Geneva. Geneva, Switzerland. Davidsson, P. (1995). Culture, structure and regional levels of entrepreneurship. Entrepreneurship & Regional Development, 7(1), 41–62. Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319 –340. Davis, F. D., & Venkatesh, V. (2004). Toward preprototype user acceptance testing of new information systems: Implications for software project management. Engineering management. IEEE Transactions on Engineering Management, 51(1), 31– 46. 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. Department of Statistics Malaysia. (2007). Census of Establishments and Enterprises 2005. Retrieved from http://www.statistics.gov.my/portal. Duncombe, R., & Heeks, R. (2002). Enterprise across the digital divide: Information systems and rural microenterprise in Botswana. Journal of International Development, 14(1), 61–74. Eggleston, K., Jensen, R., & Zeckhauser, R. (2002). Information and communication technologies, markets, and economic development (Discussion Paper No. 0203). Department of Economics, Tufts University. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. Gallego, M. D., Luna, P., & Bueno, S. (2008). User acceptance model of open source software. Computers in Human Behavior, 24(5), 2199–2216.



23

Grimes, S. (2000). Rural areas in the information society: Diminishing distance or increasing learning capacity? Journal of Rural Studies, 16(1), 13–21. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2010). Multivariate data analysis (7th ed.). Englewood Cliffs, NJ: Prentice Hall. Haji Idris, N. A. (2008). Women and economic development. Kuala Lumpur: Dewan Bahasa dan Pustak. Harper, D. A. (1996). Entrepreneurship and the market process. London: Routledge. Hassan, M. S., Shaffril, H. A. M., Abu Hassan, M., & D’Silva, J. L. (2009). Developing agriculture in Malaysia: Internet utilization among Malaysian youth agro-businessman. European Journal of Social Sciences, 11(2), 215 –224. Hassan, M. S., Shaffril, H. A. M., & Bahaman, A. S. (2012). Factors influencing the perception of youth agrobased entrepreneurs towards the role of ICT in increasing agro-business productivity. Journal of Agriculture and Social Sciences, 50(4), 539–553. Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76(4), 408– 420. Homer, P. M., & Kahle, L. R. (1988). A structural equation test of the value-attitude-behavior hierarchy. Journal of Personality and Social Psychology, 54(4), 638–646. Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. Kaplan, D. (2000). Structural equation modeling: Foundations and extensions. Thousand Oaks, CA: Sage. Khurshid, A., & Ghani, J. (2001). Lahore: A public-private partnership to enter the global IT market (Background paper). Geneva: ILO. Kleijnen, M., Wetzels, M., & de Ruyter, K. (2004). Consumer acceptance of wireless finance. Journal of Financial Services Marketing, 8(3), 206 –217. Kline, R. B. (2005). Principles and practice of structural equation modeling. New York, NY: The Guilford Press. Krueger, N. F., Reilly, M. D., & Carsrud, A. L. (2000). Competing models of entrepreneurial intentions. Journal of Business Venturing, 15(5–6), 411 –432. Lee, H.-H., Fiore, A. M., & Kim, J. (2006). The role of the technology acceptance model in explaining effects of image interactivity technology on consumer responses. International Journal of Retail & Distribution Management, 34(8), 621 –644. Li, D., & Qiu, L. (2008). Applying TAM in B2C e-commerce research: An extended model. Tsinghua Science and Technology, 13(3), 265 –272. Mallinckrodt, B., Abraham, W. T., Wei, M., & Russell, D. W. (2006). Advances in testing the statistical significance of mediation effects. Journal of Counseling Psychology, 53(3), 372– 378. Moon, J.-W., & Kim, Y.-G. (2001). Extending the TAM for a World-Wide-Web context. Information & Management, 38(4), 217 –230. Nah, F. F. H., Tan, X., & Teh, S.-H. (2004). An empirical investigation on end-users’ acceptance of enterprise systems. Information Resources Management Journal, 17(3), 32 –53. Ndubisi, N. O., & Jantan, M. (2003). Evaluating IS usage in Malaysian small and medium-sized firms using the technology acceptance model. Logistics Information Management, 16(6), 440–450. Norsida, M. (2007). The agriculture community. In M. A. Fatimah, R. A. Nik Mustapha, K. Bisant, & M. A. Amin (Eds.), 50 years of Malaysian agriculture: Transformational issues, challenges and direction (pp. 128 –143). Serdang: UPM. Othman, M. N., Ghazali, E., & Cheng, O. C. (2005). Demographics and personal characteristics of urban Malaysian entrepreneurs: An ethnic comparison. International Journal of Entrepreneurship and Innovation Management, 5(5–6), 421 –440. Othman, N. H., & Ishak, S. (2009). Attitude towards choosing a career in entrepreneurship amongst graduates. European Journal of Social Sciences, 10(3), 419–434. Pan, C. C., Sivo, S. A., & Brophy, J. (2003). Students’ attitude in a web-enhanced hybrid course: A structural equation modeling inquiry. Journal of Educational Media and Library Sciences, 41(2), 181– 194. Pick, J. B., Gollakota, K., & Singh, M. (2014). Technology for development: Understanding influences on use of rural telecenters in India. Information Technology for Development, 20(4), 1 –28. Qureshi, S. (2005). How does information technology effect development? Integrating theory and practice into a process model. In Proceedings of the 11th Americas Conference on Information Systems (pp. 500 –509), Omaha. Retrieved from http://aisel.aisnet.org/amcis2005/261/.


24


Rahman, M. A., Mahfuz, M. U., Ahmed, K. M., & Rajatheva, R. (2005). ICT based sustainable rural business opportunities in developing countries: A wireless-networked RCP-RAP approach. American Journal of Applied Sciences, 2(8), 1256–1260. Ramayah, T., & Lo, M. C. (2007). Impact of shared beliefs on “perceived usefulness” and “ease of use” in the implementation of an enterprise resource planning system. Management Research News, 30(6), 420– 431. Ramdani, B., Kawalek, P., & Lorenzo, O. (2009). Predicting SMEs’ adoption of enterprise systems. Journal of Enterprise Information Management, 22(1–2), 10–24. Ramsey, E., Ibbotson, P., Bell, J., & Gray, B. (2003). E-opportunities of service sector SMEs: An Irish crossborder study. Journal of Small Business and Enterprise Development, 10(3), 250–264. Rogers, E. M. (2003). Diffusion of innovations (5th ed.). New York, NY: Free Press. Samah, B. A., Shaffril, H. A. M., Hassan, M. S., Hassan, M. A., Ismail, N., & Others. (2009). Contribution of information and communication technology in increasing agro-based entrepreneurs productivity in Malaysia. Journal of Agriculture & Social Sciences, 5(3), 93– 97. Schepers, J., & Wetzels, M. (2007). A meta-analysis of the technology acceptance model: Investigating subjective norm and moderation effects. Information & Management, 44(1), 90–103. Shaffril, H. A. M., Hassan, M. S., & Samah, B. A. (2009). Level of agro-based website surfing among Malaysian agricultural entrepreneurs: A case of Malaysia. Journal of Agriculture & Social Sciences, 5(3), 55 –60. Sidhu, K., & Kaur, S. (2006). Development of entrepreneurship among rural women. Journal of Social Sciences, 13(2), 147 –149. Straub, D. W., Limayemm, M., & Karahanna, E. (1995). Measuring system usage: Implications for IS theory testing. Management Science, 41(8), 1328– 1342. Tapscott, D. (1998). Growing up digital: The rise of the net generation. New York, NY: McGraw-Hill. Taylor, S., & Todd, P. (1995). Assessing IT usage: The role of prior experience. MIS Quarterly, 19(4), 561– 570. Teo, T., Lee, C. B., & Chai, C. S. (2008). Understanding pre-service teachers’ computer attitudes: Applying and extending the technology acceptance model. Journal of Computer Assisted Learning, 24(2), 128– 143. Tuyon, J., Mohammad, S., Junaidi, S., & Ali, R. (2011). The role of microfinance in development of micro enterprises in Malaysia. Business Management Quarterly Review, 2(3), 47–57. Valentine, G., & Holloway, S. L. (2002). Cyberkids? Exploring children’s identities and social networks in online and off-line worlds. Annals of the Association of American Geographers, 92(2), 302–319. Veciana, J. M., Aponte, M., & Urbano, D. (2005). University students’ attitudes towards entrepreneurship: A two countries comparison. The International Entrepreneurship and Management Journal, 1(2), 165– 182. Venkatesh, V., & Bala, H. (2008). Technology acceptance model 3 and a research agenda on interventions. Decision Sciences, 39(2), 273 –315. Venkatesh, V., & Davis, F. (2000). A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management Science, 46(2), 186–204. Yusoff, M. N. H., Yaacob, M. R., & Ibrahim, M. D. (2010). Business advisory: A study on selected microsized SMEs in Kelantan, Malaysia. International Journal of Marketing Studies, 2(2), 245–257. Zaremohzzabieh, Z., Samah, B. A., Muhammad, M., Omar, S. Z., Bolong, J., Hassan, M. S., & Shaffril, H. A. M. (2015). A test of the technology acceptance model for understanding the ICT adoption behavior of rural young entrepreneurs. International Journal of Business and Management, 10(2), 158– 169. Zaremohzzabieh, Z., Samah, B. A., Muhammad, M., Omar, S. Z., Bolong, J., & Shaffril, H. A. M. (in press). An investigation into factors influencing rural youth entrepreneurs intentions to use ICT: A case of Malaysia. International Journal of Entrepreneurship and Small Business.