Factors affecting ICT adoption among rural users: A case study of ICT Center in Iran

Telecommunications Policy 37 (2013) 1083–1094

Contents lists available at ScienceDirect

Telecommunications Policy URL: www.elsevier.com/locate/telpol

Factors affecting ICT adoption among rural users: A case study of ICT Center in Iran B. Khalil Moghaddam a, A. Khatoon-Abadi b,n a b

Cultural Heritage, Handicrafts and Tourism Department of Isfahan Province, Isfahan, Iran Department of Rural Development, College of Agriculture, Isfahan University of Technology, Isfahan, Iran

a r t i c l e i n f o

abstract

Available online 10 October 2013

Rural ICT centers are the initiative of the third millennium and widen the accessibility horizons of information and Communication Technology among disadvantaged groups of societies, and play a significant role in rural development processes. Adoption of new technology in rural Iran has been the main challenge and focal point of all agricultural extension activities since the modernization era of the 1950s. Consequently the rapidly growing gap between urban and rural economy has reinforced the critical role of ICT in creating an equal society. Identifying the factors which foster adoption of ICT is among the important challenges of alleviating digital divide. ICT centers attract different groups within rural communities and create a forum for unprivileged rural settlers to learn about and to use computer and internet. This paper attempts to identify the factors influencing the adoption of ICT in rural Gharn Abad's ICT center of Golestan Province. The sample included 218 individuals, who were selected by stratified random sampling method. Survey method was used, and data was analyzed by correlation as well as multiple regression techniques. Based on the results, the existence of ICT center itself, with various funding sources, reinforced the adoption regardless of the users' economic status. At the same time, the other factors such as individual, social, the households' informative & communicative, as well as the innovation related factors were found influential. This case study could be used as a sample for planning, establishing, and developing the ICT centers in the other similar situations. & 2013 Elsevier Ltd. All rights reserved.

Keywords: ICT Adoption ICT Center Rural development Iran

1. Introduction Information and communication Technology in the first decade of the third millennium proved its vital role in poverty alleviation and sustainable development processes. ICT as a third millennium's phenomenon cannot meet the third millennium development goals unless being adopted by different groups within societies. Research projects which explore issues relating to ICT adoption may contribute to decreasing the ongoing digital gap in a society. Based on review of the literature, it has strengthened livelihoods, has especially ensured food security and has increased income opportunities for local communities. ICT has been regarded as a livelihood intervention and community development tool. For example, Australia, India and Malaysia have integrated ICT within their rural development processes effectively. Within Iran, the access of urban young people to internet is increasing dramatically and many government organizations are

n

Corresponding author. Tel.: +98 311 391 3443; fax: +98 311 391 2254. E-mail addresses: [email protected] (B. Khalil Moghaddam), [email protected] (A. Khatoon-Abadi).

0308-5961/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.telpol.2013.02.005

1084

B. Khalil Moghaddam, A. Khatoon-Abadi / Telecommunications Policy 37 (2013) 1083–1094

offering online access to their links. Nevertheless young people in rural areas of Iran suffer from the widening digital gap (UNCTAD, 2006). Iran's ranking of the ICT Development Index (IDI) (including: access, use and skills) in 2010 was 87 out of 152 countries whereas Korea's ranking was 1 and Chad's ranking was 152. Comparing with some other countries such as United Arab Emirates, Qatar, Bahrain, Saudi Arabia, Turkey and Oman with the respectively rankings of 32, 44, 45, 46, 59 and 60, Iran suffers a good position in ICT. Furthermore, the percentage of households with internet access in Iran has been reported 20.8% at 2010, which in comparison with other countries such as Korea Republic (96.8%), Qatar (84%), Bahrain (74%), United Arab Emirates (65%), Saudi Arabia (54.4%), Turkey (41.6%), and Oman (27.7%) (ITU, 2011) is in a lower status. The rural population of Iran involves 31.4% of the whole population (Statistical Center of Iran, 2007) and is conventionally characterized by illiteracy, poverty and powerlessness. The percentage of the rural settings with a population of over 100 people, which are connected to the information networks, is 16.6% (Ministry of Information and Communication Technology, 2008). The remoteness of official organizations from rural settings and insufficient income in companion with a lack of communication technology has resulted in huge rural-to-urban migration. The research context: the department of ICT in Iran has targeted the year 2015 for full internet coverage throughout the country. However, with due attention to rural poverty, the households cannot afford owning personal computers, that illuminates the crucial role of ICT centers in bringing about justice, equity, and fostering sustainability. Consequently, two challenges exist: (1) the scattering of villages throughout Iran with the average of 23 households per village, and (2) the adoption of ICT by the rural settlers themselves. Towards achieving sustainable human development goals, establishment of ICT centers in rural areas of developing countries has been encouraged by United Nations Development Program. In the year 2002 for the first time, as a pilot project, Gharn Abad village alongside of Caspian Sea in Northern Iran, was selected. In this area farming is the prevalent occupation, whereas some are engaged in off-farm and seasonal jobs. This center is a two-floor building with 560 m2, located in Golestan Province which includes the following operational units: Conference and theatre room, ISP unit, instructional classes, café-net, and Tele working unit. The activities engage internet and email search, climate news, agricultural information, e-business, and English training courses. This center receives different financial aids from various sources such as: local peoples’ donations, the government, private sector, and the Iranians who live overseas. The users (who are between 9 and 31 years old) consist of the primary/secondary school and university students, farmers, housewives and Tele workers who participate in the Center's activities. The participant members belong to six villages being located between 2 and 10 km from the center. Gharn Abad is the first e-village in Iran (Jalali, 2007). In 2003 three more centers were established in the other rural settings through a joint project operated by both UNDP and Management and Planning Organization of Iran (MPO). The villages included Maranak village in Damavand Township of Tehran Province, village of Tiss in Chah-Bahar Township of Sistan-Baloochestan as well as Mahabad of Ardestan Township in Isfahan provinces (MPO, 2005). However, based on the field observations, the only successful ICT center which maintained its momentum was Gharn Abad. This center was the winner of ‘E. Asia 2007 Award’ for the best project, with regard to its creativity and technology orientations in Asia Pacific (Jalali, 2007). Iran's lower status in the economic and telecommunications indices (compared with the Southeast Asian countries) illuminates the need for researches on the adoption of ICT, with due respect to the following: high percentage of rural population and the alarming digital gap between urban and rural societies. Accordingly, new technological innovations often fail because too much attention is still given to technical-related features without taking into account the most important parameters which directly relates to the users’ adoption phenomenon (Verdegem & Marez, 2011). The main objective of this research is to study the parameters which affect the ICT adoption in Gharn Abad ICT Center to finally generalize the outputs for establishing several ICT centers throughout the country. 2. Literature review The review of the literature illustrates almost a similar social context for the adoption of ICT in general, in most of which the common issue is economical factor. The contribution of this study to the literature is its specific focus on a rural forum in which information and communication technology is operational with its specific characteristics. Krysa (1998) has determined the constraints of using computers by teachers in schools: time, hardware and software access difficulties, attitudes of administrators towards computer, teachers’ attitudes, educational problems, teacher training and their personal computer skills (1998). Based on Krysa's study, personal computer skill as an individual factor was included in the conceptual model. Luchetti and Sterlacchini (2004) through an econometric analysis on a sample of Italian Small and Medium Enterprises (SMEs) showed that the adoption as well as the effective use of ICTs depended, firstly on the types of ICT, and secondly on different firm characteristics (2004). The relevance of the mentioned study is its emphasis on characteristics of ICT (that is, the innovation) and firm characteristics (that is, the adopter's). Consequently, characteristics of individuals (as the adopter's characteristics) and of the ICT (as Innovation's characteristics) are considered in the conceptual model. Cheong (2002), studied the characteristics of the users and nonusers of internet, and demonstrated the users as being younger, more educated, and with more annual income. Besides, gender, income, and the years of experiences were stimulators of the internet users. Cheong, in the second part of the paper has applied regression and correlation analyses. The independent variables were: (1) the use of media; (2) family functions such as having meal or watching TV with the family; (3) assessment of media credibility; (4) perceived value of the Internet; and (5) demographic factors, annual households’ income, occupation and educational levels. The dependent variable was ‘the internet use based on the weekly

B. Khalil Moghaddam, A. Khatoon-Abadi / Telecommunications Policy 37 (2013) 1083–1094

1085

hours’. The results showed that there was a low-level relationship between using the other media, and the internet application. Also there was a significant relationship between the ‘positive perceptions of user towards internet’ and the ‘internet application’. Moreover, the issues like, gender, annual household's income and the educational level proved to be statistically significant in relation to the ‘internet application’. There was not any relationship between ‘age’ and ‘internet usage’ (2002). According to Cheong’s study, the independent variables such as demographic factors, i.e.: ‘the amount of using mass media’ and the ‘users’ perception towards importance of ICT’, had affected the amount of ICT application as indices for determining adoption of ICT. This also has been reflected in our final conceptual model. Thrane’s (2003) study of ICT adoption in Norwegian teenage homes, revealed two major hypotheses: regarding the prerequisite for adopting ICT: There are correlations between adoption and: (1) the ICT characteristics (2) existence of willingness among potential users, and also access to facilities (Thrane, 2003). In Thrane's work, the ‘relative advantage’ of the technology was considered as important factor for the ICT user, which had been already mentioned as ‘Innovation's Characteristics’ by Rogers and Shoemaker through the study of innovation. In the present study this factor was included as one of technological/innovation related factors within the conceptual model. Taylor, Zhu, Dekkers and Marshall (2003a) identified the excluded people from internet and showed that people who were the younger aged, more educated, married with child, owning of one house, with higher income, and employed, had more access to internet in comparison with people in the opposite side. Additionally in their work, the results of logistic regression analysis showed that a few factors such as educational level, marital status, having child, and income level were statistically significant in ‘use of internet at home’ (2003a). Also, Taylor, Zhu, Dekkers and Marshall (2003b), in another paper determined the impact of socio-economic factors which affected home internet patterns in Queensland. According to them, there were 8 patterns for applying internet at home, as well as 9 users’ characteristics. The internet usage patterns were introduced as for job, educational level, entertainment, email, home finance, on-line purchasing, and the community networking. They also identified correlation between certain economic and demographic factors (as independent variables) and the internet, in home usage patterns (2003b). Ssewanyana and Busler (2007) by studying the adoption and usage of ICT in Ugandan firms have shown that the adoption of ICT in developing countries by firms follows the same pattern as in developed countries, and the only difference is in the level of adoption. The usage of computers and internet is high in medium and large firms. The small firms which are mainly locally owned have low usage due to the high cost of required investment, limited knowledge and skills (2007). Based on the mentioned work, knowledge and skills of ICT will be considered as an independent variable in the present paper. All of the ICT adoption studies can be categorized in three different levels including Macro (e.g. countries), Mezzo (e.g. organizations) and Micro (e.g. individuals/households) (Kovacic & Vukmirovic, 2008). In the present literature review, the studies of Thrane (2003), Taylor et al. (2003a, 2003b), Wahid (2007) and Cheong (2002) can be categorized to the microlevel, whereas Uzoka and his co researcher's study (2007) belongs to the macro-level. Krysa's research (1998), as well as the studies of Ssewanyana and Busler (2007) and Kaynak, Tatoglu, and Kula (2005) are three examples for the mezzo level. The study of adoption in Gharn Abad ICT center is a research type which is either similar to macro-level or mezzo-level. In the ICT center as a public domain (which is similar to SMEs) the individuals use and adopt the ICT similar to micro-level. Generally, adoption phenomenon in the ICT center follows the main above discussed principles. Based on Waarts and VanEverdingen (2005) the majority of influential factors are common in the three mentioned levels. However, according to the level in which ICT adoption is being studied, and also to the context of adoption phenomenon (i.e., public, private or personal), therefore orientation (positive or negative) and amount of their effectiveness would be different. 3. Theoretical framework Adoption of ICT by communities is considered as one of the four prerequisites for the sustainability of ICT centers, that is, financial viability, staff capability, community acceptance and service delivery (Harris, 2004). There have been significant researches on ICT adoption. However majority of them have focused on ICT adoption in firms, Small and Medium size Enterprises (SMEs), schools and homes without paying enough attention to factors affecting the adoption in the rural centers. The findings from some of these studies have been demonstrated in theories and models in the following. Also, a conceptual model is proposed based on the theoretical framework section. It begins with the ‘Theory of innovation/ adoption’ (Rogers, 1983), suggesting that the technological characteristic of innovation is an important factor influencing ICT adoption. Another theory involves ‘Technology Acceptance Model’ (TAM) based on Theory of Reasoned Action (TRA) (Davis, Bagozzi, & Warsaw, 1989). TRA presented by Fishbein and Ajzen (1975), explains and predicts behavioral intention in many general settings. TAM model illuminates how certain beliefs and attitudes convert into appropriate behavior, further to emphasize subjective norms. According to this model, there are two kinds of beliefs that is, the ‘perception of usefulness’ and the ‘perception of ease of use’ which through interaction with external stimuli, play a significant role in adoption of ICT (Davis et al., 1989). Although, the theory has not determined the external stimuli (variables), it highlights the important role of further studies on socio-economical, innovative and other external factors. As a result of Wahid's study (2007): (1) the internet adoption among women is affected by perceived ease of use rather than perceived usefulness, (2) the internet adoption among men is affected only by perceived usefulness rather than perceived ease of use, and (3) high cost access and lack of English proficiency are identified to be most severe obstacles of adopting internet in Indonesia. Theory of Planned Behavior (TPB) (Ajzen, 1991) identifies three influential factors on

1086

B. Khalil Moghaddam, A. Khatoon-Abadi / Telecommunications Policy 37 (2013) 1083–1094

individuals’ intention toward adoption of innovation; namely attitude, subjective norms and perceived behavioral control. Based on Uzoka, Shemi, and Seleka (2007) adopting e-commerce in Botswana is affected by perceived advantages and complexity. Although this confirms Theory of Planned Behavior (TPB), but in the above work, attitude seems to weigh more than the subjective norm and the perceived behavioral control. Similarly, in the present study attitude has been considered as one of the main elements. Furthermore, based on ‘TAM2’ model, known as the improved TAM (Venkatesh, 2000), the attitude component does not affect adoption; rather the perceived technology characteristics directly influenced the individual's intention to use the new technology. Additionally, social factors have been reconsidered in the model. Both TAM and TAM2 have been applied in different forms to explain technology adoption in a wide variety of contexts, ranging from consumer to intra organizational adoption processes (Raaij & Scheepers, 2006). The other theory is Rashid and colleague’s model (2001) which was presented for e-commerce technology adoption by New Zealand SMEs at 2001. In this model 4 factors (technological, organizational, environmental and individual) are identified as influential factors of ICT adoption. In the present study, the impact of governmental factor (i.e., a sub-environmental factor) will not be examined, since; the ICT center is governed by its own rules. In contrast, due to differences of ‘family characteristics’ (including household's socioeconomic, communicative and informative factors), these variables were considered in the model as one of the contextual factors. Also, with respect to Rashid and colleague’s work, the users’ ICT skills and knowledge were studied in the present research as the personal factors which affected adoption of ICT. Contrasting to Rogers’ adoption theory, Rashid et al. in their study have focused on the technical factors (e.g., relative advantage, complexity, compatibility, cost, and the image surrounding the innovation). However, since in Gharn Abad ICT center, the services were provided in an equal term and free of charge, hence, in relation to Rashid and colleague’s model, the variable of ‘ICT cost’, and the organizational factors were omitted. Therefore, it can be concluded that the organizational factors in this research (in comparison with SMEs in Rashid and colleague’s model) refer mainly to the ICT center as the context of supplying and adopting the innovation. Also, Unified Theory of Acceptance and Use of Technology (UTAUT) resulted from unification of models such as TRA, TAM, TPB, and Innovation/Diffusion Theory (Venkatesh and Davis, 2000). According to TAM and TPB, every intention for action, leads to concrete experience without limitations. However, this may not occur in the actual world. So UTAUT model was presented to improve TAM and TPB models. On the other hand, based on TAM2, the adapted model for ICT adoption in Iran presented by Movahed and Abesi (2003), the external variables which impacted directly on adoption were: perceived as: ease of use, perceived usefulness of ICT, social norms and feelings, and social outputs. In the present study, only the ‘perceived usefulness of ICT’ and ‘perceived ease of use’ have been considered as the effective factors which affect the adoption of ICT. Kaynak et al. (2005) by focusing on influential factors on adopting e-commerce within SMEs in Turkey showed sympathy towards validating the mentioned theory. The results showed that e-commerce adoption was significantly influenced by its ‘Perceived usefulness’ (Kaynak et al., 2005). Another theory belongs to Kovacic and his colleagues (2008) who focused on ICT adoption at individual level in Serbia. Their model was based on three factors: socio-demographical, economical and e-skills. Although in the theoretical framework, there is rarely direct studies on adoption in ICT centers, but the basis of all theories are applicable for the adoption process within ICT centers. Based on the adoption theories and models, as well as some of the case studies, the structure of adoption theoretical framework, is generally adaptive in different contexts. Hence in this paper, due to significance of ICT center, elements of the conceptual model (Fig. 1) have been resulted from a combination of the mentioned theories, models and case studies. It is notable that the degrees of the importance of these theories, models and case studies in forming the conceptual model are different. For example, Rashid and colleague’s model (2001) has been more focused. Some of the contextual factors (for example language, religion, culture and the national ICT policies) have not been included in the model due to their commonalities. The following adjusted model to be used in this paper (Fig. 1) has clustered the influencing factors into; three main categories including: (1) individual, (2) contextual (involving social related characteristics of the users’ households’, users’ households’ economic characteristics, as well as users’ households’ informative & communicative characteristics) and (3) technological (innovation related characteristics) aspects. Accordingly, after the data is analyzed, the conceptual model will be converted into a new adapted model of adoption in rural ICT center of Iran. The following research hypotheses are resulted from the conceptual model (details of the related variables of entire hypotheses are shown in Fig. 1). Rashid and colleague’s model (2001) as the main theory has played a significant role in making the conceptual model and hypotheses. Hypothesis 1. There is a relationship between the user's individual characteristics, and ICT adoption. Hypothesis 2. The individual user's characteristics have impacts on ICT adoption. The first and second hypotheses have been derived from TAM (Davis et al., 1989), TPB (Ajzen, 1991), Rashid and colleague’s model (2001), and Kovacic and colleague’s model (2008), as well as resulted from the studies such as Krysa (1998), Cheong (2002), Taylor et al. (2003a, 2003b), Movahed and Abesi, 2003, Kaynak et al. (2005), Ssewanyana and Busler (2007), Wahid (2007), and Uzoka et al. (2007). Hypothesis 3. There is a relationship between the users’ households’ economic characteristics, and the adoption of ICT. Hypothesis 4. The users’ households’ economic characteristics have impacts on ICT adoption.

B. Khalil Moghaddam, A. Khatoon-Abadi / Telecommunications Policy 37 (2013) 1083–1094

Users’ households’ social factor (as an Environmental factor):

● Number of household’s members ● Percent of household’s literates ● Number of household’s members with the knowledge of computer

● The household’s knowledge level of computer, before of establishing the center

●

Number of household’s members with formal occupation

● Household’s view of the ICT center’s

Users’ households’ informative & communicative factor (as an Environmental factor):

1087

Technological (innovation -related) factor:

● The amount of audiovisual use

● The amount of informative & communicative facilities use

● The amount of printed media use

● ● ● ●

Relative advantage Compatibility Complexity Image surrounding the ICT

● Observe ability

Individual factor : ● Age ● knowledge and skill

usefulness

in computer

● Household’s satisfaction towards the

● Satisfaction of the

user’s ICT use

center’s management

● Household’s attitude by the user’s ICT use for education

● Interest towards

● Household’s attitude by user’s ICT use

ICT adoption

for work

adopting innovations

● Gender

● The users’ households’ attitude towards

● Marital Status

ICT as entertainment

● Father’s job

Users’ households’ economic factor (as an Environmental factor):

● Mother’s job ● Spouse’s job ● Father’s educational level ● Mother’s educational level

● Ownership of farmland

● living location ● Educational level ● User’s vocation

● The amount of land under cultivation

● Older brother’s educational level

● Number of owning livestock

● Older sister’s educational level

● Gross annual income ● Ownership of more than one

● Spouse’s educational level

● Study area ● Life style

house

Fig. 1. Conceptual model for ICT adoption in rural ICT centers.

The third and fourth hypotheses are according to Rashid and colleague’s model (2001), Kovacic and colleague’s model (2008), and the studies of Cheong (2002) and Taylor et al. (2003b, 2003a). Hypothesis 5. There is a relationship between the users’ households’ social characteristics, and ICT adoption. Hypothesis 6. The users’ households’ social characteristics have impacts on ICT adoption. TAM2 (Venkatesh, 2000), Rashid and colleague’s model (2001), and Kovacic and colleague’s model (2008), as well as the studies of Cheong (2002), and Taylor et al. (2003b), are the basis of the fifth and the sixth hypotheses. Hypothesis 7. There is a relationship between the innovation-related characteristics of the users, and ICT adoption. Hypothesis 8. The innovation-related characteristics of the users have impacts on ICT adoption. Hypotheses 7 and 8 are resulted from Rogers’ (1983), Rashid and colleague’s model (2001), and Kovacic and colleague’s model (2008), and also the studies of Thrane's (2003), and Luchetti and Sterlacchini (2004). Hypothesis 9. There is a relationship between the informative & communicative characteristics of the users’ household, and ICT adoption. Hypothesis 10. The informative & communicative characteristics of the users’ household have impacts on ICT adoption. At the end, Hypotheses 9 and 10 are based on Rashid and colleague’s model (2001), and Kovacic and colleague’s model (2008), and also the works of Cheong (2002), and Thrane's (2003). 4. Material and methods The methodology included two different sections: the first part involved exploratory field study through qualitative technique of semi-structured interview among the Gharn Abad technical staffs, and the development communication experts. As a result, the usage patterns of ICT in the center (Appendix), were defined, set out and weighted based on the

1088

B. Khalil Moghaddam, A. Khatoon-Abadi / Telecommunications Policy 37 (2013) 1083–1094

amounts of the complexity and the difficulties (between 1 and 19 or 0.05 and 2 or from easiest to the most difficult). Also 5 levels assumed for measuring the amount of familiarity & skill with respect to the usage patterns (Appendix). The second part comprised of survey research accompanied with correlation and regression analyses. A questionnaire based on Likert spectrum (that is sequential multiple level responses) was designed and used for the data gathering. The similar studies which have been mentioned in the literature have used quantitative methods to determine and prioritize the factors affecting adoption of ICT. The interpretative statistics has been used to determine the influencing extent of the parameters. The qualitative variables were converted into the semi quantitative data (through statistical and weighting techniques in the Factor Analysis). The analyses: statistically it is justified to use the analytical methods and tests for semi quantitative variables as similar to quantitative variables. Also the different correlation analyses along with the multiple regression with stepwise method were used for each variable couples depending on the type of data (Kalantari, 2012). Studies such as Cheong's (2002), measured ICT adoption as the dependent variable with ‘the time spent for using internet per week’ (Cheong, 2002). Due to the fact that the same number of hours of the ICT usage patterns with different levels of convenience, show the different adoption levels. Therefore in this paper, another index (in addition to the quantity of the spent hours as an index) that is the users’ level of ICT knowledge and skill was also considered. The important issue in determining the amount of ICT adoption was the fact that the indices’ assessment units were different and for summing up, they needed to be scaled free. In this paper by using linear scale free method, the two mentioned indices, became scale free. Based on a community expert consensus, these two indices were estimated and each of the users’ usage patterns was weighted between 1 and 19 and or 0.05 and 2. By applying the scale free and weighting techniques of the quantities of the two indices, the sum of the scale free quantities represented the amount of ICT adoption (the maximum quantity of the final adoption index was 1.91, and the minimum was 0.014). So the following equation is presented for estimating the amount of ICT adoption for every individual user. 19

AICT i ¼ ∑ Aupji j¼1

Aupji ¼ WSFST ji þ WSFLKSji Amount of Adoption of ICT for every user: AICT i Amount of user’s Adoption regarding each Usage Pattern: Aupji Weighted Scale Free amount of users’ Spent Time regarding every usage pattern: WSFST ji Weighted Scale Free amount of users’ Level of ICT Knowledge and Skill regarding every usage pattern: WSFLKSji i: Number or users (1, 2, 3, …, 218) j: Number or Weight of usage patterns (1, 2, 3, …, 19) The statistical society included 342 Gharn Abad residents and the surrounding villages who use ICT center once a week at least (45 individuals as Tele workers, and the rest of 297 individuals as general users (café net users)). The probability stratified random sampling method (proportionate to size) was used to provide context for generalizing data. Because, the differences between these two strata (as sample unit), as well as the similarities between the users of each separate stratum (as observational unit) regarding the type and the amount of utilizing the patterns. The sample size determined by Cochran's formula (Saraie, 1993) is 218 (190 in the stratum of the café net users and 28 in the stratum of the Tele workers). In Table 1, KMO (Kaiser–Meyer–Olkin) figures for the validity of the used concepts have been demonstrated. KMO is a statistical criterion by which the external validity of the applied concepts (Variables) in the questionnaire is measured. The KMO figure representing %64 and more shows that a phrase or a word would have a homogenous concept among the respondents. 5. Results 5.1. Descriptive data Gharn Abad ICT Center consists of the internet users (café net) and Tele workers among which 10.7% of Tele workers and 76.8% of café net users are men. In total 68.3% of users are males. The user's age ranges between 9 and 31 years among whom 92.7% are singles. The maximum percent of married users belongs to Tele-workers (%28.6). Also %89.3 of Tele workers is with higher educational level (either having a college degree or attended college). Users with higher educational level include 26.1% in total. Only 4.6% have had computer related formal education. Among the users, 83.2% are students without a job and 96.8% live with their parents. 83.5% of all users are Gharn Abad residents, whereas the rest live in the surrounding areas, and also 75% of Tele worker, live in Gharn Abad which highlights the fact that the spread of ICT in rural settings demands establishment of more centers. 5.2. Two variable analysis To test the relationship between the independent and the dependent variables (ICT adoption), the Pearson, Spearman, Teta Cramer and Phi Lambda analyses could be applied (Kalantari, 2012). Based on Table 2, there is a significant relationship between the adoption of ICT and, the entire individual user's characteristics, excluding the variable of ‘living location’. Thus, Hypothesis 1 is accepted for all of the individual user's variables, excluding the ‘living location’. According to Table 2, out of

B. Khalil Moghaddam, A. Khatoon-Abadi / Telecommunications Policy 37 (2013) 1083–1094

1089

Table 1 The levels of reliability and the KMO figures for the applied concepts validity. Source: the research findings. Factors

Items

The Cronbach Alpha

K.M.O. (%)

The individual (user)

The user's knowledge of the computer before the ICT center was established The level of user's satisfaction towards the center and its management The user's interest to adopt innovation

0.86

68

654.87

0.81 0.62

80 64

461.071 0.000 124.479 0.000

The user's household knowledge level regarding the computer before the 0.97 establishment of the center The parents’ attitude regarding their children's use of the ICT center 0.95 The spouses’ attitude regarding their partners’ use of the ICT center 0.99

80

1546.527 0.000

85 91

3979.121 0.000 3156.032 0.000

The Social (user's household)

Bartlett's test

Sig

0.000

Innovation related

The innovation characteristics

0.86

79

1239.513

The Users’ knowledge and skills of the ICT

The level of the users’ awareness and skills about the ICT usage patterns

0.94

92

3042.785 0.000

0.000

the users’ households’ economic factor, the dummy variable of ‘ownership of more than one house’ has a significant relationship with ‘ICT adoption’. Therefore Hypothesis 3 only for the mentioned variable is accepted. With regard to the correlation quotients (Pearson, Spearman, and Phi Lambda), there is a significant relationship between: the variable of ‘computer based knowledge, and the skill level of the users’ households before the establishment of the ICT center’, ‘the users’ households’ attitude towards ICT as leisure and entertainment’ as well as ‘educational level of the users’ mothers’, ‘the number of households’ members who are familiar with the knowledge of computer’, ‘the households’ members with an official job’, and ‘the users’ households’ attitude towards ICT as leisure and entertainment’ as well as ‘the older sister’s educational level’, and ‘ICT adoption’. Therefore, for theses variable Hypothesis 5 is accepted, but for the other independent variables (being indicated in Table 2), is rejected. According to the Pearson correlation coefficient, there is a significant relationship between the entire innovation related variables, excluding of ‘ICT Compatibility’, and ‘ICT adoption’. Hence, Hypothesis 7 is accepted for all of the variables excluding ‘ICT Compatibility’. Based on the data presented in Table 2, according to the Pearson correlation quotients, there is a significant relationship between all of the variables of ‘Users’ households’ informative & communicative factors’, and ‘ICT adoption’ excluding ‘usage of audio visual facilities’. Hence, Hypothesis 9 is rejected for the mere mentioned variable. 5.3. Multiple variable analysis As a result of the two variable analyses, 24 independent variables proved to have a significant relationship with ICT adoption. Only these 24 variables were entered in the regression model. The results of the multiple regression analysis with stepwise method have been shown in Tables 3 and 4. Based on Table 3, out of 24 independent variables, only 6 variables showed to have significant impact on ICT adoption. The variables of: ‘Users’ elementary and middle school education’, ‘the users’ households’ attitude towards ICT as leisure and entertainment’ and ‘observe ability of the ICT’ have had negative impacts on ICT adoption. It seems the following logics can be respectively the suitable reasons for negative effects of 3 mentioned variables on ICT adoption: (1) The elementary and middle school students had low skill level of using computer. (2) ICT in the families with the above teenage users is used mostly for the entertainment purposes. (3) The rural inhabitants lack sufficient motivation towards the ICT. The other 3 variables (presented in Table 3) showed positive impacts on ICT adoption. According to the amount of Beta coefficients (Table 3) the most influential variable was ‘user as a Tele worker’ (i.e., Tele working as main job for user), and the least effective factor was ‘the users’ households’ attitude towards ICT as leisure and entertainment’. The underlying reason for the effectiveness of the independent variable of the ‘user as a Tele worker’ on ‘ICT adoption’ could be interpreted as the skill level and the frequent time period of the ICT use by the users. The orientation of the entire 6 independent variables in the regression model, determine 77.2% of the variance of ICT adoption as dependent variable (R2 ¼ 0.772). Based on the amounts of B coefficients for independent variables (Table 3) that predict variances of dependent variable assuming constant the other independent variables, it’s clear that the variables of ‘User as a Tele worker’ and ‘Household's attitude towards the user's ICT use as an entertainment’ have respectively the most and the least power of prediction for variances of ICT adoption. With attention to the low figures of S.E (Standard Error for estimation) (Table 3), the accuracy level of line regression in estimating the amount of ICT adoption will be suitable. The significance of a high amount of R2 of regression model (Table 4) as well as proving the significance of regression via statistic of F (Amount of F for degrees of freedom of k¼6 and n−k−1¼ 211 in 95% and 99% respectively are 2.8 and 2.1, that F resulted from ANOVA table, as shown in Table 4, is greater), imply the ‘goodness of fit’ of the regression model of research.

1090

B. Khalil Moghaddam, A. Khatoon-Abadi / Telecommunications Policy 37 (2013) 1083–1094

Table 2 Correlation between the independent variables and the amount of ICT adoption. Source: the research data. Factors

Variable

The correlation value

Sig

Innovation related characteristics

Relative advantage Compatibility Complexity Image surrounding the ICT Observe ability

n

0.174 (Pearson) 0.115 (Pearson) nn −0.352 (Pearson) nn 0.353 (Pearson) nn 0.468 (Pearson)

0.010 0.090 0.000 0.000 0.000

Informative & communicative characteristics of the users

The amount of audio visual media use The amount of informative & communicative media's use The amount of printed media's use

Ns

0.104 (Pearson) 0.255 (Pearson) nn 0.361 (Pearson)

0.125 0.000 0.000

User's individual

Age Knowledge and skill in computer Satisfaction of the center's management Interest towards adopting innovations Gender

nn

0.652 (Pearson) 0.559 (Pearson) −0.159 (Pearson) n 0.158 (Pearson) nn 0.489 (TetaCramer) nn 0.282 (TetaCramer) n 0.230 (TetaCramer) nn 0.364 (TetaCramer) Ns 0.207 (TetaCramer) nn 0.673 (Spearman) nn 0.181 (Phi– Lambda)

0.000 0.000 0.019 0.019 0.000

Ns

Marital status Study area Life style Living location Educational level User's vocation

Users’ households’ economic characteristics

Users’ households’ social related characteristics

nn n

0.004 0.042 0.000 0.098 0.000 0.000

−0.003 (Pearson) −0.023 (Pearson) −0.094 (Pearson) Ns 0.008 (Pearson) n 0.232 (Teta– Cramer)

0.969 0.731 0.161 0.904 0.039

Number of household's members Percent of household's literates Number of household's members with the knowledge of computer The household's knowledge level of computer before of establishing the center Number of household's members with formal occupation Household's view of the ICT center's usefulness Household's satisfaction towards the user's ICT use Household's attitude with respect to user's ICT use for education Household's attitude with regard to user's ICT use as a job Household's attitude towards ICT use for entertainment Father's job

Ns

−0.060 (Pearson) −0.061 (Pearson) 0.298 (Pearson) n 0.162 (Pearson)

0.374 0.370 0.000 0.017

nn

0.001 0.311 0.155 0.855 0.000 0.000 0.368

Spouse's job Father's educational level Mother's educational level Older brother's educational level Older sister's educational level Spouse's educational level

5% significance level, Ns: nonlinear significant relationship. 1% significance level.

nn

nn

The amount of land holding (ha) The amount of land under cultivation Number of owning livestock Household's gross annual income Ownership of more than one house

Mother's job

n

Ns

Ns Ns

Ns nn

0.221 (Pearson) 0.069 (Pearson) 0.097 (Pearson) Ns −0.012 (Pearson) nn 0.525 (Pearson) nn −0.291 (Pearson) Ns 0.043 (PhiLambda) Ns 0.014 (PhiLambda) Ns 0.000 (PhiLambda) Ns −0.093 (Spearman) nn −0.213 (Spearman) Ns 0.173 (Spearman) nn 0.314 (Spearman) Ns 0.350 (Spearman) Ns Ns

0.155 1.000 0.193 0.002 0.057 0.000 0.220

B. Khalil Moghaddam, A. Khatoon-Abadi / Telecommunications Policy 37 (2013) 1083–1094

1091

Table 3 Multiple regression analysis to identify the effective factors on the amount of ICT adoption. Source: the research data. Variable

Beta

S.E

B

Sig

Constant coefficient X1: Be Tele worker (Main job), as a Dummy variable X2: Be user with elementary and middle school education, as a Dummy variable X3: Amount of user's household's use of informative & communicative facilities X4: Observe ability of ICT in a low level or value, as a Dummy variable X5: User with higher educational level, as a Dummy variable X6: The users’ households’ attitude towards ICT as entertainment

–

0.062 0.057 0.038 0.070 0.040 0.046 0.019

0.597 0.882 −0.226 0.322 −0.111 0.118 −0.043

0.000 0.000 0.000 0.000 0.006 0.011 0.026

0.627 −0.213 0.163 −0.105 0.110 −0.078

Table 4 The result of the stepwise multiple regression model assessment to identify the ICT adoption’s effective factors. Source: the research data R

R2

Adjusted R2

F

Sig. F

0.879

0.772

0.766

119.301

0.000

6. Discussion Based on the objective of the research, results of the Hypotheses 2, 4, 6, 8 and 10 (the regression analysis) is discussed in the following, and the Hypotheses 1, 3, 5, 7 and 9 relating to correlation quotient is discussed in the two-variable analysis section. (1) Only the three variables related to individual characteristics (including: ‘the elementary and middle school education of user’, ‘the user's higher educational level’ and ‘the user as a Tele worker’) were influential on ICT adoption. Therefore, Hypothesis 2 is accepted regarding the mentioned three variables, but is rejected with respect to the other individual characteristics’ variables (Tables 2 and 3). The expansion and extension of computer science in school’s educational programs, and popularity of internet-based occupations among Iranian young generation, are underlying factors in ICT adoption. (2) Through the regression analysis, among the five main effective factors regarding the amount of ICT adoption, only the economic factor did not prove any impact. Therefore, Hypothesis 4 is rejected (Tables 2 and 3). Taylor et al. (2003a, 2003b) and Cheong (2002) had already studied ICT adoption at homes in which they paid for the services. Whereas, in the context of this research the expenses were covered by the center itself. This made the economic factor as the least influential. In Iranian context, most villagers’ priority on their spending refers to food and other survival needs. (3) The negative effect of ‘users’ households’ attitude towards ICT as an entertainment’ on ICT adoption has made Hypothesis 6 acceptable (Tables 2 and 3). Rural households in Iran prefer their children spend their leisure time under the family’s supervision, to protect them and to make sure their spare time would not end up with addiction. They regard ICT as a threat to morality. (4) The positive effect of the variable of ‘observe ability of the ICT relative advantages’ on ICT adoption is similar to the findings of Luchetti and Sterlacchini (2004), Thrane (2003) and Rashid and Al-Qirim (2001). Therefore, Hypothesis 8 is accepted merely emphasis on this variable (Tables 2 and 3). In Iranian rural settings, people have more sympathy with tangible benefits of the new technologies. Moreover in Gharn Abad, the ICT skill is prestigious and potentially a chance for an alternative income. (5) At the end, the positive impact of the dummy variable of ‘users with higher educational level’ on ICT adoption approves the study of Taylor et al. (2003a) and Luchetti and Sterlacchini (2004). On the other hand, ‘the application amount of telecommunication devices by the families of users’ was an influencing variable on ICT adoption, which was similar to the findings of Cheong (2002) and Dyson (2004). Hence, regarding the mentioned variable, Hypothesis 10 is accepted (Tables 2 and 3). Not only the young people’s educational level, but also their access to communication media provided a fruitful context in Gharn Abad and had led them to adopt to put into practice their knowledge of the new technology (Fig. 2).

7. Conclusion and recommendations The adoption of ICT was influenced by some individual characteristics such as gender, age, educational level, computer and internet skills. In this research the economic factor did not affect adoption of ICT, in general. It could be concluded that in public centers in which funding and subsidies provided financial support, the impact of economic factor would be undermined. Nevertheless, the adoption of ICT was dependent highly on the pre-existing computer skills among the users.

1092

B. Khalil Moghaddam, A. Khatoon-Abadi / Telecommunications Policy 37 (2013) 1083–1094

Users’ households’ social factor (as an Environmental factor): ● Number of household’s members with the knowledge of computer

● The household’s knowledge level of computer, before of establishing the center

● Number of household’s members with formal occupation

Users’ households’ informative & communicative factor (as an Environmental factor): ● The amount of informative & communicative facilities use

● The amount of printed media use

● Household’s attitude by user’s application of ICT for work

ICT adoption

● The users’ households’ attitude

● Older sister’s education

● Age ● Knowledge and skill in computer

● Satisfaction of the center’s management

● Interest towards adopting innovations

● Gender ● Marital status ● Study area ● Life style ● Educational level

towards ICT as entertainment

● Mother’s education

Individual factor:

Technological (innovation -related) factor:

● User’s vocation

● Relative advantage ● Complexity ● Image surrounding the innovation

Users’ households’ economic factor (as an Environmental factor):

● Observe ability

● Ownership of more than one house Fig. 2. Proposed model for ICT adoption in rural ICT centers of Iran.

With due attention to extreme poverty and a lack of systemic and holistic development program, most of the rural people are excluded from new technologies such as ICT. Since the modernization era of the 1950 s, rural people of Iran have approached and adopted the new technologies with clear and observable economic inputs (Khatoon-Abadi, 1995). Nevertheless, the application of innovations in rural development processes succeeded mainly when rural people could had access to innovation and benefited from its tangible outcomes. Based on the research findings, the following recommendations are proposed to further developing of ICT within rural communities and fostering more researches in the area: (1) The regression analysis showed that ICT adoption in Gharn Abad ICT center had not been affected by the economic variables of the users’ households. Therefore, if within the similar situations and a public funded center, the users’ individual and household characteristics are encountered, then the adoption of ICT could be encouraged. (2) There was not any correlation between the users’ formal education, and the adoption of ICT. This might be due to the performance of computer and internet training programs in the surrounding villages, before the establishment of Gharn Abad ICT center. It is, then, recommended that before establishing a center, different training programs be financially supported and directed in the rural regions by different stakeholders. (3) This study shows the impact of the ‘observe ability of ICT advantages’ on adoption process. It can be recommended that field visits be organized for dissemination of ICT among rural people. It is because through different training visits, people learn about the successful centers in terms of the advantages, as well as the accountability of ICT. (4) There was a negative correlation between ‘the users’ households’ attitude towards ICT as leisure and entertainment’, from one hand, and the adoption of ICT by the users on the other. Based on the field visits and the qualitative interviews, the main reason is the parents’ concern about the immoral function of ICT by their children. Based on their attitudes, it could be harmful for their children to have access to certain websites. It is recommended that educational workshops be facilitated for rural households to inform them of the other useful aspects of ICT such as interactive learning, audio visual aid, e-marketing and socio-cultural networking. (5) There is a correlation between informative & communication tools by the families of ICT users from one hand and the adoption of ICT on the other hand. It can be concluded that application of informative & communication devices at home affect development of ICT centers in rural areas. It is recommended that rural ICT centers be established in parallel to the public financial aids and subsidies for equipping of rural households with relevant facilities and hard wares. Finally it is assumed that the adoption of ICT within a rural forum or center would affect the general ICT adoption (inside and outside of the center) upon which further inquiries is recommended.

B. Khalil Moghaddam, A. Khatoon-Abadi / Telecommunications Policy 37 (2013) 1083–1094

1093

Appendix. (A part of the questionnaire for measuring the dependent variable): level of familiarity & skill and amount of use in ICT See Table A1. Table A1 (A part of the questionnaire for measuring the dependent variable): level of familiarity & skill and amount of use regarding ICT. Weight of usage patterns

Usage patterns

Amount of familiarity and skill Without Awareness (hearing the Interesting Having skill familiarity name of usage patterns) in application 0 1 2 3

Applying

Amount of use (hours/ week)

4

0

Basic familiarity (Turn on & Turn off, Familiarity with desktop) 1 or (0.05) Painting 2 or Using computer for amusement (0.1583) (Game, Film and Music) 3 or Using internet for amusement (0.2666) (Game, Film and Music) 4 or (0.375) Chatting 5 or Type (Mere in word, not chatting or sending (0.4833) email) 6 or Email (Mere personals e mails, not e mails (0.5916) for Tele working) 7 or (0.7) Searching for general information such as pictures 8 or Using the tutorial CDs through computer (0.8083) 9 or Design and update the weblogs or websites (0.9166) 10 or Completion of the tutorial projects and (1.025) group activities through computer and internet 11 or Use of the Government services through the (1.1333) internet 12 or Using the non-tutorial CDs through (1.2416) computer 13 or (1.35) Using the special software such as Photoshop 15 or (1.5666) 16 or (1.675) 17 or (1.7833) 18 or (1.8916) 19 or (2)

On line tutorial learning On line non-tutorial learning Job seeking through the internet Tele working Programming

References Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(1), 179–211. Cheong, W. H. (2002). Internet adoption in Macao. Journal of Computer-Mediated Communication, 7(2), 11–18. Davis, F. D., Bagozzi, R. P., & Warsaw, P. R. (1989). User acceptance of computer technology: A comparison of two theoretical models. Management Science, 35, 982–1003. Dyson, L. E. (2004). Cultural issues in the adoption of information and communication technologies by indigenous Australians. In F. Sudweeks, & C. Ess (Eds.), Proceedings of cultural attitudes towards communication and technology (pp. 58–71). Australia: Mardoch University. Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intentions and behavior: An introduction to theory and research, Reading (p. 335)MA: Addison-Wesley335. Harris, R. W. (2004). Information and communication technologies for poverty alleviation. United Nations Development Program, Asia-Pacific Development Information Program (UNDPAPDIP). International Telecommunication Union (ITU). (2011). Measuring the Information Society. Switzerland. Geneva, 4 and 153. Jalali, A. A. (2007). Interview in: 〈http://ictgolestan.blogfa.com〉. 2012.09.003. Kalantari, K. (2012). Data processing and analysis in socio-economic research [book in Persian] (5th ed.). Tehran: Farhange Saba Press. Kaynak, E., Tatoglu, E., & Kula, V. (2005). An analysis of the factors affecting the adoption of electronic commerce by SMEs: Evidence from an emerging market. International Marketing Review, 22(6), 623–640. Khatoon-Abadi, A. (1995). Radio and development communication (book in Persian). Tehran: Soroush Press.

1094

B. Khalil Moghaddam, A. Khatoon-Abadi / Telecommunications Policy 37 (2013) 1083–1094

Kovacic, Z. J., & Vukmirovic, D. (2008). ICT adoption and the digital divide in Serbia: Factors and policy implications. Proceedings of the Informing Science & IT Education Conference, 363–387. Krysa, R. (1998). Factors affecting the adoption and use of computer technology in schools. Available from: 〈http://www.usask.ca/education/coursework/ 802papers/Krysa/Krysa.PDF〉, 5–8. Luchetti, R., & Sterlacchini, A. (2004). The adoption of ICT among SMEs: evidence from an Italian survey. Small Business Economics, 23. Springer 151–168. Management and Planning Organization of Iran (MPO)., & UNDP. (2005). Information Technology for achieving sustainable human development (book in Persian). Isfahan: Danesh Pajoohan Barin Scientific Institute Publication. Ministry of Information and Communication Technology (2008).The report of the goals and performance of the IT section in 4th National Development Program, Iran. Available from: 〈http://www.ict.gov.ir/ictcontent/media/image/2008/10/2154_orig.pdf〉. Movahed, M., & Abesi, M (2003). A model of technology adoption. Journal of Development and Technology (in Persian), 3 and 4, 92–97. Raaij, E. M., & Scheepers, J. J. L. (2006). The acceptance and use of a virtual learning environment in china. Computers and Education, 838–852. Rashid, A. M., & Al-Qirim, N. A. (2001). E-Commerce Technology Adoption Framework by New Zealand Small to Medium Enterprises. Research Letters Information Mathematical Science, 2(1), 63–70. Rogers, E. M. (1983). Diffusion of innovation (3rd Ed.). New York: The Free Press. Saraie, H. (1993). Introduction to Sampling Methods, in Research. (book in Persian). Samt Press. Ssewanyana, J., & Busler, M. (2007). Adoption and usage of ICT in developing countries: Case of Ugandan firms. International Journal of Education and Development using Information and Communication Technology (IJEDICT), 3(3), 49–59. Statistical Center of Iran. (2007). Iran statistical year book 2006. Available from: 〈http://salnameh.sci.org.ir/LoginPageEN.aspx〉. Taylor, W. J., Zhu, G. X., Dekkers, J., & Marshall, S. (2003a). Factors affecting home internet use in central Queensland. Informing Science Journal, 6(1), 573–588. Taylor, W. J., Zhu, G. X., Dekkers, J., & Marshall, S. (2003b). Socio-economic factors affecting home internet usage patterns in central Queensland. Informing Science Journal, 6(1), 233–242. Thrane, K. (2003). Adoption of ICT in Norwegian teenage homes. Telenor R&D Report. Available from: 〈http://www.telenor.com/rd/pub/rep03/R_20_2003.pdf〉. United Nations Conference on Trade and Development (UNCTAD). (2006). The digital divide report: ICT diffusion index 2005. Newyork and Geneva: United Nations, 49–52. Uzoka, F. M. E., Shemi, A. P., & Seleka, G. G. (2007). Behavioral influences on e-commerce adoption in a developing country context. The Electronic Journal on Information Systems in Developing Countries, 31(4), 1–15. Venkatesh, V. (2000). Determinants of the perceived ease of use: Integrating control, intrinsic motivation and emotion into the technology acceptance model. Information Systems Research, 342–365. Venkatesh, V., & Davis, F. (2000). A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management Science, 46(2), 186–204. Verdegem, P., & Marez, L. D. (2011). Rethinking determinants of ICT acceptance: Towards an integrated and comprehensive overview. Journal of Technovation, 31(8), 411–423. Waarts, E., & Van-Everdingen, Y. M. (2005). The influence of national culture on the adoption status of innovations: An empirical study of firms across Europe. European Management Journal, 23(6), 601–610. Wahid, F. (2007). Using the technology adoption model to analyze internet adoption and use among men and women in Indonesia. The Electronic Journal on Information Systems in Developing Countries, 32(6), 1–8.