E-inclusion or Digital Divide: An Integrated Model of Digital Inequality

E-inclusion or Digital Divide: An Integrated Model of Digital Inequality ABSTRACT To address the dearth of integrative theories on the digital divide problem, this study proposes an integrative model including both measurements and causes of the digital divide. In order to test the applicability of this model, the authors applied secondary analysis to a 2014 dataset comprised of survey responses from 398 Chinese migrant workers. The results showed how the integrative model can be applied as a potential analytical framework to understand the digital divide. The model suggests that Chinese migrant worker partially accept ICT. However, due to their lack of motivational, intellectual, and social access they remain minimally e-included. Enhanced industrial and public resources may be required to address the challenge of the digital divide. Socio-economic disparity remains the main force that inhibits their e-inclusion and acceptance of ICT. Keywords E-inclusion, ICT acceptance, access, digital divide, Chinese migrant workers INTRODUCTION The challenge of ICT inequality has been widely addressed by researchers from various disciplines. This interdisciplinarity has resulted in longstanding divided approaches addressing the same social phenomena. As Yu (2011) pointed out in her review on digital divide literature, the current body of research reflects epistemic dichotomous divisions in regard to perspectives on the individual as opposed to society. When discussing explanations and determinants of the digital divide, these divisive research traditions differ in their choices of research objects (societal influencers versus individual predispositions), primary determinants (social structure versus individual agency), and interpretation of reality (observable, objective social facts versus abstract, subjective constructs). Differing macro and micro perspectives on the digital divide or ICT inequality hindered comprehensive understanding, measurement, and solutions of this complicated social problem (Cetina & {This is the space reserved for copyright notices.] ASIST 2016, October 14-18, 2016, Copenhagen, Denmark. [Author Retains Copyright. Insert personal or institutional copyright notice here.] Cicourel, 2014). Therefore, researchers have championed the notion of integrative theories of the digital divide (Dijk, 2005; Yu, 2011). The present study applied the proposed integrative model to analyze the lived experiences of Chinese migrant workers. Based on the model, the researchers aimed to comprehensively analyze the digital divide problem. MODEL CONCEPTUALIZATION E-inclusion and Digital Divide Digital divide scholars have largely posited that the nature of the digital divide is participation inequality in this digital society (Barzilai-Nahon, 2006). Borrowed from Warschauer s (2004) theory of social inclusion, the concept of e-inclusion can be defined as the effective participation of individuals and communities in all dimensions of the knowledge-based society and economy through their access to ICT. Further, e-inclusion refers to the degree to which ICT contribute to equalizing and promoting participation in society at all levels (i.e. social life, jobs, political participation, health, recreation, etc.). (eeurope Advisory Group, 2005, p. 7). Conversely, the digital divide represents the gap between those who are empowered to substantially participate in an information and knowledge-based society and economy, and those who are not. (eeurope Advisory Group, 2005, p. 7). It is argued that the digital divide has demonstrated widening, rather than narrowing, in the recent years despite the fact that people all over the world generally have better access to computers, smartphones, and the Internet (Dijk, 2005). E-inclusion and digital divide are contrasting phenomena which represent the desired and actual situations as it relates to universal digital participation. The more the digital divide is rectified, the closer we are to an e-inclusive society. As long as individuals and groups are excluded from participation at any level in our current digital society (e.g., political participation, social life), we are still in a digital divided society. ICT Acceptance Previous studies heavily criticized the traditional binary measurement of use or non-use of ICT, which is insufficient when it comes to capturing the progressive nature of human- ICT interactions. The technology acceptance model (TAM), however, defined ICT acceptance as a continuous process consisting of intention of use, actual adoption, and

continuous use of specific technologies to satisfy personal or organizational needs and empower the individual or the group (Chen, 2007; Venkatesh, Morris, Davis, & Davis, 2003). ICT acceptance can thus be understood as a behavioral measurement that delineates the relationship between ICT and individuals, whereas e-inclusion is an effect measurement intended to explain the ways in which individuals are empowered by ICT in life and work (e.g., personal health records, social media, organizational management systems, computer-supported cooperative work, or distance learning). Access Access describes the freedom or ability to obtain or make use an abstract or tangible thing (Webster s Dictionary, 2016). Access therefore requires possession of resources, not actions, in order for one to make usage of ICT. ICT access subsequently describes resources that directly relate to making meaningful use of ICT. The challenge of inadequate ICT acceptance or e-inclusion (that is, digital divide) results from lacking access to some degree. In other words, access is proposed as a theoretical construct that describes the micro causes of ICT acceptance and e-inclusion. According to previous studies (Bucy, 2000; Dijk, 2005), ICT access is a four-fold typology: material access, intellectual access, motivational access, and social access. Generally, each of these types of access is required to experience ICT acceptance and eventual e-inclusion. Material access describes the possession of computers, Internet connection and other technology or the permission to use them. Intellectual access refers to possession of direct intellectual abilities and digital literacy to support ICT acceptance. Motivational access is the desire to accept and use ICT through the actions of purchasing, possessing, and continuously learning or advancing skills. Social access denotes to the possession of necessary social identities, social relations, or social conditions that are directly needed for ICT acceptance (e.g., digital library membership, a techsavvy daughter, or creation of online profiles). Resources The research on digital divide also pinpointed tangible items which facilitate various types of access (Dijk, 2005). A resource is a theoretical construct referring to financial capital, material, staff, and other assets that can be drawn on by individuals in order to obtain ICT access. When compared to access (micro cause), resources describe the meso causes of digital divide. Previous literature yielded at least seven types of resources (e.g., Dijk, 2005): Forces Material resources refer to money and properties that can be exchanged for equipment, services, and other items which tangibly support ICT access. For instance, the possession of a laptop represents material access, whereas the availability of money to purchase a laptop equates to material resource. Intellectual resources denote general literacy, knowledge, and cognitive abilities that provide support for intellectual access and motivational access. For example, English literacy represents the general literacy, which is a critical asset (intellectual resources) required to understand online health materials in English (intellectual access). Psychological resources are perceptions and attitudes such as self-efficacy, confidence, and other psychological and subjective elements which mainly facilitate motivational access. Social resources involve social capital or social identities that can be exchanged for various types of access, especially social access. For example, if an individual belongs to a social networks whether online or offline (social resources), they are more likely to get help when using ICT (social access). Spatial-temporal resources represent the available time and space to support social access and ICT acceptance. Time and space are socially constructed factors which greatly shape the challenge of the digital divide (Yu, 2012). For instance, when compared to those who work in construction, those who work in education have greater time and opportunity to access ICT. Industry resources describe the resources provided by ICT vendors which are necessary to support various access. These include product tutoring and customer service. Public resources denote the viability of broadband, public digital services and similar infrastructure which support resources and various types of access, especially material and intellectual access. A plethora of studies address the forces that perpetuate the digital divide (e.g., Hsieh, Rai, & Keil, 2008). Different from a resource, a force is a theoretical construct representing a higher order influential mechanism that causes a direct or indirect change in ICT acceptance and empowerment. Forces reflect the macro and dynamic perspective and describe the powers that can change the quantity, quality, and structural distribution of resources (meso perspective) needed for ICT access (micro perspective) of individuals, people groups, or organizations. Forces either facilitate or inhibit ICT acceptance or e-inclusion and can include socio-economic

forces, political forces or cultural forces (Borgida et al., 2002; Dijk, 2005; Hsieh et al., 2008). An Integrative Model of Digital Divide As seen in Fig. 1, e-inclusion and ICT acceptance are measurements that evaluate the degree of the digital divide (i.e., the extent to which people experience the digital divide). Essentially, e-inclusion is an effect measurement to evaluate how people are empowered in the digital society at various levels in life and work, whereas ICT acceptance is a behavioral measurement to assess how people adopt and use ICT. By comparison, forces, resources, and access represent the macro, meso, and micro causes or influencers of the digital divide (i.e., the reasons why people experience the digital divide). Forces can change the quantity, quality, and structural distribution of resources possessed by people, which in turn shape the access needed for e-inclusion or acceptance of certain technologies. Figure 1. An Integrative Model of Digital Divide Ideally, the challenge of the digital divide will be solved when all individuals and communities equally participate in this digital society. However, with ever-evolving technologies and other forces leading to inequality, the digital divide will persist and thus require lasting efforts to evaluate and solve the problem. This theoretical model can be applied as the analytical and measurement tool to solve the real-world ramifications of the digital divide. RESEARCH QUESTIONS In order to investigate how our integrative model can be applied to analyze the implications of the digital divide, we used an existing dataset of Chinese migrant workers to answer following questions: RQ1: Do Chinese migrant workers accept ICT and experience e-inclusion? RQ2: If so, what types of access and resources do Chinese workers have? RQ3: If so, which forces influence their ICT acceptance or e-inclusion? METHODS Data Collection To test the applicability of the integrative model, the authors used existing survey data pertaining to the experiences of Chinese migrant workers who migrate from rural regions to perform low-level jobs in cities. Since they are on the margins of Chinese society, this population is considered more susceptible to the digital divide when compared to dominant groups (Wong, Li, & Song, 2007). The survey was conducted in 2014 using the snowball sampling technique. A total of 398 responses from migrant workers throughout China was collected. Coding and Analysis Two researchers recoded the questionnaire according to the typology presented in the integrative model, and achieved 78% agreement. Discrepancies were discussed with the third researcher until final agreement was achieved. Descriptive analysis was conducted to investigate the distribution of variables related to e-inclusion, ICT acceptance, access and resources. Finally, multiple iterations of stepwise regression modeling was applied to identify forces influencing e- inclusion and ICT acceptance. FINDINGS E-inclusion and ICT Acceptance E-inclusion refers to full participation in the information society. In the survey, it was measured as what actitivities do you use ICT for? The results showed that 357 subjects (91.8%) were used for social reasons (e.g., communication, wechat, QQ), 318 (81.7%) for entertainment (e.g., games, music, novel), 301 (77.4%) for information (e.g., news, weather, work-related information), and only 173 (44.5%) of migrant workers use ICT for personal business (e.g., online shopping). In other words, Chinese migrant workers are partially e-included in terms of communication, entertainment and everyday life information activities. Few people use ICT for intense research or education. Actual adoption and continuous usage of certain ICT equipment were measured to indicate ICT acceptance. For actual adoption, 347 (89.2%) and 234 (60.2%) migrant workers have used cellphones and computers respectively. Only 83 (21.3%) have used tablets and 17 (4.4%) used wearable technologies. Interestingly, only 78 (20.1%) of migrant workers have used publicly available technologies (e.g. libraries). For ICT continuous usage, they spend 2-6 hours per day using ICT. On average, they have used Internet for 3.85 years, cellphones for 4.18 years, and mobile Internet for 3.38 years. The results indicate that Chinese migrant workers mainly rely on their cellphones and computers for digital activities. ICT Access Distribution Based on the four dimensions suggested by the model, we found the following patterns in terms of access: Device ownership and broadband access were assessed for material access. 365 (93.8%) workers have their own cellphones, 272 (69.9%) own computers, 101 (26.0%) tablets, and 14 (3.6%)

wearable technologies. In terms of broadband access, only 280 (72.0%) subjects can get access to broadband conveniently. Intellectual access was measured by adoption skills (general ICT ability) and meaningful usage skills (ability to use ICT for specific tasks). Around half of migrant workers indicated that they have little barriers when it comes to learning ICT (49.6%), and only 91 (23.4%) users indicated that they experienced no difficulty in learning ICT. For meaningful usage skills, 285 (73.3%) workers know how to pay utilities and purchase online, 200 (51.4%) can participate socially online, and 193 (49.6%) know how to find useful information through ICT. Social access was measured as social capital in the survey. Friends or colleagues (50.6%), ICT vendors (29.0%), and families (23.7%) are sources that provide ICT support when migrant workers encounter difficulty. Motivational access can be reflected in motivations and attitudes related to ICT usage. For general motivation to use ICT, 279 (71.7%) workers cannot live without cellphone, and 172 (44.2%) like most functions of their smartphones. For purchasing motivations, 161 (41.4%) would like to buy devices even at a price of sacrificing other aspects of life. Around half of the workers prefer digital devices to interpersonal sources (53.7%) and print media (49.6%) for information search, and even for emotional expression (33.7%). When faced with technical difficulties, 287 (73.8%) users will not give up, 254 (65.2%) will seek help, and 170 (43.9%) will attend training programs. Access distribution analysis indicates that Chinese migrant workers have better material access than the three other types of access. Around half of them appear to face challenges in terms of intellectual, social, and/or motivational access. Resources Distribution Resource distribution analysis can help to identify factors that result in lack of certain types of access. The survey results yielded the following six attributed resources: General availability and quality of social capital were measured to refer to the social resources. Most workers have reliable social support for information (63.5%) and other needs (68.6%). For available money for ICT usage (indicator of material resources), 308 (79.2%) workers spent less than $15 on Internet and communication monthly, and less than $615 on devices annually. For spatial-temporal resources, 271 (69.7%) have at least two places (home, public spaces, etc.) for ICT usage, and 229 (58.9%) have enough time to use ICT for information. Intellectual resources refer to the general intellectual abilities and other skills related to digital literacy. General information literacy (the ability to find, evaluate, receive, understand, express and make usage of information) was measured in the survey using 7-point Likert scale. On average, migrant workers measured their information proficiency as 4.9 out of 7 points. Public and industry resources are resources that are intended for the general public. These types of resources were represented as publicly provided spaces, devices, infrastructure, education, etc. A total of 299 (76.9%) and 207 (53.2%) workers use nearby cyber cafes or libraries. Only 185 (47.6%) workers believe that their broadband infrastructure is not problematic. In terms of public education, there are very few community groups (16.2%), public library programs (11.3%), volunteer teaching services (7.2%), or peer training opportunities (5.7%) for migrant workers. Industry resources are also relatively unavailable for migrant workers. Customer service or vendor assistance is not available to 154 (39.6%) workers. Meanwhile, customer or employee training are available to only 31 (8.0%) and 103 (26.5%) subjects. The resource distribution analysis indicates that when compared to personal or micro resources, the lack of macro-level resources (industry and public) is more responsible for migrant workers digital divide problems. Forces Among various possible forces measured in the survey, personal socio-economic status (e.g., marriage, wage, education) is the only force positively related to both e- inclusion and ICT acceptance of migrant workers. The regression results for e-inclusion (types of ICT related activities) and acceptance (types of digital devices that have been used) indicated that: migrant workers with higher education, younger age, and higher monthly income had more experience with using devices; but construction workers showed poorer e-inclusion and acceptance level than those doing other jobs (e-inclusion regression model: R 2 =0.124, F=14.7, p<0.001, Durbin-Waston=1.65; ICT acceptance regression model: R 2 =0.149, F=17.9, p<0.001, Durbin-Waston=1.75). The regression results also indicated that there are other forces of e-inclusion and ICT acceptance besides socio-economic force which need to be addressed further in the future. CONCLUSIONS AND FUTURE WORK This study proposed a theoretical framework to analyze the challenge of the digital divide and test it empirically using an existing dataset of Chinese migrant workers. The results showed how the integrative model can be applied as an analytical tool to understand the digital divide. According to the typologies suggested by the model, Chinese migrant workers have moderately accepted ICT and have been partially e-included. Most have material access to ICT, but almost half of them lack of social, intellectual, or motivational access. These gaps predominantly result from the lack of public and industry resources, especially ICT education. Since the empirical data was not driven by the model, some aspects suggested by the model were not reflected in data. For future studies, additional empirical data will be collected to test and operationalize the proposed model. In the long term, we would like to develop the theoretical framework

and relative instruments to guide further digital divide research. REFERENCES Barzilai-Nahon, K. (2006). Gaps and bits: Conceptualizing measurements for digital divide/s. The Information Society, 22(5), 269 278. http://doi.org/10.1080/01972240600903953 Borgida, E., Sullivan, J. L., Oxendine, A., Jackson, M. S., Riedel, E., & Gangl, A. (2002). Civic culture meets the digital divide: The role of community electronic networks. Journal of Social Issues, 58(1), 125 141. http://doi.org/10.1111/1540-4560.00252 Bucy, E. P. (2000). Social access to the internet. The Harvard International Journal of Press/Politics, 5(1), 50 61. http://doi.org/10.1177/1081180x00005001005 Cetina, K. K., & Cicourel, A. V. (2014). Introduction. In Advances in social theory and methodology (RLE Social Theory): Toward an integration of micro- and macrosociologies (pp. 1 48). Routledge. Chen, I. Y. (2007). The factors influencing members continuance intentions in professional virtual communities - a longitudinal study. Journal of Information Science. http://doi.org/10.1177/0165551506075323 Dijk, J. A. G. M. van. (2005). The deepening divide: Inequality in the information society. SAGE Publications. eeurope Advisory Group. (2005). e-inclusion: New challenges and policy recommendations. eeurope Advisory Group. Hsieh, J. J. P.-A., Rai, A., & Keil, M. (2008). Understanding digital inequality: Comparing continued use behavioral models of the socio-economically advantaged and disadvantaged. MIS Quarterly, 32(1), 97 126. Wong, D. F., Li, C. Y., & Song, H. X. (2007). Rural migrant workers in urban China: living a marginalised life. International Journal of Social Welfare, 16(1), 32 40. http://doi.org/10.1111/j.1468-2397.2007.00475.x Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of information technology: Toward a unified view. MIS Q., 27(3), 425 478. Yu, L. (2011). The divided views of the information and digital divides: A call for integrative theories of information inequality. Journal of Information Science, 37(6), 660 679. http://doi.org/10.1177/0165551511426246 Yu, L. (2012). Towards a reconceptualization of the information worlds of individuals. Journal of Librarianship and Information Science, 44(1), 3 18. http://doi.org/10.1177/0961000611424586