Digital Capital and Segregated Education Spaces: Measuring Socio-Economic Determinants of Digital Divide in India

• In Event: Marginalization and technology: Issues, solutions, and new approaches

Proposal

Digital capital, often defined as the “accumulation of digital competencies and digital technology” (Ragnedda 2018), is gradually becoming a critical tool for getting ahead in a “meritocratic society" (Park 2017). Recent studies thus have indicated the importance of examining not only the aspect of access to technology but also assessing how well digital related skills have been diffused across various sub groups of the population. Lack of access to internet or device or lack of adequate skills can give rise to digital divide which further perpetuates educational inequalities among social and economically marginalized groups. These two aspects of digital divide have garnered a lot of attention owing to the outbreak of Covid-19, which consequently led to the shutting down of in person schooling. Additionally, prominent international agencies have endorsed and emphasized the importance of ICT access and know how to ensure inclusive growth. The Sustainable Development Goals (SDGs) have highlighted the need to expand ICT access and skills at an affordable cost especially in less developed countries. Poverty, disparate economic wellbeing and labor market outcomes has also been attributed to digital divide (unequal access to ICT and inadequate ICT know-how). Further, this is how OECD defines digital divide – “the gap between individuals, households, businesses and geographic areas at different socioeconomic levels with regard both to their opportunities to access information and communication technologies (ICTs) and to their use of the Internet for a wide variety of activities” (OECD, 2001, p. 5).
In India, the most recent report on education released by National Sample Survey (NSS 2017, which gathered information on access to the internet, shockingly reveals that merely 23.8 per cent of households have some kind of access (which has gone down from 27 percent in 2014, as per NSS 71st round). However, the situation is even more dismal when it comes to the proportion of households owning computers or any form of device (viz. smartphone, tablet etc.) – only 10.7 per cent households own any kind of device (which according to NSS 71st round in 2014 was 14 percent).

Significant disparities exist in access to computers and digital resources among different social strata in India. Merely 4% of Scheduled Caste (SC) and Scheduled Tribe (ST) students possess the means to access computers with internet connectivity, as opposed to the notably higher 21% within the general category (Oxfam India 2022). Disparities are also evident in internet accessibility, with a mere 14% of the STs having access, while this figure rises to 41% for the general category (Rajam et al 2021).
Given the disproportionate diffusion of computer related know-how among various disaggregated categories of population, it is pertinent to explore the extent of these inequalities between the haves and have nots across markers of gender, caste, region, income levels. The present study would make an attempt to capture the marginal effects from probit regressions on the probability of individuals having a computer at home, Internet access, computer literacy, Internet literacy and Internet use and to what extent are these outcome variables impacted by an individual’s demographic, locational, family background related variables. Also, the study would make an attempt to capture the computer literacy skills across various disaggregation - gender, caste, location, income levels.
The study aims to examine how digital capital intersects with the social background of the learners in India and how digital capital and lack of it can reinforce existing socio-economic inequalities in segregated educational spaces.

Dataset

The present study would draw from nationally representative two large scale surveys conducted by Government of India in 2017-18 and 2020-21 on Education and Multiple Indicators, respectively. The first dataset provides information on educational details of students in the age group of 5-29 years, For each currently attending child, the survey collects information on type of education, level of current attendance, class/grade/year of study, type of management of educational institution, details of benefits received, if any, such as fee waiver, scholarship, free study materials and free mid-day meals, and detailed break-up of private educational expenditure.
The second data set (Multiple Indicators survey) provides a comprehensive set of information related to ICT skills (which we have termed computer literacy). This survey collected information from each individual above the age of 15 years in the form of nine questions that asked their abilities with respect to their computer literacy. The number of households surveyed in the 75th round were 113,757, (68,254 in rural areas and 45,503 in urban areas). NSS 78th round had 11,75,542 respondents (from 2,76,000 households).
The present study focuses on individuals belonging to the age group 15-24 years, who are currently attending secondary, higher secondary and higher education in India. It aims to estimate the extent of digital access (first level digital capital) and digital literacy (digital skill) amongst the younger population. The study further seeks to examine how gender, social capital (class, caste, parental educational background) along with types of educational spaces (public vis-a-vis private educational institutions) can act as critical determining factors for digital divide.
Estimation procedure
Since the dependent variable is categorical and dichotomous in nature, linear regression model cannot be used and so the paper intends to employ binary probit regression model.
Y_i^*= X_i β_i+ µ ………. (1)
Y_i^*= β_0+β_1 X_1+β_2 X_2+⋯……..β_i X_i+ µ……... (2)
where Yi = underlying latent propensity of individual ‘i’ who have access to computer/internet and have computer literacy that is Y = 1 . Xi is the vector which includes all the independent variables including socio- economic, demographic, locational, income and family background variables of individuals. β_i is the corresponding coefficient vector for each outcome i. µ is the random error term which follows normal distribution.
The above probability is derived through maximum likelihood estimates of given by β_i which are the partial derivatives of the estimated probit index function X_i β_i with respect to individual regressors. β_i^* helps to capture the direction and magnitude of impact of the set of independent variables on the probability of access and computer based skills of individuals.

Author

• Anuneeta Mitra, Oakton College