Digital divide and access to online education: new evidence from Tamil Nadu, India

The COVID-19 pandemic resulted in partial or full closure of educational institutions and forced them to limit their regular teaching activities to online modes of education. The contagious nature of the virus and the emergence of new variants may cause disruptions and weaken educational practices, learning outcomes, and impact the future of students in the coming years. The prolonged closure of educational institutions and massive, unplanned shift towards the digital mode throws new challenges for the educational system, teachers, students, and parents. Among other factors, the digital divide^{Footnote 1} plays a key role in driving this transition and quality of learning in any context. Given the financial and technical implications, the pandemic is likely to keep a large number of students out of learning and their schools. As more than two-thirds of the student population faced disruption of schooling in 2020, a steady increase in the number of students that might fall below the Minimum Proficiency Level (MPL) in reading (UNESCO 2021a, b, c) is expected. Effective policy interventions to bring them back and address the issue of access to the new mode of education for students from all regions, classes, gender, and socio-economic status are very important. Global development agencies and leaders call for prioritising education in the stimulus packages and addressing challenges that have emerged in the new context.

Based on the evidence collected through three rounds of the Tamil Nadu Covid Pulse Survey (TNCPS),^{Footnote 2} this article maps Tamil Nadu’s experience with a new mode of education. The results offer new insights on specific interventions made by the state, and challenges faced during this transition. The article briefly reviews the global and national response to the disturbance caused by the pandemic on education and the challenges faced while continuing with the online mode of education. It then discusses Tamil Nadu’s experience in coping with online education and concludes the discussion by establishing interrelations with the larger trends and directions for policy interventions.

The response to pandemic

Globally, the national education responses to COVID-19 indicate that most education systems followed the mixed-media approach with distance learning through traditional media (like TV, Radio) and interactive classes by teachers during the pandemic (UNESCO 2020a, b). The ‘majority of governments (133 countries) provided a mix of online, television, and radio solutions to ensure learning continuity, providing strong ground to build more resilient education systems and bridge the digital divide’ (UNESCO 2021a, b, c). Most governments used smartphones to improve access to online learning platforms (UNESCO et al. 2020).

At the national level, the Union HRD Minister asked the students to continue their learning by making full use of the available digital e-Learning platforms like DIKSHA, ePathshala, National Repository of Open Educational Resources (NROER), Swayam, Swayam Prabha, etc. well before the nationwide lockdown and ‘Janata Curfew’ on 22nd March 2020 (Government of India 2020a, b). Later, ‘PRAGYATA Guidelines on Digital Education’ was released online on 14 July 2020 and the government continued to promote its digital initiatives to complement this transition (Government of India 2020a, b). Though India has emerged as one of the largest e-learning markets, its digital infrastructure is still not sufficient to meet these requirements. At the national level too, the hybrid model (combination of online and offline) was followed in several contexts (Mishra et al. 2020).

At the state level, Tamil Nadu also followed multiple strategies to promote online education and the dissemination of digital content or learning material through various platforms. This includes broadcasting lessons on Kalvi TV^{Footnote 3} for school students, TN-DIKSHA digital learning platform for the children, Veettupalli programme (School at home) for children enrolled in schools functioning under Tamil Nadu State Board, and introduction of other innovative practices (Government of Tamil Nadu 2020). Other initiatives include Tamil Nadu Teachers Platform (TNTP); TNSCERT YouTube channel; Facebook—Workplace; programme for student mental wellbeing (with UNICEF); Tamil Nadu Vagupparai Nokkin (Classroom Observation module); NEET—Online Practice tests and Crash Course (MHRD 2020). The state also modified the existing programmes to meet the local requirements.

Challenges with new mode of education

With the large-scale introduction of online teaching and learning as the new norm for education, not as an optional or complementary tool, the pandemic helped in addressing the reluctance towards the adoption of modern technological solutions. However, this was not accompanied by new instructional and evaluation methods to address the challenges caused by the changes in distance, scale, and personalised teaching and learning in online mode. The existing solutions may not address all concerns or may fail in compensating the traditional educational practices (Dhawan 2020).

Even before the pandemic, students’ access to digital technologies, resources, and competence show sharp differences across the regions and communities. The pandemic exposed the existing digital divide and how its deepening could further expand the inequalities (Andrew et al. 2020; Jæger & Blaabæk, 2020; Katz et al. 2017). The pre-existing social and digital divides make the marginalised groups more vulnerable and put them at risk of falling further behind in learning (ILO & World Bank 2021). Studies estimate significant losses in learning and an increase in school dropout rates due to the pandemic across poor countries and also indicate that the pre-existing socioeconomic inequalities may amplify the vulnerability of young people. In some contexts, the intersection of gender, poverty, and disability may result in deepening the social inequalities and access to digital resources (Jones et al. 2021). Due to limited access to the devices, girls face challenges in developing Information and Communication Technology (ICT) skills and engagement with distance learning (Amaro et al. 2020). The pandemic is likely to increase the risk of a large number of girls dropping out of their schools permanently; one of the estimates suggests that around 11 million girls will not return to school when the crisis is over. Most of them are in their adolescence and are from low- and lower-middle-income countries (Azevedo et al. 2020; De Paz Nieves et al. 2021). In many countries, students with special needs did not receive additional support to continue their education (UNESCO 2021a, b, c). The pandemic has contributed to rising inequalities in learning outcomes in different countries (Parolin, 2021; UNESCO 2020a, b).

Globally, the introduction of distance-learning programmes has exposed several issues including unequal access to ICT infrastructure, insufficient internet, power, and skill gaps in using distance-learning platforms (UNESCO 2020a, b). Though recent developments in educational technologies and digital applications provided a strong base for adapting to the situation, educational institutions faced a wide range of logistic, technical, financial, and social problems with the large-scale adoption of online education during the pandemic (Chatterjee & Chakraborty 2021; Oyedotun 2020). Access to appropriate devices and a comfortable place to access uninterrupted classes, and competency to use the applications, etc. are crucial.

The introduction of online education imposes an additional financial burden on households and puts the students at risk of discontinuing education. Due to the economic impact of the crisis, around 24 million students from pre-primary to tertiary education globally are at risk of not returning to learning. They are largely found at the primary and secondary levels (11 million) and concentrated in South and West Asian (5.9 million) and sub-Saharan African (5.3 million) regions (UNESCO 2021a, b, c). Various projections map the likely impact of the pandemic on proficiency and indicate a steady decline in learning outcomes and MPL (Azevedo et al. 2020; Kaffenberger 2021; Kuhfeld et al. 2020; UNESCO et al. 2020; World Bank 2020). In the absence of adequate support, students from countries with weaker infrastructure and longer school closures are likely to face larger learning losses and are likely to see a reversal of gains made over the past decades (Engzell et al. 2021). The experience reported from different countries highlighted the concerns on the psychological impact on students due to social isolation, confinement and lack of physical exercise, reduced access to nutrition, and school meals (UNESCO 2020a, b).

India’s experience as reported from different contexts also follows many of these broad patterns. The country is home to the world’s second-largest internet user base (with 49.8% of the population), but more than half of the population does not have access to the internet. In this transition, underprivileged communities are still lagging behind in the digital race (Devara 2020). Even before the pandemic, India experienced a deep digital divide between rural and urban areas. Nearly 66% of Indians live in villages but only a little over 15% of rural households have access to internet services (as against 42% in urban) as reported by NSSO data on social consumption of education in 2017–18. Only 17% of the students enrolled in rural areas (against 44% in urban) have access to the internet in the country. Currently, enrolled students from advantaged socio-economic groups have relatively better access to digital infrastructure, compared to only a smaller share of students from the poorest income groups (only 2% have access to a computer with internet) and marginalised groups (Reddy et al. 2020). It has been found that the ‘trend of ICT asset ownership and use is a factor of caste, highest adult education, main source of income, and income group. Lower caste groups and lower educated households own and use the ICT assets less than higher caste groups and higher educated households’ (Tewathia et al. 2020: 8). The exclusionary nature of digital access and differences in penetration of digital technologies varies across the regions and social groups are likely to worsen the learning opportunities further.

Studies have highlighted the impact of technology-enabled learning programs on improving the learning outcomes, and productivity of teachers by complementing rather than substituting teachers in India (Muralidharan et al. 2019). The response of students to this shift, the emergence of digital classrooms and applications as the only mode of education (more than an option or supplementary/aid) indicates that it cannot replace traditional classroom learning (Parthiban et al. 2021).

Long hours of online classes were found to be detrimental to children’s health and well-being (including mental health), social life, and learning outcomes (Cao et al. 2020; Kumar et al. 2021). Reports from different parts of the country also highlight other issues related to long hours of online classes faced by students, teachers, and parents. Continued exposure to digital gadgets harms children’s physical and mental health, personal and social life by increasing stress, anxiety, uncertainty, and other behavioural issues (Dhawan 2020; Singh et al. 2020). The pandemic has made a significant impact on the mental health, education, and daily routine of students (Chaturvedi et al. 2021). Majority of students use mobile phones to attend online classes where students’ concentration during long hours of classes can be affected (Bhattacharya 2020). The following sections present the evidence collected from TNCPS to map Tamil Nadu’s experience with online and digital learning initiatives.

The TNCPS adopted a quantitative methodology with the support of the Computer-Assisted Telephonic Interviews (CATI) system to administer the interview schedule. TNCPS can be defined as a ‘panel survey with refreshment’; the stratified sampling design identifies the district as a stratum and identified villages as the Primary Sampling Units (PSUs) in rural areas and Urban Frame Survey Blocks (UFSBs) in urban areas. The households sampled for the TNCPS were a sub-sample of the households which were already contacted for the Tamil Nadu Household Panel Survey (TNHPS).^{Footnote 4} The sampling frame for the first stage units was drawn from the list of 2,12,282 households contacted for TNHPS Pre-Baseline Survey (PBS) in 2018–2019. From this, 1,51,830 households with contact information (landline/mobile) were grouped into 3 broad strata based on the number of families belonging to different social groups (SC-ST; OBC; and Others). A total of 13,749 rural and urban samples were selected for the first round of the TNCPS survey and this formed the panel for the subsequent rounds. For this paper, we look at only data from 3 rounds (2, 3, and 4) as online education was introduced only after the round I survey was completed (Table 1).

Being a telephonic panel survey, efforts were made to ensure a better response rate, quality control, verification and adding refreshment samples to address attrition and balance between rural and urban areas.^{Footnote 5} Compared to similar initiatives reported elsewhere in the country, TNCPS panel data have a relatively larger sample size and good response rates(Pramanik, et al. 2022).^{Footnote 6} Round IV followed the strategy of selecting a proportional sample of rural households using the same stratification design used to select the initial sample for addressing the low representation of rural samples in previous rounds. During the survey, information was collected from the key informants mostly parents and heads of the households. As mentioned earlier, the analysis presented in the paper covers three recent rounds of TNCPS.

Along with the descriptive statistics, the paper also presents the results emerged from regression analysis attempted on relevant variables. We use a nonlinear panel regression model, with binary dependent variable, in a generic panel regression model to analyse the determinants of access to online education during the time of COVID-19.

where\({y}_{it}\) is the binary dependent variable of interest, i.e. access to online education and the values are 1, if households have accessed online education, and 0, otherwise.\({x}_{it}\) is a vector of explanatory variables for individual (household, in this study) “i” in the time period “t” (TNCPS waves II, III, and IV in this study).\({\alpha }_{i}\) is a time-invariant, individual (household, in this study)-specific effect in the model, which is considered to be an unobservable explanatory variable.\(\theta\) is the vector of parameters to be estimated.

This model analyses the observation of many individuals (households, in this study) observed in a few time periods. The descriptive statistics for both dependent and independent variables are given in the annexure (Table 12).

The fact that the majority of students currently enrolled do not have access to internet and computer with internet (respectively 75% and 91%) in India indicates the nature of challenges in making online education inclusive. Access to digital infrastructure varies sharply across the states. Only a smaller share of students currently enrolled in Tamil Nadu has access to a computer with internet or computer or internet (14%, 19%, and 23%, respectively). The sharp difference in the share of students having access to a computer with internet in rural and urban Tamil Nadu (9% and 20%, respectively) demonstrates the nature of the digital divide in the state (Reddy et al. 2020).

Improving digital divide?

More than half of the households covered under three rounds of the survey had at least one student attending school or college (52.8%, 53.9%, and 51.8%, respectively, in rounds II, III, and IV) and this pattern remains similar in rural and urban areas. Around 10% of the households have three or more students in the family reported in all three rounds. The total number of students reported varies marginally across the three rounds (Table 2). The gender composition indicates that around 52% of the students reported across the three rounds are males.

Around 20% of the students are studying in colleges while the remaining are attending school. The share of college-going students remains low in rural areas (18.7% against 21.6% in urban areas as reported in round II). At the school level, 55.6% of the students are enrolled in government or government-aided schools and this share was found high in rural areas (68.3% against 43.4% in urban areas during round II). In the case of higher education, we notice a reverse trend as 75.2% of the students are enrolled in private self-financing colleges and this share was found high in urban areas (77.4% against 72.1% in rural areas during round II). The details indicate that more students from the rural areas are enrolled in diploma and vocational /ITI courses (11.3% and 7.9%, respectively, against 6.8% and 4.4% in urban areas). The share of students enrolled in undergraduate and postgraduate courses remains high in urban areas (78.1% and 9.0%, respectively, in urban areas against 73.7% and 5.3% in rural areas during round II).

During round II, nearly 60% of the students had accessed online education in Tamil Nadu. The third round reported a marginal decline in this share (from 59.3% to 52.5%) and the decline may be linked to the relaxations in lockdown restrictions and partial reopening of educational institutions during this period. This was followed by a sharp increase in the percentage of students accessing education through digital services during round IV and the same could be attributed to the onset of the 2nd wave of COVID-19.

While reporting a gradual improvement in the overall access, all three rounds of the survey report differences in the share of students accessing online education in rural and urban areas (Table 2). This digital divide marginally increased between round II and round III (from 14.7% to 16.2%) but declined in the later period (12.8% as reported in round IV). Within Tamil Nadu, the pattern of access to online education varies across districts. The top three districts with higher access to online education were Ramanathapuram followed by Erode and Kanniyakumari, while the bottom three districts were Thiruvarur, Cuddalore, and Viluppuram.

The pattern of access may vary across different levels of education. Compared to round II, round IV reports a steady improvement in share of students accessing online education at all levels (Table 3). Compared to school students, share of students accessing online education found high among the college students. Students from urban areas have better access to online education in schools (74.7% against 60.1% in rural areas) and colleges (91.1% against 88.9% in rural areas). Nearly 90% of students enrolled in colleges and schools use mobile phones to access online classes, while the use of tablets and desktops/laptops is found high among college students.

Given the differences in social and economic status of the households, challenges faced in accessing online education during the pandemics may vary across the communities. Vulnerable groups with limited resource bases and access to digital resources are likely to be excluded from digital learning. The social background of students and their access to online education indicates that students from vulnerable groups have low access to online education. While TNCPS reports a gradual improvement in access to online education for students from all social backgrounds in the state but a significant variation across the social groups (Table 4).

Devices used for online education

Across three rounds, majority of students (83.7%, 83.5% and 90.4%, respectively, in rounds II, III and IV) who accessed online education were using mobile phones to access online classes followed by other gadgets including tablets, laptops/desktops, and neighbour’s devices. During round II, 3.9% of the students (4.5% and 3.5%, respectively, in rural and urban areas) reported that they used the laptops distributed by the government of Tamil Nadu. Compared to urban areas, rural areas have fewer students using tablets, computers, or laptops to access online education. A significant share of students, especially in rural areas followed classes broadcasted through Kalvi TV during three rounds of the survey. It also indicates a marginal decline in the use of laptops/desktops and neigbour’s device or no device used for accessing the online classes during three rounds (Table 5).

Overall, more than 85% of the students use mobile internet data to access online education in the state compared to other means like wifi modem, others’ wifi/hotspot, etc. Nearly 90% of the total students have access to study materials shared through different forms while this share is found marginally low in rural Tamil Nadu. Students use multiple options (including combinations) to collect the study materials (Table 6). Between rounds III and IV, the share of students who received these materials by visiting their school/college increased sharply from 55.3% to 76.7%. At the same time, the share of students who used WhatsApp/ Telegram/Google Classroom to collect the materials declined sharply (from 54.6% to 36.6% against a marginal decline in the use of email (from 7.6% to 6.9%) and preloaded material in laptops^{Footnote 7} (from 3.1% to 1.8%).

Many students reported that they received additional support from teachers after online classes and this has improved over time (increased from 46.1% to 55.6% between rounds II and III). Notably, more students from urban areas received such additional support (57.6% against 52.4% in rural areas) during the third round. Around 83% of students from both rural and urban areas received this additional support through phone calls/emails. During round III, share of students who received support through additional classes or school/college visits was found marginally high in rural areas (20.6% against 17.6% in urban areas).

Coping with new mode of education: challenges

While looking into the reasons reported for not attending online classes across the three rounds, we notice some changes in the share of students reporting that their schools or colleges do not conduct online classes or they were attending in-person classes (Table 7). The government guidelines for the educational institutions to facilitate online mode of teaching and distribution of digital content, changes in the spread of reported cases, relaxations and emergence of new waves and variants of COVID-19, etc. also could have contributed to this pattern.^{Footnote 8} Information on ‘Returning to school/college for regular classes’ may be looked with the context (reported in round 3) and other factors which could contributing to decrease in usage of online education during these periods.

The respondents reported various reasons for their children not attending online classes. Some students could not attend online classes on time as the households had more students using a single device for accessing online education. It may be noted that many households (around 10%) had three or more students attending classes online. Others reported the lack of internet network connectivity and lack of computer knowledge of the parents/students as the reason for not attending online classes. The share of students facing these issues remains high in rural areas. Compared to the earlier round, we notice some positive changes in round III but they do not seem to continue in the later period (Table 7).

Like elsewhere, Tamil Nadu also reported loss of job and income due to the lockdown restrictions and economic stress caused during the pandemic. The evidence shows differences in the pattern of job loss reported in rural and urban areas and their reversal over different rounds of the survey. In the early rounds, rural areas reported higher job loss, while urban areas reported more job loss in the later rounds (the recovery took longer period). Similarly, job loss reported by casual workers remain high in all key sectors where as this share remained low for salaried workers in the state (Ananthpur et al. 2022). As more households from marginalised sections engage vulnerable livelihoods like casual wage work, the worst hit sector, students from these backgrounds face more challenges in accessing online education. Many parents could not afford the cost of electronic devices or wifi modem and data charges or mobile data charges to support their children’s online education (Table 7).

The additional financial burdens and technical challenges adversely affect the enrolment and drop-out rates at various stages of education. While looking into the reasons for not attending online education, the survey finds that some students (6.3%, 3.8% and 4.2% in round II, III and round IV respectively) dropped out of their school/college due to financial crisis. Compared to rural areas, share of students who dropped out of schools and colleges due to this reason remains high in urban areas (8.6%, 4.2% and 5.0%, respectively, in rounds II, III and IV against 4.0%, 3.6% and 3.7% in rural areas). Given the nature of the crisis and the possibility of under-reporting, the government may adopt effective strategies to monitor these trends and control further rise in the dropouts and bring back the out-of-school children.

Many students currently accessing online education face technical and network issues during online classes (39.3%, 33.3%, and 28.5% during rounds II, III and IV, respectively). The survey reports a gradual decline in the share of students facing technical issues in both rural and urban areas over three rounds. Some students face challenges in availing a separate room or a quiet place to attend the online classes. Similarly, lack of computer knowledge makes online education more challenging for some students (Table 8). Gradual growth in the share of students who did not face any issues while attending online classes and a decline in the gap between rural and urban areas indicate improved access to digital learning in the state.

The survey also looked into the duration of online classes and its impact on the learning and well-being of students. During the round II survey, 16.9% of the students spend 4 h and above (average) on online education per day (against 30.1% who spent 2 h and 19.4% 3 h per day). Similarly, 55.1% of the students spent 5 days per week on online education (49% and 59.2%, respectively, in rural and urban areas), while 14.7% were attending classes for 6 days. The time spent on online education by students (in terms of average hours per day and average days per week) was found low in the rural areas. A large number of students (36.8%, 32.5% and 40.4%, respectively, during rounds II, III and IV) reported that online classes caused health issues like dryness/irritation of the eyes (Table 8). Households also reported headaches, back/neck pain experienced by their children due to online classes.

Students and teachers may require adequate time and exposure to adapt to the new mode of education. Many students (12.8%, 13.5% and 13.5%, respectively, during rounds II, III and IV) were unable to comprehend the teaching (Table 8). Round II reports the level of understanding of topics taught in the online classes and found that 73.8% of the students understand partially (against 18% fully understand and 8.2% unable to understand). Thus, 54.7% of the informants reported that students need additional support in this regard. As the information was reported by the key respondent (often parents), the reality can go beyond this reported figure. Issues like students surfing on the internet, and odd teaching hours also make online education more challenging as reported by many respondents during the survey (Table 8).

Educational institutions follow different methods and strategies to conduct regular classes and examinations. Reports and public discussions have highlighted the anxiety shared by students over their examinations and career aspirations. Mandatory house confinement missed classes, and examinations are associated with the stress reported among the students in Tamil Nadu (Rajendran et al. 2021). The uncertainty has adversely affected their career choices, preparation, and participation in some of the popular entrance examinations (like JEE, NEET) in the country (Roy and Roy 2021).

Round III collected information on students’ preferences for the mode of final semester/term examination and found that 68.4% of students prefer in-school examinations (72.5% and 65.8% respectively in rural and urban areas). Others preferred getting questions online and taking the examinations from home (22.6%) or full online examinations (9%). During the round IV survey, about 70.1% of the students stated that their final examinations/semesters were not conducted. Among those (29.9%) who wrote final examinations, majority received the questions online and wrote their examinations at home (Table 9).

Kalvi TV: bridging digital divides in rural Tamil Nadu

The fact that a large number of students from the rural areas have accessed classes broadcast through the Kalvi TV during the survey (Table 5) highlights the potential for strengthening Kalvi TV. Given that the majority of the households own television (94.6% and 96.3%, respectively, in rural and urban areas as reported during round III), Kalvi TV has great potential for complementing digital learning and new modes of schooling in the state. Rural areas continue to have a larger share of students accessing Kalvi TV but we notice a decline in its share across the rounds. Out of 6590 students attending schools during round IV survey, 41.4% of the students accessed Kalvi TV (respectively 45.9% and 36.6% in rural and urban areas). Technical issues related to the devices or services, lack of awareness about the schedule/telecast time, channel details and type of classes covered, etc., or poor support (related to power supply, network, battery, etc.) could have contributed to this trend. Promoting Kalvi TV among the public and capitalising on its earlier welfare interventions (like distribution of free televisions to households) can be a critical strategy to improve students’ access to online classes and digital content during the pandemic.

The results from nonlinear panel i.e. Conditional fixed-effects logistic regression, show low access to online education for urban and rural households who faced job loss during the pandemic. For urban households, an increase in job loss by a unit reduces the likelihood of online education access by 0.64 units for their children attending school or college. While the scenario marginally changes for rural households but indicates a positive relation between job loss and exclusion from online education in rural and urban contexts (Table 10). The results follow the larger pattern of digital divide and challenges in accessing online during the pandemic in different contexts. This has resulted in higher drop-out rate, especially in rural areas and among the students from marginalised groups and this digital divide could be a reason for this negative sign. The evidence from early phase of the pandemic indicates an increase in the share of children (6–14 years) out of school (from 2.5% to 4.6% during 2018 to 2021), especially in rural India (Banerji & Wadhwa 2021).

In the case of reasons for not attending online education category, both the independent variables are negatively significant. If households face technical issues and knowledge gaps for children or parents, their access to online education decreases by 1.56 times and 2.0 times, respectively. The result is consistent with findings of ASER 2022; the non-availability of gadgets such as smartphones and phones for children to use, and network or connectivity issues were the challenges faced by children. Considering the differences in the source or medium used in accessing online education during the crisis, we tried to analyse the pattern against their common characteristics. The result shows variations across the groups. If Television’s usage increases by a unit, then access to online education increases by 6.54 times where one-unit increase in usage of mobile phone leads to an increase in online education by 6.28 times. Both laptop/desktop and tablet increase access to online education by 5.49 times and 4.46 times, respectively, if their usage increases by a unit, respectively. Since many school students are studying in government schools, government-run Kalvi TV might be a reason for the most use of Television.

Our study employed the Conditional fixed-effects logistic regression model over the Random—effects logistic regression model based on the results from the Hausman specification test (Table 11). We find the P-value of the test as highly significant (0.004) and reject the hypothesis that Random-effect model is better than the Conditional fixed-effects. A lower AIC value (306.882) of the Conditional-effect logistic regression model also confirms that this model is better suited to this analysis than the Random-effect logistic model (2778.942). The finding emerging from the descriptive statistics and regression indicate the nature of the digital divide and challenges in accessing online education in the state. The results also indicate that students from rural areas and socially marginalised groups (socially or economically) have lower access to online education. Many parents engaged in vulnerable livelihoods and casual wage work lost their jobs during the pandemic and unable to meet the additional expense for ensuring their children’s access to online education.

While government interventions have improved access to online education and digital content over time, many students especially in rural areas still do not have access to online education in the state. This digital divide and differences in access to digital infrastructure need to be addressed by systematic and focused interventions. Government should initiate viable and effective policies to ensure that no student is left without access to online classes and digital content. Government interventions like ‘Aakash tablet project’ or other state-level schemes for distributing laptops and tablets for students play a critical role in bridging the digital divide, especially among students from marginalised groups (Bapna et al.2020). Focusing on specific socio-economic groups and their levels of education could be more effective.

Government can promote Kalvi TV through regular updates on its schedule and content to the larger public and repeat the telecast for better coverage. This would make Kalvi TV more accessible and reliable to act as a viable medium of education, particularly in rural areas which had reported lower access to online education. Similarly, tracking the potential rise in the drop-out at all levels of schooling (especially among the marginalised groups) and encouraging their re-enrolment and ensuring proactive support to students with special needs, etc. may help the government to ensure that no child is left behind in accessing education in the state. Prioritising the existing measures like distribution of free laptops, mobile data, etc. to the students at a higher level will be more effective. The state government introduced a provision of 2 GB of free mobile data to the students from the government and government-aided colleges during the pandemic. While this has great potential in addressing the issue faced by many students, a low usage was reported in round III. Making the requirements and guidelines more flexible (instead of centralised implementation) could have encouraged more students to access such schemes.

Recent developments from the new waves of the pandemic have led to more uncertainties around education and extend the pressure of new mode of education. This is likely to have far-reaching implications for quality learning, teaching, and career transition in the coming years. Ensuring free access to the educational resources in any context, building a sustainable approach to teachers’ professional development and adapting teaching to the new context (through crash-course on remote learning pedagogies and blended learning approach), protecting child nutrition and maximizing contact time, etc. are crucial in minimising the disruption caused by the pandemic (Gustafsson 2021; Oyedotun 2020; Saboowala and Mishra 2020; UNESCO 2021a, b, c). The government should use this opportunity to devise effective policies to make the educational system more resilient and inclusive by addressing the digital divide and inequalities that exist at all levels and explore the possibilities of digital transformation to improve the quality of teaching and learning in the state.