The World Bank’s Indonesia Covid-19 Observatory is a Indonesia Brief multi-GP partnership that aims to generate usable and Covid–19 Observatory №8 near-real-time insights on the impact of the pandemic to inform the Government of Indonesia’s policy response to the crisis. PL E ASE D IRECT INQUIRIES TO RIRIN SALWA PURNAM ASARI – RPURN AMASARI@WORLDBAN K .ORG 25 January 2021 Page 1/2 RAB IA AL I – RALI1@WORLDBAN K .ORG Education services Updates from the fourth round of during the COVID-19 T HE WORLD B ANK (WB ) HIG H- F REQ UE NCY M ONITORING OF COVID- 19 IM PACTS ON HOUSE HOLDS (HIF Y) pandemic 0 3 – 1 5 NOVEM BER 20 20 T HE WB COVID- 19 ONLINE SURVEY 27 NOVEM BER – 20 DECEM BER 20 20 → This brief presents an update on delivery of education services in Indonesia during the COVID-19 pandemic using findings from the fourth rounds of (i) the HiFy Survey, a nationally representative panel survey of about 4,000 households across 40 districts and 35 cities in 27 provinces of Indonesia; and (ii) the Online Survey, dis- tributed through Facebook, that collected data from 2,421 respondents. Ex-post survey weights were used to make the samples nationally representative. W hile most schools across Indone- Figure 1 Share of students for whom sia remained closed for face-to- school was open for face-to-face face learning, the vast majority of learning students were engaged in distance learning and a small minority had dropped out. In the November round of the HiFy survey, schools were open for face-to-face learning in the week prior to the survey for just over a third of those enrolled in primary and secondary schools1 (figure 1). Most of these students attended school for only a few days a week, with only 10 percent attending full time. Further, 93 percent of all students continued to engage in distance learning in the week preceding the survey.1 Most Indonesian schools thus adapted to the pandemic by implementing some form of distance learning, and 25 percent of students were covered by a “hybrid” learning model that combined face-to-face and distance learning. Note: 90% Confidence Interval Finally, a small minority had dropped out: 2 percent of children aged 5 – 18 years who had been enrolled in school up to March Figure 2 Constraints faced 2020 were no longer enrolled in November.2 27 percent of these in distance learning had delayed further education temporarily, while nearly one in five did not wish to continue schooling. Just under a third were Learned Offline Learned Online out of school due to lack of money for tuition. 1 Even 84 percent among those whose schools were still closed Distance learners faced differing challenges, while the time for face-to-face learning they spent on learning varied significantly across regions and had communicated with their teachers, the vast socioeconomic groups. 71 percent of students engaged in distance majority via WhatsApp. learning faced at least one challenge while using these services. 2 The numerator for the The most frequently reported constraints were limited internet dropout rate includes those who were no lon- access (no internet access was a constraint for very few), difficul- ger enrolled at the time of the survey but did not ties in focusing/concentrating, lack of devices needed to use the graduate in June 2020 from any level of school- services, and lack of adult guidance (figure 2). In this context, it ing, as well as those who is not surprising that the average number of hours spent daily on graduated from kinder- garten, primary or junior distance learning activities varied widely from nearly 3.5 hours secondary school. Those who graduated from per day in DKI Jakarta to 2.3 and 2.2 hours elsewhere on Java and senior high school but were not currently en- outside Java respectively. Primary school students, those in rural rolled were excluded, as areas, and those in the bottom 40 percent spent on average less they had completed the 12 years of compulsory time daily on distance learning than others (figure 3). education. Indonesia Covid–19 Observatory 25 January 2021 Page 2/2 The share of students using online/mobile modalities for dis- Figure 3 Average number of hours tance learning did not change between late May and November, spent per day on distance learning hovering at just over 40 percent of distance learners. In August activities 2020, the Ministry of Education and Culture allocated IDR 7.2 tril- lion in social assistance in the form of internet data packages to support use of online/mobile learning modalities among students and teachers conducting long-distance learning over the course of four months.3 By November, the data packages had reached 51 percent of students, of whom 82 percent reported successfully us- ing them. However, the share of distance learners who used mo- bile learning applications or participated in online schooling in the week prior to the survey did not change significantly and hov- ered around 40 percent between late May/early June and Novem- ber. At the same time, only half of those who received the pack- ages were using an online/mobile modality for distance learning in November, and those who received the packages spent on av- erage 30 minutes more on distance learning per day than those who didn’t. One hypothesis to be examined in further work is that while the program may not have impacted adoption of online/mo- bile learning among students who weren’t already using it, it may have intensified use among existing users.4 The Indonesian government needs to ensure implementation of infection control measures in schools, sensitizing the public on measures taken and their results in a transparent and time- ly manner. While the rise of COVID-19 cases remained largely Note: 90% Confidence Interval uncontrolled across Indonesia, public concern about infection in schools was very high. 74 percent of online survey respondents >40% Figure 4 Share of distance learners were very worried that children in their households would con- using online/mobile technologies, tract the coronavirus while at school; another 22 percent were November 2020 somewhat worried. Not surprisingly, thus, over half ranked part- time face-to-face learning combined with distance learning as their preferred schooling method until the pandemic was over; of students a quarter preferred full-time distance learning, and only 13 per- are using cent preferred full-time face-to-face learning. In this light, while online/mobile reopening of schools is a positive sign, it is imperative for GOI modalities to ensure implementation of rigorous infection prevention mea- for distance sures in schools and sensitization of the general public on mea- learning sures taken and their results. During 2021, most schools will likely need to offer part-time face-to-face learning combined with distance learning; the scope of impacts on learning remains hard to quantify. Distance learning is not a substitute for face-to-face learning, in terms of the time students are currently spending on it and given that it is a completely new teaching modality for teachers and students. As such, it is likely that it will only partially mitigate the learning “lost” due to school closures, the extent of which will be revealed once national learning assessments are reinstituted. Similarly, the extent to which inequalities in learning outcomes that predate the pandemic will be exacerbated in its wake is also not yet clear. In the November round of the HiFy survey, schools were likelier to be open for face-to-face learning for students who have historical- Note: 90% Confidence Interval ly lagged behind (those in rural areas or in the bottom 40 percent) but the differences were not statistically significant. At the same time, those who were wealthier or in urban areas like DKI Jakar- ta – groups also benefiting disproportionately from GOI’s phone/ data package subsidy – were better positioned to make up for the losses through distance learning enabled by digital technologies (figure 4). Provision of distance learning needs to be customized to the needs of rural populations, for whom adoption of digital tech- nologies may not be feasible in the short run. The share of dis- tance learners in rural areas using online/mobile technologies stood stagnant at 36 percent between late May and November. In contrast, 51 percent of these students had been learning at home by watching an educational TV program in late May. It is unlikely 3 https://www.the- that the long-term issues (low and unaffordable connectivity and jakartapost.com/ news/2020/08/27/minis- poor digital skills) that have kept rural populations from adopt- try-provides-rp-7-2-tril- ing digital technologies can be overcome quickly in these places lion-in-phone-credit-da- ta-packages-to-support- to rapidly scale up usage of digital solutions to mitigate learning distance-learning.html losses. It is thus critical for GOI to focus in the short run on get- 4 This also seems con- ting high-quality learning content to these students through the sistent with the finding that the better off were distance learning modalities most easily accessible by them, e.g. more likely to receive the data packages i.e. those TV programming, rather than relying primarily on online/mobile in Java, in urban areas, and not in the bottom solutions that remain inaccessible to most in these areas over 10 40 percent, groups that months into the pandemic. were already more likely to participate in online learning back in late May.