SMS GIRL DATA INSIGHTS: How has COVID-19 affected support for girls’ education in Punjab, Pakistan? April 2021 Amer Hasan, Koen Geven, and Ayesha Tahir SMS GIRL: DATA INSIGHTS | 2 KEY TAKE-AWAYS This brief presents initial findings from  OVID-19 has hurt households’ ability to earn. Over a third 1. C of households report that they had lost a substantial share 36% REPORT LOSS an ongoing phone survey of families in of their income. OF INCOME Punjab, Pakistan designed to assess  hildren may not return to school. Of female students 2. C what is happening to girls’ elementary who were enrolled in schools pre-COVID, 2.5 percent in our school education during COVID-19. The data used in this brief describe sample may not return to school after schools reopen. An even greater percentage of boys (4.9 percent) may not return 3.5% OF CHILDREN to school. WILL NOT RETURN the experiences of 5,898 families in Punjab between August and October  nowledge of available options for remote learning is 3. K 2020. Data have been weighted to very low. Less than a third of the families we spoke to had heard of available opportunities for remote learning. make the sample representative of all schools in Punjab. Even fewer families had heard of options provided by the government since the start of the pandemic: Teleschool or Taleem Ghar. Even many parents who said they knew of 82% DO NOT KNOW OF available options reported that their children did not take TELESCHOOL OR advantage of opportunities for remote learning. TALEEM GHAR Parents report that both boys and girls spend almost three 4.  BOYS & GIRLS hours on average on academic learning. Girls’ parents are SPENT ABOUT more likely to report that their children are spending time on 3 HOURS ON household chores than boys’ parents. In contrast, boys’ parents ACADEMIC report almost twice as much time spent by their children on LEARNING This brief provides information from an on-going work outside the household compared to girls’ parents. survey. Further data is being collected and analyzed. Subsequent briefs will provide updates  hile parents report that they have few concerns about 5. W on these families as we learn more about their school reopening, they expressed lower expectations about experiences. Unless otherwise noted, statistics their children’s education. A quarter of households said are based on the full sample of households that their expectations for the highest grade their child will PARENTS’ contacted—90% of which are families with complete have fallen. Overwhelmingly parents cited the fear EXPECTATIONS girls in grades 5–7 before the pandemic. that their children may not be able to catch up on the learning FOR THEIR CHILD’S Statistics are weighted to make the sample they have missed as a result of school closures. SCHOOLING HAVE representative of all schools in Punjab and to DECREASED allow comparisons between boys and girls. SMS GIRL: DATA INSIGHTS | 3 INTRODUCTION Policymakers are concerned about the effect Figure 1: A Timeline of School Closures and the SMS GIRL Evaluation of school lockdowns on girls’ enrollment and well-being. This concern is particularly salient SMS GIRL Evaluation Milestones in Pakistan where there is a high out-of-school population and children seldom return to TV for remote Baseline SMS GIRL school once they drop out. learning survey intervention starts starts starts The data reported in this brief were collected as part of an effort to understand how 02 21 26 COVID-19 and the resulting school closures are affecting families and how they are Mar Apr May Jun–Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr responding. 2020 2020 2020 2020 2020 2020 2020 2020 2020 2021 2021 2021 2021 In Punjab, schools closed on March 15, 2020— just two weeks after the 2020 school year 01 15 01 09 15 26 18 01 19 began (figure 1). This coincided with the start Start of Total Countrywide School School high primary School of exponential growth in COVID-19 cases school lockdown lockdown reopening closures schools & middle closures ends schools across the country. As part of its support to year (9 districts) students, the government of Punjab launched School School closures reopenings the Taleem Ghar program on April 2, 2020. On April 13, this was complemented by an initiative of the Federal government called School Closures Teleschool. These initiatives provide students in grades 1–8 with lectures and have been rolled out on television and through online means including an app for Android-based devices, a website, and via YouTube. SMS GIRL: DATA INSIGHTS | 4 THE STUDY SETTING This brief presents results from a phone survey Figure 2: Phone Survey Respondents carried out between August 21 and October 27, 2020.1 The survey reached 5,898 families— predominantly those with girls ages 10–14 enrolled in grades 5–7. Approximately 10 percent (536 Phone Survey families) had boys in those grades (figure 2). The survey respondents are from 6 districts in Punjab.2 The students about whom we present information below are predominantly enrolled in public, private and Punjab Education Foundation schools. This phone survey is the first in a series of such Public schools surveys planned to evaluate whether phone- based messages can help keep girls learning during school closures and whether information 6 districts Private schools shared with parents via these messages might in Punjab encourage them to re-enroll their girls when schools reopen. Punjab Education Foundation schools Phone surveys were necessitated by the need to adhere to the government’s Standard Operating Procedures (SOPs) for student safety. Phone- based messages were chosen as the medium 5,898 families for this intervention given the high feasibility of being able to reach our key respondents and low cost of doing so (approximately PKR 0.25 per 10% boys message sent).3 1 The first round of the survey took several months to complete. 1. Can phone-based messages help keep girls learning during school closures? This reflects the time needed to obtain all phone numbers to call  an information shared with parents via these messages encourage them to 2. C families, as well as low initial response rates. re-enroll their girls when schools reopen? 2 Attock, Faisalabad, Lahore, Muzaffargarh, Rahim Yar Khan and Sargodha. Weights are used to adjust for survey design. 3 This is equivalent to US$0.0015 per message at prevailing ex- change rates. SMS GIRL: DATA INSIGHTS | 5 5 KEY FINDINGS How did COVID-19 affect 1 households between August SOMEONE LOSS OF and October? Less than 1 percent of households in this SICK/DEAD INCOME < 1% 36% sample report someone in the household having gotten sick from COVID-19 or having died. Thirty-six percent of parents in the sample reported losing a substantial share of income. This suggests that COVID-19 had predominantly OF HOUSEHOLDS OF PARENTS manifested itself as an income shock among these households by the time the survey was conducted.4 Figure 3: Few Parents Know of Available Options for Remote Learning*— Do households have the means Even If They Own a TV or Have Access to the Internet 2 and knowledge to access remote learning opportunities? While two-thirds of the households have access 31% to TV or internet, most parents do not know about available options for remote learning. When asked specifically about Teleschool or Taleem Ghar, only 69% WITH TV OR WITHOUT TV OR INTERNET 22 percent of those with a TV reported that they INTERNET Of these, ONLY 10% were familiar with these programs (figure 3). On average, few families reported that their children have knowledge specifically of Taleem take advantage of remote learning. However, Of these, ONLY 22% Ghar and/or Teleschool among those who knew of available remote have knowledge learning opportunities, take up was higher. specifically of Taleem Ghar and/or Teleschool No knowledge of any remote learning programs 4 This is in line with evidence in other surveys such as those con- Knowledge of any remote learning programs ducted by the Center for Economic Research in Pakistan (CERP) Knowledge of Taleem Ghar and/or Teleschool and the Center for Global Development (CGD). *Note: Remote Learning Programs include Taleem Ghar, Teleschool, any learning programs on YouTube/ Whatsapp/Zoom, Khan Academy, Radio-My Best Friend (UNESCO)/Listen to Learn. SMS GIRL: DATA INSIGHTS | 6 Will households send Figure 4: The Likelihood That Children Will Return to School Varies by Age— 3 their children back to for Both Boys and Girls school? 0.20 Of female students who were enrolled in schools pre-COVID, 2.5 percent in our sample Boys may not return to school after schools reopen. An even greater percentage of boys (4.9 percent) in our sample who were previously 0.15 Girls enrolled may not return to schools. When we inquire about the siblings of these children, the Pr(not returning to school = 1) numbers are higher for older children (7 percent on average for those aged 13–17) than for younger children (2 percent on average those 0.10 aged 5–12) (figure 4). Taken together, it seems that a substantial share of students may be lost from the education system, particularly in adolescence. 0.05 0.00 -0.05 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Age Note: Areas denote 95% Confidence Intervals SMS GIRL: DATA INSIGHTS | 7 How are children spending Figure 5: The Average Number of Minutes Girls and Boys Spend on 4 their time while schools Activities at Home Varies—Sometimes in Unexpected Ways are closed? 175 We asked respondents to report how children Academic learning 187 in their household spend their time on a range of activities while schools are closed. Parents 148 report that both boys and girls spend almost Non-academic learning 109 three hours on average on academic learning- with girls’ parents reporting about twelve 151 Time spent on activity minutes more on average. The findings also Leisure 122 suggest that time use during the lockdowns is somewhat gendered (figure 5). Girls’ families 148 report that their children spend about an hour Work outside household and twenty minutes per day on household 75 chores, while boys’ parents report that their 105 children spend an hour and ten minutes on Family care chores. While girls’ and boys’ families report 103 that their children spend roughly equal amounts Unpaid work 108 of time on family care (looking after siblings or within household Boys elders)—there is a distinct gendered pattern. 68 Girls’ parents are more likely to report that Girls their children are spending time looking after 67 Household chores siblings, while boys’ parents are more likely 79 to report that their children are spending time 0 50 100 150 200 looking after elders. Boys’ parents are likely to report that their children spend more time Mean (minutes) on leisure activities than girls’ parents do— approximately two and a half hours compared to Note: Academic learning includes time spent on remote learning, self-study and time with tutor; leisure two hours. The families of boys report that their includes time spent watching television and playing; unpaid work inside the home includes making food, children do two and a half hours of work outside toys, or other materials that are sold by members of the family; work outside the house includes paid the household compared to those of girls who and unpaid work such as farm work, livestock, etc.; non-academic learning includes leisure reading, religious learning/activities and learning new skills; family care includes child care and elderly care; and report their children spending half as much time household chores include cooking and cleaning. on this—about one hour and fifteen minutes. SMS GIRL: DATA INSIGHTS | 8 What concerns do parents 5 have about sending their children back to school and 42% & 21% PARENTS REPORT how have these impacted their THEIR EXPECTATIONS expectations? FOR THE HIGHEST GRADE THEIR CHILD Eighty-eight percent of parents had no concern WILL COMPLETE OF PARENTS OF PARENTS about schools reopening. However, parents of HAVE FALLEN OF BOYS OF GIRLS girls were slightly more likely to report being concerned about the health of their child (1.7 percent) compared to parents of boys Figure 6: Overwhelmingly Parents Think Children Won’t Be Able to (0.5 percent). Catch up with Learning as a Primary Reason for Why Their Expectations for the Child’s Education Have Changed Despite these low levels of concern about safety, the pandemic has drastically affected Child will not catch 49 parents’ expectations about the future of their up with learning 80 children’s education. Twenty-one percent of girls’ parents and 42 percent of boys’ parents Decrease in 25 report that their expectations for the highest household income 15 grade their child will complete have fallen as a result of school closures. Households who 26 Any other reason report having lost income during the school Reasons 8 closures are more likely to say their expectations are reduced due to a decrease in household Fewer income 11 opportunities for the child income—25 percent—compared to 15 percent 6 for households that did not lose income. Yet this Household has lost income is not the most common reason: both groups of Child will generate 2 income for the family Yes households are overwhelmingly concerned that 4 their children will not be able to catch up with No 0 learning (49 percent and 80 percent) (figure 6). Child helps out at home 1 0 10 20 30 40 50 60 70 80 Percentage SMS GIRL: DATA INSIGHTS | 9 CONCLUSION This brief presents five key findings from the first in a series of phone surveys that are being conducted to assess the impact school closures are having on girls’ education in Punjab, Pakistan. Additional rounds of phone interviews will be carried out to test the effectiveness of phone-based text messages to ensure that girls continue to learn and re-enroll in school once schools reopen. In subsequent rounds, we seek to engage households in semi-structured interviews to understand the dynamics driving some of the patterns presented here. We will also assess how the patterns presented here evolve as the crisis stretches into 2021. SMS GIRL DATA INSIGHTS This brief is part of a series produced by the team managing the SMS Girl Impact Evaluation. This includes Tazeen Fasih, Koen Geven, Amer Hasan, Naveed Hussain, Kevin MacDonald, Rabea Malik, Javaeria Qureshi, Ayesha Tahir, and Najaf Zahra. We would like to thank Rohit Chhabra and Sheena Fazili for research assistance. This impact evaluation is supported by the School Education Department (SED) and the Punjab Information Technology Board (PITB), under the Third Punjab Education Support Project (PESP-3). Funding from the Strategic Impact Evaluation Fund and the South Asia Gender Innovation Lab is gratefully acknowledged. Figures Authors’ calculations based on SMS Girl baseline survey data. Photo credits cover: Caroline Suzman/World Bank page 3: courtesy of the World Bank page 9: STARS/Kristian Buus