1. Introduction

The lockdowns put in place to control the spread of the COVID-19 virus during the period March-June 2020 represented a sudden, dramatic and unexpected disruption to all components of social and economic life. The combination of the closure of schools and the broader lockdown/confinement measures such as the restrictions on movement and the administrative closure of many businesses and other organisations affected the life of children and their families. It transformed the educational experience of children over a period of 2-3 months and, sometimes, more.

School systems had to rapidly improvise to ensure some continuity in the education of children and adapt their teaching methods to a situation in which, in the space of a day, the setting in which education took place moved from the school to the home for most children and the mode of instruction shifted from face-to-face contact between pupils and their teachers/instructors to some form of remote or distance learning, often supervised by parents.

The home and social environment of children was also affected in many ways, which, in their turn, affected the educational experience of children. In-person contact with people other than household members was severely restricted. The working arrangements of many parents changed, often dramatically. Many were laid-off on a temporary basis or had to work from home. In addition, parents faced a range of stresses associated with the pandemic: concerns for friends, relatives and family who were sick, fear that they themselves, their children, members of their wider family and friends would catch the virus, financial worries associated with job losses, business closures and temporary lay-offs, frustrations about reduced interactions with friends, relatives and family, reduced freedom of movement, etc.

This report offers an initial overview of the available information regarding the circumstances, nature and outcomes of the education of schoolchildren during the first wave of COVID-19 lockdowns of March-April 2020. Its purpose is primarily descriptive: it presents information from high quality quantitative studies on the experience of learning during this period in order to ground the examination and discussion of these issues in empirical examples.

Information on the response of education systems to the COVID-19 pandemic in different countries is available in a number of publications [see OECD (2021[1]; 2020[2]; UNESCO, UNICEF and the World Bank, 2020[3])]. The approach taken here complements the picture described by these studies. Its focus is the circumstances and experience of the education of children rather than the actions taken and policies implemented by educational systems at the various stages of the crisis.

Information is presented on three interrelated topics. Chapter 2 covers the nature of the educational experience during the period of lockdowns and school closures. Chapter 3 provides information on the home environment (the setting in which education took place for the vast majority of school age children) and Chapter 4 presents information on the mental health and learning outcomes for children during this period. Finally, Chapter 5 offers a summary and some concluding comments.

Much of the information available regarding the activities, behaviour and opinions of the students, their parents and teachers (together with that of the general population) during the first wave of lockdowns comes from surveys that use non-probability sampling designs, most often volunteer samples1 or quota samples.2 The reasons for this are understandable: a perceived need to gain information quickly (Huber and Helm, 2020[4]) and the absence of easily accessible sample frames covering the target populations of interest. Such approaches do not provide a secure basis for making valid inferences about the populations and groups of interest from the responses collected.3

In contrast, this report draws on a corpus of information that is restricted to studies that meet a minimum standard of statistical quality. For survey-based studies, the requirement is that they based on probability-based samples – that is, from surveys designed to be representative of clearly defined target populations (e.g. the adult population, school students, parents of school age children, etc.). The statistics are thus based on samples in which each member of the target population has a known and non-zero probability of selection. The corpus of studies meeting this condition is small: they were published in English, Finnish, French, German and Italian. In May 2021, all OECD member countries were invited to provide additional studies meeting these conditions that could cast light on the different sections highlighted in the report, whatever their language of publication. Table 4.3 and Annex A contain the details of the main studies from which information has been taken.

The principal source of the information is surveys that have collected information regarding topics related to the experience of schooling during the first wave of COVID-19 lockdowns – i.e. during the period March-June 2020. For the most part, the respondents are adults, either the parents/guardians of school age children, an adult resident in a sampled household or, where relevant, teachers. Few surveys collected information directly from children.

The data come primarily from 10 countries: Australia, Belgium (Flanders), Canada, France, Germany, Ireland, Italy, the Netherlands, the United Kingdom and the United States. This reflects the (unfortunate) reality that high quality data on the educational experience of schoolchildren and the life circumstances of these children and their families during lockdowns were collected in a small number of countries.

The studies relied on in this paper also have their limitations, but of a different nature from those based on “convenience” samples. The need to rely on telephone interviews and web-based surveys (in-person interviews being impossible) meant that coverage of the target populations was sometimes reduced (not all members of certain target populations will have Internet access or a known phone number) and response rates were often low. In some cases, sample sizes were small. In others, experimental designs were used. An example is the US Census Bureau’s Household Pulse survey which was based on a design that involved very large samples and assumed very low response rates. While all studies retained in the report were designed to be representative of some population, and thus provide more reliable information than “convenience” surveys, the error associated with lower than usual response rates should be taken into account when interpreting results.


[4] Huber, S. and C. Helm (2020), “COVID-19 and schooling: Evaluation, assessment and accountability in times of crises—reacting quickly to explore key issues for policy, practice and research with the school barometer”, Educational Assessment, Evaluation and Accountability, Vol. 32/2, pp. 237-270, https://doi.org/10.1007/s11092-020-09322-y.

[1] OECD (2021), The State of School Education: One Year into the COVID Pandemic, OECD Publishing, Paris, https://dx.doi.org/10.1787/201dde84-en.

[2] OECD (2020), “Schooling disrupted, schooling rethought: How the Covid-19 pandemic is changing education”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/68b11faf-en.

[3] UNESCO, UNICEF and the World Bank (2020), What have we learnt? Overview of findings from a survey of ministries of education on national responses to COVID-19, UNESCO, UNICEF, World Bank, Paris, New York, Washington DC., http://uis.unesco.org/sites/default/files/documents/national-education-responses-to-covid-19-web-final_en_0.pdf.


← 1. A common approach is the use of “participative” or open access on-line surveys in which a survey form is made available via a website for any interested person to complete. Alternatively, an invitation is sent to a contact person who, in addition to being asked to complete the survey, is asked to pass on the invitation to others (e.g. other teachers in their school).

← 2. A form of non-probability sampling in which targets are defined for the numbers of respondents with particular characteristics such as sex, age and educational attainment.

← 3. While some studies weight respondents to known population totals, these types of approach cannot compensate for response bias due to the self-selection of respondents and ensure the representativeness of samples.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2021

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.