Survey data on the impact of COVID-19 on parental engagement across 23 countries

This data article describes the dataset of the International COVID-19 Impact on Parental Engagement Study (ICIPES). ICIPES is a collaborative effort of more than 20 institutions to investigate the ways in which, parents and caregivers built capacity engaged with children's learning during the period of social distancing arising from global COVID-19 pandemic. A series of data were collected using an online survey conducted in 23 countries and had a total sample of 4,658 parents/caregivers. The description of the data contained in this article is divided into two main parts. The first part is a descriptive analysis of all the items included in the survey and was performed using tables and figures. The second part refers to the construction of scales. Three scales were constructed and included in the dataset: ‘parental acceptance and confidence in the use of technology’, ‘parental engagement in children's learning’ and ‘socioeconomic status’. The scales were created using Confirmatory Factor Analysis (CFA) and Multi-Group Confirmatory Analysis (MG-CFA) and were adopted to evaluate their cross-cultural comparability (i.e., measurement invariance) across countries and within sub-groups. This dataset will be relevant for researchers in different fields, particularly for those interested in international comparative education.


Description of data collection
A series of data were collected via online distributed questionnaires in all participating countries (23 countries). The questionnaire was created in an international English version and subsequently translated and adapted to the official languages and localisms of the participating countries. After the first translation, questionnaires were back-translated into English, the equivalence of the questionnaire in the target languages was evaluated and relevant adjustments made. The questionnaires were then distributed through the networks of the participating institutions in each country. The ICIPES target population was parents/caregivers of children between 6 and 16 years old, living with their child and between grade 1 and 13 that represents between 1 and 13 years of schooling, counting from the beginning of Level

Value of the Data
• The database offers first hand valuable information about parental engagement, school support for parents and children, home-schooling and family life balance and parental acceptance and confidence in the use of technology from 23 countries around the world. • The international database provides a rich environment for examining how parents and caregivers relate to children's learning in this period of social distancing caused by the global COVID-19 outbreak. • The international database offers data comparable on parental practices during the lockdown across 23 countries and five regions (America, South-Asia, East-Asia, Africa and Europe), allowing investigations on aspects of specific relevance in each of these geographic regions. • The international dataset contains scales such as parental engagement, parental acceptance and confidence in the use of technology scale and family socioeconomic status, which allow testing hypothesis about the interactions of these and other variables across and within the participating countries. • The international database involves considerable information for the researchers, analysts, policymakers and education stakeholders to take steps and measures to improve the quality of parental engagement in children's education during and after the lockdown period.

Data Description
With the advent of the detection of the first case of COVID-19 in the late of November in China and later in the beginning of March in the other countries, an urgent governance step has been initiated by the Ministries of National Education to carry on various educational activities remotely since schools have experienced compulsory shut downs until the end of April-June, depending in which country you are in, to prevent spreading the virus across countries [17] . The pandemic has shown countless barriers that families face daily in their goal of educating their children. It is a unique historical opportunity for researchers and policymakers to understand all the lessons from this global emergency and work closely with parents/caregivers to support them in engaging with children's learning as they are the best partners in mitigating both short and long-term impacts of COVID-19 on children's learning.
Research connects children social and cognitive development to parents' educational practices at home [9] . Mostly, to parental practices that have the potential to provide learning experiences for children, such as: reading to children, using complex language, responsiveness and warmth in interactions and conversations, playing with numbers, painting and drawing, learning about numbers and letters and going to the library [5 , 4 , 12] .
In the current pandemic, parents have spent more time with their children. Moreover, the primary responsibility for enforcing and maintaining young people's educational engagement lies with them. While there is a substantial body of literature which explores parental engagement in education (e.g., [2] ), the uniqueness of the current circumstances demands more investigation of how parents are building capacity at home, what activities are they developing with their children, what kind of support they have received from the schools, and how parents have shaped and built their roles and IT skills.
The data provided in this study allows researchers to embark on investigations to the above and other related areas and questions.

Identification variables in the dataset
All ICIPES 2020 data files contain several identification variables that provide information to identify the participants' important characteristics. URN This variable identifies the specific questionnaire that was administered to each parent. This number was automatically provided by the Online Surveys tool.
In this study, the online survey was conducted with semi-structured questionnaires. Online survey is one of the best ways to reduce the cost when conducting a study, but it is also an effective way to get real data from the online population [13] . A total of 4658 respondents (parents) answered questionnaires from the participating countries: Cameroon, Ethiopia, Ghana, Tanzania, China (i.e., Mainland, Hong Kong, and Macao), Japan, Belgium, Italy, Spain, Turkey, United Kingdom, India, Pakistan, Sri Lanka, Chile, Colombia, Costa Rica, El Salvador, Honduras, Mexico, Peru, Uruguay, the United States. Later, the countries split into five regions: Africa, East Asia, Europe, South Asia, America. Tables 1 to 12 present some characteristics information about countries, regions, and respondents participating in this study.
The following section provides information about the procedure followed to construct three scales in ICIPES 2020. 4658 * Concerns about the extremely low response rates (less than 10%) for the parents surveys led to a decision not to include the corresponding data in the international database.     Social cognitive learning theory [3] and the theory of acceptance and use of technology [14-16 , 1] formed the conceptual framework for these scales. The social cognitive learning theory provides a socially appropriate framework for understanding how parents learn to deal with technology at home from their observations and interactions with other parents, teachers and their children. The second explains the factors associated with parental acceptance and confidence in the use of technology.
Before constructing the three scales, we constructed and implemented normalised weights (also known as senate weights) (SENWT in the dataset) to make sure that when constructing these three scales, all countries are represented equally regardless of their sample sizes. SENWT can also be used when analysing the pooled sample (all countries) to ensure the equal contribution of each country to the results.
Q5 What do you do in your main job? (e.g., teach high school students, help the cook prepare meals in a restaurant, manage a sales team). This was an open question that was recorded into an ordinal variable following the list of occupations described in the one-digit ISCO (International Standard Classification of Occupations).
Q7 In a normal month, what is your total household income? This variable was recorded by grouping the income level reported in deciles of income within each country.
Q13N is composed of How many usable devices are there in the house? (Smartphones, tablets or iPads, laptops, desktops).
Q14 How many computers per child have you got at home?

Parental acceptance and confidence in the use of technology
Parental engagement scale was constructed as a second-order construct, with constructs measuring the parents' level of parental acceptance and confidence in the use of technology as 'tools', 'for social purposes' and 'self-perceived capacity'. The items asked parents about the frequency with which they carry out different activities using technology (response options: Always, Often, Occasionally, Rarely Never), and how confident they felt carrying out these activities (response options: Not at all confident, Slightly confident, Moderately confident, Quite confident, Extremely confident).
Parental acceptance and confidence in the use of technology = tool + social + capacity.

Confirmatory factor analysis (CFA)
Confirmatory Factor Analysis (CFA) was used to estimate the model for the three scales and for each country using maximum likelihood (ML). Missing data was handled with listwise deletion. Model fit was evaluated using the Comparative Fit Index (CFI) and the Tucker-Lewis Index (TLI) as the goodness of fit statistics, and the root-mean-squared error of approximation (RM-SEA) and the standardized root mean squared residual (SRMR) as residual fit statistics. Acceptable model fit was guided by the cut-offs (CFI > 0.90; TLI > 0.90; RMSEA < 0.10; and SRMR < 0.08) as suggested by [8] .
Internal Consistency After constructing three scales, in order to evaluate reliability (internal consistency), we used Cronbach's alpha coefficient [6] .
Multi-Group Confirmatory Factor Analysis (MG-CFA) In order to evaluate the extent to which the scales can be validly compared across countries and geographical areas, we ran Multi-Group Confirmatory Factor Analysis (MG-CFA) first for the pooled sample including all participating countries, and later for countries within sub-groups (America, South Asia, East Asia, Africa and Europe) [10] . Here, we adopted the same strategy as [11] and [7] to conduct analysis and to interpret the results (for more information about procedure see these two papers [11] and [7] ).

Important information for potential users
The following tables include important information for potential users to be able to interpret the scales correctly.

MG-CFA result for parental engagement scale
Tables 15-20

MG-CFA result for socioeconomic status scale
Tables 23-28

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.