Conference Agenda

The paper is part of national and international studies aimed at defining and assessing digital competencies, with particular attention to those of students at different school levels.

Digital competencies are at the top of the European political agenda, aiming to improve them for digital transformation. The European Skills Agenda 2020 promotes digital competencies and supports the goals of the Digital Education Action Plan for the development of a high-performance digital education system. The Digital Compass and the European Pillar of Social Rights Action Plan set targets to reach at least 80% of the population with basic digital skills and 20 million specialists in information and communication technologies by 2030. In Italy, the current legislation (National Plan for Recovery and Resilience) provides for the country to equip itself with a system of certification of digital competencies from 2025.

It is therefore necessary to define what is meant by digital skills and to measure them. In this perspective, INVALSI, the Italian National Institute for the Evaluation of the Education and Training System, is launching the DIGCOMP.MIS project to define a prototype model to attest digital competencies, applicable for spring 2025; reference are students of secondary II second grade but with the prospect of observing the evolution of digital skills from the end of secondary first grade to the end of secondary second grade.

The framework assumed by INVALSI and by this paper is DIGCOMP 2.2. (Digital Competence Framework for Citizens) developed by the European Commission to describe and assess the digital skills of citizens aged 16 and over.

From 2013 to today, DIGCOMP has found application in the context of employment, education, training, and lifelong learning; it has been adopted at the European level to build the Digital Skills Indicator and to monitor the Digital Economy and Society Index.

Specifically, this project deals with the definition of the levels of digital competence and the adequacy thresholds corresponding to the different school grades.

In large-scale educational surveys, the variables considered consist of skills, knowledge, or skills possessed at a stage of the school career or in a given age group, constructs not directly observable, but defined based on a theoretical reference framework and operationalized to administer standardized tests.

An outcome in terms of numerical score, however, is not directly informative of what students with a given score know and can do concerning the investigated domain; this is a limit for those interested in interpreting the results of a survey and obtaining information for interventions or teaching practices.

The attribution of an explicitly described level allows students, families, and teachers to have significant feedback, which can be integrated by the students into their perception of competence and useful for teachers teaching. Many national and international surveys combine a score result with a description of the corresponding level; similarly, INVALSI does.

The aim of the project will therefore be to define the type of target levels and their identification.

The paper aims to give an account of the first phase of the project, particularly the analysis of scientific literature and models tested or in use in other European contexts that allow linking of the elaboration of the model to the most authoritative and updated studies of national and international research; This, together with the reference to the DIGCOMP framework, allows the proposed model to be modular also given future comparative developments of digital competences surveys.

Social-emotional competences are crucial for children’s development, especially in middle childhood. Numerous frameworks describe social-emotional competences (Soto et al., 2019), often focus on the acquisition of social-emotional skills: i.e., social-emotional learning (SEL). SEL aims to enhance five interrelated individual competences: self-awareness, self-management, social-awareness, relationship skills, and responsible decision-making (CASEL, 2024). Promoting SEL not only helps to prevent behavioral disorders but also positively influences other areas, such as prosocial behavior, well-being, and academic skills (Durlak et al., 2022).

To assess these competences and measure the effectiveness of SEL interventions, accurate assessment instruments are needed, for both research and educational purposes (McKnown et al., 2017; Soto et al., 2019). These instruments should identify risks to social-emotional well-being and then ease the search for appropriate interventions supporting each child’s individual development (Denham et al., 2016). Especially concerning children during their early school years, developing and/or improving assessment tools in the area of social-emotional competences is essential (e.g., Abrahams et al., 2019; Halle & Darling-Churchill, 2015).

Addressing this need for accurate assessments for children (Soto et al., 2019), we developed a screening to assess social-emotional skills. In addition to the dimensions proposed in the CASEL model, we also considered four subdomains that were identified by Halle and Darling-Churchill (2016) as frequently being part of social-emotional assessments: social competence, emotional competence, behavior problems, and self-regulation. Based on these models and proposed sub-competences, we developed a digital screening taking into consideration Emotion Recognition, Prosocial Behavior, Emotion Regulation, and Social Situations. As our target group is children at an early stage of schooling (6 to 8 years), we used a digital approach via tablets and provided all questions and instructions as audios and in a written form. This enabled children to work at their own pace using headphones. Another advantage of the digital assessment is a higher motivation of the children (Blumenthal & Blumenthal, 2020).

For measuring Prosocial Behavior and Emotion Regulation, we used a 5-point Likert scale with the word-based response format rating ‘never’ to ‘very often’ to achieve better scale properties and more differentiated results than with the traditional yes-no format (Mellor & Moore, 2014). The subtest Social Situations is a situational judgement test (SJT), consisting of descriptions of challenging school scenarios (e.g., someone laughing at the child) supported by a graphical representation. The test offers four different behavioral options describing reactions in the scenario. Children rate these options on a 5-point Likert scale referring to whether or not they would react like proposed in the option. The scale is anchored in “certainly would” and “certainly would not” react like this, as used by Murano et al. (2020). This subtest requires social-cognitive information processing (Crick & Dodge, 1994), and SJT is a promising approach assessing social skills (Soto et al., 2019). The subtest Emotion Recognition is a performance measurement, therefore, very robust against attempts of faking good (Abrahams et al., 2019). Pictures of facial expressions (produced with the support of artificial intelligence) representing emotions are presented and children choose the fitting emotion out of five options.

Currently, we are piloting this assessment tool in two pilot studies with second graders. First, using a participatory approach, we ask children how they liked the screening and where they experienced challenges. The collected feedback is then used to adapt the instrument before implementing it with a larger group. Our presentation will focus on the following two research questions:

• Is the assessment user-friendly and intuitive for Grade 2 students in individual and group settings?
• Do the items of the introduced screening meet criteria like difficulty and discriminatory power?
• Does the proposed structure fit the findings in the factor analyses?