NUMBER 12 JANUARY 2021
the digital literacy of its participants with a set of tasks requiring the students to use their actual skills. They are then assigned numeric scores reflecting their proficiency levels. Following relevant studies, high socioeconomic status is measured employing two different proxies: (a)whether at least oneof theparents has a tertiarydegree, and (b)whetherat least one of the parents is employed in a professional or specialist occupation (i.e., one-digit ISCO-08 1, 2, and 3 occupations 8 ). Figure 2 shows socioeconomic gaps in CIL and CT test scores for several countries. In this figure, the dots correspond to a difference in average test scores between a high-status group (i.e., students whose parents have a tertiary degree or students whose parents work in a professional or specialist occupation) and a low-status group (i.e., students whose parents have below tertiary education or students whose parents do not work in professional or specialist occupations). The higher the dots, the larger is the magnitude of the socioeconomic gap in test scores. Red dots correspond to gaps on the CIL test, whereas blue dots refer to gaps in the CT test. One should also note that CT was an optional component of ICILS 2018, which was taken up by 9 of the 14 ICILS 2018 countries. In general, the size of both gaps varies significantly across countries and across both proxies for family background. For instance, the CIL test score gap based on parental education is equal to 30 points in Finland and nearly 60 points in the Republic of Korea (hereafter Korea, for ease of reading) or Luxembourg. 9 Similarly, the corresponding CT test score gap is less than 40 points in Denmark and Finland, but it exceeds 60 points in Luxembourg and in the United States. Furthermore, with respect to parental occupation, the gaps in CIL test scores range from 20 points in Korea to more than 50 points in Luxembourg. In a similar vein, the gaps in CT test scores vary from a little over 20 points in Korea to more than 60 points in Luxembourg.
When addressing the importance of digital skills in the labor market, the conventional approach is to look at digital competences as an encompassing concept. However, a growing literature on routinization and job polarization shows that it is increasingly relevant to separate abstract/ cognitive skills from routine (and manual) skills. Labor market returns are very high for abstract and cognitive skills, whereas routine skills are less and less in demand (Autor and Dorn 2013; Goos et al. 2014; Spitz-Oener 2006). This consideration underscores the importance of distinguishing between general digital competences on the one hand, and computational thinking (CT) on the other. The former refers to one’s ability to use computers to search for, acquire, and process information, to create content, and to communicate with others 4 (Fraillon et al. 2020, Chap. 2). CT instead refers to one’s ability to identify, test, and implement possible algorithmic solutions to the problem at hand and to analogous problems that might arise in a new context or situation 5 (Fraillon et al. 2020, Chap. 3). CT is consistently regarded as one of the most important competences that individuals need to possess in order to be able to cope with future changes in the labor market (Czaja and Urbaniec 2019; Rakowska and Cichorzewska 2016; Slavinskis et al. 2015).
DATA AND MAIN RESULTS
The aim of this brief is to look at how the socioeconomic gap in general digital competences differs from the corresponding one in CT. For this purpose, data from the International Association for the Evaluation of Educational Achievement (IEA) recently released 2018 wave of the International Computer and Information Literacy Study (ICILS) 6 are used. ICILS 2018 tests grade 8 students from various countries in two areas, (a) computer and information literacy (CIL) and (b) CT 7 , using a task-based approach. That is, besides a set of self-assessment questions asking students about how often they use ICT and for what purposes, ICILS measures
4. With reference to the DIGCOMP framework, it is close to the competence areas 1, 2, and 3. 5. This is close to the DIGCOMP competence area 5. 6. ICILS is an international comparative survey targeted at grade 8 students (or grade 9 in some countries) and aiming to measure their ICT skills. 7. This is the first time that students participating in ICILS were tested in CT. 8. ISCO = International Standard Classification of Occupations; see https:/ www.ilo.org/public/english/bureau/stat/isco/ 9. Both CIL and CT tests were scaled to have a mean of 500 points and a standard deviation of 100 points. Thus, a gap of 50 points on a given test corresponds to a difference of half the standard deviation. Note, however, that individual scores on these tests cannot be directly compared, as the scores were scaled with respect to different populations. That is, a student’s score on the CT test reflects that student’s position relative to the mean of the nine countries (equally weighted) which participated in the test in 2018. Similarly, a student’s score on the CIL test reflects her position relative to the mean of all the countries (equally weighted) which participated in the CIL test. Because the underlying populations are different, individual test scores cannot be compared directly across the tests.
3
Made with FlippingBook - Online catalogs