Measuring Digital Capital in Italy 2023 PDF

TYPE Original Research PUBLISHED 19 May 2023...

TYPE Original Research PUBLISHED 19 May 2023 DOI 10.3389/fsoc.2023.1144657 Measuring digital capital in Italy OPEN ACCESS Felice Addeo1*† , Valentina D’Auria1† , Angela Delli Paoli2† , EDITED BY Hannah Bradby, Gabriella Punziano3† , Massimo Ragnedda4† and Maria Laura Ruiu5 Uppsala University, Sweden 1 Department of Politic and Communication Sciences, University of Salerno, Fisciano, Italy, 2 Department REVIEWED BY of Humanities, Philosophy and Education, University of Salerno, Fisciano, Italy, 3 Department of Social Maria Paola Faggiano, Sciences, University of Naples–Federico II, Naples, Italy, 4 Department of Arts, Northumbria University, Sapienza University of Rome, Italy Newcastle, United Kingdom, 5 Department of Social Sciences, Northumbria University, Newcastle, Gianmaria Bottoni, United Kingdom City University of London, United Kingdom Linda Lombi, Catholic University of the Sacred Heart, Italy Introduction: This paper aims to theoretically and empirically investigate the *CORRESPONDENCE Felice Addeo concept of digital capital in the Italian context. Digital Capital can be conceived [email protected] as independent individual capital whose lack within a population can be a cause † of digital inequality. Our paper draws from recent works that have measured the These authors have contributed equally to this work Digital Capital as a combination of digital access and digital competences, and RECEIVED 14 January 2023 have tested this operational definition through an online survey on a UK sample. ACCEPTED 03 May 2023 The results of such research proved the construct validity of the operational PUBLISHED 19 May 2023 definition, thus showing that Digital Capital could be empirically measured. CITATION However, a measurement model needs to be tested and validated over time and Addeo F, D’Auria V, Delli Paoli A, Punziano G, in different socio-cultural contexts in order to be refined and strengthened, and Ragnedda M and Ruiu ML (2023) Measuring digital capital in Italy. Front. Sociol. 8:1144657. eventually disseminated on a large scale. doi: 10.3389/fsoc.2023.1144657 Method: This is the reason why this paper will show the results of a funded COPYRIGHT research project (named DigiCapItaly) carried out to test the validity of the Digital © 2023 Addeo, D’Auria, Delli Paoli, Punziano, Ragnedda and Ruiu. This is an open-access Capital measure in a different country, i.e., Italy. The data were collected with an article distributed under the terms of the online survey using a representative sample (by age, gender and geographical area) Creative Commons Attribution License (CC BY). of individuals living in Italy aged 18 years or more. The creation of a composite The use, distribution or reproduction in other forums is permitted, provided the original index to measure Digital Capital followed a two-stage Principal Component author(s) and the copyright owner(s) are Analysis approach. credited and that the original publication in this journal is cited, in accordance with accepted Results: First, the paper provides a methodological framework for facing academic practice. No use, distribution or challenges and pitfalls in operationalizing and assessing a complex concept in reproduction is permitted which does not social research. Secondly, results show that Digital Capital operational definition comply with these terms. works in Italy as well as in the UK, thus legitimizing its recognition as an independent capital. KEYWORDS digital capital, Italy, digital divide, validation, composite index 1. Introduction The growing use of digital solutions in everyday life offers multiple questions and food for thoughts. Unfortunately, the spread of technology is still not the same in every country, due to the digital divide. Following Hilbert (2015, 2016), Van Deursen and Van Dijk (2015, 2019), digital divide is meant as inequality in access to and use of new digital technologies. In these studies, online and offline inequalities are highly linked and influence each other: inequalities seem to extend also to many other aspects of individuals’ offline lives and vice versa (Wei et al., 2011; Ragnedda, 2017). Indeed, digital inequalities are not a strictly geographical concept, only linked to the evolution of broadband infrastructure (Warf, 2001; Crang et al., 2006; Riddlesden and Singleton, 2016). Instead, digital inequalities seem to be linked to other aspects, such as age (Neves et al., 2018), working position (Van Deursen and Van Dijk, 2019), income (Martin and Robinson, 2007; Jung et al., 2010), belonging to specific social groups such as the disabled, racial minority groups (Clark and Gorski, 2001, 2002; Wilson et al., 2003), and many others. Frontiers in Sociology 01 frontiersin.org Addeo et al. 10.3389/fsoc.2023.1144657 Most of the research in this area have shown strong links users’ access to the internet (Hoffman and Novak, 1999; DiMaggio between social exclusion and digital engagement, which tend to et al., 2001), on purely informational aspects (Hamelink, 2000) influence each other: those with fewer economic resources, lower or technical and IT aspects (Rojas et al., 2004). In the second social position, and less cultural capital are also affected by this type stage, some studies go beyond mere accessibility by recognizing of inequality. differences in the uses of the Internet according to digital skills However, previous studies on digital divide are very far and competencies (Peter and Valkenburg, 2006; Van Dijk, 2006). from analyzing all the implications of the phenomenon. Initial The third stage is well summarized by Ragnedda (2017, 2018), who definitions of the concept have focused on purely informational linked the concept to the tangible and intangible outcomes and the aspects (Hamelink, 2000) or technical and IT aspects (Rojas et al., benefits of using technological devices, that is exploitation of the 2004) and circumscribed it to the possession of technologies advantages of the Internet and the changes in people’s life that could (Katz and Aspden, 1997; Hoffman and Novak, 1999; DiMaggio improve their living conditions. et al., 2001). Despite the limited scope of these studies, they Thus, digital inequalities may be conceived as the consequence deserve the credit of having underlined how the unequal of the different availability and accumulation of digital resources, distribution of resources among the population underpins digital both material (such as technological devices and digital inequalities (Helsper et al., 2015). It is with the extension of the infrastructure) and immaterial (digital skills, problem solving, concept of digital divide to the use of digital devices (Hargittai, content-creation, etc.), and the different distribution of benefits 2002; Peter and Valkenburg, 2006; Van Dijk, 2006), that the that users are able to achieve. importance of users’ knowledge and technological skills begins to This recognition of the possibility that digital assets-in their be acknowledged. These research streams consider also the set of technological and capability aspects-are socially valuable and can skills and competencies developed through engagement with IT improve life chances makes it possible to conceptualize them as a as a constituent of the digital divide (Hamelink, 2000; Prieur and distinct form of capital, referred to as digital capital (Ragnedda, Savage, 2013). Within this perspective, digital divide has a dual 2018). The concept of capital here is meant in Bourdieu’s (1983) nature based on digital competencies and digital resources. terms as transcending economic aspects and involving internalized Thus, digital inequalities are conceived as the consequence of and externalized resources able to produce benefits in other arenas. the different accumulation and availability of digital resources, both Although independent, digital capital is strongly intertwined material (such as technological devices and digital infrastructure) with other types of capital (e.g., economic, social, cultural, etc.) and immaterial (digital skills, problem-solving capability, content- (Ragnedda, 2018). This reinforces the idea of a dual process (offline creation capacities, etc.). → online → offline) in which offline inequalities produce digital From this point of view, they can be understood as another inequalities, which in turn could reinforce inequalities present in form of capital–the Digital Capital-which is the object of this offline contexts (social, political, economic, personal) (Ragnedda, paper. The paper leverages on the recent work by Ragnedda et al. 2017, 2018; Ragnedda et al., 2022a). (2019), who have operationalized the concept of Digital Capital and In this sense Ragnedda (2018) defines Digital Capital as “a measured it through an online survey based on a representative set of internalized abilities and aptitudes” (known also as “digital sample of UK citizens. The results proved the construct validity of competencies”) as well as “externalized resources” (also called the operational definition, thus showing that Digital Capital could “digital technology”) “that can be historically accumulated and be empirically measured. transferred from one arena to another.” Within this perspective, By leveraging on these conceptual and empirical definitions of digital capital contributes to life opportunities enhancement by Digital Capital, the paper supports the empirical aim to explore creating a bridge between online and offline realms. Online and test the validity of the Digital Capital and measure it for the activities produce social benefits such as opportunities for first time in Italy. Specifically, it aims to validate the operational socialization, for creating weak ties and reinforcing strong ties; definition of digital capital used in the UK in Italy and to economic benefits such as opportunities in finding employment explore how it is correlated with the socio-economic and socio- and better jobs, in accessing online services, in online shopping; demographic variables in such a context. political benefits in reinforcing citizenship and participation The paper is organized as below. The next sections (The in deliberative democracy; personal benefits contributing to concept of digital capital-The research context) provide a entertainment, fitness and health; cultural benefits in enhancing theoretical overview of the concept of digital capital and the cultural engagement and cultural activities (Ragnedda et al., rational for choosing Italy as research context. Sections Research 2022b). design, research method, sample and data collection, Operational There are previous attempts to define the concept of Digital definition and measures, and Results deal with the research design, Capital (Morgan, 2010; Seale, 2012). For example, Seale (2012) the operationalization procedures and the statistical results. Finally, defines it as the technological know-how, the informal time in the last section findings are discussed and conclusions are drawn. invested in enhancing technological skills, the formal time spent in ICT education, the online social network. She is interested in how and whether digital capital promotes the inclusion 2. The concept of digital capital of disabled students. Morgan (2010) conceptualizes it as a new literacy for today’s students unconnected with print-based We can distinguish three stages in the research on digital literacies. In the majority of cases, the concept of digital capital inequalities. The first one focuses merely on the differences in is used at a firm level to indicate the set of resources of the Frontiers in Sociology 02 frontiersin.org Addeo et al. 10.3389/fsoc.2023.1144657 digital economy (Tapscott et al., 2000; Roberts and Townsend, Haight et al., 2014; Hargittai and Shaw, 2015). Thus, men 2015). are supposed to be more likely to have a higher level of However, these studies are theoretical at hearth. On the Digital Capital; contrary, Ragnedda (2018) aims to measure digital capital as a - Education: Previous studies have shown that the educational specific and independent form of capital and to construct and level positively influences the level of Digital Capital (Attewell, validate it with sociodemographic and socioeconomic variables. 2001; Clark and Gorski, 2001, 2002; Mossberger et al., 2007; Our work leverages precisely on the operational definition provided Shelley, 2009), establishing a positive relationship: users with and validated by Ragnedda et al. (2019) which articulates the higher education are more successful in online activities and concept of digital capital into two components: digital access and have better management skills (Van Deursen and Van Dijk, digital competence. 2013; White and Selwyn, 2013; Blank and Groselj, 2014). Digital access includes the digital equipment (devices used to - Place: The literature shows that urbanization influences the access the Internet), connectivity (quality of access to the internet), level of digital capital. In central areas the presence of the time spent online and the support and training in using infrastructures allows for a greater internet penetration than in the Internet. rural areas (Crang et al., 2006; Mardis, 2013; Townsend et al., Digital competence follows the competences framework defined 2013; Ashmore et al., 2015; Philip et al., 2017). This would in the European Digital Competence Framework for Citizens make geographical differences significant: people from the city (Carretero et al., 2017), so including the individual abilities ranging (more exposed to technology), are assumed to have a higher from the capability in browsing, searching, filtering and verifying level of Digital Capital than those living in peripheral areas. information to the capability of creating online content and protect privacy: information and data literacy, communication and collaboration, digital content creation, safety and problem-solving (see Figure 1). 3. The research context There are already several studies in the literature that demonstrate how sociodemographic and socioeconomic features The rationale of a study on Digital Capital in Italy is are intertwined with online dynamics such as how the Internet is consequence of the low level of digital skills and knowledge, accessed, how well it is managed, the level of engagement in ICT as testified by the Digital Economy and Society Index (DESI)1 and the breath of online activities (DiMaggio et al., 2004; Zillien and developed by the European Commission on an annual basis Hargittai, 2009; Robinson et al., 2015). The relationship between from 2014. The DESI describes the digital performance and sociodemographic and socioeconomic preconditions and online tracks the progress of each member states, with the aim to life cannot be considered linear and unidirectional. Instead, it is an help them to improve their own weaknesses. The results show interactive and continuously evolving cause-and-effect relationship significant gaps in both basic and advanced digital skills, in (Park, 2017). This circular process allows socio-demographic terms of the four dimensions of the DESI Index: Connectivity and socioeconomic preconditions to compromise every aspect of (fixed and mobile broadband coverage), Human capital (Internet online life, starting from online access (Tsatsou, 2011), to digital user skills and advanced skills), Integration of digital technology skills (Jones-Kavalier and Flannigan, 2008) and online activities (business digitization and e-commerce), Digital public services (e- (Hargittai and Hinnant, 2008; Zickuhr and Smith, 2012). government). According to the literature, there are five socio-demographic The DESI 2022 report, based on 2021 data, shows Italy among and socio-economic variables related to Digital Capital: the poorest performers. Italy ranks 18th out of 27 Member States, with a score of 49.3 out of 80. What seems to be relevant is that only - Income: As shown in several previous studies (Witte and the 46% of the 16–74 years old have at least basic digital skills (54% Mannon, 2010; Talukdar and Gauri, 2011; Mardis, 2013; in the EU) and only 23% have more than basic digital skills (26% Ragnedda and Muschert, 2013, 2015; Van Deursen et al., in the EU). Generally, the area where greater progress is required is 2017), income is supposed to affect various aspects of online the Human Capital dimension, “Internet user skills” and “advanced life, in particular the possession of technological devices. skills.” In this area Italy tend to be near the bottom of the ranking For this reason, it should positively affects the level of (Italy is ranked 25th out of 27 European countries). This shows Digital Capital; that digital inequalities are much more complex and refer to not - Age: many studies have found a negative association between only the possession of devices, but also the possibility of developing age and digital skills, engagement and activities (Lenhart et al., skills in order to benefit from them. The third dimension of the 2008; Lee et al., 2011; Dutton et al., 2013; Blank and Groselj, DESI index, Digital public services, also shows major challenges: 2014), so it is assumed that digital capital is negatively related while most Europeans engage in a wide range of online activities, to age; Italy is ranked 18th out of 27. This shows that once again the - Gender: Although in many developed countries with a high Italian position is below the European average (58.5 Italian vs. Internet penetration, gender digital divide has been reduced 67.3 European out of 80). Italy demonstrates to perform better or even bridged (Blank and Groselj, 2015;), there are many in the Integration of digital technology showing a score above the other developing countries where we can still find a gender gap in terms of frequency, intensity and type of internet use 1 For information and details refer to https://ec.europa.eu/digital-single- (Wasserman and Richmond-Abbott, 2005; Hargittai, 2010; market/en/desi. Frontiers in Sociology 03 frontiersin.org Addeo et al. 10.3389/fsoc.2023.1144657 FIGURE 1 The concept map of digital capital. Source: Authors’ own elaboration based on Ragnedda et al. (2019). European average (40.7 vs. 36.1 for the European average) and However, despite the low growth of digitalization progress in placing 8th out of 27 countries. Among the most interesting results, the last 5 years, in 2022 Italy has climbed a few places in the Italy is in a good position regarding SMEs with at least a basic level European ranking (moving from a DESI index score of 45.5 to of digital intensity (60%, well above the EU average of 55%), and the 49.3). This small comeback gives hope that the level of digitization use of electronic invoicing by enterprises (95% of Italian companies in Italy will improve over time, also thanks to scientific and compared to 32% of the European average). However, the use of academic research that could draw attention to digital issues. big data is still low (with an Italian average of 9% compared to 14% at the European level). These mostly positive results are probably related to policy aimed at the digital growth of Italian businesses. 4. Research design, research method, Whereby, the growth cannot be considered entirely natural but it sample and data collection is mainly an induced process. Indeed, such policy is included in a more comprehensive five-year plan called “Italia 2025” by the The main objective of our study is to measure digital capital Minister for Technological Innovation and Digitization aimed at following the research approach developed by Ragnedda et al. a digital transformation of the country. Moreover, one of the most (2019) and implemented in the United Kingdom, with the encouraging news about the Connectivity dimension is that Italy necessary adaptations for the Italian context. Thus, the paper aims is one of the most advantaged countries in the 4G replacement to answer the following research questions. project, in accordance with the 5G Action Plan for Europe. On the RQ1: Is it possible to implement and validate the operational other hand, the Italian position regarding mobile connectivity is definition used in the UK to measure Digital Capital, conceived as quite controversial: while mobile connectivity coverage is still one a distinct and independent capital, in the Italian context? of the worst in Europe, Italy is one of the leader countries in mobile RQ2: Does Digital Capital measured in Italy behave in the same only access (with 23% of households compared to 11% in Europe). way as in the UK? In other words, through a construct validation Results on Digital public services also seems controversial: although procedure, has the Digital Capital in Italy similar statistical below the European average (67 vs. 72%), Italy performs well in relationship with the socio-economic and the relevant socio- terms of offering digital and open data services. The low score demographic variables (age, gender, level of education, income, area seems due to the poor interaction both between users and online of residence)? services and between users and public authorities that should Our research questions required different statistical techniques encourage the use of public data and digital services. Once again, to be properly addressed. More specifically, RQ1 was addressed there is evidence that the digital culture in Italy is weak in terms of by conducting Exploratory Factor Analysis (EFA) with a two-stage the availability of services and infrastructures. Principal Component Analysis approach (Di Franco and Marradi, Frontiers in Sociology 04 frontiersin.org Addeo et al. 10.3389/fsoc.2023.1144657 2013), in order to build and validate the Digital Capital Index section collects sociographic information about the interviewee (DCI). To answer RQ2 and construct validate the DCI, bivariate (such as gender, age, educational qualification, etc.); the second analysis was carried out using five sociodemographic variables: section focuses on information about Internet use (e.g., which gender; age; education; income and occupation. Data analysis devices one usually uses, from which places one generally connects was carried out using the statistical analysis software, IBM SPSS to the Internet, whether one has enrolled in online courses, etc.); Statistics 25 R. the third section delves into the sociocultural and socioeconomic The data were collected on December 2021 through a web background of the respondent, with a set of questions about survey involving Italian people aged 18 and over. As with the UK different life domains (economic, social, political, cultural, and study, the research team opted for online-only administration of personal) (e.g., the respondent’s the degree of political interest the questionnaires, as this meets the gnoseological need to include and participation; leisure time occupation; sociocultural status of only those who (whether for age reasons or not) have a minimum birth family; types of activities carried out online, etc.); finally, level of technical competence and access. The questionnaire is the fourth section deals with life satisfaction (with a focus on composed by four sections with a total of fifty questions. The first economic conditions). The average time required to complete the questionnaire was 25 min. The survey was pre-tested on 20 Internet users in two rounds. Based on the feedback, changes were made. TABLE 1 Sample-population comparison. The sample was built by extracting respondents from an online panel provided by Toluna, a professional organization for market Sample Population research. The web survey collected 1,100 full responses with a Count % % response rate of approximately 8% of contacted people. The sample Gender Male 534 48.5 48.3 has been built to be representative of the Italian population with reference to gender, age, and geographical area (as shown in Female 566 51.5 51.7 Table 1). The sample size was calculated with a 2.95% margin of Age 18–29 159 14.4 14.2 error at 95% confidence level. 30–44 258 23.5 21.1 45–65 400 36.4 38.0 Over 65 283 25.7 26.7 5. Operational definition and measures Geographical area North-West 295 26.8 26.8 Following Ragnedda et al. (2019), the construction of DCI was North-East 229 20.8 19.6 carried out in three steps: First, a univariate analysis was used to Center 207 18.9 19.9 check for the data quality and provide an overview of the results; then, a multivariate analysis was conducted to create the DCI; and South 262 23.8 22.9 finally, a bivariate analysis was conducted to test the validation of Islands 107 9.7 10.9 the DCI. Source for comparison: ISTAT (2021). The analysis was performed as described in detail below. TABLE 2 Digital access: operational definition. Sub- Description Items or modalities Collection Measure component Digital equipment Devices used to access -Mobile phone or smartphone Multiple response Nominal the Internet -Laptop or netbook -Tablet computer -Desktop Computer -Media or game players -Smart Tv -Other devises (e.g., e-book reader, Smartwatch) Connectivity Quality and Place of In which of the following settings do you most frequently access Multiple response Nominal access the Internet? Time spent online First time using the How old were you when you used the Internet for the very first Open question Scale Internet time? Support and Request for help, formal Have you ever had any formal training in using Internet? Multiple response Nominal Training training received, and help offered If you needed help, would there be someone who could help you Closed question Nominal with using the Internet? Have you looked or asked for help to use the Internet in the past 3 months? Have you helped someone use the Internet in the past 3 months? Frontiers in Sociology 05 frontiersin.org Addeo et al. 10.3389/fsoc.2023.1144657 TABLE 3 Digital competences: operational definition. Sub-component Description Items or modalities Collection Measure Information and data literacy Browsing, searching, filtering data, I am confident in browsing, searching Likert Scale Scale information and digital content and filtering data, information and -Not at all true of me-Not very true digital content of me-Neither true nor-untrue-Mostly true of me-Very true of me Evaluating data, information and I regularly verify the sources of the digital content information I find Managing data, information and I regularly use cloud information digital content storage services or external hard drives to save or store files or content Communication and Interacting through digital I actively use a wide range of Likert Scale Scale collaboration technologies communication tools (e-mail, chat, -Not at all true of me-Not very true SMS, instant messaging, blogs, of me-Neither true micro-blogs, social networks) for online nor-untrue-Mostly true of communication me-Very true of me Sharing through digital I know when and which information I technologies should and should not share online Engaging in citizenship through I actively participate in online spaces digital technologies and use several online services (e.g., public services, e-banking, online shopping) Managing digital identity I have developed strategies to address cyberbullying and to identify inappropriate behaviors Digital content creation Developing digital content I can produce complex digital content in Likert Scale Scale different formats (e.g., images, audio -Not at all true of me-Not very true files, text, tables) of me-Neither true nor-untrue-Mostly true of me-Very true of me Integrating and re-elaborating I can apply advanced formatting digital content functions of different tools (e.g., mail merge, merging documents of different formats) to the content I or others have produced Copyright and licenses I respect copyright and licenses rules and I know how to apply them to digital information and content Programming I am able to apply advanced settings to some software and programs Safety Protecting devices I periodically check my privacy setting Likert Scale Scale and update my security programs (e.g., -Not at all true of me-Not very true antivirus, firewall) on the device(s) that I of me-Neither true use to access the Internet nor-untrue-Mostly true of me-Very true of me Protecting personal data and I use different passwords to access privacy equipment, devices and digital services Protecting health and wellbeing I am able to select safe and suitable digital media, which are efficient and cost-effective in comparison to others Problem-solving Solving technical problems I am able to solve a technical problem or Likert Scale Scale decide what to do when technology does -Not at all true of me-Not very true not work of me-Neither true nor-untrue-Mostly true of me-Very true of me Identifying needs and I can use digital technologies (devices, technological responses applications, software or services) to solve (non-technical) problems Creatively using digital I am able to use varied media to express technologies myself creatively (text, images, audio and video) Identifying digital competence I frequently update my knowledge on gaps the availability of digital tools Frontiers in Sociology 06 frontiersin.org Addeo et al. 10.3389/fsoc.2023.1144657 To create the DCI, we first built the two sub-indices of Considering the Kaiser (1960) criterion of retaining only those Digital Access (by combining the set of digital access’ questions factors having eigenvalues of 1 or more, results showed that shown in Table 2) and Digital Competences (by considering the extraction of one factor was appropriate to represent the the sub-components shown in Table 3). A two-stage Principal factorial solution; moreover, the average size of the factor Component Analysis approach was run to develop a Digital loadings (over ± 0.6) is good (Comrey and Lee, 1992), this Capital Index from Digital Access and Digital Competences suggests that all the selected variables contribute to define indices, able to synthesize a high number of items and to the factor. simplify the interpretation of the results (Di Franco and Marradi, In the final stage, we adjusted index score to a range from 0 to 2013). After the first extraction, each factor (and the loading 100 to simplify its interpretation. variables) was independently analyzed to remove those variables Indeed, to answer the RQ2, we carried out a bivariate that were not strictly connected to the concepts under analysis. analysis between the DCI and socio-demographic variables considered crucial (age, gender, level of education, income, place of residence) to test the validation of DCI. Specifically, TABLE 4 Factor loadings of the variables used for the digital access index. we used one-way analysis of variance (ANOVA) to explore the relationships between Digital Capital and gender, income, Digital access index educational level and place of residence. Meanwhile, we used Digital equipment 0.746 a correlation analysis to test the statistical relationship between Connectivity 0.756 Digital Capital and age. Finally, the relationship between Digital Capital and gender was addressed by performing an independent Time spent online 0.606 samples t-test. Support and training 0.646 Kaiser–Meyer–Olkin (KMO) test = 0.701; Bartlett’s test, Sig. < 0.000. TABLE 5 Factor loadings of the digital competence items. Factors Problem solving Content creation Safety I am confident in browsing, searching and filtering data, information and digital content I regularly use cloud information storage services or external hard drives to save or store files or 0.553 content I regularly verify the sources of the information I find I actively use a wide range of communication tools (e-mail, chat, SMS, instant messaging, blogs, 0.698 micro-blogs, social networks) for online communication I know when and which information I should and should not share online 0.601 I actively participate in online spaces and use several online services (e.g., public services, 0.678 e-banking, online shopping) I have developed strategies to address cyberbullying and to identify inappropriate behaviors 0.501 I can produce complex digital content in different formats (e.g., images, audio files, text, tables) 0.706 I can apply advanced formatting functions of different tools (e.g., mail merge, merging 0.764 documents of different formats) to the content I or others have produced I respect copyright and licenses rules and I know how to apply them to digital information and 0.632 content I am able to apply advanced settings to some software and programs 0.828 I periodically check my privacy setting and update my security programs (e.g., antivirus, 0.626 firewall) on the device(s) that I use to access the Internet I use different passwords to access equipment, devices and digital services 0.757 I am able to select safe and suitable digital media, which are efficient and cost-effective in 0.519 comparison to others I am able to solve a technical problem or decide what to do when technology does not work 0.775 I can use digital technologies (devices, applications, software or services) to solve 0.761 (non-technical) problems I am able to use varied media to express myself creatively (text, images, audio and video) 0.651 I frequently update my knowledge on the availability of digital tools 0.712 Kaiser–Meyer–Olkin (KMO) test = 0.946; Bartlett’s test, Sig. < 0.000. Frontiers in Sociology 07 frontiersin.org Addeo et al. 10.3389/fsoc.2023.1144657 TABLE 6 Factor loadings of the variables used for the digital access index. TABLE 7 Component matrix of the combination between digital access index and the digital competence index. Digital competences index Component 1 Problem solving 0.869 Digital access 0.853 Content creation 0.822 Digital competences 0.853 Safety 0.866 Kaiser–Meyer–Olkin (KMO) test = 0.707; Bartlett’s test, Sig. < 0.000. TABLE 8 Relationship between qualification and DCI. Qualification Mean Standard dev. 6. Results Some high school, no diploma 43.0 16.9 6.1. RQ1 High school graduate 50.7 14.6 Some college credit, no degree 51.2 13.5 As mentioned above, the first stage of analysis focused on the Bachelor’s degree 51.9 14.9 building of the Digital Capital Index by combining the Digital Access Index and the Digital Competences Index. Master’s degree 60.6 14.7 In order to create the Digital Access Index the multiple Postgraduate qualification 62.5 9.5 responses related to each sub-component of Digital Access were F = 13.692; Sig. < 0.000. conceived as dummy variables and summarized into single variables. The four variables were included in the EFA to test the TABLE 9 Correlation analysis between age and DCI. operational definition and develop the Digital Access Index (see Table 4). The factor scores were saved using the regression method. Digital capital index Indeed, the Digital Competence Index was built by directly Age −0.404∗∗ applying the EFA to the set of items shown in Table 5. The first step ∗∗ Correlation is significant at the 0.000 level (two-tailed). of the factor analysis provided a three-factor solution that explained the 58% variance and are named after “Problem-solving,” “Content creation” and “Safety” competencies. shows a positive and statistically significant impact of education on By implementing the two-step factor analysis approach (Di digital capital (see Table 8). Franco and Marradi, 2013), the three factors were converted into The literature shows that younger people have a higher level variables by considering only those items with high factor loadings of access and use (in terms of digital skills, types of online on each component. Then, we performed a second EFA on the activities, engagement, etc.) than the older ones (Lee et al., 2011; three variables, representing “Problem-solving,” “Content creation” Dutton and Blank, 2013; Blank and Groselj, 2014; Ragnedda and “Safety,” in order to extract a single factor representing et al., 2019). In this regard, we performed a correlation analysis Digital Competences (see Table 6). This double step improved the between age and Digital Capital (shown in Table 9). The correlation interpretation of the latent dimension by “refining” the results and coefficient shows a statistically significant negative relationship isolating those features that strongly contribute to the factors. (−0.404, p < 0.000). The last step of our process was to combine the Digital Digital divide literature has shown differences between men Access Index and the Digital Competence Index through a further and women about digital experience and digital knowledge: several extraction of a single factor representing the DCI (as shown in studies show that men are more likely to use digital devices and Table 7). The factor analysis provided a one-factor solution that to develop better digital skills (Ono and Zavodny, 2007; Blank and explained the 72.8% variance and shows that the two components Groselj, 2014). However, in Ragnedda et al. (2019) this evidence is have the same weight in determining Digital Capital. not supported. Digital gender inequalities are very likely to decrease over the years, especially in more developed societies, as shown by 6.2. RQ2 Blank and Groselj (2014). In our analysis, a small gender difference emerged from the one-way ANOVA, with men being slightly better The second research question addressed the validation of digital than women, with a positive deviation of 3.6 are. However, the t-test capital in a different social and cultural context, namely the Italian (shown in Table 10) returned coefficients about a non-statistically one. To achieve this, we applied a construct validation procedure significant relationship (with F = 0.025). in which we tested the DCI with the sociodemographic and Similarly, several scholars (Witte and Mannon, 2010; Talukdar socioeconomic variables considered in the literature by previous and Gauri, 2011; Ragnedda and Muschert, 2013) have suggested studies (see Section The concept of digital capital): Education, Age, that the level of economic resources can have a major impact on Gender, Income, and Place of Residence. the ability to access digital devices and/or develop specific skills Previous studies have shown that education has a positive in this regard. In other words, recent literature shows that those impact on the digital experience, e.g., levels of access, usage, who have more resources are also more likely to have access, skills, digital skills, etc. (Clark and Gorski, 2001, 2002; Shelley, 2009; Van and engagement related to technology. The results of the one-way Deursen and Van Dijk, 2013; Blank and Groselj, 2014; Ragnedda ANOVA shown in Table 11 shows a positive relationship between et al., 2019). Consistent with these findings, the one-way ANOVA income and DCI, highlighting a 6.3 gap between those with high Frontiers in Sociology 08 frontiersin.org Addeo et al. 10.3389/fsoc.2023.1144657 TABLE 11 Relationship between income and DCI. Income Mean Standard dev. 95% confidence interval Upper 5.369 5.369