The concept of digital literacy (DL) appears for the first time in the works of Zurkowski (1974), for whom it is an ability to identify, locate, and examine information. However, despite its novelty, the conceptions it encompasses have been changing (Lim and Newby, 2021). Proof of this are the contributions of Gilster (1997) who combines the idea that DL is also closely linked to skills such as access, evaluation, and management of information used in learning processes. Digital learning is understood as the set of technical-procedural, cognitive, and socio-emotional skills necessary to live, learn, and work in a digital society (Eshet-Alkalai, 2012; European Commission, 2018). It is related to reading, writing, calculation skills, and effective use of technology in personal, social, and professional areas. It is also considered inseparable from the social and educational needs of the society in which we live (Larraz, 2013; Brata et al., 2022). Therefore, we refer to a concept that has several aspects including the technological aspect, the informative and multimedia aspect, and the communicative aspect. It involves a complete process and multiple literacies (Gisbert and Esteve, 2011; Lázaro, 2015; Valverde et al., 2022). It requires mastery of certain competencies related to the identification of training needs, access to information in digital environments, the use of ICT tools to manage information, interpretation, and representation of information, and the evaluation of information and the transmission of information (Covello and Lei, 2010; Walsh et al., 2022).

The systematic review is composed of four phases, following the model of Miller et al. (2016) and Scott et al. (2018).

PHASE 1: Search terms: In this phase, we developed a schematic of search terms from Web of Science and Scopus databases. We also accessed the databases to locate specific studies that were referenced in the publications that we found in the databases during our initial search. The schematic of terms and thematic axes that were used as a starting point for scanning both databases for anything related to the descriptor “digital” and the descriptor “literacy” is presented in Figure 1.

PHASE 2: Selection process based on inclusion and exclusion criteria. The following selection criteria were applied: year of publication between 2010 and 2021, availability of full text, and language of publication in English, Portuguese, or Spanish. Once the first results were obtained, they were selected based on title, abstract, and the use of standardized instruments in their methodology. We rejected the studies that used “ad hoc” instruments to measure digital competence.

In addition, the selection indicators provided by Cooper and Hedges (1994) and Cooper (2009) were used, such as peer-reviewed journals, referenced databases, and citation indexes.

PHASE 3: Analysis of methodological quality and indicators based on scientific evidence. Following Torgerson (2007) and Risko et al. (2008) and taking into consideration the MQQn (Risko et al., 2008), we used seven indicators to analyze the quality and effectiveness of the studies (Acosta and Garza, 2011). These were: alignment of theory, findings, reliability and validity, descriptive details of participants and the study, sample, and consistency of findings and conclusions with the data (Risko et al., 2008). Alternatively, evidence-based indicators were also used along with study effect sizes (Díaz and García, 2016; Canedo-García et al., 2017).

PHASE 4: Reliability and outcomes. Reliability was established for both the selection criteria and the coding criteria during each phase, to evidence the replicability of the results. In addition, the results entailed a qualitative analysis of the selected studies, the central arguments, and the evidence provided in a modulated way to address the research questions.

Therefore, the procedure to be followed was documented and charted according to the PRISMA statement (Moher et al., 2009; Page et al., 2021) (see Figure 2). Likewise, an analysis was undertaken of the key foci in the various studies to highlight the relevant findings and evidence they provided in this regard. The key focus of our work was: first, to analyze the documents related to the digital literacy of university students; second, to identify which variables affect digital literacy; and third, to undertake a comparative analysis between the different variables that were analyzed.

All the selected studies had as samples university students who were pursuing some type of degree or postgraduate degree related to education, and therefore, studying to become future teachers. An intervention design was presented that corresponds to a pre-intervention, the intervention itself, and a post-intervention using techniques such as the activation of prior knowledge, instructions, emulation, and subsequent tests. We also found studies that had an experimental design assessing control groups and experimental groups (Kajee and Balfour, 2011; Kuhn, 2017; Pequeño et al., 2017; Sharp, 2018; Lerdpornkulrat et al., 2019).

In the case of those responsible for the intervention, practically in all cases, the teacher acts as such, with one or two of them taking the lead. Although the presence of specialized personnel should also be highlighted, as is the case of the work elaborated by Alfonzo and Batson (2014) and Elliott et al. (2018) in which a professional librarian also intervened. Or, in the work detailed by Ball (2019), where a consultant who is not a teacher but a professional expert in the use of digital devices and trained for such an occasion by a responsible brand (Apple) carried out the training at the center.

If we examine the constructs or competencies covered by the works selected in our search, we find that all of them, in addition to dealing with digital literacy, also focus on self-efficacy perceived and developed through digital literacy.

The results of our study could be understood under different themes.

First, we found studies that referred to digital competence and other educational issues. Within them, we found a series of competencies that are emphasized such as digital writing and reading. Research developed from digital media, such as databases, web, or applications aimed at the treatment of digital literacy was noted as emerging pedagogies and educational innovation. The digital design of content and materials and the skills to edit, publish or share them, and competencies related to mathematics and its digital literacy, formed part of digital literacy.

Second, we found studies related to digital competence and the use and employment of the Internet, social networks, web 2.0, and the treatment of digital risks and their relationship with digital literacy.

Third, we found works that in addition to focusing on digital literacy, also focused on different psychological constructs such as motivation, commitment, attitudes, or satisfaction (Tables 1, 2).

Regarding instructional literature, we found a large number of results on mass training programs or courses in which digital literacy was the focus. Examples include a course offered in which students could sign up to, or modules taught during the teaching of a subject. We also found investigations on interventions that had been carried out through different subjects in the study program from where the sample was taken. In this case, the samples were taken on an ad hoc basis from a specific student body which the researcher intentionally decided based on a previous intervention experience with them (Ata and Yildirim, 2019; Ball, 2019; Campbell and Kapp, 2020; Domingo-Coscolla et al., 2020; Tomczyk et al., 2020; Vinokurova et al., 2021).

In terms of material resources, all the studies used some type of documentation (digital or not) with instructions on the development of the activities, in which the students were provided with what to do and the steps to follow. In this case, the development scenario was both online and face-to-face, based on different activities given through workshops or seminars for their development.

It should also be noted that in those investigations in which the intervention itself required a specific application or program, the same was used, specifically, and even the intervention had a specific scenario since it was carried out in person in specialized laboratories where experts and specific material was available for this purpose. As an example of these specific materials, in our results, we found the use of the Photo Story 3, Dashboard, and Wikipedia, as well as the EMODO program or the SELI platform (Kajee and Balfour, 2011; Robertson et al., 2012; Ball, 2019; Hamutoglu et al., 2019; Tomczyk et al., 2020).

Regardless of the setting and the program or application employed, we can classify the duration of these interventions into two broad groups: those that had a duration of <1 semester, and those that had an intervention whose duration ranged from one semester to one academic year.

Regarding the instruments used, it should be noted that most of them used survey forms as an evaluation instrument, either by the researcher or by the students. In addition, it is usually used as a resource to collect information of a personal nature and about one's own experience throughout the intervention. We must also highlight the fact that in many of the results found, this form was used digitally or virtually, abandoning the old paper forms (Kajee and Balfour, 2011; Robertson et al., 2012; Carl and Strydom, 2017; Elliott et al., 2018; Ball, 2019; Lerdpornkulrat et al., 2019; Campbell and Kapp, 2020).

Regarding the use of questionnaires, scales or self-reports, we found several works that used participants' digital literacy histories as instruments. Through them, the researcher could learn first-hand about the sample's personal experience of digital literacy, the previous knowledge they possess, the digital skills they had mastered, those they lack, or those they consider they should improve. It also included the sample's vision regarding the use and employment of digital resources in teaching practice (Kajee and Balfour, 2011; Robertson et al., 2012; Pequeño et al., 2017; Elliott et al., 2018).

In the case of scales, we found two papers that employed a Likert-scale elaborated ad hoc. We also found studies that employed standardized scales like the Information Literacy Assessment Scale for Education (ILAS-ED), the Digital Literacy Scale, or the E-Learning Attitudes Scale.

Some of the studies we reviewed used semi-structured interviews as a means of monitoring and providing feedback to the students Table 3; (Kajee and Balfour, 2011; Alfonzo and Batson, 2014; Gill et al., 2015; Carl and Strydom, 2017; Elliott et al., 2018; Elphick, 2018; Ata and Yildirim, 2019; Campbell and Kapp, 2020).

As for the sequence through which the different interventions were developed, we found two types—first, those that divided the contents in time, as is the case of the work of Kajee and Balfour (2011), who covered a first semester digital writing from online classes, self-instructions and face-to-face classes in a specific laboratory, and in a second semester was exposed to different digital research techniques, following the same methodology. In contrast, we spotted the second type, where the same technique was followed throughout the study, as is the case of Robertson et al. (2012). They applied digital stories as a tool for the development of the activity, but also the evaluation of the competency. In the research carried out by Lerdpornkulrat et al. (2019), it is apparent that with the use of the rubric, the teacher gave them an example of the work and asked them all to practice evaluating and grading this work. In this way, they could check if they understood how to use a rubric. They then used the rubric to self-assess their work. After receiving feedback, both groups of students revised and resubmitted their completed projects again.

In the investigation by Elliott et al. (2018), the intervention was structured in work modules with the following sequence of sessions: they were introduced in the first session with opportunities for group discussions and questions. Essential module reading was provided in weekly online study units and module workshops integrated academic reading and writing activities, such as paraphrasing and referencing, with module content.

In the study by Ball (2019), in the first year, the students took modules on publishing history, culture, markets, and media. In the second year, the intervention was based on their publishing skills, reading for writing development, and grammar and general literacy.

Hamutoglu et al. (2019) organized their intervention in different weeks, such that during the first week of the 14-week semester, the instructor oriented the students for the course and administered pre-tests. In the following week, students were provided with a session on the Edmodo platform and orientation training on the course content.

In the work of Gabriele et al. (2019), the experimental research plan (i.e., activities to be performed, methodology to be adopted) was established over 4 months followed by the organization of the reading material (power point presentations, introductory videos of the software, handouts, ad hoc created applications as examples).

We also found interventions that had very short time durations, but provide daily detail of the contents and interventions. Similarly, Alfonzo and Batson (2014) dedicate 1 day to the search and orientation in digital resources, 1 day to the APA standards, and 3 days to develop and use a specific application.

In the research by Istenic et al. (2016), the intervention was based on six different types of tasks related to a variety of mathematical problems, including problems with redundant data, problems with multiple solutions, problems with multiple paths to the solution, problems with no solution, mathematical problems in logic, and problems with insufficient information.

In some interventions, the sequence through which they are developed is the very development of the subject of the degree course from which they are implemented, as is the case of the work of Gill et al. (2015).

In the work of Carl and Strydom (2017), students were first familiarized with the devices and then introduced to electronic portfolios, which helped them to create blogs that serve as platforms for electronic portfolios, and guided them on how to collect artifacts and how to reflect and share content.

In one work we found narrative was used as a technique so that the students could later present their work, analyze it in groups, rework it and present it again to their classmates. Kuhn (2017), Pequeño et al. (2017), and Elphick (2018) followed this model.

Adopting a novel consultative approach, Botturi (2019) co-designed the intervention with his students in two steps: they were surveyed 4 weeks before the start of the course and asked to choose between two options: an overview of different topics/methods/experiences, or an in-depth exploration of one or two topics/methods/experiences. All respondents indicated a preference for the first option and provided indications of the topics they wished to cover (see Tables 4, 5).

The limitations of our search are listed in Table 6. At the theoretical level, we encountered studies that were not very current, missing research questions or hypotheses, or even missing objectives. At the statistical level, we found several studies had a small or unrepresentative sample.

Analyzing the interventions themselves, we identified a few limitations, especially in those studies that neither indicates the tasks, record the entire process, or lack key information to replicate the intervention. In some studies, key information relating to the person carrying out the intervention was missing, particularly on whether they had the specific training for this purpose. Another limitation that was identified was that very few evaluation strategies were in place to evaluate the interventions (see Table 7).

Similarly, gaps were found regarding ethical controls, where in some studies the main limitation was that ethical controls were non-existent or not specified (Robertson et al., 2012; Istenic et al., 2016; Kuhn, 2017; Elphick, 2018; Ata and Yildirim, 2019; Tomczyk et al., 2020).

Figure 3 shows the evolution over the years of the samples used in each of the studies from 2011 to 2020.

Figure 4 shows the evolution over the years of the controls used in each of the studies from 2011 to 2021.

This work aimed to analyze the empirical evidence found in international studies between 2011 to 2021 related to the digital literacy of university students, including those pursuing degrees in education. This objective has been met.

Regarding the first focus related to literacy, this paper highlighted the fact that studies from the West are the most prevalent in this field (Çoklar et al., 2017; Ata and Yildirim, 2019; Hamutoglu et al., 2019; Sujarwo et al., 2022), which correspond to cross-sectional studies, mostly employing instruments such as “the Digital Literacy Scale” developed by Ng (2012), and “the information literacy self-efficacy scale (ILS)” developed by Kurbanoglu et al. (2006). Regarding the level of mastery, the results showed an upper intermediate level of competence in information and digital literacy, communication, and collaboration, but a low intermediate level in terms of digital content creation, particularly in the creation and dissemination of multimedia content using different tools (López-Meneses et al., 2020; Moreno et al., 2020).

Regarding the second focus, digital literacy in university students, this study reviewed the various contributions of other works and found the presence of a competent group in this field, which makes efficient use of both the Internet and digital media (Çoklar et al., 2016; Ata and Yildirim, 2019; Lim and Newby, 2021). However, differences were also found in this collective relating to gender, where women were more competent than men in digital literacy, information literacy, technological literacy, and communicative literacy (Hamutoglu et al., 2019; López-Meneses et al., 2020; Navarro, 2020). However, on the other hand, we lso found studies that revealed particular gender gaps where men showed a higher propensity for DL, while women outperform men in the overall digital literacy test (Ata and Yildirim, 2019). Ata and Yildirim (2019) also found differences in DL between students where university students studying science or mathematics-related majors had higher levels of digital literacy than students majoring in social sciences or psychology fields (Ata and Yildirim, 2019; Chow and Wong, 2020).

And as for the third focus, digital literacy in future teachers, we found a dual use of digital literacy, in its social and leisure aspect (searching or maintaining friendships through social networks, sharing digital content, downloading content, or playing online games), and in its academic aspect (searching in search engines, working through online documents, organizing or synthesizing information from different processors, using computer programs to make presentations, edit images or content, or create audiovisual content (López-Meneses et al., 2020).

The main contribution of this review lies in its comparison between pre/post-pandemic studies, which show a great increase in the use of technologies in the educational world (across the curriculum), and research work focused on measuring the competencies of these devices (Baber et al., 2022). These new investigations have not only followed the line of previous ones but focused on the measurement of digital literacy and its influence on it by variables such as the degree of origin, gender, age, or being a digital native or immigrant (Castañeda-Peña et al., 2015; Çoklar et al., 2016; Castañeda et al., 2018; Ata and Yildirim, 2019; Gür et al., 2019; Hamutoglu et al., 2019; Lerdpornkulrat et al., 2019; González et al., 2020; Navarro, 2020; De Sixte et al., 2021). But there has been an expansion of the topics and variables that are studied in conjunction with digital literacy, among which we find as a novelty, the study of psycho-educational variables such as academic motivation (Chow and Wong, 2020), self-efficacy and motivation (Lerdpornkulrat et al., 2019), effort expectations (Nikou and Aavakare, 2021), and self-concept as a student and as a teacher (Yeşilyurt et al., 2016). The importance attached to the educational field, the identification of different roles or behaviors within the concept of digital literacy that is delimited, or even the types of uses within the concept of digital literacy (López-Meneses et al., 2020; Moreno et al., 2020; Navarro, 2020; Lim and Newby, 2021) are new trends.

Therefore, we can affirm that in this study the research predictions are fulfilled, in that the results found show relevant differences from international studies pre-post pandemic; and by different cultural backgrounds (Spanish Latin, Portuguese, Finnish...), gender, and personal digital resources. In terms of applications for educational practice, these results do not indicate that university students are competent in terms of digital literacy, although they demonstrate some competencies like online information search, information evaluation, information processing, information communication, and dissemination skills (Çoklar et al., 2016; Lerdpornkulrat et al., 2019). Therefore, there is the risk of training an incomplete student body in digital competence. For complete and comprehensive digital literacy for university students, especially future teachers, there is an urgent need to invest in digital literacy programs. This will ensure that the comprehensive digital competence of students corresponds to the use and employment of the Internet and digital devices in their teaching tasks (Gisbert et al., 2016), and be a guarantee of their integration into teaching practice (Aslan and Zhu, 2016; Nikou and Aavakare, 2021).

As for the limitations of this work, they are closely related to the seven indicators for analyzing study quality and effectiveness (Acosta and Garza, 2011), which are: alignment of theory, findings, reliability and validity, descriptive details of participants, and the study, sample, and consistency of findings and conclusions with the data (Risko et al., 2008). Along with evidence-based indicators, and effect sizes of studies (Díaz and García, 2016; Canedo-García et al., 2017). So future lines of research or work, should take into account overcoming these limitations, and embrace them in the face of their development.

The number of studies found in the systematic review is comparable to what is usual in this type of study and even higher. For example, in the exemplary systematic review by Scott et al. (2018), they identified only 29 studies that met the quality criteria, reviewing 50 years of studies published in the US, and of these, only four were quantitative. In the study by Borgi et al. (2020), they only found ten studies that fit the criteria in a very good analysis. Other systematic reviews go along the same lines, and in the same journal and section Frontiers in Psychology. For example, Dickson and Schubert (2020) and Liu et al. (2022) found only six studies in a review of great interest; the study by Nguyen et al. (2021) identified 18 eligible articles; Shou et al. (2022) with 12 studies included; or Tarchi et al. (2021); Huang (2022) found seven studies for quantitative analysis and eight for indirect evidence; Coxen et al. (2021) with 21 articles included in the focal analyzes of the systematic review. The number of studies to be representative is not defined by the number but by the existence of such studies. In a systematic review, all studies are reviewed, thus the population of published studies that fit the indicated criteria. With these studies, it was possible to do an analysis of objective indicators in a general comparison between studies; assessing the instruments used; examining the characteristics of the interventions such as strategies, instructional procedure, and psychological variables considered; comparing the fidelity controls of the treatments, which guarantees their rigor and their application in the terms prescribed by the empirical validation of the interventions; and reviewing the limitations of the studies and their contributions by years. These contributions were based on objective data from the studies and have been represented in tables and figures. In addition, a qualitative analysis is provided that highlights the value of intervention studies in relation to digital competence, and the key psychological variables that have been used. It is true that the studies published since 2010 were used, and that there could have been more studies before, but considering the evolution of this type of focus in relation to digital competence and the psychological variables involved, it is evident that the most interesting thing is to consider the recent years which is when its need and use has been generalized throughout the population.

In general, the results show that university students are digitally literate and make efficient use of both the Internet and digital media. In this sense, we found an intermediate or higher level in skills related to communication and collaboration, such as through different chat rooms, platforms, and communication applications. But an intermediate-low level in terms of digital content creation, especially in the creation and dissemination of multimedia content. So, this should be one of the future competencies to increase in this group. Although there are differences according to gender, age, or degree of origin.

We have to invest in comprehensive digital literacy programs for teachers in initial training, which appears implicit in the training plans of their official studies. Digital literacy needs to be a part of the official curriculum, and be developed rather quickly as a separate subject but in an interdisciplinary manner throughout their training. In this way, they become digitally literate people capable of creating and generating digital content and possessing the necessary competencies and skills to use and share such content.

We must also invest in assessing teachers' self-perception. Only by knowing their opinion, skills, and shortcomings, can digital training programs be designed. Digital literacy is a predictor of good digital use and a predictor of the good use and employment of digital devices and the Internet in the future when they would be teaching.

The findings of this study compel us to consider the following: first, we need to rethink the form and manner in which future teachers are capacitated in digital literacy, if we are doing it in the best way, or if on the contrary there are gaps that should be solved. Second, we should take into account the contributions of the results found and their consequences to formulate effective intervention designs and strategies to effectively capacitate pre-service teachers in digital literacy.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

J-NS-G, NG-Á, IM-R, JG-M, and SB-C: conceptualization, methodology, software, writing—review and editing, visualization, supervision, and validation. NG-A: formal analysis, investigation, and resources: UAL, ULE, USAL, IPC, data curation, writing—original draft preparation, and funding acquisition. J-NS-G and NG-A: project administration. All authors contributed to the article and approved the submitted version.

The generalx operating funds of the universities have been used Universidad de León (Spain), Universidad de Almería (Spain), Universidad de Salamanca (Spain), Instituto Politécnico de Coimbra and NICSH (Portugal).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.