This systematic review identified various types of digital learning designs integrating different tools, software’s and learning platforms conceptualized as flipped classroom, blended learning, remote online- learning, online/e-learning courses and CAI designs and dominated by asynchronous online learning. Regarding digital technology, educational video games, virtual reality (VR), Mobile learning applications (apps), audio files (podcasts), simulation programs and various video formats are the most-used tools in the learning designs.
The main findings showed statistically significant differences or equal results in favour of digital learning designs in 18 of the 20 studies compared to traditional classroom teaching. Two of our meta-analysis showed a statistically significant improvement for flipped classrooms (39–41) and websites/apps (31–33). The meta-analysis of flipped classrooms showed effects favouring the intervention for student learning outcomes and grades. This is in line with another systematic review (18), but in contrast to the conclusions drawn by another review (46). Further, the effects of using a website/app on practical skills showed statistically significant differences in favour of digital websites (31–33). This is supported by another systematic review (3). On the other hand, the meta-analysis of using websites/apps on acquisition of knowledge indicated better grades on MCQs among students who received traditional classroom teaching compared to those who visited a website as part of the digital learning design. However, the results were not statistically significant, and the meta-analysis included fewer than 100 participants, so the results need to be interpreted carefully. Regarding students’ evaluations of the statement ‘I was able to apply what I learned’ in two studies (31, 32), the meta-analysis showed that the findings are not consistent. This is in line with the results from student’s evaluation in the studies that were not included in the meta-analysis.
Six of twelve studies (27, 34–37, 42) that were not included in the meta-analysis showed statistically significant effects of the intervention. Five of these twelve studies (26, 38, 43–45) showed results equal to the control group. One of these twelve studies showed results in favour to traditional classroom teaching (30).This leaves a somewhat mixed impression of the effects of different learning designs and is in line with another review regarding different learning designs (3, 21).
Ten studies (26, 28–30, 33–36, 38, 45) that were not included in the meta-analysis reported students’ self-reported learning. These studies showed, overall, a positive experience and high satisfaction with the digital learning design. One study showed lower satisfaction (30) and another showed high variance in students’ satisfaction between the cohorts in the study (27). On the other hand, the outcomes show significant differences in favour of digital learning designs in these two designs. These two studies are in line with another paper regarding the relationship between students self-reported learning and grades and “where students in the active classroom learned more, but they felt like they learned less” (53, p. 1).
One explanation of the results in the flipped classroom designs is that the pedagogical possibilities within the flipped classroom setting facilitate pre-class learning activities and enhance self-regulative abilities among students, flexibility and transparency in the learning process (47). These pedagogical possibilities can be explained and thereby lead us to expect the model to be effective and to enhance students’ learning outcomes.
Regarding the results from the website/app (31, 33) on practical skills, several explanations can be offered. Interactive apps can be a useful tool because they are flexible, accessible, interactive and give the students the ability to observe how to perform practical skills. Demanding aspects of mobile learning “are the links between and the need to facilitate different sustainable pedagogical and learning strategies by integration, support, interactive use and appropriate choice of tools” (3, p. 32). Another explanation is that learning is promoted when learners are engaged in applications on their newly acquired knowledge of skills (5).
In these studies, student in the intervention group were given free access to a website/app immediately after the traditional classroom teaching finished. An aspect and explanation regarding the significant differences between m-learning (app) and traditional learning resources and their context (books, articles and classroom teaching or laboratories) in learning practical skills is that the mobile learning activities can be designed in alignment with the learning outcomes and the assessment form (48). Another explanation of the results showing no significant effect on knowledge acquisition is that, if the app is mainly designed to learn practical skills, then the knowledge aspect might not be supported or facilitated as a learning outcome. On the other hand, to learn skills it is necessary to acquire knowledge, although this might not be expressed as an expectation in the app’s design, or it might not be facilitated by reading about or completing tasks that can support knowledge acquisition. The behaviourist learning approach is often used in mobile learning designs in higher education where teachers are the content deliverers, and this seems to dominate higher education teaching practices (3, 49). In a critical perspective, apps must be integrated into the system of learning for different learning materials (e.g. books, articles), and the content, learning activities and technology must be designed in such a way that they are aware of one another and complement each other to achieve the learning outcomes (52).
Overall, the effects of the digital learning designs in the studies that were not included in the meta-analysis, indicated that planned learning designs that integrate digital learning technology, have a great potential to enhance learning or to be equally effective for achieving learning outcomes (e.g. grades). These findings can be explained by and are in line with other studies showing results regarding increased student involvement, critical discussions and increased student–teacher contact (50). However, there is also a criticism here, which raises questions about technology optimism and the uncritical belief that the use of technology leads to learning in itself (51). It is how digital tools are implemented and used pedagogically that matters for students’ learning outcomes, not the technology itself (3).
There is the potential to complete meaningful learning activities in digital learning designs to facilitate physiotherapy students’ learning, for example, critical thinking and thereby higher-order learning activities, and to support the shaping of an independent decision-making authority as a professional physiotherapist once they graduate from a physiotherapy program (11). Facilitating digital learning designs also challenges and has implications for the professional educator’s role and teaching practices in physiotherapy education when moving from a transmission to a transformative teaching practice (e.g. flipped classroom designs; 4, 12, 52).
Strengths and limitations
We consider several aspects as strengths in this systematic review. First, we developed a rigorous and comprehensive search strategy on digital learning technology in learning design developed by two of the authors (NBØ, YR) together with two information specialists at the campus (MWG, EK). Second, we find it possible to be able to synthesise and conduct a meta-analysis even though the included studies had varied interventions, small sample sizes and varied forms of effect size.
However, the review has some limitations. Several factors, including weak study designs (single cohorts), underreporting of statistical methods and educational intervention details and non-validated measures methods (multiple choice questions and self-report questionnaires), may limit the potential to influence current practice. Studies also had methodological challenges regarding assessing the risk of bias for all the included studies. The results are also limited because only one study included long-term (2-semester) follow-up to assess learning retention.
The included studies used various conceptions of digital learning designs. This generates an unclear terminology and can make it difficult to compare designs and synthesize the results.
There is also a need for more robust studies, such as experimental designs, control variables and the use of statistical methods and reporting of the results, especially in the flipped classroom designs. More in-depth and follow-up research studies assessing learning retention, students’ approaches to learning and preferences to studying in a digital learning environment would also be beneficial. Furthermore, scholar should study the experiences and attitudes of teachers towards developing and implementing digital learning designs in physiotherapy education.