This research was conducted through a quasi-experimental design with a posttest as the only evaluation measure. The design followed the SPIRIT (Standard Protocol Items: Recommendations for Interventional Trials) declaration where the protocols to be followed were established. The initial sample consisted of 203 students of health sciences who were enrolled in the Physiotherapy degree course from the Universities of Málaga and Cádiz. However, 4 participants over the age of 30 years were excluded from the analysis to homogenize the study. The inclusion criteria were students who were attending the Physiotherapy degree course and were enrolled in the subject of General Procedures in Physiotherapy II. The results of the study were measured in the 2021-2022 academic year.

The legal ethical aspects followed the principles of the World Medial Association and the Declaration of Helsinki for ensuring the principles and well-being of the study participants. Before the execution of the research intervention, participants signed an informed consent form in which they were provided with detailed information about the research. Their confidentiality and data protection were guaranteed at all times. Ethical approval for this educational research was obtained from the Andalusian Human Research Ethics Committee (approval number 001/2022). In addition, this study was conducted according to the ethical guidelines for educational research [36]. Participants in this research did not receive compensation.

The study was carried out in the subject of General Procedures in Physiotherapy II at the Universities of Málaga and Cádiz. Each teacher in charge of the subject explained to the students the possibility of participation in the study and, therefore, the use of the AR applications, with which the general and transversal competences of the subject would be worked on. The topics of the AR program were about electrotherapy, phototherapy, ultrasounds, shock waves, and tecartherapy.

The students were informed and filled in the informed consent form before creating the control group and the experimental group, which would use the AR applications. A control and an experimental group were formed for both universities.

All students belonging to the experimental and control groups participated in the same lecture on electrotherapy, which was part of the second year of the Physiotherapy degree, both at the University of Cádiz and the University of Málaga. The experimental groups used AR applications to complement learning, integrating interactive 3D models that allowed the visualization of anatomical structures and simulations of electrode placement to apply electrotherapy procedures. At the University of Cádiz, the Aumentaty platform was used, while at the University of Málaga, Zapworks was used, although the content was the same in both cases, with the markers being adjusted to each technology. In addition, the applications provided step-by-step instructions for the correct configuration and application of electrotherapy.

In the experimental group, students received a lesson that included the use of AR applications. This complemented the subject content by integrating interactive elements such as links, videos, and 3D models. Students interacted with bookmarks that allowed them to access additional information, which enriched their learning experience.

Both applications used markers (physical or digital) to trigger AR experiences. They are compatible with mobile devices (iOS and Android), allowing users to interact with the content on their own phones. Both allow the visualization of 3D models and additional multimedia content (audio and text). They are used to enrich educational experiences. Nevertheless, there are some differences between Zapworks and Aumentaty; Zapworks is more versatile and advanced, with a visual interface that allows the creation of more customized and complex experiences, while Aumentaty is simpler and oriented to users with less technical experience. Zapworks relies on internet connection to access experiences through a browser or app, while Aumentaty can work offline. Aumentaty is more focused on the educational and cultural environments, with a more restricted use in terms of interactive possibilities, while Zapworks offers more flexibility to create different types of interactive experiences.

The control group received the same content via a traditional lesson without AR. The control group participated in a traditional lecture that consisted of a PowerPoint presentation covering the same content that was presented to the experimental group. This approach allowed for a direct comparison of the effectiveness of traditional teaching with AR-enriched teaching.

The data gathering procedure was carried out in the same way at both universities in order not to influence the results obtained. The System Usability Scale (SUS) [37] was used for this purpose. It is a methodological tool similar to the Likert Scale that measures the usability of an object, device, and application. The scale consists of 10 questions that can be scored from 1 to 5, where 1 means “Strongly disagree” and 5 means “Strongly agree” (Multimedia Appendix 1). First, the student was introduced to the two AR applications and was then asked a series of questions, which were the ones translated from the original test on Usability.gov [37]. The objective was to involve the students in the application of AR as well as to see the degree of agreement or disagreement with it in order to use it correctly and thus guarantee their agreement with the software and their correct learning.

A descriptive analysis of the control and the experimental group was conducted, considering the variables of age and sex. In addition, an individual analysis was conducted for the Rating and Usability variables in each of the Universities. In this analysis, the results of each population were obtained, and subsequently, analysis and interpretation of both variables were carried out together in order to contrast the previously mentioned hypotheses.

The main aim was to compare the degree of usability of the two AR applications used during teaching. Analysis was performed using the R software (version 4.1.2; R Foundation for Statistical Computing). In addition, we also aimed to compare the academic performance of the groups using AR versus the control group and between the two experimental groups, which used two different AR applications.

It was accepted that the control group and the experimental group were composed of different people based on the sex and age variables determined by means of the contingency tables. As the dropout rate was 0% in the usability survey, owing to the same dates of birth, it was not possible to assign a value in the surveys and it was given as a lost value. Therefore, a total sample size of 199 participants was used for the design. First, the hypothesis (H0: μ1=μ2) of the existence of significant differences between the control group and the experimental group in academic performance was accepted as a general idea, given the significant differences in the variables of Rating and Usability due to the use of AR applications. Otherwise, the hypothesis (H1: μ1≠μ2) that there are no significant differences in academic performance due to the use of AR was assumed.

To evaluate the main hypothesis, the probability μ2 was set to the distribution μ1. In this case, the mean μ2 was within the 95% confidence limits, noting that the mean μ3 and μ4 were not within the limits. This indicated that there were significant differences at 95%.

For the analysis of the data obtained from the Rating and Usability variables, the data were checked for incompleteness, normality, and homoscedasticity. In the event that there was no normality or homoscedasticity, the pertinent analyses would be carried out for subsequent interpretation in order to be able to contrast the main hypothesis put forward.

An ANOVA with the Welch correction was performed for the variables age and rating. Two-by-two comparisons were performed using the Holm significance correction because it is the most robust with respect to homoscedasticity. In addition, comparisons of the ratings obtained by the four groups were performed using the Kruskal-Wallis test. For the statistical analysis of the Usability variable, the rating obtained and the usability survey (SUS) were used. A parametric T-Welch test analysis was performed for the Usability variable. Finally, a multiple regression analysis was conducted on the scores obtained by the participants, including sex and age as variables.

The initial sample consisted of 203 participants. However, 4 participants over the age of 30 years were excluded from the analysis to homogenize the study. The total sample analyzed was 199 participants. Out of the total number of participants, 103 belonged to the experimental group and therefore had access to the AR applications. A total of 49 students from the University of Cádiz used the Aumentaty AR artifact (ie, the Aumentaty group), and 54 students from the University of Málaga tested the Zapworks AR artifact (ie, the Zapworks group). A total of 96 participants belonged to the control group, of which 37 were from the University of Cádiz and 59 were from the University of Málaga.

Regarding the population sample from the University of Cádiz, the mean (SD) age was 21.5 (1.2) years, with no significant difference between the control group (mean 21.4, SD 1.0) and the experimental group (mean 21.6, SD 1.3).

In the sample from the University of Málaga, there was a significant difference in the population, with a total mean (SD) age of 20.5 (1.2) years. The control group had a mean (SD) age of 20.9 (0.9) years, while the experimental group had a mean (SD) age of 20.2 (1.3) years. The demographic data are presented in Table 1.

Figure 1 shows the distribution of the sexes according to the group, and there were no differences in the distribution of sex among the 4 groups. There were slightly more men in the Aumentaty group, but the Cramer V indicated that the effect size was very small.

The rating variables showed a significant difference individually in each of the samples in both Universities (P<.001), as presented in Table 2. In addition, Figure 2 shows the comparisons of the rating variables obtained by the 4 groups using the Kruskal-Wallis test. As shown in Figure 2, the ages of the control groups were similar, while the ages of both the control groups differed from those of the intervention groups. Additionally, the ages of the intervention groups differed from each other.

Figure 3 shows the data on the difference between the Usability scores between groups. A moderate difference was observed between the University of Cádiz (mean 63.8, SD 6.15) and the University of Málaga (mean 61.6, SD 3.99), with a Hedges' g of 0.43. However, this difference was marginally non-significant (P=.05)

Finally, a multiple regression analysis was conducted on the scores obtained by the participants, including sex and age as variables. The results from both parametric and nonparametric analyses indicated that neither sex nor age had a significant effect on the participants’ scores. However, significant differences were found between students from the University of Málaga and the University of Cádiz (P<.001), with a beta coefficient of 1.4 favoring the Aumentaty group. Additionally, significant differences were observed between the control-A groups and the Zapworks group (P<.001), with a beta coefficient of 0.92 for the experimental group from Málaga, as presented in Table 3.

In Table 4, a multiple regression analysis was performed for the Usability variable, including the variable of sex. In this case, the differences between groups, which were on the borderline of significance, became more clearly significant. The Usability variable showed a significant difference (P=.03) in favor of the Aumentaty group as shown in Table 4.

Figure 4 shows how age affects the scores; in terms of scores, age differences between individuals did not matter. A similar performance in scores was observed among participants from the control groups of both the universities. However, it should be noted that including students over the age of 26 years in the analysis might have influenced the scores obtained, depending on the AR technique used.

The aim of this quasi-experimental design with the posttest was only to assess the use of two AR applications in two university populations and their influence on academic performance. It also sought to compare both populations in terms of the applications used and the academic performance.

The results of this study confirm the theoretical hypothesis that AR can enhance the learning experience compared to traditional methods. AR allows for greater interactivity with the content, which, as indicated by previous studies [39], can foster greater retention of information and increased student motivation. In contrast, the control group, who used traditional PowerPoint presentations, showed a more passive approach to learning, which supports previous research pointing to the limitations of traditional methodologies in capturing students’ continued interest [40].

From a theoretical point of view, comparison with the control group is fundamental to understand how AR changes the dynamics of learning. By offering a more immersive and participatory experience, AR can be aligned with constructivist theories of learning [41], which emphasize the active role of the learner in the construction of knowledge. This study provides evidence that AR not only enriches visual and contextual content, but can also alter the role of the learner within the learning process, making it more dynamic and proactive. However, it is important to note that the effectiveness of AR may depend on factors such as lesson design and students’ prior technological experience, which raises areas for future research.

When specifically examining the studies that utilized AR, positive results were found. First, the projection condition in physiotherapy education showed superior performance across various cognitive load dimensions compared to traditional teaching and tablet conditions; additionally, the use of AR and the projection of anatomical images resulted in a significant increase in the questionnaire score for the experimental group [42].

Second, significant differences were noted in students’ self-confidence regarding their learning abilities [39]. In another study, Kandasamy et al [43] implemented an AR application for teaching spine anatomy, which resulted in significant differences in students’ perceived comprehension and level of engagement [43].

Other studies have also yielded positive results in physiotherapy students; the researchers utilized a questionnaire based on the DeLone and McLeanís model [44] to evaluate the effectiveness of the educational experience. The main findings revealed that participants had a positive perception of the Augmented Studio experience, with high ratings in satisfaction, learner engagement, and system effectiveness [27].

In another study conducted by Pérez et al [28], the majority of students (89.8%) considered AR to be highly relevant in their educational processes. Regarding its usefulness and ease of use, 51.3% of respondents highlighted initial difficulties with familiarizing themselves with the AR app design and digital object creation. However, 82.1% viewed the incorporation of AR as highly positive for learning, finding it interesting (83.4%) and engaging, fostering active participation and reflection. Despite potential challenges, 79.5% of students recommended the use of AR and its apps as valuable resources for learning, promoting creativity and innovation [28]. Regarding the limitations of AR, a study conducted by Bacca et al [30] highlighted the significant challenge of maintaining superimposed information, particularly in marker-based AR applications; this limitation, reported at a rate of 9.3%, can result in student frustration and necessitates the need for improvements.

In a previous study about AR in anatomy learning, it was shown that there were significant differences (P<.0001) in the grades of students who belonged to the group that had used AR, showing an improvement in knowledge about anatomy [7]. This truly relevant fact is one of the results obtained in our design, where it was observed that the grades obtained in each sample (Universities of Málaga and Cádiz) were also significant (P<.001). These results showed a significant improvement in grades when AR is used in university teaching.

In a study similar to ours [31], which used Aumentaty with students from the University of Cádiz in the Physiotherapy degree, satisfactory results were observed in the improvement of knowledge comprehension, which translated into an improvement in the grades. The same happened in this study, when an interpretation of the Rating variable was conducted jointly for both universities through univariate and multiple regression analyses [31]. In the previous study, on the first univariate analysis for rating, the significant differences (P<.001) with a beta of 1.4 were observed in the sample from the University of Cádiz [31]. Furthermore, when comparing the control group with the Zapworks group between the two universities, the significant differences (P<.001) with a beta of .92 was observed, similar to the study where the significant results were in the group that used Aumentaty [31].

In another study that evaluated the use of AR in learning, a significant improvement in the visualization of the content was observed (P<.0001) in the group that had used AR. In addition, 3D compression, which is fundamental for the acquisition of knowledge in anatomy, was significantly greater in the experimental group with higher percentages compared to the control group [33]. Therefore, the significant differences were found in the group using AR and the results found were similar to the results (P<.001) observed in this study where an improvement in ratings was observed.

In a previous study where the usability was studied, the use of AR, one of the variables related to our design, was addressed [32]. The sample was taken from the participants who were obtaining a Degree in Nursing at the University of Valladolid, where 85% of the students who were familiar with AR believed that it offers many educational possibilities and, therefore, 86.67% thought that it could be implemented in the Faculty of Nursing [32]. Using these data, 93.33% responded that it would enhance their learning, which is why it is important to analyze and interpret the usability of this tool in Health Sciences students [32]. However, in our design, no significant differences were obtained (P=.11) in the degree of usability, which is why it was not possible to obtain conclusive facts about this variable.

This design has a number of strengths. During the development of the quasi-experimental design, the sample size was adequate and there were similarities in the subject matter at both universities. The main limitation was that the control group received conventional teaching and therefore cannot be considered a pure control group.

Significant differences are observed between the group using AR applications compared to the control group in both universities and in the two univariate contrasts. In the multiple regression analysis, which explains the greatest variability regarding the main effects, a greater effect is detected in the use of AR at the University of Cádiz compared to the University of Málaga. Finally, the results of the multiple regression analysis indicates that there are no effects of sex and age on the scores obtained by the participants in the study.