Introduction

The education landscape in the 21st century unfolds within a social context permeated by digital technologies, specifically those so-called “technologies of the fourth industrial revolution”, among which artificial intelligence, the internet of things, 3D printing, or educational robotics stand out (Ahmad et al., 2022). The pervasive influence of these technologies on daily lives is undeniable, and nowhere is this more needed to deepen and understand them properly than in the realm of education, under the careful considerations of what is commonly referred to as “Education 4.0” (Ciolacu et al., 2021).

As mentioned by Verma and Singh (2021), the profound changes brought about by these technologies necessitate a deeper exploration and a keen understanding to ensure that they are effectively integrated to enhance teaching and learning experiences. By doing so, educators and learners can be empowered to navigate the digital age with confidence and competence, ultimately shaping a brighter future for education in the 21st century. Regarding the above, the integration of technology into education has evolved beyond being a mere trend; it has become an imperative, underscored by the urgency revealed in recent global events. The seismic impact of the global pandemic and other societal phenomena, which have significantly curtailed in-person interactions, has made this need abundantly clear (Fidalgo-Blanco et al., 2022; Latinovic & Sikman, 2023). In the wake of these unprecedented challenges, educators and educational institutions have faced the daunting task of swiftly adapting to digital tools. This transformation, driven by the necessity for social distancing and the transition to remote learning, has unfolded against a backdrop of both opportunities and challenges.

According to Bizami et al. (2023), the opportunities are evident in the newfound flexibility and accessibility that digital tools provide, allowing education to persist even in the face of physical constraints. However, these opportunities are accompanied by a host of challenges, including issues related to digital equity, the necessity for effective online pedagogies, and the significance of maintaining the quality of education in virtual environments (Shenkoya & Kim, 2023). As educators navigate this digital frontier, it becomes increasingly vital for them not only to seize the opportunities presented by technology but also to address the associated challenges with innovation and resilience. By doing so, they can continue to provide high-quality education while preparing learners for the dynamic and technology-driven landscape of the future.

In the context of education, the effective incorporation of technology is a multifaceted challenge that extends beyond the mere introduction of digital tools into classrooms (Huk, 2021). It involves navigating a complex web of interconnected factors, each influencing the other in a delicate articulation of pedagogy, technology, institutional culture, teacher training, and individual preferences (Chituc, 2022). As mentioned by Dymek (2023), pedagogically, technology should not be treated as an add-on but as an enabler of enriched learning experiences. It demands a thoughtful consideration of instructional design, the alignment of technology with educational goals, and the creation of engaging digital learning environments.

In relation to the previous points, it’s important to highlight that educators need digital literacy skills to effectively utilize technological tools. Simultaneously, institutions should invest in robust infrastructure and support mechanisms (Ramírez-Montoya et al., 2021). The institutional culture plays a crucial role in shaping technology adoption, with policies and attitudes toward innovation impacting the overall digital learning environment. Additionally, accommodating the diverse preferences and needs of both educators and students is essential for ensuring equitable access and a seamless user experience (Sangole et al., 2022). To further elaborate on the previous statements, achieving successful educational integration of technologies, as emphasized by Kommers (2000), necessitates a harmonious alignment of human, institutional, and technological elements. Instructors must possess the necessary digital literacy skills, and students may have varying levels of digital competence and access to technology, impacting their engagement with digital learning resources (Schmid & Petko, 2019). Furthermore, institutional culture and policies play a pivotal role in shaping the adoption and use of technology in education. They influence decisions related to infrastructure, training, support, and the overall digital learning environment (Blashki & Isaias, 2014; Hrytsenchuk & Trubachev, 2021). Consequently, the commitment of educational institutions to invest in technology and promote a culture of innovation can significantly impact the success of technology integration.

In the context of higher education, Chaka (2022) and Asad and Malik (2023) point out that the technologies employed to support teaching and learning must be carefully selected and implemented with a keen eye on their pedagogical suitability and their potential to enhance the overall learning experience. Also, in this context, the digital-promoted transformation has brought about profound changes in the way instructors and students interact with course materials, engage in collaborative activities, and assess learning outcomes. So, traditional classroom settings have been complemented, and in some cases supplanted, by virtual learning environments where technology plays a central role (Ayub et al., 2019).

Within the realm of educational technology, there exists a vast array of tools and resources that professors can use to enhance their teaching and engage students. These tools encompass a broad spectrum, ranging from simple web-based applications to sophisticated learning management systems (Díaz et al., 2020). As an example, they may choose to integrate video conferencing platforms or virtual reality applications into their teaching, or they may leverage social media platforms, online discussion forums, or content creation tools to enrich the learning experience (Cherrstrom et al., 2019). The selection of these tools is often influenced by the specific learning objectives, the nature of the subject matter, and the preferences of both teachers and students.

Within the diverse spectrum, Learning Management Systems (LMS) have emerged as one of the most extensively utilized resources in higher education. This is owing to their functionalities, which enable the provision of a virtual environment facilitating interaction, sharing, and monitoring of academic progress among educators and students (Zhao, 2018). Additionally, they contribute the greater promise of flexibility and accessibility, making them particularly valuable in a world where remote and blended learning has become more and more frequent in the educational environment at all levels (Bervell et al., 2020). Moreover, these platforms have become integral to the educational landscape, providing a centralized and digitally accessible space for course materials, communication, and assessment (Mohd et al., 2021).

Given the central role of LMS platforms in higher education, it is not surprising that they have garnered significant attention from researchers. Numerous studies have explored various aspects of LMS usage, including adoption rates, pedagogical practices, student engagement, and the impact on learning outcomes. Research on LMS platforms in higher education began in the late 90s and has steadily grown throughout the new century. While the number of articles produced per year may not be notably significant, research in this area has remained ongoing (Fig. 1).

Fig. 1
figure 1

Research about LMS and Higher Education published in Scopus-indexed journals

Full size image

From a practical standpoint, research has shed light on the challenges teachers encounter when using LMS platforms, encompassing technical difficulties, time constraints, and concerns about the pedagogical effectiveness of online tools (Lazarinis et al., 2011). Additionally, studies like Pardamean et al. (2021) and Kuk et al. (2016), have examined the strategies employed by instructors to overcome these challenges and enhance their use of LMS platforms.

Furthermore, research has explored the perspectives of students regarding the use of LMS platforms, student satisfaction, engagement, and perceptions of the impact of LMS usage on their learning experiences. According to Kats (2013), despite the increasing adoption of LMS platforms in higher education institutions, numerous challenges persist, mainly related to their use by both educators and students. To attain a comprehensive understanding of these challenges and to identify suitable solutions, it is crucial to delve into the theoretical models that underlie the use of these platforms by higher education teachers (AL-Nuaimi et al., 2022; Malanga et al., 2022).

Regarding the above, Zareravasan and Ashrafi (2019) mention that the successful integration of technology into higher education hinges on understanding and addressing diverse complexities, and this is where theoretical models come into play. In this sense, theoretical models provide a structured framework for comprehending the multifaceted interactions among pedagogy, technology, culture, and individuals in the context of education (Jiang et al., 2022). Moreover, these models offer a lens through which one can analyze and interpret the adoption and use of technology in higher education. They help in identifying the factors that influence instructors’ decisions to use LMS platforms, the challenges they encounter, and the strategies they employ to overcome these challenges (Nilo & Pinto, 2022).

Furthermore, the application of theoretical models allows for a systematic examination of the factors contributing to the effective integration of LMS platforms in higher education. By identifying and analyzing these models, insights into the underlying dynamics of technology adoption and utilization among university instructors can be gained (Ashrafi et al., 2022).

Considering all the aforementioned factors, there arises a practical urgency for conducting this review. This urgency is based on the increasingly prominent role that LMS platforms play in higher education, alongside the persistent challenges faced by instructors, teachers, and institutions in effectively utilizing these platforms. As the digital transformation of education progresses unabated, gaining a deep understanding of the theoretical underpinnings that shape LMS adoption and utilization becomes imperative.

Method

The review has been conducted following the phases initially proposed by Kitchenham and Charters (2007), which have subsequently been reviewed and expanded by other experts in the field, such as Clear in 2015. Following their approach, these stages have been summarized and organized into three broad categories: the Review Planning stage, the Review Execution stage, and the Results Dissemination stage, which details can be observed in Fig. 2.

For the Review Execution stage, the method of Systematic Literature Review (hereafter referred to as SLR) was employed, following the guidelines and directives of PRISMA (Page et al., 2021).

Fig. 2
figure 2

Phases of the RSL process

Full size image

Review Planning

In the initial phase of the research, the central purpose of the review was outlined, which focuses which revolves around identifying the factors influencing the use of Learning Management Systems (LMS) by university instructors. To achieve this objective, three fundamental guiding questions were proposed:

RQ1. What are the characteristics of studies on the intention or use of LMS by university professors concerning objectives, sample and limitations?

RQ2. What theoretical models are used by studies to explore the determinants of the use of LMS by university instructors?

RQ3. What specific factors are included in the empirical models for the intention or use of LMS?

RQ4. What type of instruments are used to measure the intention and use of LMS?

RQ5. What are the future lines of research suggested in the studies?

To address these questions, several keyword searching strings were proposed and applied in the most recognized academic databases such as WOS, Scopus, EBSCOhost, and SciElo.

To address the search for “use of LMS,” the following terms were used: “usability of Learning Management Systems” OR “adoption of learning management systems” OR “adoption of LMS” OR “usability of LMS” OR “acceptance of Learning management systems” OR “acceptance of LMS” OR “use learning management systems” OR “use LMS” OR “use a learning management system” OR “accepting learning management system” OR “LMS adoption” OR “LMS acceptance.” Furthermore, to address the search for “university instructors”, “higher education” OR “university” OR “universities” OR “post-secondary” OR “tertiary education” OR “university” OR “professor” OR “instructor” OR “teacher” was utilized. To ensure the relevance and adequacy of the selection of these keywords, 3 strategies were considered (a) algorithms of previous systematic reviews on this topic were reviewed; (b) synonyms of these words in Thesaurus were identified; and (c) a validation of the selected keywords was carried out by two experts in this line of research.

Review execution

Subsequently, filters accessible in the respective databases were implemented, refining the search and generating a syntax that facilitates the replicability of the entire study identification process (see Table 1). The research period covered studies from 2013 to 2023, with the final search conducted on September 18, 2023. All identified records were exported to the Rayyan Software ().

Table 1 Study search syntax
Full size table

Also in this phase, using the Rayyan software, 21 duplicates were identified, meaning those records that were repeated in more than one database, were eliminated. Therefore, out of the original 113 records, after removing the 21 duplicates, 92 unique records remained. In the screening or eligibility process, a review of the titles and abstracts of the selected unique records was conducted to ensure they were directly related to the objectives of this SLR.

For this purpose, some keywords were considered to facilitate the decision to include (university professor, professor, empirical, quantitative, mixed design) or exclude the study (secondary school, theoretical review, narrative review, review, systematic review, meta-analysis, student, freshman, undergraduate student, qualitative research). For this procedure, the filter available in the Rayyan software utilized these words, detecting records containing the specified words. 66 records that contained at least one of the exclusion words were eliminated, resulting in 26 remaining records. Subsequently, the PDFs of the selected studies were downloaded to complete a full reading and apply exclusion criteria, which in this case are 3 (Reason (1) Qualitative studies; Reason (2) Descriptive and correlational quantitative designs; Reason (3) Sample differs from university professors). After completing this stage, 13 research studies were chosen (see Table 2).

Table 2 Codification of the studies included in the review
Full size table

Finally, to control and evaluate the review’s bias, an independent reviewer was included to validate the final sample of included studies.

Results dissemination

Once the list of included studies in this SLR was available, the systematic extraction of information from each study proceeded. To address the established objectives, a data extraction matrix was constructed with the following columns: (1) ID, (2) citation, (3) Aims, (4) Variable measuring intention and/or use, (5) Theoretical model, (6) Measurement of intention and/or use of LMS, (6) Empirical model variables, (6) Sample, (7) Participant variables, (8) Explained variance of behavioral Intention (BI), and (9) limitations.

Data examination occurred through two simultaneous and complementary procedures. The first involved data categorization based on their similarity or direct connection, with the application of the term normalization or standardization. The second process encompassed descriptive statistical analysis, including counts, frequency analysis, percentages, and graphical representations.

Results

Result of RQ1. Characteristics of the studies included in this systematic review regarding objectives, sample, and limitations

Results on the objectives of the analyzed studies

Regarding the stated objectives in the analyzed studies, the most frequently declared was to determine the factors influencing LMS usage (n = 8 studies). The next most declared objective was to expand the TAM model by adding determining factors in LMS usage (n = 2 studies). One study had a similar objective related to expanding the UTAUT model, another sought to determine how anxiety influences LMS usage, and finally, another study proposed a new perspective for LMS usage (professional identity) using the theoretical framework of change management (Meyer et al., 2007) (see Fig. 3).

Fig. 3
figure 3

Objectives of studies included in the review

Full size image

Results on the characteristics of the participants included in the studies (sample size and world region)

Concerning the sample, the average was 239 teachers, with the smallest sample consisting of 70 teachers and the largest comprising 560 teachers. Regarding the continent where these studies were conducted, considering Turkey as part of Asia, this continent contributes to 61.5% of the studies. On the other hand, it is observed that in America, only the United States contributes to one study, and no studies were found in Latin America (see Table 3).

Table 3 Size and geographical data of studies included in the review
Full size table

Result of the declared limitations in the studies

Regarding the limitations stated by the authors of the analyzed studies, 12 out of 13 studies specified limitations. The most reiterated limitation was the sample, mentioned in 10 studies. This limitation was associated, for example, with teachers being from only one university, the use of convenience sampling, or having a small sample. Another limitation was the omission of sociodemographic variables such as age, gender, experience, etc. (n = 6 studies). Moreover, none of the studies used moderating effects to assess the usage or intention to use the LMS. Other limitations included the use of a cross-sectional design (n = 5 studies), the use of self-reporting in measurement (n = 4 studies), and even for measuring the current use of the LMS. Finally, other limitations declared only once in the studies were the predictive power of the model, the theoretical model used, the existence of other unmeasured factors influencing LMS usage, reliance on a specific LMS, failure to consider mediation effects, and the type of data analysis conducted (see Fig. 4).

Fig. 4
figure 4

Declared limitations in the studies

Full size image

Results of RQ2. Theoretical models used in studies to explore determining factors of LMS usage by university teachers

To address the first question guiding this article, which pertains to the theoretical framework authors employed for the predictive models in their studies, 9 of them declared using the Technology Acceptance Model (TAM) as their foundational basis. Additionally, 2 studies stated using the Unified Theory of Acceptance and Use of Technology (UTAUT), 1 study utilized Change Management theory, and another study employed the Technology Related Stimulus-Response (TR-SR-TF) theory (see Fig. 5).

Fig. 5
figure 5

Theoretical models used to explore the determinants of LMS usage

Full size image

It is also important to highlight the variables that these models include, which are described and conceptualized in Table 4.

Table 4 Technology acceptance models identified in this review
Full size table

Although four different models were identified, the TAM model was the most recurrently included in the studies (69%), which originated from the Theory of Reasoned Action (TRA) (see Fig. 6). TRA explains behaviors, in this case, those of teachers, based on their beliefs and subjective norms (Ajzen & Fishbein, 1980). Thus, Davis (1989) employed TRA and stated that the use of a specific technology (in this case, the LMS) depended primarily on the mediation of the perception of ease of use and the perception of usefulness. These factors influence the attitude toward usage, which, in turn, determines the intention to use and ultimately, the actual use of the LMS.

Fig. 6
figure 6

Theory of Reason Action

Full size image

Results of RQ3. Variables included in empirical models on the intention or adoption of LMS

Concerning the variables used in the models, considering all those measured, including usage, intention to use, or both, six studies considered using 6 variables, three studies considered 8 and 9 variables, respectively, and one study considered 11 variables.

In addition to the variables corresponding to the TAM model (AU, BI, PE, PEOU, ATT), researchers included other determining factors related to the use or intention to use the LMS. These were grouped according to classifications made in previous studies corresponding to: Personal Factors, Technological Factors, Social Factors, and Institutional (Abdallah et al., 2016; Ziraba et al., 2020). Specifically, to predict the intention to use or the usage of the LMS, all 13 studies considered personal factors, 11 studies considered technological factors, 7 studies considered social factors, and 6 studies considered institutional factors (see Fig. 7).

Fig. 7
figure 7

Factors included in empirical models on intention or use of LMS

Full size image

Concerning the predictive power of the models tested in the studies, the predictive capability for intention to use (BI) and actual usage (AU) was separately analyzed. On average, the explained variance for BI was r2 = 0.55, with the study with the lowest explained variance at r2 = 0.41 and the one with the highest explained variance at r2 = 0.73. The explained variance for AU was r2 = 0.33, with the study with the lowest explained variance at r2 = 0.21 and the study with the highest explained variance at r2 = 0.54 (see Fig. 8).

Fig. 8
figure 8

Variance explained in the empirical models on BI and AU

Full size image

Results of RQ4. Type of instruments used to measure the intention and usage of LMS

Regarding how measurements of determining factors are conducted, all analyzed studies applied Likert-type self-report scales for assessment. Concerning the dependent variable when evaluating LMS usage, 7 studies assessed only the intention to use the LMS, 2 studies assessed only the current usage of the LMS, and 4 studies assessed both (intention of use and current usage of the LMS) (see Fig. 9).

Fig. 9
figure 9

Dependent variables included in empirical models

Full size image

Results of RQ5. Future lines of research declared in the studies

From the 13 articles analyzed, 34 recommendations for future research were identified and grouped into 9 categories. The most frequent mentioned in 62% of the articles was to advance in the improvement of the models by including other factors (ID: 2, 3, 6, 8, 9, 10, 11, 13). The next suggestion for future lines of research was to broaden the characteristics of the sample, with a frequency of 39% (ID: 1, 2, 6, 9, 12). Three other suggestions were to replicate the studies to confirm the findings (ID: 2, 3, 6, 10); Include moderating variables in the models that influence teachers’ intention to use LMS (ID: 1, 2, 6, 8); and propose new models on teacher adoption of LMS (ID: 5, 8, 9, 12) which had a frequency of 31% in the studies. Two other suggestions had a frequency of 23% in the studies, corresponding to conducting longitudinal studies (ID: 4, 5, 6) and implementing qualitative or mixed designs (ID: 10, 11, 12). With a frequency of 15% of the studies, it was suggested that progress be made in improving the instruments for measuring the actual use of the LMS (ID: 3, 7); and finally, 8% of the studies suggested considering mediation effects in the models (ID: 2).

Discussion

Hereafter, we present the discussion of the results associated with the four research questions of the studies. Additionally, limitations and future research directions in area of LMS acceptance by university teachers are presented.

Discussion of the results of RQ1. Characteristics of the studies regarding objectives, sample, and limitations

The most reiterated objectives of the studies were to determine the factors influencing the usage of LMS by university professors. Undoubtedly, this represents a significant advancement in research as it enables researchers to identify variables that are determinative and influence the adoption and usage of LMS. However, as some researchers have pointed out, there is a need to progress towards experimental studies, such as the development of interventions that promote the effective use of LMS in everyday teaching. This should go beyond the current usage primarily referring to a material repository and progress to other uses that involve maximizing its potential and benefits in teaching (Lobos et al., 2022; Mella-Norambuena et al., 2022).

Regarding the characteristics of the participants included in the studies, the average was 239 teachers, which could be considered a limitation of the studies. This aligns with the most frequently declared limitation by authors in the studies analyzed in this RSL. This is noteworthy considering the potential inference of findings to the population, as it could lead to unstable and inaccurate parameter estimates in the models. While various approaches (such as Monte Carlo methods) exist to estimate sample sizes, researchers still caution that studies often employ insufficient sample sizes for optimal estimation of predictive models (Sim et al., 2022).

Concerning the continent where these studies were conducted, Asia contributes the most productivity (61.5%), while no studies were identified in Latin America. This aligns with prior research indicating that South Korea is one of the pioneering countries in this field (Altinpulluk & Kesim, 2021).

Discussion of RQ2. Theoretical models used in studies to explore determinants of LMS usage in university professors

This review identified that the analyzed studies considered four different models as a basis for explaining the acceptance of technologies by faculty, with TAM being the most frequently used (69%) due to its simplicity and high validity. However, this model exhibits some limitations: (a) it focuses on predicting the use of technologies rather than the enhancement of user performance; (b) its ability to predict the actual use of technology has been questioned, considering that studies are based on self-report, assuming the limitations of instruments based on user self-perceptions; (c) research has been conducted by measuring TAM variables in relatively homogenous groups, limiting the possibility of generalizing results to real environments that tend to be more heterogeneous; (d) the predominantly quantitative nature of studies related to the model; (e) difficulty in reliably quantifying behavior in an observed investigation; (f) incapacity to address issues such as cost and structural imperatives that drive users to adopt innovation; (g) controversial heuristic value, and (h) limited practical value (Chuttur, 2009; Malatji et al., 2020). Consequently, its founders have attempted to reexplain it on numerous occasions, and various extensions (TAM 1, TAM 2, TAM3) have been developed, with continuous progress in revealing and incorporating new factors into the model’s core variables.

Finally, it is essential to highlight that this review identified that 57% of the studies tested a model to explain the intention to use the LMS and not the actual usage. In specialized literature, various theories have sought to explain human behavior, with the predominant focus being on the conscious cognitive aspect of this behavior (formation of intentions) and its impact on subsequent technology usage behavior, almost to the exclusion of other factors (Bosnjak et al., 2020; Lai, 2017). However, empirical evidence reveals that the effect size of this association is between low and moderate, an inconsistency termed the “intention-behavior gap” (Bhattacherjee & Sanford, 2009). A study focused on predicting technology usage behavior aimed to highlight an alternative perspective to the intention-to-use variable, emphasizing the equally important role of the automatic response known as a habit (Limayem et al., 2001). The authors suggest that people’s intentions during the early stages of adopting a particular technology may not have the same effect over time because it is habits, not intentions, that “govern” a person’s behavior. Therefore, it would be crucial to advance by considering other factors that explain behavior or directly measuring it.

Discussion of RQ3. Variables included in empirical models on the intention or adoption usage of LMS

This review systematized a variety of personal, technological, social, and institutional factors used in models to attempt to predict the intention to use or the usage of LMS. This aligns with findings from other studies; for instance, a study on university professors in Ghana concluded that institutional factors (responsibility of university authorities in providing ICT infrastructure), personal factors (periodic training), social factors (other incentives for faculty to appreciate and engage with the electronic learning environment) are crucial for the successful adoption of educational technologies like LMS (Asad & Malik, 2023). Another systematic literature review aimed at examining factors influencing instructors in the use and adoption of the Moodle platform for teaching university courses analyzed 58 publications, revealing that technological factors, social factors, human factors, and reinforcement factors were determinants in understanding this phenomenon (Ziraba et al., 2020).

The result of this review identified that all studies used structural equation modeling (SEM) for predictive models. SEM models have traditionally been used in the social sciences to test theories. Given the growing development and importance of predictive analysis and the capacity of SEM models to maintain theoretical plausibility in the context of predictive modeling, researchers have demonstrated the high predictive performance of estimators in SEM models (Evermann & Tate, 2016). Therefore, modeling through SEM data analysis techniques has gained popularity, and researchers have described its strengths and limitations (Tomarken & Waller, 2005). This provides a balanced perception of its characteristics. Among its strengths, SEM models enable the development of innovative analysis techniques such as latent growth, multilevel models, approaches to handle missing data, and those for normality, standing out as a broad framework for data analysis with flexible and unique capabilities. On the other hand, it is important to recognize potential limitations of SEM, such as the problem of omitted variables, the importance of lower-order components of the model, possible difficulties when considering well-fitted models, the inaccuracy of some commonly used empirical rules, and the importance of study design and its data (Sharma et al., 2021; Tarka, 2018). However, it is crucial to mention that this type of analysis has proven effective for predictive models, specifically for the intention to use LMS by teachers. Nonetheless, researchers may consider other data analysis techniques to predict actual usage, extracting data from the same LMS to account for usage behavior trajectories. In this case, data mining analyses, which allow for extracting pedagogically significant information from tracking data generated by the LMS, would be relevant (Macfadyen & Dawson, 2010).

Another noteworthy aspect of the presented results is that no study considered sociodemographic variables within predictive models. According to available literature, this might be due to the generalization of approaches in studies with large samples or substantial amounts of data. Methods cover not only normal continuous variables but also non-normal continuous and dichotomous variables. However, while methods considering the complex structures and characteristics of samples exist, traditionally, those for continuous data are predominantly used (Muthén & Satorra, 1995).

Discussion of RQ4. Type of instruments used to measure the intention and usage of LMS

All analyzed studies that measured intention to use, actual usage, or both used self-report scales, and no studies presented empirical measures (LMS analytics). Data obtained in research are an essential element in science communication, making it important to reflect on the limitations involved when using self-report instruments. Various procedures and validation standards have been proposed for tests measuring different constructs (Brutus et al., 2013). Nevertheless, a significant limitation of self-report measures relates to potential biases in the responses of study participants, in this case, university professors, due to social desirability (Bensch et al., 2019; Vésteinsdóttir et al., 2019).

A systematic literature review and meta-analysis addressed discrepancies between recorded digital media usage and self-report usage (Parry et al., 2021). The authors note that, in line with the results of this study, researchers primarily use self-report measures over more objective measures such as the quantity or duration of media usage, even though the validity of these self-reports remains unclear. The meta-analysis aimed to evaluate the agreement between these measures and based on 106 effect sizes, demonstrated that self-reported media usage correlates only moderately with recorded measures, self-reports were rarely an exact reflection of recorded media usage, and measures of problematic media usage show an even weaker association with usage records. These results cast doubt on the validity of findings based solely on self-reported measures of technological media usage.

Therefore, the findings highlight a significant discrepancy between self-reports on media usage and equivalent measures obtained through usage tracking techniques. The authors’ conclusion signals a clear limitation regarding the validity of self-report measures on media usage. In response to this, the authors suggest that those interested in measuring media usage face the question of how to measure media usage and how to proceed. Based on this, they offer the following recommendations: (a) researchers should stop assuming that self-reports are accurate indicators of actual usage behavior and adjust their inferences and conclusions accordingly; (b) the use of measures that closely approximate the targeted behavior, such as using tracking or logging services to measure media usage, is encouraged; (c) assuming error when research can identify factors influencing media usage.

Discussion of RQ5. Future lines of research declared in the studies

In the studies that were analyzed, although 9 categories of recommendations for future research were identified, the ones with the highest frequency reported in the studies (< 30%) were 5. First, the authors suggest the incorporation of other factors or dimensions as predictors within the models that may affect the acceptance of LMS, for example, flexibility of LMS use, support for LMS use, attitudes towards technology, previous experience with technology-based distance learning, system quality, perceived self-efficacy, facilitating conditions, personal innovation, access to technology, anxiety with the use of technologies, among other new variables; This coincides with recent studies that have pointed out the emphasis of further investigating the dominant factors that predict teachers’ adoption and use of LMS (Asamoah et al., 2024). Secondly, broadening the characteristics of the sample, for example, it would be interesting for future studies to consider integrating professors from private and public universities, full-time and part-time, teaching in conventional and distance modalities, experienced and inexperienced teachers in the use of LMS, from different countries and from different disciplinary areas. This future line has relevance with respect to the generalization of findings, a process that involves making inferences from particular observations, and, therefore, is recognized as a quality standard in quantitative research and the detailed consideration of participant characteristics is valued (Polit & Beck, 2010). Thirdly, researchers suggest replication of the studies, considering other technological contexts, other educational levels, other LMS and other countries. Coincidentally, scientific research in general has reinforced the idea that replicability is a prerequisite for determining valid conclusions, testing hypotheses with some unique contextual variants to evidence similar or dissimilar results (Borgsted & Scholz, 2021). The fourth future line highlighted by the authors, consisted of the suggestion to include moderating variables in the models that seek to predict the use of LMS, for example, the variables gender, the course taught by the teachers, age and experience of the teacher. This future line has been highlighted by expert researchers in methodology, since, the main objective of a moderation analysis, is to evidence the existence of differential effects of the independent variable on the dependent variable according to the moderator included, allowing more accurate explanations of the study phenomenon (Memon et al., 2019).

Limitations and future directions based on the findings of this study

The review was conducted considering specific methodological aspects, including the search period (from 2013 to 2023), and databases such as Scopus, WOS, EBSCOhost, and ERIC were explored. Additionally, studies in both English and Spanish were included. Therefore, it’s possible that studies included in other databases, outside the time range of this review, and articles developed in other languages could have been excluded. Furthermore, the conclusions are limited to those of theoretical scope.

Regarding future lines of research, this review identified various factors affecting the adoption and use of LMS. Therefore, it is recommended that researchers, with the necessary training, focus on promoting the use of LMS considering the identified factors. Additionally, there is a need to increase research productivity in Latin America, as this review did not find any studies from this region. Furthermore, addressing the limitations identified in the studies, particularly the sample sizes, is crucial. Future studies should consider the minimum required sample sizes for predictive models using techniques such as Monte Carlo simulations under different conditions, including simulations related to effect sizes, the number of indicators, the magnitude of factor loadings, and the proportion of missing data (Sim et al., 2022). For further research and deeper understanding of the phenomenon, both quantitative and qualitative studies should be conducted to investigate the impact of LMS on learning and user perceptions. Building upon the synthesis of studies that have analyzed the factors influencing the intention and use of LMS, interventions could be proposed to promote the use of this important technological tool among faculty in their everyday teaching.

Recommendations of good teaching practices to effectively integrate LMS platforms in teaching

Based on the findings and discussion of this study, the practices recommended to effectively integrate LMS into teaching are the following: (a) Design a syllabus that integrates combined and articulated activities between the LMS and those that are carried out with the teacher during classes, or in the modality that the course is being implemented (e-learning, b-learning, other); (b) Selecting an LMS tool that is relevant to the curricular activity, this practice is very useful because the teacher will find a valuable resource to improve student learning in accordance with the characteristics of the subject, contents and students, for example, a discussion forum will be useful for some pedagogical purposes, while for others a test with automated feedback will be more pertinent; and (c) Progression of the use of the LMS by the teacher, this is a practice that allows discovering and improving the use of the resources, for example, starting by applying formative evaluations in the LMS until reaching the use of more sophisticated tools.

Policy recommendations based on the findings of this review

Regarding policy recommendations at the institutional level, first, it is required to provide a solid structure of technological resources to ensure proper functioning and use of LMS, avoiding system failures that generate demotivation, frustration of teachers, and even lead them to give up using the LMS. For this, it is essential to periodically evaluate the LMS to meet the growing demands of teaching and ensure that it meets the functionality of the learning environment (Alshira’h, 2021). Secondly, a support system for teachers as users of the LMS is required, for example, a team of specialists to provide support to teachers on the use of the LMS; it can also be a checklist or list of frequently asked questions, useful to solve general or common doubts among users (Wade et al., 2024); or also generate new supports using artificial intelligence, such as chatbots to generate immediate responses to teacher queries related to the use of LMS, offering explanations and providing additional resources, they are even considered as virtual teaching assistants (Labadze et al., 2023). Thirdly, it is imperative that those responsible in educational institutions constantly offer training for teachers on the use of LMS, showing the potential they have, and it is desirable that these trainings include the application of some tool in the LMS of some curricular activities that each teacher is performing, thus providing a real experience (Putra et al., 2023; Turnbull et al., 2020). Fourth, it is recommended that educational institutions provide an LMS architecture with a simple navigation interface, with diverse didactic resources, facilitating communication and interaction, evaluation, feedback, flexibility, data analysis and meeting the needs of a quality pedagogical approach (Kilag, 2023; Oliveira et al., 2016). Fifth, that institutions incorporate into their policies the use of teaching and learning analytics generated by the digital footprint left by the use of the LMS, allowing the generation of dashboards for both teachers and students, which will allow the identification of study behaviors and process feedback for timely decision making. As a final note, in Higher Education Institutions it is recognized that educational paradigms are constantly changing and demands for new flexible, challenging, accessible and engaging learning experiences persist, and therefore it is required to successfully navigate the current landscape where the strengths of technology and pedagogy inevitably converge, for the same purpose, improves student learning performance (Granić, 2023), and quality of education (Kilag, 2023). In this context, universities are challenged in designing an innovative educational plan or framework on how to integrate the use of LMS to enhance teaching (Mershad et al., 2020). Consider an internal diagnosis that identifies the main hindrances to the use of LMS that includes phases to advance to achieve maximum efficiency in the implementation and use of LMS by their professors (Oliveira et al., 2016).

Conclusion

The background information presented in this study leads to the conclusion that: (a) There is a need to increase research that considers ‘real use’ and is not limited to investigating only the ‘intention to use’ this technological tool by university teachers; (b) Although there is consensus on the use of the TAM, its level of prediction warns of the need to propose new models to advance predictive power and challenges researchers to explore the integration of other possible factors and/or new proposals for theoretical models; (c) It is also necessary to specify models that differentiate the type of use of the LMS by the teacher, since they are platforms that group together different tools for communication, evaluation, repository, among others; (d) Undoubtedly, it is also necessary to advance in institutional proposals that support teachers by identifying profiles and promoting policies for their permanent and effective use, enhancing the success of their implementation in their subjects. All these aspects become essential, considering that many universities are rapidly migrating or installing blended learning and/or online teaching modalities, where the effective use of the LMS of these courses will be needed, as a protagonist scenario of the teaching and learning processes.