1 Introduction

Multimedia learning is a well-established field with the main goal of studying the design of multimedia documents that are more likely to support learning (Mayer, 2021). Therefore, studies of multimedia learning typically focus on learners. However, multimedia documents are not necessarily used autonomously, and they are often selected by teachers. Although there has been research on the integration of technologies into the classroom by teachers (e.g., Agyei & Voogt, 2011; Farjon et al., 2019; Sasota et al., 2021), there is little research on how teachers are in alignment with multimedia principles when using these kinds of materials and whether this intuitive or informed knowledge contributes to the use of technologies to create instructional material.

1.1 Cognitive Theory of Multimedia Learning

In the field of multimedia learning, Mayer (2005, 2014, 2021) and Schnotz (2005, 2014, 2021) developed two prominent theories. The integrated text and picture comprehension model by Schnotz (2021) is preferred when written text and static or animated pictures are involved, whereas the cognitive theory of multimedia learning (CTML) by Mayer (2021) is the one most commonly referred to for digital multimedia documents – including animations or other types of videos. In the present study, we are interested in such instructional materials, and thus present the later model in more detail.

CTML describes learners’ cognitive processing of information based on three assumptions derived from previous empirical and theoretical research: (1) information is processed through two separate cognitive channels – a visual and an auditory (Clark & Paivio, 1991), (2) the working memory has limited processing capacity (Baddeley, 1992), and (3) the learner has an active role in the processing of information (Moreno & Mayer, 2000). The dual channel assumption applied to multimedia learning by Mayer (2021) corresponds to the assumption that information is processed by learners based on the type of representation (i.e., verbal, or non-verbal format) rather than the sensory modality triggered by the stimulus (i.e., auditory, or visual). The latter are still included in the model but not as part of the actual information processing undertaken by learners. From the limited capacity assumption, Mayer (2021) draws limitations on the amount of information that can be processed by each channel. This limitation in cognitive load, in turn, affects how learners direct their attention toward information. The core of CTML lies in the assumption that learners “engage in cognitive processing to construct a coherent mental representation of their experiences” (Mayer, 2021, p. 61). Indeed, following CTML, active learning is achieved gradually through three cognitive processes: selecting (attending to relevant information in the media), organizing (mentally organizing the selected information in working memory), and integrating (connecting this information with prior knowledge stored in long-term memory). Building on these assumptions regarding cognitive processing, the CTML also presents a series of design principles to support the creation of instructional multimedia.

Multimedia design principles developed in the CTML are meant to support learners’ active processing by (1) minimizing extraneous processing (e.g., coherence, signaling – five principles), (2) managing essential processing (e.g., segmenting, modality – three principles), and (3) fostering generative processing (e.g., multimedia, voice, emotional design – eleven principles). Briefly, the first two sets of design recommendations go hand in hand with the assumption of dual channel and limited capacity underlying the CTML. These two assumptions are considered to be the backbone of the CTML. The dual channel assumption, developed by Paivio (1986) corresponds to the idea that there are two separate information processing channels for instructional materials. One channel is dedicated to visual and spatial information, and correspond to pictorial media, and one channel is dedicated to auditory and verbal media. The limited capacity assumption is derived from the working memory model of Baddeley (1992, 2003), and states that a limited amount of information can be processed simultaneously, by each of the channel. Their objective is to ensure that information is distributed between channels and does not overload them by guiding information processing. The third set targets learners’ motivation, and these design principles aim to foster deeper processing of information. Although there is little overall review of the effectiveness of the multimedia design principles derived from the CTML, a meta-analysis by Noetel et al. (2021) reveals that 11 design principles (most of which derived from CTML) positively influence learning outcomes, particularly when the materials are complex. Targeted meta-analyses for design principles applicable to video have been conducted as well, particularly for coherence—also referred to as the reduction of seductive details (Sundararajan & Adesope, 2020), and the implementation of meaningful segmentation (Rey et al., 2019). In line with CTML, a review by Sundararajan and Adesope (2020) shows that including extraneous detail in instructional multimedia is detrimental to learning, with a larger effect size when learners have little prior knowledge of the subject. A meta-analysis of segmenting conducted by Rey et al. (2019) also confirms the positive effect of user-paced segmentation in instructional animation. Again, an expertise reversal effect of prior knowledge was observed. To sum up, research on instructional multimedia points out that the importance of learners’ individual differences for both coherence and segmentation design principles. Indeed, learners with the least expertise in the subject are those most likely to benefit from instructional multimedia following the coherence and segmentation design principles, which aim to respect the dual channel and the limited capacity assumptions.

1.2 Investigating the Role of Teachers’ Alignment with Multimedia Learning Principles

Whether teachers are aware of the principles of the CTML and whether they apply them when selecting instructional multimedia materials in their teaching has rarely been investigated. As one of the few existing studies, Eitel et al. (2021) developed a questionnaire to assess whether in-service and pre-service teachers endorse common misconceptions about multimedia principles, including learning style myths (see the review from Kirschner, 2017). In this study, teachers’ misconceptions overlapped with multimedia principles, such as the principle itself (i.e., learning from text and pictures is better than from text alone) and the modality principle (i.e., it is best to present text orally when there is a picture). Eitel et al. (2021) found that more than 40% of in-service and pre-service teachers had misconceptions regarding learning styles, hemispheric isolation, and naïve summation; and above 95% had misconceptions in at least one of those three. However, the study does not provide information regarding teachers’ awareness and agreement with the assumptions forming the basis of the multimedia theories, or with actual design principles.

When it comes to instructional digital multimedia materials, some research (Oh et al., 2019, La Torre & Désiron, 2024) also show that they rarely are designed in accordance with multimedia principles. However, it is still unclear how teachers’ alignment with multimedia principles affect their frequency of use of these tools in their teaching. Indeed, few researches looked into how teacher select (Licht et al., 2024) and create (Ring & Brahm, 2022) instructional video through the scope of the multimedia learning principles framework. Based on empirical study, we could expect that lack of alignment could lead to misuse or inadequate creation of multimedia materials, which could eventually be detrimental for learners. A first step toward addressing this question lies in investigating the role of alignment with multimedia principles on frequency of digital multimedia creation.

1.3 Technology Integration to Support Multimedia Creation

Since available digital multimedia materials are not often following multimedia principles, teachers may, as with non-digital instructional materials, need to create their own. This process would likely require them to have the motivation, technical skills and technological tools to do so.

The integration of technologies into the classroom by teachers for their practice or that of their students has been the subject of several theoretical approaches, such as the “will-skill-tool” model (WST model) developed by Christensen and Knezek (2001). According to this model, teachers’ beliefs and attitudes toward digital technologies, their technical skills, and the technical equipment of the school are significant and positive predictors of ICT integration. Numerous studies have shown that these three enablers predict a high degree of variance in technology integration (e.g., Agyei & Voogt, 2011; Farjon et al., 2019; Grant, 2019; Knezek & Christensen, 2008, 2016; Petko, 2012; Pozas & Letzel, 2021; Sasota et al., 2021). However, the importance of the predictors varies depending on the context. Knezek and Christensen (2016) point out that the independent variable in which the local environment has the greatest variability is usually the strongest predictor. For example, Velazquez (2007) found different results depending on the region, in that skill was the strongest predictor in the US—a digitally more advanced country where there are few training opportunities for teachers—whereas tool was most important in Mexico, a digitally less advanced country where there was little access to technology. From these results, Velazquez (2007) concludes that in contexts with a low stage of technology adoption, will and skill are likely to replace tool as the most important predictor of integration. In line with this conclusion, a more recent study by Schmitz et al. (2022) shows that in technologically highly developed countries, will seems to be important for technology integration, whereas tool is the most relevant predictor of technology integration in technologically less developed countries.

Apart from the discussion surrounding the most important predictor of the original three enablers of the WST model, several extensions of the model were proposed to enhance its predictive power. For in-service teachers, the most important model extension was prompted by Petko’s (2012a) statement that pedagogy could be another relevant predictor of the model, paving the way for the “will, skill, tool, pedagogy” model (WSTP model; Knezek & Christensen, 2016). Originally, Christensen and Knezek (2001) considered the skill component to be teachers’ ability regarding the use of digital technologies. To operationalize the skill component of the WST model they used measures covering teachers’ self-reported technical skills. With the addition of the pedagogy component Knezek and Christensen (2016) argue that pedagogy, in contrast to skill, is not merely technical ability, but the ability to teach with digital technologies. They suggest that the pedagogy component can be measured with technological pedagogical content knowledge (TPCK) as depicted in the TPACK framework by Mishra and Koehler (2006). In this framework, Mishra and Koehler (2006) distinguish between three types of knowledge: technological knowledge (TK), pedagogical knowledge, and content knowledge. Furthermore, the overlaps between these knowledge domains are considered as well, with TPCK as the overlap of all three knowledge domains measuring the pedagogy component of the WSTP model. Consequently, skill mirroring purely technical ability according to Christensen and Knezek (2016) should be measured with TK from the TPACK framework (see for example Farjon et al., 2019). Although some studies indicate that objective performance tests would be helpful for testing the TPACK framework, most studies operationalized TK (skill) and TPCK (pedagogy) as teachers’ self-reported knowledge to handle computer functions (skill) and to teach with digital technologies (pedagogy; Drummond & Sweeney, 2017). Adding pedagogy operationalized as teachers’ self-reported ability to teach with digital technology substantially advanced the predictive power of the model (Knezek & Christensen, 2016). However, apart from investigating the effects of teachers` technical self-reported knowledge (TK) and their self-reported knowledge in teaching with digital technology (TPCK) it might also be worthwhile to extend the predictive power of the model by adding factual knowledge about for example the straightforward multimedia principles.

Combining the WSTP model with teachers’ alignment with multimedia learning principles could advance both research strands in two ways: (1) it could provide the negative of the use of digital multimedia in the classroom, with a teacher perspective, and (2) it could advance research on ICT by gaining insights on factors linked to digital instructional materials design as complementary to those related to technologies in general. For example, Backfisch et al. (2021) differentiated between quantitative aspects of technology integration, which refer to the frequency of students’ or teachers’ technology use in class, and qualitative aspects of technology integration. The qualitative aspects of technology integration distinguish between two dimensions: the level of teaching quality and the level of technology exploitation, which includes teachers’ capability to implement the distinct potential of educational technologies and to design an innovative learning environment. Regarding the level of technology exploitation, they argue that an innovative learning environment is, among other things, characterized by multimedia resources; thus, information can be presented through multiple and simultaneous representations, such as text, pictures, audio, or video. In other words, teachers need to know multimedia principles to use or create innovative learning materials with multiple representations (Backfisch et al., 2021).

1.4 Present Study and Research Questions

Following the argument of Backfisch et al. (2021), the present study investigated teachers’ alignment with multimedia principles, and considered the effects of will, skill, tool, and pedagogy scores together with this alignment on the creation of digital multimedia materials by teachers. We particularly focused on the following research questions: (RQ1) How well do teachers align with core multimedia principles? (RQ2) Which of WSTP factors and multimedia design alignment come into play for teachers who regularly create digital multimedia for their lessons?

2 Materials and Methods

2.1 Participants

Our analyses were based on data collected for the [anonymized for review] large-scale study. The target sample was all upper secondary school teachers in Switzerland, recruited through single-phase sampling. After eliminating one case that represented one school and could therefore not be used for further multilevel analyses requiring a minimum of three participants per cluster, the effective sample consisted of 2247 teachers (47.3% female, 50.5% male and 2.2% other) from 112 schools, of the three languages regions (DE = 1780, FR = 213, IT = 254). The mean age was 46 years (SD = 10 years), whereas the mean years of teaching experience were 16 years (SD = 10). Given that all teachers at upper secondary level were the target population of the study, the response rate for teachers was 6%, while the response rate for schools was 21.5% (for further information about the sample see Schmitz et al., 2024).

2.2 Measures

2.2.1 Instructional Digital Multimedia Documents

The outcome variable was a single item on the self-reported frequency of use of computer for teaching duties “For the creation of multimedia teaching materials (e.g. presentations, videos, animations)”. Answer scale ranged from “never” to “Almost every day”.

2.2.2 Alignment with Multimedia Processing and Design

Based on the literature presented above, four items were constructed to determine teacher’s alignment with core multimedia assumptions and design principles derived from the CTML (Mayer, 2021). Specifically, we were interested in those applicable to instructional documents, including videos (e.g., animations, and recorded presentations). Therefore, we opted to test for alignments with dual channel and limited capacity assumptions, together with the coherence and segmentation principles. The items were first constructed by the first and third author before being reviewed by an external researcher in the domain of multimedia learning. Minor changes in the phrasing were then implemented. Consequently, teacher’s alignment with multimedia processing and design was measured by asking teachers: “Do you agree with the following statements about learning with instructional multimedia materials (e.g., video, animated images)?”. Examples of materials were included to ensure that respondents had in mind the expected representation of multimedia materials. The four statements: “Learning is better when the auditory and visual senses are combined” (dual channels assumption), “Students can easily be overwhelmed by the amount of information in multimedia.” (limited capacity assumption), “Decorative details in instructional multimedia materials are detrimental to learning” (coherence principle), and “It is usually best to present multimedia content (e.g., videos) step-by-step when prior knowledge is low” (segmentation principle). The answer options were binary.

2.2.3 Will, Skill, Tool, and Pedagogy

Based on the theoretical model of WSTP (Knezek & Christensen, 2016; Petko, 2012), the “will” factor was operationalized as teacher’s general positive beliefs toward the use of technologies in class and measured with four statements (e.g., Students’ performance can be improved by using digital technologies) and a 5-point agreement Likert scale (1-not at all; 5-completely).

As the next component of the WSTP model, teachers’ self-reported technological “skill” was measured with questions on technological knowledge (TK) derived from the TPACK.xs scale (Schmid et al., 2020). This scale consists of three items covering overall familiarity with digital technology (e.g., I know about a lot of different technologies, with a 5-point rating scale ranging from 1 – not at all to 5 – completely).

The “tool” component of the WSTP model was measured with three items covering the quality of the available hardware and software for teaching and learning (e.g., Reliability of the network and of internet, 1 – very bad to 5 – very good). This scale derived from that used by Petko et al. (2018).

The “pedagogy” component of the WSTP model was assessed with the TPCK scale from the TPACK.xs instrument by Schmid et al. (2020; e.g.can choose technologies that enhance the content of a lesson, with a 5-point rating scale ranging from 1-not at all; 5-completely).

2.3 Data Analyses

Our study aims to provide an overview of teachers’ alignment with multimedia processing and design, as well as to determine the factors of alignment, based on the WSTP model from Knezek and Christensen (2016). Following previous studies, we also controlled for potential moderator variables: gender (e.g., Backfisch et al., 2021), teaching experience (Farjon et al., 2019; Knezek & Christensen, 2023) and school type, due to the nature of Switzerland’s schooling system (i.e., pre-university vs. vocational).

We used the data from a national survey in Switzerland in the context of the [DigiTraS II] project and conducted descriptive statistics with weighted data, since a full population sample was not achieved, and we attempted to consider the uneven distribution of the sample with regard to participants per school, school type and language region (Meinck, 2015). More detailed information and the weighting formula can be found in the article by Schmitz et al. (2024). However, for the inferential statistics, we used unweighted data, since weighted data tend to distort results in regression analysis (see Gelman, 2007). To consider the clustering of the data multilevel linear modeling analysis was performed with maximum likelihood estimation and the frequency of teachers` multimedia document creation as the dependent variable (null model). After that, another MLM analysis with maximum likelihood estimation was performed including the enablers of the WSTP model, multimedia principles, age, gender, teaching experience in years and school type as predictors into the analysis (model 1). Apart from the categorical variables, all variables (enablers of the WSTP model, gender, age and teaching experience in years) were group-mean centered as the major interest was to investigate teacher-level effects rather than school-level effects (Enders & Tofighi, 2007). Finally, we compared the null model`s fit with the fit of model 1. Analyses were computed using Jamovi (v.2.2 for Microsoft Windows) statistics and the R package (car).

3 Results

3.1 Teachers’ Alignment with the CTML

Descriptive analyses of the teachers’ answers to the alignment with multimedia questions (with “aligned” for yes and “unaligned” for no) from weighted data (Table 1) showed that about one-third of the teachers were in complete agreement with the assumptions and multimedia principles (28.90%), and one-third agreed with all but the coherence principle (31.17%). The remainder of the teachers showed diverse levels of alignment, and only 0.58% disagreed with all the principles. Descriptive per principle (Table 2) show that while teachers mostly are in alignment with the multimedia learning processing assumptions (dual channels and limited capacity), there are less so with the design principles, particularly that of coherence (42.2% aligned).

Table 1 Teachers’ alignment patterns with CTML (weighted data)
Full size table
Table 2 Teachers’ alignment, by multimedia item (weighted data)
Full size table

3.2 Predictors of Teachers’ Creation of Digital Multimedia

All rating scales used in the analysis showed excellent reliability (“will” Cronbach’s α = 0.89, “skill” α = 0.86, “tool” α = 0.72, “pedagogy” α = 0.79). Mean score index values were computed for all these variables. Descriptive for the creation of digital multimedia single-item and index for the WSTP are available in appendix (Table 4 and 5).

Teachers’ creation of digital multimedia materials can be guided not only by their ability and possibility to use digital technologies, but also by their knowledge of processing assumptions and design principles from the CTML. Therefore, we examined the predictors of creation of digital multimedia materials through the scope of the WSTP models and teachers’ alignment with the CTML. We performed MLM analysis to predict the creation of digital multimedia while considering the effect of the school intercept (random component). Based on our exploratory hypothesis, we compared a null model with one including teachers’ will (beliefs), teachers’ self-reported technological knowledge (skill, TK), and teachers’ access to tool (quality of access to infrastructure and uses in class), as well as pedagogy (TPCK) – following the WSTP model, and the four measures of CTML alignment (dual coding, limited capacity, coherence and segmenting). We also controlled for the effects of age, gender, teaching experience in years and school type (Fig. 1; Table 3).

Fig. 1
figure 1

Summary of Model 1 (* p < 0.050, *** p < 0.001)

Full size image
Table 3 Multilevel models explaining teachers’ frequency of creation of digital multimedia
Full size table

The intraclass coefficient (ICC) indicated that about 2.3% of the total variance was associated with differences between schools. The parameter (b = 3.87, SE b = 0.03) falls within the 95% confidence interval. Further, CI is above 0 and thus we can reject the null hypothesis. We observe that will, skill, pedagogy, and dual coding, as well as teaching experience in years and gender, are positively associated with digital multimedia creation. Female teachers with less teaching experience holding positive beliefs toward technologies, who report to be familiar with technologies and to know which technologies to integrate in their teaching, and are in alignment with the dual coding assumption, are likely more frequently creating their own digital multimedia materials.

4 Discussion

Our first aim (RQ1) was to provide an overview of upper secondary teachers’ alignment with a selection of multimedia learning processing and design principles. Descriptive statistics show that teachers differed according to their alignment with multimedia learning theoretical underpinnings. Almost one-third of the teachers are fully aligned with the four processing assumptions and design principles derived from Mayer’s CTML (2021), while another third agrees with all but the coherence principle. This suggests that the understanding of instructional multimedia materials of the majority of the teachers is rather close to the scientific theory of multimedia learning. Even though this might not be due to their true familiarity with this theory, there seem to be widespread common knowledge that – when presented with relevant statements – teachers can mostly distinguish the appropriate ones from others. Our study is therefore a first step toward understanding how knowledge of adequate use influences actual use. Nonetheless, it remains unclear whether this knowledge arises from experience and wisdom of practice using instructional multimedia materials or whether it can also come from other sources, such as teacher education or professional development. Moreover, this study employed binary variables to explore whether teachers are aligned with multimedia assumptions and design principles, rather than assessing the depth of their beliefs in these principles. Despite the clear definition of multimedia principles, this methodological approach leaves open the possibility that correct responses to binary items may have occurred by chance, with a 25% likelihood of guessing at least two items correctly. To address this limitation, future research could incorporate measures of teachers’ certainty, akin to the misconception scale introduced by Eitel et al. (2021).

The second goal of our research (RQ2) was to determine whether the WSTP model and teachers’ alignment with multimedia processing and design principles could predict teachers’ frequency of creation of digital multimedia materials. The model also revealed that variance explained includes gender and teaching experience, with female teachers with less teaching experience most likely to create digital multimedia materials. Surprisingly, it is not so much teachers’ knowledge about the multimedia design and processing principles that influence the creation of multimedia as much as it is their positive beliefs toward technologies, self-reported knowledge to use them and to integrate them relevantly in their lesson that guides the creation frequency. In other words, what seems to matter to teachers is not to know if what they are doing is beneficial to learners as much as whether they perceive that they can do it. The alignment with multimedia principles might predict the quality of created multimedia material or student learning instead of the increased frequency of multimedia material creation. The results of this study could thus also hint at a potential misuse of digital multimedia resources. However, more in-depth research is needed to assess this element, and its effect on learners.

These results are surprising, but it is important to stress they may be an artifact of the way the items were worded. Teachers who are more skeptical about digital technologies seemed to be more willing to agree with critically worded items such as those on limited capacity and coherence, whereas teachers who are more positive seemed to reject those items more frequently, even if they are correct from the CTML point of view. Whether this kind of alignment – expressed as agreement with questionnaire items in our study – is enacted in practice remains uncertain. Such self-report questionnaires also leave open the question of the background for such alignment, and future research could investigate how much of this alignment is informed and how much is intuitive knowledge of multimedia material design and processing.

Indeed, although there is little research on multimedia learning material available to or used by teachers, Oh et al.’s (2019) study on videos in MOOCS revealed that those rarely follow multimedia design principles. Overall, our results are also in line with previous research showing that personal factors, such as motivation or beliefs, can affect ICT use by teachers (Backfisch et al., 2021; Petko et al., 2017). Furthermore, Ertmer (2005) state that successful technology integration relies more on beliefs than on the availability of tools. Our findings seem to confirm this notion, as tools is not a significant predictor of digital multimedia creation. In high-tech countries such as Switzerland beliefs and self-reported knowledge in using digital technology and teaching with digital technology are reportedly more important than tools (Petko & Prasse, 2018). This is in line with the postulation of Velazquez (2007) that in contexts with a high stage of technology adoption, will and skill are likely to replace tool as the most important predictor of integration. Fittingly, the study by Schmitz et al. (2022) confirmed that in high-tech countries, will seems to be important for technology integration, whereas tool is the most relevant predictor of technology integration in countries at a pioneering stage of technology adoption. This supports the claim that while tools are necessary for technology use in general, there are not sufficient for frequent in high-tech environments. This element should be considered by policy makers, in that only providing tools does not secure more frequent technology use. However, the role of teachers’ alignment with multimedia principles also does not play much of a role, with only the dual channels assumption reaching significance.

Additionally, it’s crucial to emphasize that the belief measure used in this study concentrated solely on positive beliefs toward technology integration in the classroom. However, belief encompasses multiple dimensions (Tondeur et al., 2017), and employing a more comprehensive measure could yield different findings. For example, Antonietti and Giorgetti (2006) investigated beliefs toward computer-supported instructional tools by distinguishing beliefs regarding thinking processes, creativity attraction, facilitating effects, impact effects, and negative effects. From a multimedia design perspective, this would require not only looking at the core assumptions and principles of multimedia learning research, but also in affective elements (Schneider et al., 2022) such as emotional design (Endres et al., 2020). Finally, it is worth noting that our results are based solely on self-report measures, some of which requiring teachers to aggregate how often they created multimedia materials over time. Although the measures had excellent reliability, the data obtained remains subjective. Future study could consider additional objective measures of the creation of digital multimedia materials. For example, by collecting instructional digital multimedia materials produced over a year. This would also provide a more fine-grained picture of the use and creation of multimedia material by teachers.

5 Conclusions

Although many studies confirmed the core principles of effective multimedia material processing and design, few attempted to examine whether teachers know and follow these principles in their practice. This study provides a new perspective on this question by studying the alignment of teachers with four multimedia learning principles. One-third of the teachers sampled were in full alignment and another third were in partial alignment with multimedia learning principles, yet this alignment was not a strong predictor of digital multimedia creation, as only alignment with the dual coding assumption contributed to the model additionally to will, skill and pedagogy components of the WSTP model. Further, teachers’ teaching experience in years and gender were also significant predictors, with female teachers with less teaching experience more prone to create their own digital multimedia materials. Because the use of instructional digital multimedia materials is not only more common in classrooms but also a keystone of the digitalization of schools, teachers need the corresponding technological knowledge. Overall, a practical implication deriving from our results is the need to support teachers in building their over instructional digital materials. Teacher training and continued education programs could incorporate modules on dual coding theory and digital multimedia design into their curriculum. This would help future and in-service teachers understand the importance of using multimedia materials in their teaching practices when considering learning outcome, and provide them with the skills to create such materials effectively. These knowledge and skills are particularly important to allow then to design multimedia materials that best support learners processing, as it may not always be so strait forward (Fiorella et al., 2018; Renkl et al., 2017).

With the changes and growth of digital technologies, instructional multimedia documents cover a broader range of formats, with a particular increase in the use of video. There is surprisingly little evidence of teachers’ knowledge and practice in the use of instructional multimedia for learning (Oh et al., 2019), and more research is needed to define and address the gap between research and practice. In line with the work of Bétrancourt et al. (2012) and La Torre & Désiron (2024), there is a need for an in-depth content review of authentic material used by teachers, from which to build guides on how to adapt existing instructional material, or create new one, following multimedia design principles. This would provide researchers and designers with a better picture of how teachers’ alignment with multimedia principles affects their practice. Further, this study hints toward a need for more research looking into instructional multimedia design from the perspective of their informed use by teachers. Indeed, it may be that adequate integration of multimedia as ICT requires not only skills with technologies, but design skills that include a deeper understanding of multimedia learning. Moreover, there is need to only look into the frequency of multimedia integration in teaching, but also the quality of this integration.