Behavioral intention and media richness in the UTAUT
As noted in Sect. 2.5, the original model of the UTAUT includes PE, EE, SI, and FC. PE is similar to the perceived usefulness of the TAM; it explains how students can use MOOCs to improve their learning performance. EE is similar to the perceived ease of use in the TAM and describes the degree to which individuals believe that MOOCs are easy to use . SI includes opinions and pressure from classmates, instructors, social groups, and friends regarding the use of MOOCs, which may increase students’ adoption tendencies and use of MOOCs.
Since MOOC platforms depend on multimedia technology, investigating the role played by media richness theory in the acceptance and adoption of MOOCs can provide greater insight into the factors that influence individuals’ intention to adopt MOOCs. Hew and Kadir  reported that media richness has a significant effect on behavioral intention. Rich media help users communicate more quickly and improve their performance. Media richness also supports users during e-learning. Therefore, we hypothesize that media richness affects learners’ perceptions of the possibility of using MOOCs and propose the following hypotheses:
The behavioral intention to utilize health informatics MOOCs is influenced by performance expectancy.
The behavioral intention to utili-ze health informatics MOOCs is influenced by effort expectancy.
The behavioral intention to utilize health informatics MOOCs is influenced by perceived media richness.
The actual adoption of health informatics MOOCs is influenced by perceived media richness.
Behavioral intention and the actual use of MOOCs
A positive and significant relationship between behavioral intention and learners’ use of MOOCs has been reported . According to , once learners are convinced to adopt a system, they are likely to do so. This relationship has been confirmed by  and . Furthermore, behavioral intention mediates the relationship between SDT and the actual use of MOOCs . Wan et al.  identified the mediating effect of behavioral intention and continued use intention. Previous studies have also reported the mediating role of behavioral intention in the relationships between self-determination, UTAUT, and task technology fit models and the intention to use MOOCs [10, 39]. We posit a positive relationship between behavioral intention and the actual use of MOOCs and thus propose the following hypothesis:
Behavioral intention has a positive effect on the actual use of health informatics MOOCs.
SDT, behavioral intention, and media richness
As noted in Sect. 2.6, the three attributes of SDT are autonomy, relatedness, and competence. Perceived relatedness may increase the motivation of users in a context featuring a supportive culture, a supportive environment and autonomy. Relatedness enhances belongingness and leads to a state of enjoyment for users of technologies and systems . Learners are also influenced by people to whom they are connected, and relatedness creates bonds among learners or users in the workplace for mutual benefit. With regard to social and educational well-being, relatedness helps students investigate their behavior when using MOOCs . Perceived relatedness is linked with autonomy, and it enables learners to make decisions regarding whether to use MOOCs. Previous studies have found a positive and significant relationship between perceived relatedness and the behavioral intention to use MOOCs [1, 38, 39].
Autonomy refers to users’ feeling that they have the freedom to adopt a new technology or system independently. With regard to the current study, autonomy indicates that students have the right to decide whether to use MOOCs and to enroll in any subject of their choice that is relevant to their field. MOOCs allow learners to choose their favorite subjects without limitations due to time, schedule, or boundaries. Several previous studies have found a correlation between perceived autonomy and behavioral intention (e.g., , ).
Perceived competence refers to individuals’ perception and belief that they are capable of accomplishing a specific task. High competence leads to a high level of motivation and encourages learners to investigate and attempt new things. Perceived competence can be affected by ease of use, language difficulty, connectivity, and digital skills. In e-learning environments, learners should be familiar with how to use digital platforms and interact via different communication channels, which is explained by CET [36, 38]. Based on the preceding discussion, we propose the following hypotheses:
The behavioral intention to utilize health informatics MOOCs is influenced by perceived autonomy.
The behavioral intention to utilize health informatics MOOCs is influenced by perceived relatedness.
The behavioral intention to utilize health informatics MOOCs is influenced by perceived competence.
This study followed a quantitative cross-sectional approach, and data were collected via an online survey. A cross-sectional design is considered a time- and cost-effective approach. A structured questionnaire was adopted from previous studies to collect the primary data from the respondents. The questionnaire consisted of items related to constructs of the UTAUT, self-determination theory and channel expansion theory. All items were measured on a seven-point Likert scale. Details are given in Sect. 3.5. Prior to data collection, ethical approval was obtained for data collection. Informed consent was also obtained from the respondents. The aim of the study was explained to the respondents, who were informed that the data would be used only for academic purposes and that the identity of individuals and organizations would be kept confidential. Furthermore, the reputation of individuals and organizations would not be harmed. The respondents were given three to four days to complete the questionnaire. A total of 170 questionnaires were distributed to students and teachers, and 145 completed responses were received and used in the analysis.
The participants in the survey were faculty members and students from public and private universities in Saudi Arabia. The data collection took place from January to February 2022. A nonprobability convenience sampling technique was used to select the study sample. This sampling technique has been widely used in the social, management, and learning sciences [39, 42]. Related studies investigating MOOCs [1, 6, 17] have also used a convenience sampling technique to collect data. Prior to data collection, ethical approval was obtained from the scientific research ethical committee of Qassim University (Protocol # 21-14-11).
To test the framework and hypotheses proposed in the current study, the first part of the questionnaire obtained the participants’ responses to items pertaining to each theory. The items were measured on a seven-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). The UTAUT comprises six constructs: PE (4 items), EE (4 items), SI (5 items), FC (5 items), behavioral intention (3 items), and actual use (4 items). SDT consists of three constructs, perceived relatedness (7 items), perceived autonomy (7 items), and perceived competence (6 items), which were adapted from  and . Media richness (6 items) was adapted from . The second part of the questionnaire focused on demographic information, such as gender, age, role as a teacher or student, sector, and level of education.
The data were analyzed using structural equation modeling (SEM). Partial least squares SEM (PLS-SEM) was used to examine the reliability and validity of the scales. For this purpose, measurement models were developed and tested using PLS-SEM, and a structural model was developed and tested to test the hypotheses. The purpose of the measurement model was to evaluate the composite reliability, average variance extracted, and discriminant validity using heterotrait-monotrait (HTMT) ratios and Cronbach’s alpha coefficients . The threshold criteria were as follows: composite reliability (CR) > 0.70, average variance extracted (AVE) > 0.50, Cronbach’s alpha > 0.70, factor loadings > 0.70, and HTMT ratios < 1. The structural model was tested using the bootstrapping method.