Clicking with Confidence: Influence of a Student Co-Designed MOOC on Students' Emotions and Online Learning Self-Efficacy

The COVID-19 pandemic has taught us that being able to learn online is now a crucial life skill and cannot be left to chance. Pedagogical interventions are critical to support students in building their digital skills and confidence, given identified links between online learning readiness and academic success. Based on this premise, the purpose of this study is to investigate the influence of an online learning preparatory MOOC on students’ emotions and levels of online learning self-efficacy (OLSE). The paper begins by illustrating how the design and delivery of the MOOC had the potential to provide participants with the necessary mastery experiences, vicarious experiences, verbal persuasion, and affective regulation opportunities to evaluate and develop their online learning self-efficacy beliefs. Students starting or continuing their higher education online because of COVID-19 were encouraged to take the MOOC as part of their development. Students’ self-reports of their emotion and OLSE were compared pre-and post-MOOC. Paired sample t-tests found significant differences in all four dimensions of OLSE following participation in the MOOC. Participants reported higher levels of Self-Efficacy Navigating Technology, Self-Efficacy Managing Time, Self-Efficacy Learning at a Distance, and Self-Efficacy Communicating Online. Additionally, changes in learners’ emotions were observed post-MOOC. Most participants reported feeling anxious about online learning before the MOOC. This changed, however, post-MOOC, with the majority feeling positive and hopeful about online learning following the two-week course. The paper concludes by discussing the implications for practice.

While most educators want to click beyond the COVID-19 pandemic, it remains a challenging and valuable learning experience.The pandemic tested our resilience and contributed to renewed interest in the role of emotion in teaching, learning, and assessment.The need to support learners with a "pedagogy of care" has been paramount throughout the pandemic (Buckley-Marudas & Rose, 2021; Burke & Larmar, 2020).However, a kinder and softer form of online pedagogy by itself does not address the anxiety, trepidation, and very real challenges students face in becoming effective online learners.Learning online is not the same as learning in a traditional classroom and requires mastery of a different skill set.The pandemic has taught us that these skills cannot be assumed, and that carefully designed interventions are required to acknowledge and support students' affective experiences as they move to a new modality and develop their digital capacity to be successful online learners.This paper reports the impact of a MOOC co-designed with students to support the development of online learning self-efficacy (OLSE) in response to the COVID-19 crisis.The findings demonstrate how the MOOC as a support intervention, anchored in a wider life-skill framework, played a valuable role in the context of the pandemic where most students learned online out of necessity rather than by choice.

Literature Review
In the past, education research has focused primarily on the role of cognition and the rational brain in learning at the expense of affect (Beirne, 2020).Fortunately, in more recent years, research has taken an "affective turn" (Zembylas, 2021).In this section, a scoping review will briefly explore the current understanding of the role of emotion and self-efficacy in educational contexts, and previous intervention research in these areas will also be examined.

Emotions in Education
Awareness of the role emotions play in academic settings (i.e., academic emotions) has been growing for several years (O'Regan 2003;Pekrun et al., 2002).However, the global health pandemic has produced greater interest and created more questions about the relationship between emotions and learning in higher education contexts (Raccanello et al., 2022;Katzman & Stanton, 2020).An expanding body of literature overwhelmingly recognises academic emotion as a crucial factor that can constrain or facilitate the learning experience.Studies have shown that emotion can have an impact on the learning process through attention, memory, motivation, and self-regulation (Pekrun, 2011).
Emotions are frequently classified by their valence as either positive (e.g., hope, excitement) or negative (e.g., anxiety, frustration) and generally, positive emotions are seen to be more conducive to learning (Tan et al., 2021).However, the situation is often more complex and can depend more on whether the emotion is activating or deactivating (Pekrun, 2006).In some cases, negative activating emotions such as frustration or anxiety can enable learning and achievement (Pekrun & Perry, 2014;Rowe & Fitness, 2018).
The current study aligns with a dynamic perspective of emotions, which views emotions to be under constant change, varying situationally and over time (Dörnyei, 2009), and emerging from person-environment interactions (Pekrun et al., 2011;Schutz et al., 2006).It can thus be deduced that a change in the mode of learning, especially under the unique external conditions of the COVID-19 pandemic, had the potential to be emotionally overwhelming, eliciting strong emotional responses among students that may influence their participation and performance.
Previous research indicates that anxiety is a problem faced by online learners, especially firsttime online students (St Clair, 2015) and that other negative emotions such as fear, anger, and helplessness have been found to be higher in online students compared to students in traditional classes (Butz et al., 2015).
Understanding learners' emotions under the unique and challenging conditions of transitioning to higher education and an online mode of delivery is an important step in supporting students during and after this process.Studies focusing on emotion and well-being in higher education contexts during the pandemic are not uncommon (e.g., Raccanello et al., 2022;Visser & Law-van Wyk, 2021), however, the current study seeks not only to identify students' emotions but also track how they change following participation in an online learning preparatory MOOC.

Self-Efficacy in Education
Self-efficacy is posited to be an important component of learning success.As a key element of social cognitive theory, self-efficacy "refers to beliefs in one's capabilities to organize and execute the courses of action required to produce given attainments" (Bandura, 1997, p. 3).Notably, it is an individual's belief in their own ability and does not necessarily equate to the reality of actual ability.Self-efficacy is hypothesised to play a key role in human agency, affecting decision making, effort, perseverance, and resilience (Bandura, 1997).People with high self-efficacy for a task are more likely to participate, expend more effort, and persist at that task in the face of difficulties (Bandura, 1997).
The current study focuses more specifically on online learning self-efficacy (OLSE) which refers to one's confidence to perform academic tasks successfully in an online environment.While less advanced than research in relation to self-efficacy for face-to-face learning, existing studies indicate that OLSE is associated with successful online learning experiences.Studies have shown that OLSE can be a predictor of achievement (Ergul, 2004;Joo et al., 2013), retention (Holder, 2007;Yukselturk et al., 2014), perceived learning (Alqurashi, 2019;Wright et al., 2006), satisfaction (Artino, 2008;Landrum, 2020;Shen et al., 2013), and engagement (Pellas, 2014;Prior et al., 2016).The ability to self-regulate and motivate oneself to engage in the learning process is important for all students, independent of the learning environment.The importance, though, is amplified in an online or distance learning environment where the instructor is not always visually or synchronously present and learners have to take greater responsibility for the management and control of their own academic progress (Milligan & Littlejohn, 2014;Stephen et al., 2020;Terras & Ramsay, 2015).
Like emotion, self-efficacy beliefs are not fixed traits but vary across activity domains and situational conditions (Bandura, 1997).The multifaceted nature of online learning suggests that OLSE will vary across the different tasks or situations associated with the online learning context.Many researchers, however, have noted that a large proportion of studies have focused on computer self-efficacy or self-efficacy in relation to technology use, disregarding other aspects of the online learning experience (Alqurashi, 2019;Shen et al., 2013;Zimmerman & Kulikowich, 2016).Shen et al. (2013) demonstrated that OLSE is multi-dimensional by identifying five dimensions: (i) self-efficacy to complete an online course, (ii) self-efficacy to interact socially with classmates, (iii) self-efficacy to handle tools in a Course Management System (CMS), (iv) self-efficacy to interact with instructors in an online course, and (v) selfefficacy to interact with classmates for academic purposes.Similarly, Zimmerman and Kulikowich (2016) developed the Online Learning Self-Efficacy Scale (OLSES) which includes other aspects of online learning, such as self-directedness, communication skills, timemanagement skills, and technology use.An adapted version of the OLSES is the scale used in the current study because it was deemed to be more holistic and relevant to the learning context.

Developing Self-Efficacy Through Pedagogical Interventions
The important links between self-efficacy and learning success has meant that improving self-efficacy via teaching, learning support, and curriculum design has been the focus of several studies.For example, Bartimote-Aufflick et al. (2016) identified 17 intervention studies published since 2000 that compared self-efficacy under different conditions or monitored selfefficacy over time.Findings from the studies identified showed that self-efficacy could be improved because of completing a course, participating in a particular learning activity, or when certain teaching strategies were employed.The studies were conducted among postgraduate and undergraduate students across a wide range of countries.Strategies used to promote self-efficacy included drawing on the affordances of multimedia and e-learning material, facilitating peerinteraction, providing additional resources for challenging concepts, and encouraging students to share personal experiences (Bartimote-Aufflick et al., 2016).Collectively, the findings of this review indicate the potential of pedagogical or learning interventions in enhancing self-efficacy among higher education students.
A few studies have looked specifically at the impact of study skills courses on selfefficacy (Macaskill & Denovan, 2013;Rodriguez & Armellini, 2017;Wernersbach et al., 2013).Rodriguez and Armellini (2017) report on the role of a study skills MOOC in increasing selfefficacy among a sample of undergraduate and professional learners.Albeit a small sample (n = 32), they found statistically significant increases in both general self-efficacy and self-efficacy in relation to specific study skills upon completion of the MOOC.This raises the question of whether similar results can be found in larger scale interventions.While limited research exploring this question is available, in a brief analysis of the components of MOOCs, Hodges (2016) indicates that MOOCs can play a role in enhancing self-efficacy at scale.However, the potential of MOOCs in terms of supporting learner self-efficacy needs further investigation, which the current study seeks to address.
Underlying many of these studies is the proposition from Bandura (1986), as part of the social cognitive theory, that individuals develop self-efficacy beliefs by interpreting information regarding their own capabilities and that this information can stem from several sources.In the next section we will look more closely at these information sources and how the design of the MOOC facilitated each one.

The Learning Context The MOOC
A Digital Edge: Essentials for the Online Learner is a MOOC developed by Dublin City University's (DCU) National Institute of Digital Learning (NIDL) in collaboration with the Irish Universities Association (IUA).The MOOC was rapidly developed and first offered in September 2020 to address the challenges faced by college and university students as they began or continued their higher education online during COVID-19.Hosted on the FutureLearn platform, the two-week course requires approximately 3 hours of learning per week.The content is structured around four main themes: Ways of Thinking and Ways of Working (Week 1), and Tools for Working and Tools for Thriving (Week 2).For a more detailed breakdown of each week's themes see Table 1 and Table 2.The course is open to learners all over the world and aims to support them to learn how to learn online.A distinctive feature of the course is the codesign and facilitation by students who share their tips, advice, and first-hand experiences about effective online learning.Learning Design.The design and development of curricula and learning activities in the MOOC are aimed at improving the lifelong learning competences of students.To achieve this, the course is anchored in the LifeComp Framework (Sala et al., 2020) and the Learning Compass 2030 (OECD, 2019).Figure 1 presents a visual representation of how these models were synthesised.The instructional approach adopted in the MOOC was influenced by the design principles underlying the FutureLearn platform, which promote visual learning and learning through conversation and storytelling (FutureLearn, 2018).

Figure 1 A Digital Edge: Life Skills Framework
Central to the design of the MOOC was the need to facilitate the four information sources posited by Bandura to influence self-efficacy: i) enactive mastery experiences, ii) vicarious experiences, iii) verbal persuasion, and iv) physiological and affective sources.Table 3 describes each of these four information sources and details how the course format and content and the corresponding learner experience over the two weeks could potentially facilitate each one.(Bandura 1997).
The MOOC was an authentic online learning experience.Participation in an online course can give students tangible evidence that they can learn successfully online.It was expected that the non-formal, low risk nature of the course would encourage participation, even among less experienced learners.

Vicarious Experiences
Social comparisons, which allow individuals to perceive their abilities in relation to the successes or failures of others, such as peers and role models.Observing others, with whom they can identify, succeed at a task can provide individuals with a sense of confidence in their own ability to perform similar tasks (Bandura 1997).
The MOOC was co-designed and cofacilitated by students who had prior experience learning online.Participant's vicarious experience was encouraged through the use of real-life examples and testimonials from these students.Testimonials were included in the course content as quoted text and audio clips.Student facilitators were also available for the 2 weeks to answer questions and share their experiences in the discussion forums at the end of each step.Participants were encouraged to ask questions and draw on the knowledge and experience of the student facilitators.By observing the successes of their peers, learners can generate efficacy beliefs that they too can obtain success through persistence and effort.

Verbal Persuasion
Positive encouragement and feedback from others.Realistic affirmations from others can boost self-efficacy perceptions.Often considered a weaker source of self-efficacy as compliments can often be given loosely without substantiation (Bandura 1997).
Positive verbal persuasion was provided through the discussion forums by both the instructors and the student mentors to help participants believe that they can cope with difficult situations when learning online.

Physiological and affective states
The influence of our body's physical and emotional reactions to certain situations and tasks on self-efficacy.Experiences of anxiety, stress, arousal, fatigue, for example, and their accompanying physical manifestations, can leave a student with a low perception of their ability to persist in a task.
Well-being, emotional regulation and co-regulation were key components of the pedagogical framework of this MOOC.Polls incorporated at four points throughout the course encouraged learners to reflect and share how they were feeling about learning online (See Fig 2 .).The polls supported self-regulation among the learners and raised awareness of emotion in online learning.The poll format allowed participants to respond anonymously while also being able to see how their peers were feeling.

Research Questions
The following research questions were examined in this study: 1. Can an online learning preparatory MOOC improve OLSE among learners?2. Do students' emotions toward online learning change following participation in the MOOC?

Participants
Participants were individuals enrolled in the first iteration of the MOOC, The Digital Edge: Essentials for the Online Learner, in September 2020, who completed both a pre-course and post-course questionnaire.In total, 135 participants completed both questionnaires, forming the sample for this study.Among this sample, 98 were females (73%), 34 were males (25%), 1 person responded as "other" (0.7%), and 2 chose the option "prefer not to say" (1%).Over twothirds of participants were entering their first year of college or university but undergraduates in subsequent years and postgraduates were also included in the pool of respondents.
Overall, 6,598 individuals enrolled in this first iteration of the MOOC, but it was not possible to determine how the study's sample compared to this wider population of learners due to European data privacy rules and platform data sharing limitations.

Research Design and Procedure
A repeated measures design was adopted, using pre-and post-course questionnaires to i) identify initial levels of OLSE and emotional responses to online learning, and ii) trace changes in learners' self-efficacy beliefs and emotions related to their learning experience.Two online surveys were administered.Both surveys were embedded in the course for voluntary student response.
The pre-and post-course questionnaires were designed to capture data pertaining to participants' demographics, emotional states, and online learning self-efficacy.

Demographic Variables
Participants were asked to self-report demographic information, including gender and academic status.For academic status, participants were asked to report whether they were a new student (i.e., entering their first year of college or university) or a continuing student (i.e., in subsequent years of college or university).

Emotional States
A list of six discrete emotion states was used to collect data pertaining to student affect.The list was derived from previous studies on learning-centric emotions (Beirne, 2020;D'Mello, 2013;Pekrun et al., 2011).Participants were asked to select the emotion they experienced most strongly.Responses were subsequently used to generate a new binary variable with the categorically positive emotions coded as 1 and categorically negative emotions coded as 0.

Online Learning Self-Efficacy
Self-efficacy was measured using a scale adapted from Zimmerman and Kulikowich (2016).Adaptation involved the removal and re-wording of items to facilitate a global audience and the learning context in question.Participants were asked to rate 20 online learning-related tasks using a 5-point Likert scale.A rating of 1 signified that they believed that they would not be able to perform the task at all; a rating of 5 signified that they believed that they could perform the task extremely well.

Results
The Dimensions of Online Learning Self-Efficacy An exploratory factor analysis was conducted to identify the dimensions of OLSE.An oblique rotation method was applied because each self-efficacy dimension was assumed to be correlated with one another.Factor loadings below .30were suppressed.Four factors with eigenvalues > 1 were extracted from the factor analysis.The four-factor solution accounted for 67% of total variance and the four factors were internally consistent yielding Cronbach Alpha's of .878,.892,.848,and .900,respectively.None of the four subscale reliability analyses revealed items whose removal would increase the subscale's alpha coefficient.The Kaiser-Meyer-Olkin measure of sampling adequacy (.893) and Bartlett's Test of Sphericity (p=.001) both indicated that factor analysis is appropriate.Items loading on component 1 predominantly relate to using and navigating technology for learning.Items loading on component 2 are mainly associated with self-motivation and time management.Items loading on component 3 concern communicating using technology.The items loading on component 4 relate to learning at a distance.The component correlation matrix is shown in Table 5.The components resulting from this factor analysis were used to create four subscales: Component 1, Navigating Technology; Component 2, Time Management (this includes items relating to motivation); Component 3, Communicating Online; Component 4, Learning at a Distance.Evidence of convergent and divergent validity was examined using correlational techniques.The results of these correlation analyses are shown in Table 4 and 5.

Can a Preparatory MOOC Improve OLSE Among Learners?
To answer this research question, paired samples t-tests were conducted to compare changes between pre-and post-MOOC scores for each of the four dimensions of OLSE.Results from these analyses, presented in Table 6, reveal that there are significant differences between the means for each dimension of OLSE.The post-MOOC scores are significantly higher than the pre-MOOC scores in all cases.Effect sizes for these findings range from moderate to large (Cohen, 1988).The standardised difference between the pre-and post-course means for selfefficacy to communicate online is notably large.

Do Students' Emotions Toward Online Learning Change Following Participation in the MOOC?
To answer this research question, descriptive statistics collected pre-and post-MOOC were analysed.Before the MOOC, frequency counts for each emotion revealed that "anxiety" was the emotion felt most strongly by most participants (54%) and approximately 62% of the participants reported negative emotion overall (see Table 7).However, the distribution changed among the post-MOOC data."Hope" was the emotion felt most strongly by the plurality of learners (33%) on completing the MOOC and nearly twothirds of participants reported positive emotion overall (66%).Figure 3 further illustrates the change in participants' overall sentiment (positive or negative) towards online learning pre-and post-MOOC.

Figure 3 Overall Changes in Sentiment Pre-and Post-MOOC
Drawing on the binary sentiment data, Figure 4 presents a more detailed breakdown of individual sentiment changes following MOOC participation.These results show that 37% of participants changed their sentiment towards online learning post-MOOC.Nearly all these people changed from negative to positive sentiment.

Discussion
The purpose of this study was to explore emotion and OLSE among higher education students in the context of COVID-19 and examine the influence of an online learning preparatory MOOC on these variables.
All four dimensions of OLSE were enhanced after experiencing a two-week online learning preparatory MOOC.Following participation in the MOOC, participants reported feeling more confident in their ability to i) navigate the technological aspects of learning online, ii) communicate in an online environment, iii) manage their time while learning online, and iv) learn without being in the same room as their instructor or peers.The biggest increase was for self-efficacy to communicate online.Notably, learning through conversation is a key design principle underpinning the course and the wider FutureLearn platform (FutureLearn, 2018).These findings align with existing research that has shown that OLSE can be improved over time (Bartimote-Aufflick et al., 2016) as well as the notion that competence and confidence can be improved through authentic mastery experiences (Pajares, 1997).The question remains, however, as to whether the improvement can be attributed to participation in the MOOC.Nevertheless, the findings can be seen to extend the previous narrative exploring the potential of MOOCs (Hodges 2016) or online orientations more generally (Abdous, 2019) in enhancing selfefficacy at scale.
When reflecting on the move to online learning because of the pandemic, students experienced the full range of emotions listed.While nearly a third of students reported feeling hopeful and excited about learning online, anxiety was the strongest emotion for the majority.Other negative emotions such as hopelessness and anger were also reported, but by fewer people.The prevalence of anxiety at this time is not surprising and was also found in other studies which focused on students' emotions during the pandemic (Chien et al., 2022;Novara et al., 2022).A comparison with post-MOOC emotion reports indicates a shift in students' emotions after the two weeks.Following the MOOC, most students reported feeling positive about online learning, with hope being the emotion participants felt most strongly.For over a third of participants, their strongest emotion changed from a categorically negative emotion before the MOOC to a positive one after.These findings contribute to existing research that explores how situational factors such as course content and design can impact emotion (D'Mello et al., 2014) and, more specifically, indicate the potential of MOOCs in scaffolding the online learning experience.
Overall, these findings are important, given that higher self-efficacy and positive emotion are associated with learning success (Bartimote-Aufflick et al., 2016;Tan et al., 2021).

Implications for Online Teaching and Learning
Interventions to foster positive emotion and self-efficacy constitute an area of interest for higher education institutions and practitioners as they endeavour to encourage and support student well-being, confidence, and, ultimately, learning success.Cleveland-Innes et al. (2016) highlight the important role of online learning preparatory courses stating that: … there is a need to pay attention to the skills students need to develop to become successful online learners.It also underscored that prior levels of education and even previous course attempts may not have helped students to develop these essential skills (p.596).
While other MOOCs and short courses exist, it is important to note that our findings are interpreted in the context of COVID-19 and the MOOC described in this paper, and we would caution against making broader generalisations.In that regard, based on the findings in the current study, we recommend that the MOOC, The Digital Edge: Essentials for the Online Learner, be used as a pedagogical intervention to improve OLSE among higher education students when aiming to enhance learner's confidence communicating when learning online.Notably, since the MOOC's development, it has now become a core feature of Dublin City University's student induction program, with over 2500 students having completed the course.While there is merit to embedding or contextualising such interventions at an institutional level, the MOOC with its co-facilitation by students also shows the value of more learner-driven and self-directed initiatives.In this respect, there are two key lessons from the MOOC experience.First, the MOOC demonstrates the value and importance of incorporating a strong student voice in the learning design process.Students need to be part of the design team as co-authors rather than simply the end audience.Second, the MOOC confirmed that students should not be left to their own devices in learning how to learn online and in developing their self-regulatory dispositions and positive emotions towards studying in digital environments.Such interventions as described in this paper can play a valuable role, especially when intentionally designed and anchored in a framework to support emotion, digital well-being, and online learning readiness.

Limitations
This final section highlights some limitations that should be considered.First, as in other studies with similar themes, there is a limitation in using self-reported instruments as they are subject to measurement errors and personal biases.Second, this investigation was nonexperimental in nature which limits the interpretation and generalisability of the results.While the post-course questionnaire was embedded at the end of the MOOC, the researchers have no indication of the extent to which participants completed the MOOC.Similarly, we cannot account for any other interventions or student experiences that could have influenced their selfefficacy between completing the pre-and post-course questionnaires.Most institutions developed new resources to support their students throughout the crisis and the impact of this material is impossible to determine.In addition, the effect of the pre-test as an intervention could have had a bearing on the score of the post-test.Also, low response rates, attributable to the longitudinal nature of the study and high dropout rates in MOOCs more generally, along with the self-selecting sample, limit the representativeness and generalisability of the findings.Finally, it was not possible to track the stickiness of the positive changes reported by students over the first few weeks and beyond of their online study, which raises a methodological challenge for future researchers.

Conclusion
There is every indication that online and hybrid learning options in higher education will become more prevalent in the post-pandemic era.Furthermore, digital literacy and the ability to learn online will become increasingly relevant to life-long learning in the workplace as well as assisting students to become active contributors to society.It is our responsibility as educators to support and care for students as they strive to negotiate the demands of learning in an everevolving digital society.The strong association established in the literature between learning success and both emotion and self-efficacy indicates that these variables are important to consider in helping students be successful and persevere in the face of challenges.This study, therefore, sought to examine the effect of a MOOC on students' emotions and OLSE beliefs and the results indicated positive changes for both constructs post-MOOC.
Investigating the effects of a learning intervention on students' OLSE and emotion extends knowledge in both fields but also provides actionable learnings from the pandemic for higher education educators and policy makers.A systematic review of pandemic-related online learning readiness literature conducted by the authors in conjunction with the current study highlighted the need to bridge theory and research with practice.It showed limited knowledge of the literature and that, while many studies arising from the pandemic had headline recommendations for practice, well-designed interventions to support student success for online learning were scarce (Beirne et al., 2022).
Overall, this study offers positive new insights for practice, but there is still much to be done in this area.More longitudinal research is needed to investigate self-efficacy and emotion in online learning contexts.Understanding the temporal dynamics of these constructs and related learner and situational variables could hold important insights for pedagogical interventions.The inter-relationships between these variables and online learning readiness also warrants further inquiry (Chien et al., 2022).

Figure 4
Figure 4 Distribution of Sentiment Changes from Pre to Post-MOOC

Table 1
Overview of Course Week 1 ARTICLE 1.11 How do I work?POLL 1.5

Table 3
Sources of Self-Efficacy Development in the MOOC

Table 4
Results of Exploratory Factor Analysis Rotation Method: Promax with Kaiser Normalization. a a. Rotation converged in 6 iterations.

Table 5
Component Correlation Matrix

Table 6
Paired T-Tests and Cohen's d Statistics for the Four OLSE dimensions

Table 7
Emotion Descriptive Statistics