Student Engagement, Learning Environments and the COVID-19 Pandemic: A Comparison between Psychology and Engineering Undergraduate Students in the UK

Abstract

This study compared student learning engagement from two UK departments exploring their participation in face-to-face and synchronous online learning environments. Overall, 446 undergraduate students from Psychology (soft/non-Science, Technology, Engineering, and Mathematics (STEM) discipline) and Electrical Engineering and Electronics (EEE) (hard/STEM discipline) completed an online questionnaire over the second semester of the 2020–2021 academic year, where the teaching was mainly online. The questionnaire included validated scales regarding teaching and students’ characteristics and an open-ended question regarding the role of learning environments. There was a significant difference between the two learning environments in both departments, with most of the students believing that they were better engaged with their learning process in face-to-face environments (quantitative analysis). Additionally, the thematic analysis of student qualitative responses revealed that online student engagement was influenced by (1) Behaviour, (2) Affective, and (3) Cognitive challenges (i.e., additional workload, lack of communication and distractions in the home environment) and opportunities (i.e., the effective use of study time and online content through interactive learning environments). This study could assist academics, university policymakers, and researchers to understand student engagement alongside learning environments, reconsidering the opportunities and challenges that were gained from online learning due to the COVID-19 pandemic.

1. Introduction

Across the world, researchers in higher education have discussed over many years what “student engagement” is and how it enhances the student learning experience [1,2]. Zepke (2018) [3] has recognised that there is not one clear definition for this term, due to the diversity and complexity of this area as well as the various dimensions in which researchers have studied it. For example, the most discussed dimensions are related to cognitive (i.e., learning goals, self-efficacy, deep learning), affective (i.e., learning environment, teachers), behavioural and social (i.e., participation and interaction) student engagement [4,5]. Many researchers have discussed how student engagement may be influenced by the sociocultural (emotional, cognitive, and behavioural) perspective [6] or individuals’ characteristics, a feedback loop between teacher, peers, and the learning environment [7], or the role of reflexivity influenced by the tasks and social interactions in a specific learning environment [8]. Frequently, researchers have confused student engagement with motivation, which are both highly connected, but student engagement arises from motivation [9,10]. Other researchers have explored how student engagement was linked to academic performance [11,12,13,14] or to teachers’ involvement [15]. Finally, many researchers have aimed at conceptualising student engagement with the educational interface, including the psychosocial constructs of self-efficacy, emotions, belonging, and well-being [16], the role of feedback regarding the receiver’s and sender’s behaviour with message characteristics [17] or the impact of self-regulated learning on cognitive and socioemotional collaborative interactions supported by student-led and teacher-led tasks [18].

The integration of technology into university curricula has gradually increased in recent decades offering new opportunities and challenges for student engagement when considering teachers’ beliefs and attitudes, prior experience, and modes of transformative learning [19]. Blended learning approaches (combination of face-to-face and online activities) were mainly followed by the on-campus courses to incorporate flexibility, stimulate interaction between teachers, peers, and course material, facilitate students’ learning processes, and foster an effective learning environment [20,21] where undergraduate student behavioural and emotional engagement was affected by how various digital learning tools are used by teachers to support student learning [22]. Halverson and Graham (2019) [23] have further studied the relationships between learners’ characteristics, instructional methods, and learning outcomes with cognitive and emotional engagement in a blended learning environment proposing a further need to explore these associations. Additionally, students’ perceptions about learning activities in a blended learning environment may influence cognitive and emotional student engagement [24]. The role of educational technology was also explored following the bioecological model and its relevant recommendations for macro-, exo-, meso-, and micro-levels under the perspectives of activities, environment, peers, and teachers [2,25]. These studies discussed how two student engagement frameworks (1. bioecological model and 2. cognitive, affective/emotional, and behavioural framework) could be combined to integrate a technology-enhanced learning environment interacting with the short and long learning outcomes on a social and academic level.

The role of academic performance as a learning outcome has also been discussed by many educational researchers regarding student engagement. For example, Dunn and Kennedy (2019) [26] have explored motivations utilising the emotional, cognitive, and behavioural framework, which pointed out the importance of intrinsic motivation and social media use on student engagement and performance, whereas technology use was related to extrinsic motivation. Another model regarding technological use found a positive effect on self-directed learning (individual characteristics) and student engagement, where academic performance has been indirectly affected by technology via self-directed learning [27]. Vo et al. (2017, 2020) [28,29] have explored the difference between “soft and hard” courses (Science, Technology, Engineering, and Mathematics (STEM) versus non-STEM courses) regarding various elements, such as clear goals, feedback, instructor support, material quality, instructor facilitator, content presentation, collaborative learning, and grades. Evidence suggests that the blended learning approach has a larger contribution to STEM students’ performance than students from soft disciplines, with the factors of clear goals, material quality, and collaborative learning being amongst the most significant predictors of student performance. According to the literature, online and face-to-face activities, teacher support, and feedback processes may also play a crucial role in successful blended learning design [30,31]. Baragash and Al-Samarraie (2018) [32] have further examined the role of Learning Management System (LMS)-based and web-based learning on student experience and academic performance over the face-to-face learning process that has taken place in a formal university environment. For each of the three delivery modes, the level of interactions between teachers and students differed, whilst students over-valued the web resources to “obtain a quick and easy information to assist in their mastery of the course content” (p. 2096). A recent literature review examined the opportunities and challenges that LMS, synchronous and social media tools have provided to collaborative learning through the real-time interactions between students and self-regulated learning through the reflection process [33].

The studies on student engagement have been mostly referred to as blended learning, which involved face-to-face teaching with synchronous communication between teachers and students and the integration of online learning technologies into a physical environment [34]. Various digital tools could support blended learning approaches, supporting face-to-face and online activities [35]. Galvis (2018) [36] has explored the factors that may influence Higher Education institutions with their decision-making process on blended learning. One of the reasons is that the learning environments are expandable through the various digital resources, enhancing the interactions between teachers and students and/or among students. Blended learning approaches, including online activities, can also increase flexibility, allowing students with work and home commitments to follow courses [37]. However, by working in an online environment, students may receive less support from their teachers compared to a face-to-face environment and may be less engaged with their learning [38].

Due to the recent global COVID-19 pandemic, many Higher Education institutions across the world shifted their teaching delivery process from a face-to-face lecture in a blended learning environment to a fully online learning one, allowing students to continue learning in a safe environment [39,40]. Students could study online content asynchronously in their own time and space, while they could synchronously interact with their teachers and their peers via online conferencing platforms [41]. From a very early stage of the pandemic, researchers have pointed out several challenges and opportunities for teaching and learning related to technology, pedagogical challenges, socio-economic factor, digital competence, inadequate interactions and student self-study skills including the well-being and mental health of teachers and students [42,43,44,45]. For example, researchers have explored the role of student and teacher digital competence for the rapid change of practice by a blended learning approach, mostly delivered on campus to the integration of online conferencing tools (i.e., Zoom (5.12.0, Zoom Video Communication, Inc., San Jose, CA, USA), Microsoft Teams (1.5.00.17656, Microsoft, Washington, DC, USA) [46]. Others have questioned the teaching shift by raising concerns about its impact on student engagement [47], with synchronous and asynchronous online teaching to be reconsidered under the perspectives of learning activities, feedback, and digital platforms [48]. Zeng and Wang (2021) [49] examined the teaching elements in the design of online courses during COVID-19, such as feedback, teacher support, and social presence, that might play a role in student satisfaction (highly related to academic performance).

The Higher Education sector faced critical challenges during the COVID-19 pandemic with subsequent lockdowns, which disrupted University life [50], but there were only limited studies that have explored the behavioural, cognitive and affective student engagement dimensions. Salas-Pilco, Yang and Zhang (2022) [51] identified the main characteristics of student engagement from these tripartite dimensions in online learning were related to digital skills development, technological issues, emotional support, student self-regulation, perceived self-digital literacy and development of soft skills (i.e., communication, collaboration and teamwork skills). However, while previous studies explored student engagement focused on one learning environment (blended face-to-face or online), the teaching process and student characteristics in both environments remain unexplored.

This study aims at exploring the effects that learning environments (blended face-to-face learning before the COVID-19 pandemic and online learning over the pandemic) have on university student engagement under the behavioural, affective, and cognitive dimensions regarding student academic performance, student characteristics, and the teaching process. Students from two departments, Psychology (soft/non-STEM discipline) and Electrical Engineering and Electronics (EEE) (hard/STEM discipline), participated in this study to further explore any potential difference between students’ preferences for synchronous face-to-face or online learning environments. Figure 1 illustrates the elements that were explored in this study regarding teacher-student interactions.

In this study, students from two departments were invited to complete an online questionnaire which included Figure 1 elements. The main assumption of this study was that the students for both departments felt more engaged when they followed a face-to-face lecture rather than online. This assumption was made because these students had chosen to follow an on-campus course and the COVID-19 pandemic had disrupted their studies, forcing them to follow the teaching in an online environment.

This study includes the student academic performance as an indicator of their engagement with their courses. Therefore, it was important for the aim of this study to initially explore whether there was any significant difference between academic performance in the two departments and/or student preferences in learning environments.

Research Question 1 (RQ1): Are there any differences between academic performance and/or students’ preferences in face-to-face and synchronous online learning environments in each department?

It has been previously discussed that the teaching delivery elements (i.e., teacher support, feedback and learning goals) could influence student engagement, but it has been explored whether there is any difference between these elements, the face-to-face and online learning environments, and how students from two departments (Psychology and EEE) have been engaged. Specifically, this study assumed that the teaching process might not have a significant difference in student engagement between the two departments when similar settings were followed.

Research Question 2 (RQ2): Are there any differences between various aspects of the teaching delivery process (i.e., learning goals, feedback, teacher support) with face-to-face or online learning environments in each department?

Additionally, as has been discussed in the Introduction part, the student characteristics (i.e., self-regulation, self-efficacy, test anxiety and background) can influence the behavioural, affective and cognitive engagement dimensions. This study assumed that there was a difference between student characteristics and their preferences in learning environments for both departments, with most students being engaged in the face-to-face environment because students had weak interactions with their teachers in online environments.

Research Question 3 (RQ3): Are there any differences between students’ preferences in synchronous learning environments (face-to-face or online) and their characteristics (i.e., self-efficacy, behavioural self-regulation) in each department?

Finally, this study also assumed that students felt more engaged with the face-to-face teaching process, whereas they felt disengaged with their learning process due to the COVID-19 pandemic, as they were forced to follow online teaching. Thus, it should be explored whether the COVID-19 pandemic influenced their learning experience.

Research Question 4 (RQ4): Does the COVID-19 pandemic influence students’ learning environment preferences?

2. Methods

2.1. Experimental Conditions and Participants

This investigation took place in the Psychology and Electrical Engineering and Electronics (EEE) departments at a research-intensive university in the Northwest of England during the 2020–2021 academic year. Both departments followed similar learning approaches before and during the COVID-19 pandemic by integrating learning technology tools into their teaching approaches to support two-way live communication between students and teachers through online polling systems (i.e., Kahoot! (Kahoot!, Oslo, Norway), PollEverywhere (PollEverywhere, San Francisco, CA, USA) and Padlet (Nitesh Goel and Pranav Piyush, San Francisco, CA, USA) [52]. Over the COVID-19 pandemic period, both departments used web conferencing tools (i.e., Zoom and Microsoft Teams) to synchronously communicate with students, as well as pre-recorded videos and online discussions for asynchronous teaching and communication between students and teachers. For both departments, students worked on various online coursework over the semester and completed an online summative exam period at the end of the semester, whereas before the COVID-19 pandemic, the final exams took place in a physical environment.

Overall, psychology and EEE undergraduate students for the three undergraduate levels of studies (N = 446, Males = 130, Females = 316, Mean age = 20.6, and Standard Deviation (SD) age = ±3.97) fully completed an online questionnaire on student engagement over two months (March–April 2021).

Table 1. The Number of Participants and Response Rate.

Year	Psychology (Response Rate %)	EEE (Response Rate %)
1st-year undergraduate students	189 (49.1%)	37 (27%)
2nd-year undergraduate students	75 (22.7%)	36 (21%)
3rd-year undergraduate students	71 (21.8%)	38 (13.1%)
Total students responding(% of all cohort)	335 (32.2%)	111 (18.5%)

provides information about the participants in both cohorts per year of studies. In the 2020–2021 academic year, the first-year students attended online lectures with little experience with face-to-face practical classes. This was due to the COVID-19 pandemic restrictions. On the contrary, second-and third-year students had previous experience with face-to-face learning environments over the 2018–2019 and 2019–2020 academic years. An opportunity sampling design was utilised as participants were recruited via emails, social media, discussion boards, word of mouth, and the Department of Psychology recruitment website. The University of Liverpool’s Research Ethics Committee approved this study.

The survey was conducted online, using a 10–15-min-long questionnaire hosted on the Qualtrics web-based survey platform (www.qualtrics.com, accessed on 11 February 2021). A participant information sheet and a consent form were provided to the participants before anonymously completing the online questionnaire. They also received a debrief after the questionnaire completion.

2.2. Questionnaire

The questionnaire had four sections consisting of 70 multiple choice questions and one open-ended question. The first section (Section 1) included 6 questions about student preferences in learning environments. The answer choices reflected their learning preferences and their interactions with teachers and/or their peers. For example, participants were asked whether they mostly preferred to attend lectures in a physical environment (i.e., lecture theatre, classroom) supplemented with technologies such as lecture recordings and online voting systems (i.e., PollEveryWhere, Kahoot!) or synchronously in an online environment (i.e., Zoom/Microsoft Teams) in their own space supplemented with asynchronous online activities such as watching pre-recorded videos at their own time.

The second section (Section 2) consisted of 38 questions that aimed to assess students’ engagement with their learning processes, including factors such as clear goals, teacher support, teacher feedback, teacher facilitation, online activities, synchronous session, collaborative learning, and learning outcome. The work of Vo, Zhu, and Diep (2020) [29] inspired this part of the questionnaire, which explored the difference in student engagement in blended learning environments for soft and hard disciplines.

The third section (Section 3) included 25 questions that assessed students’ learning characteristics and habits, such as course utility, self-efficacy, test anxiety, surface strategy, source diversity, and negative habits. This questionnaire has been inspired by the Motivation for Strategies and Learning Questionnaire (MSLQ), as it measures both attitude and behaviour [53]. The updated version, entitled “The Digital Strategies and Motivated Learning (DSML)”, released by Cho and Summers (2012) [54] incorporated the changes in current blended learning environments, and a shorter version has been used in many other studies [46,55]. Participants responded to Sections B and C on a 7-point Likert Scale (1 = Strongly disagree/Not at all and 7 = Strongly agree/Very great extent).

The last section (Section D) included two questions to assess the effects of the COVID-19 pandemic on students’ learning process. The first item assessed to what extent (5-point Likert-scale, 1 = A great deal to 5 = Not at all) participants felt that they had developed new learning habits to cope with the changes of moving to an online learning environment due to the pandemic. This question would be used to explore whether the COVID-19 pandemic was a moderator for student preferences in learning environments. The second item was an open-ended question that allowed students to provide their comments on their behaviour/attitudes towards their current learning experience and how they felt this affects/impacts their learning process (e.g., how their learning motivation/engagement was affected, and how their ability to work from home and the COVID-19 restrictions influenced their learning). The questionnaire has been uploaded to the ZENODO online repository platform (Organisation Européenne pour la Recherche Nucléaire, CERN, Switzerland) (https://doi.org/10.5281/zenodo.6982919, accessed on 20 August 2022).

3. Results

A one-way Analysis Of Variance (ANOVA) showed that there was a statistically significant difference between the two departments in student performance (F(1, 444) = 20.24, p ≤ 0.001) (

Table 2. Participants’ average grades per year per discipline (Mean (±SD)).

Year of Studies	Psychology	EEE
1st Year	60.2 (±9.42)	62.7 (±10.77)
2nd Year	58.4 (±6.62)	65.7 (±12.38)
3rd Year	60.1 (±9.56)	64.8 (±8.73)
Total average	59.8 (±8.91)	64.4 (±10.67)

). However, ANOVA statistical analysis and Turkey post hoc revealed no statistically significant difference between the three years of studies for the Psychology students (F(2, 332) = 1.17, p = 0.311), or between the three years of studies for the EEE students (F(2, 108) = 0.77, p = 0.465). Thus, we could assume that the students from the three years of studies for each department had a similar level of engagement with their undergraduate studies.

In the first part of the questionnaire, the participants were asked to self-report their preferences in various learning environments regarding the delivery process, interactions with their peers, and interactions with their teachers.

Table 3. Students’ responses to questions related to their preferences when teaching was delivered into two different learning environments to support lecture sessions.

	Synchronously in A Face-to-Face Environment (i.e., Lecture Theatre, Classroom) Supplemented with Technologies such as Lecture Recordings and Online Voting Systems (i.e., PollEveryWhere, Kahoot)	Synchronously in An Online Environment (i.e., Zoom/Microsoft Teams) Supplemented with Asynchronously Online Activities such as Watching Pre-Recorded Videos at Own Time	Chi-Square (α = 0.05)
Preferences per department
Psychology	61.8%	38.2%	χ2(1, 446) = 0.37, p = 0.545
EEE	58.6%	41.4%
Preferences per Year of Studies
1st-year undergraduate students	60.6%	39.4%	χ2(2, 446) = 0.30, p = 0.862
2nd-year undergraduate students	63.1%	36.9%
3rd-year undergraduate students	59.6%	40.4%

α is the limit of the significance level, χ²(a, b) is the variance between groups, and p is the significance level.

illustrates the differences between the two departments and the years of studies regarding the students’ preferences in the two types of learning environments. A chi-square analysis revealed no statistically significant difference between the two departments and the years of studies. Additionally, the majority from each department preferred to interact with their peers (Psychology: 74.9%, EEE: 67.4%) and teachers (Psychology: 67.5%, EEE: 63.1%) in a physical learning environment.

To further explore the first research question regarding the difference between student performance and their preferences in learning environments for the two departments, a two-way ANOVA statistical analysis was conducted. Overall, there was no statistically significant interaction between the effects of department and student preferences on student performance, F(1, 442) = 0.36, p = 0.550. However, the simple main effects analysis showed that student performance was significantly influenced by student preferences in the learning environment (p < 0.001) and the difference in departments (p = 0.044).

Table 4. Participants’ average grades per year per discipline per preferences in the learning environment (Mean (±SD)).

Teaching Delivered in Two Different Learning Environments	Psychology	EEE	Total
Synchronously in a face-to-face environment (i.e., lecture theatre, classroom) supplemented with technologies such as lecture recordings and online voting systems (i.e., PollEveryWhere, Kahoot)	60.0 (±9.15)	64.1 (±11.21)	61.0 (±9.82)
Synchronously in an online environment (i.e., Zoom/Microsoft Teams) supplemented with asynchronously online activities such as watching pre-recorded videos at own time	59.5 (±8.54)	64.9 (±9.96)	60.9 (±9.22)

illustrates the student performance descriptive statistics between departments for face-to-face and online learning environments.

A two-way ANOVA statistical analysis compares the students’ responses regarding their learning environment preferences and teaching delivery elements (

Table 5. Comparisons between the two departments related to teaching delivery elements and student preferences in two types of learning environments.

Teaching Variable	Face-to-Face Environment (i.e., Lecture Theatre, Classroom) Supplemented with Technologies such as Online Voting Systems (i.e., PollEveryWhere, Kahoot)		Online Environment (i.e., Zoom/Microsoft Teams) Supplemented with Asynchronously Online Activities such as Watching Pre-Recorded Videos		ANOVA between Disciplines (α = 0.05)
	Psychology M (SD)	EEE M (SD)	Psychology M (SD)	EEE M (SD)
Cognitive Engagement
Clear goals (3 items, a = 0.839)	3.98 (±1.12)	4.26 (±1.21)	4.14 (±1.36)	4.56 (±0.93)	F(1, 442) = 0.275, p = 0.600, n2 = 0.001
Teacher support (6 items, a = 0.877)	4.17 (±1.05)	4.36 (±1.20)	4.17 (±1.19)	4.60 (±1.04)	F(1, 442) = 0.923, p = 0.337, n2 = 0.002
Learning outcome (3 items, a = 0.928)	2.06 (±1.20)	2.54 (±1.68)	3.59 (±1.89)	4.20 (±1.70)	F(1, 442) = 0.133, p = 0.715, n2 = 0.000
Affective Engagement
Synchronous session (5 items, a = 0.878)	4.19 (±1.20)	4.53 (±1.21)	4.51 (±1.24)	5.03 (±1.14)	F(1, 442) = 0.465, p = 0.496. n2 = 0.001
Teacher Facilitation (6 items, a = 0.921)	4.12 (±1.10)	4.33 (±1.33)	4.23 (±1.27)	4.58 (±1.09)	F(1, 442) = 1.867, p = 0.173, n2 = 0.001
Behavioural Engagement
Online activities (5 items, a = 0.924)	3.94 (±1.20)	3.91 (±1.38)	4.46 (±1.31)	4.68 (±1.20)	F(1, 442) = 0.823, p = 0.365, n2 = 0.002
Collaborative learning (6 items, a = 0.878)	3.49 (±1.24)	3.65 (±1.42)	3.29 (±1.18)	4.16 (±1.28)	F(1, 442) = 6.392, p = 0.012, n2 = 0.014
Teacher feedback (4 items, a = 0.869)	3.83 (±1.25)	4.25 (±1.27)	3.89 (±1.46)	4.39 (±1.37)	F(1, 442) = 0.072, p = 0.788, n2 = 0.000

a: Cronbach’s Alpha, α: the limit of the significant level, M: Mean, SD: Standard Deviation, F(a, b) is the variance value, p: significant value, n²: size effect, 7-point Likert scale (1: not at all, to 7: very great extent).

). The analysis shows a statistically significant difference in collaborative learning between the two departments. The EEE students used to work with other peers to complete laboratory and project assignments, while Psychology students mainly worked independently. For all the rest teaching delivery elements (i.e., learning outcomes, teacher facilitation, teacher support), there was any statistically significant difference in the learning environment due to the department. However, the effect sizes revealed significant differences in students’ engagement with the online learning environment. A potential explanation of this finding should be related to students’ habit of working online after the COVID-19 pandemic (Psychology students: 2.75 ± 1.095, EEE students: 2.77 ± 1.157, selecting their responses from a 5-point Likert scale where 1: A great deal to 5: Not at all).

Multiple regression analyses explore whether students’ academic performance for each department and each learning environment, which was considered as an overall indicator of student engagement with the course, was associated with the teaching elements (i.e., clear teaching goals, teacher support, teacher feedback, teacher facilitation, online activities, synchronous session, collaborative learning, and learning outcomes). The regression model predicted approximately 3% of the overall variance in total Psychology student performance for those students who had the main preference for a synchronous face-to-face environment, F(8, 325) = 0.85, p = 0.561, with none of the teaching variables to significantly contribute to academic performance. The regression model predicted approximately 7% of their academic performance for psychology students who preferred an online learning environment, F(8, 325) = 1.16, p = 0.331, with none of the teaching predictors being significant. Thus, there was not any difference in how these teaching predictors were associated with the Psychology student performance in each learning environment, keeping the level of student engagement at the same level. Based on the EEE student responses, the regression model predicted approximately 6% of the overall variance in total EEE student performance for those students who preferred a face-to-face learning environment for the teaching delivery process (F(8, 37) = 0.46, p = 0.884), with none of the teaching predictors being significant. Finally, for those EEE students who had the main preference for an online learning environment, the regression model predicted 15.5% of the overall variance in their total academic performance, F(8, 37) = 0.85, p = 0.569, with none of the teaching variables to significantly contribute to their academic performance. Thus, the teaching elements were higher associated with academic performance in online environments rather than in a face-to-face environment for the EEE undergraduate students. This is in alignment with the previous findings of this study regarding the EEE students’ preferences in the online learning environment.

A two-way ANOVA statistical analysis compares students’ responses regarding learning environment preferences and individual characteristics (

Table 6. The differences between students’ preferences in learning environments regarding the student characteristics.

Student Individual Characteristics	Face-to-Face Environment (i.e., Lecture Theatre, Classroom) Supplemented with Technologies such as Online Voting Systems (i.e., PollEveryWhere, Kahoot)		Online Environment (i.e., Zoom/Microsoft Teams) Supplemented with Asynchronously Online Activities such as Watching Pre-Recorded Videos		ANOVA Analysis between the Disciplines (α = 0.05)
	Psychology M (SD)	EEE M (SD)	Psychology M (SD)	EEE M (SD)
Cognitive Engagement
Variety of sources (4 items, a = 0.809)	2.83 (±1.04)	2.69 (±1.03)	2.63(±0.89)	2.79 (±1.11)	F(1, 442) = 1.847, p = 0.175, n2 = 0.004
Surface learning (3 items, a = 0.774)	2.92 (±1.13)	2.69 (±1.17)	2.77 (±1.04)	3.05 (±1.24)	F(1, 442) = 4.151, p = 0.042, n2 = 0.009
Self-efficacy (4 items, a = 0.800)	3.17 (±1.06)	3.16 (±1.33)	3.00 (±1.07)	3.26 (±1.07)	F(1, 441) = 1.164, p = 0.281, n2 = 0.003
Affective Engagement
Test anxiety (4 items, a = 0.871)	2.38 (±1.21)	2.51 (±1.52)	2.28 (±1.17)	2.82 (±1.42)	F(1, 442) = 2.024, p = 0.155, n2 = 0.005
Behavioural Engagement
Behavioural Self-regulation/negative habit (7 items, a = 0.818)	2.80 (±0.99)	3.04 (±1.07)	3.13 (±1.16)	3.27 (±1.15)	F(1, 441) = 0.195, p = 0.659, n2 = 0.000
Course Utility (3 items, a = 0.815)	2.25 (±1.06)	2.48 (±1.07)	2.25 (±0.96)	2.68 (±1.17)	F(1, 441) = 0.811, p = 0.368, n2 = 0.002

a: Cronbach’s Alpha, α: the limit of the significant level, M: Mean, SD: Standard Deviation, F(a, b) is the variance value, p: significant value, n²: size effect, 7-point Likert scale (1: not at all, to 7: very great extent).

). The analysis revealed a significant difference between the groups of students regarding their surface learning approach, with those students who preferred the online learning environment responding higher on the relevant to surface approach questions memorising knowledge compared to those students who preferred the face-to-face learning environment. Additionally, the size effects revealed a significant difference in students’ preferences in the online learning environment regarding a variety of sources (online material to study at their own time), behavioural self-regulation (self-evaluation and effort management for learning), self-efficacy (their capabilities to achieve academic success) and test anxiety (stress and anxiety before or during any test). The EEE students who preferred the online environment had a clear view regarding the course utility (student perceptions about the importance of the course to their future career) compared to the Psychology students and to the EEE students who preferred the face-to-face environments.

Further statistical comparison assesses the contribution of the COVID-19 pandemic to adopting a synchronous online learning environment and explores the differences between the two departments using a one-way Analysis of Covariance (ANCOVA). The question about students’ “new learning habits to cope with the University’s move to online learning over the COVID-19 lockdown period” was a covariate to control for the effects of various levels of learning environment preferences in undergraduate students. There was not any difference between the students between the two departments when they were asked to report to what extent (1: Not At All to 7: Very Great Extent) they learned better in a physical lecture/class environment rather than connecting from their home to the web-conferencing environment (e.g., Zoom, Microsoft Teams) (Psychology students: 5.2 ± 1.7, EEE students: 5.1 ± 1.8, F(1, 443) = 1.04, p = 0.308).

Deductive thematic analysis [56] was conducted to test students’ responses to the online open-ended question on the existing student engagement framework (behavioural, affective and cognitive engagement), as discussed in the Introduction part. Microsoft Excel was used for conducting a structured tabular thematic analysis [57]. Overall, 178 out of 446 students from both departments left their qualitative responses in this final part of the questionnaire (online open-ended question) and

Table 7. Breakdown of participant characteristics (department and year) leaving comments on the open-ended question.

Department	1st Year	2nd Year	3rd Year	Total Number
Psychology	68	32	38	138
EEE	14	15	11	40
Both Departments	82	47	49	178

provides a breakdown of the number of students who left a qualitative reply per department and year.

The students’ qualitative responses were split into three theme categories to explore student learning engagement and how it has been affected by the online environment (e.g., ability to work from home, and learning influenced by the COVID-19 restrictions). Based on the results of the analysis, 12 subcategories (i.e., lack of communication, heavy workload, interactive learning, authentic assessment, technical issues, lack of personalized learning) were extracted into these three main theme categories: (1) Behaviour, (2) Affective, and (3) Cognitive engagement dimensions, while the theme categories and the subcategories were defined around abstract concepts related to challenges and opportunities in online learning (

Table 8. Students’ qualitative responses regarding motivation and engagement over the COVID-19 pandemic split into various themes.

Theme and a Brief Description	Sample of Student Responses
Behavioural engagement(i.e., participation and interaction)
Challenges due to the lack of communication: Several students have lost communication with their lecturers and peers through the online learning environment.	EEE student Year 1: I have no real communication with lectures, and I don’t even know the other students on my course as anything more than another name on a zoom call.”
Challenges due to heavy workload: Many students experienced more work hours because they could not adjust themselves to the new learning conditions.	Psychology student Year 3: I fall behind sometimes and have to spend days catching up, whereas, if I was in a lecture hall, I would make all notes at the time of the lecture and not have to worry about catching up.
Opportunities for interactive learning: The interactivity of online lectures incorporating quizzes, PollEveryWhere, Padelt online discussion, or other discussion opportunities into the synchronous sessions allowed students to engage with the learning process.	Psychology student Year 2: I feel like the Padlet, and chat function allows me to be able to ask questions. Also, because the content is often a recap in the synchronous lecture, I know what I don’t understand already.
Opportunities for authentic assessment: Exams could be a venue for students to further increase their participation in the learning process.	EEE student Year 3: I much prefer online exams, as you don’t have to memorise everything which is very stressful, extremely time-consuming, and not applicable to real life.
Affective engagement (i.e., learning environment, teachers)
Challenges due to technical issues: The Internet connection prevented them from engaging with the lecture session, and several more mentioned the use of the library and other resources available in the University to gain in-depth understanding and engagement in the course	EEE student Year 2: During zoom calls, I often have issues with wifi, and considering that these live sessions are not recorded I feel I miss out on a lot of information covered during these and I have always revised in a library, and without this, I’ve really struggled, and it reflects heavily in my average grades.
Challenges due to lack of personalised process: Many students were not engaged with the teaching process because it is a very impersonal approach.	Psychology student Year 3: Due to the lack of enjoyment and personal interaction, the learning process is boring and painful, and I cannot wait for the Easter break.”
Opportunities for reducing learning environment stress: Some students who experienced stress due to the learning environment could not follow the teaching process.	EEE student Year 3: I find the lecture environment very stressful, so much so that II find it difficult to concentrate in those situations because I am so nervous. Online classes have allowed me to collect and retain more information, as I am less anxious in this ‘classroom’ learning environment.”
Opportunities for the effective design of lecture time: Students usually attended a 2-h lecture in a physical environment which did not allow them to increase their productivity.	Psychology student Year 2: 2-h lectures in person feel like a waste of time as it is too difficult to keep up with what is being said, what is on the slides, and writing this down and trying to understand this. Online lectures allow us to take our time and understand the content. Synchronous sessions make it easier to get questions answered that we may be too uncomfortable to ask in person in a lecture hall.”
Cognitive engagement (i.e., learning goals, self-regulation, deep learning)
Challenges due to distractions: Many other students have been distracted due to the home environment, which in many cases led to procrastination.	Psychology student Year 1: I live in halls, but I cannot do any work there due to noise, the walls are so thin people walking by outside are a loud distraction. It’s easier to procrastinate at home as there are always family issues to solve, children to support, and housework & chores to do!”
Challenges for further procrastination: By losing their daily “learning” routine, students could not copy their university life.	EEE student Year 2: … without the routine of travelling to campus, I find it difficult not to procrastinate. It has been a massive jump from sixth form (in which there is a very regular routine you must stick to) to university as now.”
Opportunities for effective use of time due to less commute: Students who should commute found online learning more useful.	Psychology student Year 2: Because I commute to university, learning from home has allowed me to develop a more structured schedule for the day and I feel that I can get more done because I am not spending time on the commute.”
Opportunities for studying the subject in-depth: Students were able to keep notes in their own time and space and search for help over the online synchronous session if they wished.	EEE student Year 3: Being online is easier as videos are out first and you can watch them write your notes and understand, then if there are any issues you can ask questions in the next session. Learning online is all right, I got used to it”.

).

Overall, the COVID-19 pandemic affected students’ mental health, and/or their ability to work and relax in the same environment. 104 students out of 178 mentioned the connection of their learning motivation with the COVID-19 pandemic, which led to feeling “very unmotivated” to study and learn. This seriously impacted student engagement, as they “felt less involved in” their subject. Regarding students’ behavioural engagement, the potential challenges could be related to additional workload and the lack of communication between students and teachers, whilst the potential opportunities could be regarding the interactive learning environment and assessment authenticity, allowing students to be more engaged with the course content. The qualitative responses revealed challenges and opportunities for the wide area of affective engagement. These could be related to technical issues and the use of University facilities alongside the reduction of class stress and effective teaching design. Finally, regarding cognitive engagement, the potential challenges could be related to potential distractions due to students’ study environment, which might lead to potential procrastination. On the other hand, the potential opportunities could be related to the effective use of time and online content to study in-depth their cognitive subject. Finally, many students for both departments made a clear point that online exams have significantly reduced stress and anxiety, while they prefer to follow the small classes (i.e., tutorial sessions, seminars) and laboratory sessions in a physical learning environment. Special merit should be given to students with disability issues, who provided various additional comments about their learning experience in an online environment. For example, dyslexic persons prefer a physical learning environment because “hearing others’ ideas and opinions was crucial to her/his understanding.” On the other hand, deaf students prefer synchronous online sessions with asynchronous activities because “the big lecture theatres are acoustically impractical.”

4. Discussion

This study aimed to explore students’ preferences in different learning environments regarding their learning engagement, considering individual student characteristics (i.e., self-efficacy, self-regulation, test anxiety) and teaching delivery elements (i.e., clear learning goals, collaborative learning, teacher’s role). Each of these elements is linked to behavioural, affective or cognitive student engagement, as has been discussed in the Introduction part and illustrated in Figure 1. Students from two departments (Psychology and Electrical Engineering and Electronics-EEE) over the three-year levels of undergraduate studies completed an online questionnaire, providing quantitative and qualitative responses. By inviting students from Psychology (soft discipline) and EEE (hard discipline), a secondary aim was to explore whether there was any difference in students’ learning environment preferences regarding student background. Paying attention to these differences would enable the Higher Education sector to optimize the teaching process in different environments based on the requirements for each discipline.

4.1. Difference between Academic Performance and/or Learning Environments in Each Department

As this study was conducted over the 2021–2022 academic year, all first-year students had limited experience in the face-to-face delivery process due to the COVID-19 pandemic restrictions. The second and third-year students had more experience in both (face-to-face and online) learning environments compared to the first-year students, to whom teaching delivery was through an online environment. However, this study has not found any difference in students’ preferences in learning environments over the years and within departments. On the contrary, there were significant differences between the two learning environments for both departments, with most students believing that they were engaged more in a face-to-face rather than in an online learning environment. This might be an expected finding, as these students had decided to follow an in-person course before being “forced” to adopt a new learning process due to the COVID-19 pandemic restrictions. Additionally, a study conducted before the COVID-19 pandemic explored the comparison between online and face-to-face courses with student satisfaction, concluding that “an online course cannot fully replace face-to-face learning, which offers a real-life learning experience, human interaction, and personal contacts with both tutor and fellow students” [58] (p. 43). However, by comparing students’ academic performance for the two departments regarding their preferences in learning environments, it was found that academic performance was influenced by the learning environments, with EEE students gaining higher grades in an online learning environment than psychology students and those EEE students who preferred psychical learning environments. This finding might link to what Vo, Zhu, and Diep (2020) [29] found regarding the difference in blended learning environments between students from hard (i.e., EEE) and soft (i.e., Psychology) disciplines for online learning environments. It could be potentially explained by the way that teachers employed “the strategies that best foster the acquisition of disciplinary knowledge and competencies” (p. 490), with Psychology students (soft discipline) appreciate a blended learning approach based on discussions more than EEE students (hard discipline) to which the online content (i.e., visualised material) assisted them to understand the cognitive topic [59]. A potential implication of these findings was for teachers to reconsider the presented material, allowing their students to be cognitively engaged with the online content. Furthermore, by synthesising this study’s findings with previous studies conducted during the COVID-19 pandemic, institutions and teachers must reconsider the teaching design process before adopting online elements for synchronous lectures. For example, Zeng and Wang (2021) [49] found the positive effect of synchronous learning on student satisfaction (highly related to academic performance) when the sessions were carefully designed, with teachers facilitating discussions, providing feedback to their students and facilitating student collaboration.

4.2. Differences between Various Aspects of the Teaching Delivery Process (i.e., Learning Goals, Feedback, Teacher Support) with Face-to-Face or Online Learning Environments in Each Department

Analysing the student quantitative responses regarding teaching delivery elements, a significant difference was found between the departments and the learning environments only regarding collaborative learning, with EEE students responding higher than Psychology students in both learning environments. This finding is in alignment with what Vo, Zhu, and Diep (2020) [29] found when they explored the difference between hard (i.e., EEE) and soft (i.e., Psychology) disciplines in a blended learning environment, suggesting that the teachers from the hard disciplines stressed the importance of collaborative learning and assessment allowing their students to gain the active learning experience. It might be easier for EEE teachers rather than psychology staff to integrate collaborative activities, as the engineering curriculum included more practical laboratory sessions and small class activities, and the teachers did not need to pay additional effort to design collaborative learning activities for their students. Additionally, students from both departments highly evaluated the contribution of online activities to their engagement in online learning environments as they interacted with their online content, teachers, and their peers, but they expected their teachers to provide clear goals. In an online learning environment, they were mainly students who should monitor to what extent they have reached the learning goals, employing the relevant learning strategies. Participants from the Psychology department believed that they received the same level for face-to-face and online learning environments regarding teacher support to be engaged with their courses. This is an important element for teaching, and teachers should consider it when they deliver an online course to increase student engagement cognitively by enhancing students’ higher-order thinking [60] and effectively by reducing the drop-out rate [61]. Students also believed that they received better quality feedback from their teachers in online environments, which assisted them in facilitating their learning. Therefore, they were more engaged with their courses through synchronous online sessions, reaching their learning outcomes for their course. This finding aligns with a previous study on synchronous and asynchronous settings of online teaching over the COVID-19 pandemic [62], which reported that more feedback opportunities, including emotional support, were provided in synchronous online settings, supporting student learning facilitation process. The implication of these findings is for teachers to practice on feedback delivery process in a blended learning environment by utilising their practices from their online synchronous lecture and online environment. Although many researchers have studied how the blended learning approaches could enhance the feedback delivery process and collaborative opportunities [63,64], this study has focused on the differences between online and face-to-face blended learning environments under the same experimental conditions, pointing out the need for future research on this area. A recent study has discussed how teachers could redesign their courses to increase flexibility and student engagement in non-pandemic times, promoting collaborative activities and enhancing the feedback delivery process [65].

4.3. Differences between Students’ Preferences in Synchronous Learning Environments (Face-to-Face or Online) and Their Characteristics (i.e., Self-Efficacy, Behavioural Self-Regulation) in Each Department

Based on students’ responses to the questionnaire, EEE students who preferred to work in an online learning environment felt more confident in their ability to complete academic tasks (self-efficacy) and successfully engage with online learning material (variety of sources). They have also presented a higher level of behavioural self-regulation compared to Psychology students and those EEE students who preferred to follow a teaching approach in a face-to-face learning environment. Although a previous study found engineering students had higher self-regulation skills compared to various disciplines [66]; other researchers could not support this finding [67,68] due to the complexity of self-regulation [69]. On the contrary, a recent study conducted over the COVID-19 pandemic’s first lockdown found that those students who have adapted themselves well to the new learning requirements presented a high level of self-regulation skills and confidence in their ability to use technology for learning purposes utilising the online learning opportunities [46]. Students who preferred a teaching delivery process in a face-to-face learning environment responded differently when asked about the surface learning approach compared to those who preferred online learning environments. By following surface learning approaches, students tended to minimise their commitment to understanding and their cognitive engagement in learning activities, finding ways to fix and reproduce their learning patterns [70]. Al Mamun, Lawrie, and Wright (2022) [71] have recently argued that the absence of teachers in an online learning environment highly influences cognitive and behavioural student engagement, facilitating their learning only by online learning material and embedded pedagogical approach. The students involved in this study preferred to interact with their teachers and peers in a face-to-face environment rather than in an online one. However, Schunk and Ertmer (2000) [72] pointed out that when teachers did not “dictate” to students what they needed to do and how to accomplish tasks in a face-to-face environment, they should assist students to develop the relevant self-regulate skills, feel confident in their ability for academic success; otherwise, it would be easy for students to follow a surface learning approach. The way that students coped with their learning process along with their previous experience on online tests also explained why there was a difference in test anxiety, with those EEE students who preferred the online environment to challenge their test anxiety over the COVID-19 pandemic compared to the other students [73]. Finally, as students have been “forced” to work more online over the COVID-19 pandemic, they were involved in reflecting on their learning experiences and course utility, which aligns with the recent findings on employability awareness, where students had been linking their course opportunities with employability and reflective practice in online learning environments [74]. Overall, these findings regarding the role of students’ characteristics in learning engagement in the two environments could be utilised to assist teachers to enhance their blended learning approach by keeping their students behavioural, affectively and cognitively engaged. For example, the main implications of these findings for teachers and institutions are to provide opportunities to students to connect the course content with employability (i.e., promote the connection between the course characteristics and employability, allowing students to recognise the importance of their course), support students to develop confidence in their ability as they work online, and review the assessment process by considering the role of the student test anxiety in various settings.

4.4. Students’ Learning Environment Preferences Were Influenced by the COVID-19 Pandemic

Although there was not any difference between the students from the two departments whether their preferences in learning environments have been influenced by the COVID-19 pandemic, the qualitative student response revealed challenges and opportunities in online learning environments regarding cognitive, affective, behavioural, and social engagement [5]. In summary, the challenges that students faced regarding their engagement in an online learning environment were related to lack of interactions with their students and teachers, additional workload, technical issues, including lack of the use of University facilities, and distractions over their lecture time from the home environment and family responsibilities. These findings about the challenges of the COVID-19 pandemic and university teaching and learning are in alignment with what has recently been presented by other researchers [42,75,76,77]. Additionally, the students who participated in this study mentioned the learning opportunities about the online environment, highlighting the importance of alternative teaching delivery alongside authentic assessments. For example, students felt engaged with their courses when they actively interacted with their teachers and peers in a flexible learning online environment, allowing them to move away from the knowledge memorising process following a knowledge construction approach. A recent study has discussed how a university learning environment could support teaching and learning flexibility by considering pedagogical approaches, technologies, and student learning needs in informal learning environments and without being on campus for face-to-face meetings [78]. The current study’s findings also provided the reasons why students preferred this flexibility in learning and what universities, policymakers, and teachers should consider offering students a flexible online learning environment. For example, a practical implication of these findings is regarding the online exams/assessments that the institutions could offer to reduce student test anxiety. The institutions could deliver workshops and training for students to help them with the online exam revision enhancing their critical skills and promoting deep learning. Furthermore, the institutions could practice flexible ways to support teaching from home and university lectures, allowing students to have other commitments, such as work and caring duties, without disrupting their studies and utilizing the technology opportunities. Recently, a study has explored a mixture of different types of teaching and learning techniques, offering opportunities for teachers and students to participate in a variety of class roles, interacting via online, face-to-face or blended methods [79] (McKenzie et al., 2022), but future work on this area could assist institutions and teachers to reconsider the digital learning transformation in the post-COVID-19 pandemic period.

4.5. Limitations and Further Future Work

Finally, by conducting this study on the COVID-19 pandemic, students might have provided responses that were influenced by their mental health status [80]. This factor may have affected their engagement in studying and living in the same home environment. For example, this may have impacted their preferences in an online learning environment, as they might feel isolated from others and be fatigued without “human” contact. Another limitation regarding this study might be related to the sample and the data collection method. Although the sample size was large enough to extract statistically secure results, this was a sample from only one UK University, therefore generalisability can be questioned. In future studies, a more varied sample utilising students from other Universities should take place to gain an in-depth understanding of their engagement in various learning environments. For that purpose, the questionnaire of this study has been uploaded to the Zenodo open-access repository, where other educational researchers could compare their student engagement with the findings of this study. Although it might be considered that the experimental conditions of future studies on this area will be different without students being “forced” to follow the COVID-19 pandemic restrictions, this snapshot of student engagement could be a reference study. An additional limitation regarding the student sample might be that the number of participants from the Psychology first-year cohort was higher than the rest and all the first-year undergraduate students, including EEE students, had limited experience in the face-to-face delivery process compared to the students from the other years. However, these differences in experience might not play a role as there was not any statistically significant difference in student preferences within the year of studies for each department. Furthermore, the sample size for each year for each department was more than 30 participants, which satisfied the conservative rule of thumb, where at least 30 participants are necessary per condition [81]. Another limitation of this study was related to students’ self-responses to an online questionnaire. Students had the opportunity to provide their qualitative responses to an open-ended question, but by allowing them to be involved in future focus groups, the findings of this study could be cross-checked. Finally, this study explored student engagement in the two learning environments mainly considering the teacher-student interactions. However, future work on student-student interactions in various learning environments may provide significant information about student engagement. The different types of interaction, including teacher-student, student-student, materials-student, and their effect on student engagement and motivation have been previously discussed by many educational researchers, with student-student interactions being one significant challenge in online learning environments [82,83,84].

5. Conclusions

In conclusion, this study not only discussed student engagement in an online learning environment over the COVID-19 pandemic but also attempted to explore any potential difference in face-to-face learning environments for two departments (Psychology and EEE). Student learning engagement has been discussed under the cognitive, affective, and behavioural dimensions. Student academic performance also provided valuable findings regarding student preferences in the face-to-face and online learning environments, and it was considered an indicator of student engagement with the course. Although other educational researchers have studied student engagement over the COVID-19 pandemic period, the authors have not been aware of any study which discussed this topic, exploring the role of individual characteristics and teaching delivery elements in face-to-face and online learning environments. The findings of this study could assist universities, policymakers, educational researchers and teachers in shifting teaching approaches over the post-COVID-19 pandemic era by adopting elements from online learning to a blended approach. For example, it is crucial for shifting knowledge from transition to the construction process for students to be engaged with their courses to follow the teaching way that better fits their needs (i.e., disability support, work and family commitment), being connected with their teachers over the lecture time, working on online collaborative activities, and receiving real-time feedback from their teachers. The post-pandemic era also needs to consider how teachers could further help their students with the self-regulation development process by providing clear goals and support throughout both learning environments. This area could be further explored, but this study might be a reference one for future work on student engagement in blended and online learning, as it studied student preferences in two environments before and during the COVID-19 pandemic period.