Next Article in Journal
The Enhancement of Fire Safety in Small-Scale Senior Citizen Welfare Institutions Based on Fire Protection Defense-in-Depth Strategy
Next Article in Special Issue
What Is Engineering and Who Are Engineers? Student Reflections from a Sustainability-Focused Energy Course
Previous Article in Journal
Experimental Study of the Usability of Recycling Marble Waste as Aggregate for Road Construction
Previous Article in Special Issue
Virtualizing Project-Based Learning: An Abrupt Adaptation of Active Learning in the First Days of the COVID-19 Pandemic, with Promising Outcomes
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 14
Issue 6
10.3390/su14063197
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Mobile Learning Acceptance Post Pandemic: A Behavioural Shift among Engineering Undergraduates

by
Jeya Amantha Kumar
1,*,
Sharifah Osman
2,
Mageswaran Sanmugam
1 and
Rasammal Rasappan
3
1
Centre for Instructional Technology and Multimedia, Universiti Sains Malaysia, Gelugor 11800, Pulau Pinang, Malaysia
2
School of Education, Faculty of Social Sciences and Humanities, Universiti Teknologi Malaysia, Skudai 81310, Johor, Malaysia
3
Department of Mathematics, Science and Computers, Politeknik Balik Pulau, Pinang Nirai, Balik Pulau 11000, Pulau Pinang, Malaysia
*
Author to whom correspondence should be addressed.
Sustainability 2022, 14(6), 3197; https://doi.org/10.3390/su14063197
Submission received: 31 January 2022 / Revised: 26 February 2022 / Accepted: 3 March 2022 / Published: 9 March 2022
(This article belongs to the Special Issue Strategies and Applications for Sustainable Engineering Education)
Browse Figures
Versions Notes

Abstract

:
Mobile learning has become an essential telematic tool to facilitate and compliment online teaching and learning during the pandemic. This study investigates the change of behaviour and acceptance of using mobile learning specifically for engineering undergraduates due to this shift. The data collected pre-Covid19 (n = 326) and post-pandemic (n = 349) indicated an inclination for utilizing laptops than smartphones, while Telegram prevails as a popular tool for communicating and sharing information within the learning community. Next, while video conferencing tools and online learning management systems utilization increased, educational games and reading behaviour via mobile devices declined. Concurrently, behavioural intention post-pandemic were found to reduce marginally as importance were also given towards establishing learning communities via social influence compared to perceived usefulness. The outcome of this study contributes to the limited body of literature on engineering education mobile learning acceptance, and recommendations are provided for further investigation to ensure continuous sustainable use.

1. Introduction

The technological affordance and the ubiquitous use of mobile devices have been fundamental in its applicability in today’s education paradigm. While mobile devices such as smartphones and laptops are seen as a valuable commodity in facilitating and “connecting” learning environments, mobile learning refers to the use of portable devices for learning and communicating in a learning environment. Accordingly, mobile learning or m-learning is essential for the new millennium of education technology [1,2,3] where teaching and learning methods advocate a blended approach of digital and face to face classes. The capability of these devices to be used in and out of the classroom for higher education [4,5,6] has made it adaptable for multitasking [1] when using, acquiring, producing, communicating and transforming learning contents [7] via technology and the internet [8] primarily through instant messaging and social media [9]. Hence, undoubtfully these interactions have transformed the traditional education dynamics by creating informal learning communities that support learning through applications such as WhatsApp, Facebook, and Telegram [10]. As such, the need to create these communities as off-the-record learning support [11] that provides instant access to learning [12,13], flexibility and openness [14,15] have accentuated the use of mobile devices for learning.
Conversely, smartphones have become one of the most used learning devices due to their mobility and capability to facilitate informal learning outside the classroom [16]. Hence, this advantage has exponentially orchestrated their use during the pandemic [17]. Saikat et al. [18] explained that the pandemic shifted higher education institutions (HEIs) perception of m-learning as a vital tool for education delivery. Additionally, post COVID-19 pandemic, empirical evidence have indicated positive satisfaction and behaviour in response to m-learning [8]. As the role of m-learning is mainly to complement but not substitute existing learning practices [14], the shift was conditional to changes in trends and technology [1]. All the same, while this is encouraging as HEIs struggled to adapt to the new form of distance learning [19], undoubtfully, the move to fully online teaching and learning mode radically escalated m-learning towards a new type of education system. Nevertheless, undoubtedly, m-learning during the pandemic may have negative consequences. By so, empirical evidence have indicated perceived usefulness and continuous use [20,21] were found to influence emotions and attitude negatively [2]. Concomitantly, these emotions may reduce the intention to use m-learning among HEI students [22]. On the other hand, there is much unknown on these phenomena of reduced intention post-pandemic due to scarce investigation on m- learning acceptance among HEIs students [19]. Hence, such investigations are warranted to identify factors influencing its use [20] especially post-pandemic. Furthermore, HEI students have indicated confidence in its value and the innovative approach it may provide to different modes of learning post pandemic [8].
In hindsight, the understanding of m-learning in engineering education seems contradictory [23], as scholars have indicated its value primarily for communicating, collaborating and providing access to learning content [6,24]. All the same, it’s popularity due to affordance and availability to facilitate learning [23,25], provides significant value that may reflect engineering undergraduates intention to use m-learning [26]. However, acceptance may differ based on engineering disciplines, as speculated by [7,10], which irrefutably adds to the limited acceptance studies in this context. By so, in this study, we focus on investigating the change of behaviour in regard to the acceptance of mobile learning pre and post-pandemic by focusing primarily on electrical and electronic engineering students. According to Loh et al. [22], there isa need to investigate m-learning acceptance based on a longitudinal perspective, and we speculate that evaluating pre and post-pandemic acceptance will provide insights into the shift that occurred due to the pandemic. Next, we measured m-learning acceptance as a learning behaviour that is not defined by a particular technology, mobile application nor social media application. By so, a combined Technology Acceptance Model (TAM) and Theory of Planned Behavior (TPB) was applied as done in empirical studies in mobile learning by [27,28]. According to Gómez-Ramirez et al. [28], while TPB explains behaviour based on beliefs about self-efficacy and subjective norm towards attitude (ATT) and behavioural intention (BI), TAM reflects the ATT and BI based on perceived ease of use (PE) and perceived usefulness (PU). Kumar, Bervell, et al. [10] claims that the combined TAM and TPB, known as C-TAM-TPB [29], will provide insights on continuous use of m-learning in HEI, and we gather as imperative in this study as we investigate the change of behaviour post-pandemic. Hence, this study adds to the limited literature as indicated by Lai [30], for engineering undergraduates m-learning acceptance by exploring the shift in their behaviour due to the onset of the pandemic.

2. Hypotheses and Model Development

Empirical findings have indicated TAM to be substantial in investigating the acceptance of education technologies [19]. Nevertheless, according to Oke and Fernandes [31] and Yang and Su [32] unitary theories such as TPB and TAM are not complex enough to explain behavioural intention and adoption of technologies. C-TAM-TPB considers a direct and indirect effect on behavioural intention [33] by considering experience [34], which is a substantial factor in this study to explain the shift in learning due to the pandemic. According to Al-Hamad et al. [20], the pandemic drove teaching and learning towards embracing m-learning, thus predicting higher intentions of using m-learning. While the relationships between PE, PU, ATT and BI for m-learning has been established in numerous studies [8,12,35], Almaiah and Al Mulhem [5] suggest considering the role of self-efficacy in predicting mobile learning adoption. Kumar, Bervell, et al. [10], in their study on engineering undergraduates adoption of mobile learning, conducted pre COVID indicated that while the relationships of TAM were found to be significant, mobile learning self-efficacy (MSE) did not influence PU but BI, ATT and PE while social norm (SN) influence ATT and PU. SN and MSE were measured as the two constructs associated with BI based on TPB.
By so, this study defined self-efficacy as MSE to describe students perception of using m-learning to perform learning activities via mobile devices [27]. Han and Yi [35] and Moorthy et al. [36] claimed MSE strongly influences HEI students m-learning intentions, whereas Alasmari and Zhang [37] and Briz-Ponce et al. [38] explained that SN may have a significant impact on m-learning intention. However, when explicitly considering engineering education, it has been reported otherwise for SN [10,26]. SN is usually defined as learners perception of others in their learning community, such as instructors and peers in performing a behaviour [27] or using a technology that supports learning. We theorized that as there is a change in social dynamics during the pandemic, especially concerning how learners communicate and collaborate in their learning community, this relationship may differ. Nevertheless, by using the original C-TAM-TPB, we hypothesize a change in these relationships as conceptualized in Figure 1:
Hypothesis 1 (H1):
ATT will positively influence mobile learning BI.
Hypothesis 2 (H2):
MSE will positively influence mobile learning BI.
Hypothesis 3 (H3):
PE will positively influence mobile learning ATT.
Hypothesis 4 (H4):
PE will positively influence mobile learning PU.
Hypothesis 5 (H5):
PU will positively influence mobile learning ATT.
Hypothesis 6 (H6):
PU will positively influence mobile learning BI.
Hypothesis 7 (H7):
SN will positively influence mobile learning BI.

3. Research Design

The questionnaire used in this study consist of 20 items that were adapted from [27,39] to reflect the factors of TAM and TPB. These constructs were measured using a 5-point Likert scale ranging from “Strongly agree” to “Strongly disagree”. Furthermore, demographic questions on gender, m-learning devices and m-learning communication and application preference were also explored and compared.
The survey was dispersed to the respondents of four polytechnics focusing on Electrical and Electronic engineering undergraduates in Malaysia using Google Forms in 2018. Whereas for post COVID, the data were collected between November and December 2021. The response was voluntary, and the collection of data were observed for a duration of three months. The extracted data from Google Forms were first downloaded and imported to IBM SPSS version 27 for cleaning and were analyzed for normality using Kolmogorov-Smirnova. All factors were found to be not normally distributed (p < 0.05), as reflected in Table 1. The data in CSV format were exported to SmartPLS ver. 3.2.8 to be analyzed using Partial Least Squares-Structural Equation Modelling (PLS-SEM) method. Analyses conducted referring to pre-COVID feedback are labelled as Study1 whereas post-COVID as Study2 respectively.

4. Results

4.1. Respondents Profile

The 326 respondents for the pre-COVID survey were majority male (n = 195, 59.8%), while the balance 40.2% (n = 131) were female. Similarly, post-COVID respondents (n = 349) also reflected a majority of male respondents (n = 223) at 63.90% (Table 2). The changes observed in the type of device usually assimilated for mobile learning were reflected by an increase in netbook/laptops use from 26.46% to 43.74% and a reduction in smartphone use from 70.96% to 52.59%. Next, the applications frequently used for communicating with peers pre-COVID were WhatsApp (34.26%), Telegram (15.86%), Facebook (15.39%) and Short Messaging System (SMS) (15.63%) which transitioned into namely two applications post-COVID WhatsApp (46.12%) and Telegram (35.03%). Similarly, the use of Telegram increased from post-COVID as the application most frequently used to communicate with lecturers (11.15% to 41.77%) and share information (15.37% to 41.77%) after WhatsApp. In all cases, the use of Facebook as a teaching and learning platform was reduced. Next, changes were observed in applications used for learning, namely only for online learning platforms such as learning management system (LMS), reading and video conferencing tools. Following, it was also observed that the use of personal data doubled (32.39–59.63%), while the use of institution Wi-Fi public Wi-Fi reduced significantly.

4.2. Study Power

Measuring multivariate normality based on Mardia’s Test of multivariate normality [40] at using https://webpower.psychstat.org/wiki/tools/index (accessed on 2 January 2022) [41] indicated skewness (β = 98.730, p < 0.01) and kurtosis (β = 676.760, p < 0.01) for pre-COVID and skewness (β = 129.372, p < 0.01) and kurtosis (β = 757.320, p < 0.01) for post COVID which validates the use of partial least squares (PLS) technique. Next using, GPower 3.1 with effect size f2 = 0.15, α error prob = 0.05 and power = 0.95. The minimum sample size of 138 was determined with GPower 3.1, using effect size f2 = 0.15, α error prob = 0.05 and power = 0.95. Therefore, the sample size for pre and post exceeded the minimum requirement. Concurrently, the model was measured using the measurement model for validating the constructs’ reliability and the structural model to evaluate the hypotheses.

4.3. Measurement Model Analysis

Hair et al. [42] suggested that the model’s reliability and validity must be first examined, and we report the finding as presented in Table 3. The indicator reliability was between 0.961 and 0.871, which was higher than 0.708 as suggested by [43], while composite reliability were all above 0.90 [44] except for PU for both pre and post-pandemic and in addition to ATT and PE for post-pandemic. However, as all rho_A values were >0.7 [45] internal consistency was found to be acceptable. As for convergent validity, the average variance extracted (AVE) were all above >0.5 as recommended by Hair et al. [44]. Lastly, the Heterotrait-Monotrait (HTMT) (Table 4) were all <1 as recommended by Henseler et al. [46], hence reflecting that the constructs could be distinguished accordingly. Therefore, the validity and reliability of the model are found to be acceptable.

4.4. Structural Model

Next, the structural model was analyzed to determine goodness of fit, multicollinearity, path analysis (β), coefficient of determination (R2), effect size (f2) and predictive relevance (Q2) [32,43]. First, the model fit was tested using Standardized Root Mean Residual (SRMR) and Exact fit criteria (d_ULS and d_G) (Table 5) [44,48,49]. All three values were found to be acceptable as the SRMR was below 0.08 [47], and d_ULS and d_G were found to be <95% bootstrap quantile (HI95). Variance inflation factor (VIF) values measuring the collinearity (Table 6) were found to be acceptable as they were all <5.0 [46].
Next, the path coefficients (β) determined the correlation between endogenous and exogenous variables. Therefore, the t-statistics using bootstrapping resampling of 5000 [48] were measured. The results for path coefficients (β), t-value, confidence intervals and effect size (f2) are summarized in Table 6 and followed by coefficient determination (R2) (Table 7).
For Study1, ATT (0.720) and PU(0.682) are considered as moderate and BI(0.811) as strong as categorised by [42] whereas Study2 ATT (0.678), BI (0.779) and PU(0.663) were categorized as moderate. Next, results from the bootstrap for Study 1 revealed that ATT (β = 0.443, t = 6.702, f2 = 0.262), MSE (β = 0.370, t = 5.255, f2 = 0.217) and PU (β = 0.122, t = 2.043, f2 = 0.006) positively influences BI by explaining 81.1% of behavioural intention to use mobile learning. Therefore, H1, H2 and H6 is supported whereas H7 (SN (β = 0.038, t = 0.588, f2 = 0.003)) is rejected. As for ATT, 72.0% could be explained by PE (β = 0.370, t = 5.255, f2 = 0.131) and PU (β = 0.826, t = 8.751, f2 = 0.338), hence accepting H3 and H5. Lastly, H4 was also accepted (β = 0.826, t = 6.202, f2 = 2.140) as PE was able to predict 68.2%. As for post-pandemic, ATT (β = 0.443, t = 6.702, f2 = 0.279, p = 0.000), MSE (β = 0.312, t = 5.608, f2 = 0.160, p = 0.000) and SN (β = 0.161, t = 2.818, f2 = 0.041, p = 0.005), positively influences BI by explaining 77.9% of behavioural intention to use mobile learning. Hence, H1, H2 and H7 were supported but H6 was rejected as opposed to pre-COVID findings as PU (β = 0.033, t = 0.566, f2 = 0.001, p = 0.571) was found to be non-significant in determining BI. Next, 67.8% of ATT were also explained by PE (β = 0.444, t = 7.292, f2 = 0.206, p = 0.000) and PU (β = 0.420, t = 6.457, f2 = 0.184, p = 0.000) while also indicating that 66.3% of PU could be determined by PE (β = 0.814, t = 41.986, f2 = 1.968, p = 0.000), thus accepting H3, H4 and H5.
Next, using the blindfolding procedure of OD (omission distance) of 7, the predictive relevance (Q2) (Table 8) pre-COVID for ATT (0.539) and BI (0.662) were determined as large and PU (0.450) as medium [48]. Whereas post-COVID, ATT (0.482) and PU (0.439) were medium and BI (0.616) large. As for out-of-sample predictive power (Q2predict), the model’s accuracy for new cases is predicted [49] by using PLSpredict procedures with settings of k = 10, r = 10. The PLS Q2predict for all indicators were > 0 (Table 9); nevertheless, some PLS indicators reflected higher RMSE values than the Linear Model (LM) RMSE value. Based on the suggestion of [50], both models have low predictive power as a minority of indicators have lower PLS-SEM prediction errors compared to LM for pre pandemic (BI1, BI2, PU1) and post pandemic (BI1, BI2).
Lastly, using the Importance Performance Map Analysis (IPMA) and setting the target to construct as BI, the prioritized variable based on performance and total effects is determined [50]. Based on the findings as shown in Table 10, it can be observed that PE (0.514) followed by ATT (0.463) has the most important and SN (0.039) is the lease for BI. As for performance, the highest is ATT (80.572), followed by PE (79.470), and the least is SN (72.934). However, for post-COVID (Study 2), ATT (0.458) followed by PE (0.373) has the most importance on BI while SN (0.152) remains the lease. As for performance ATT (74.574) followed by PE (70.055) and PU (69.492) had the highest performance.

5. Discussion and Conclusions

The findings revelated that there is a shift in preference to using smartphones (70.96% → 59.59%) for m-leaning towards the use of laptops and notebooks (26.46% → 43.74%). While smartphones may facilitate some online learning activities, the increased use of online learning platforms, namely LMS platforms such as Google Classroom, Moodle and video conferencing tools requires higher level of interaction functionalities which is often limited through smartphones and easier to perform using laptops. Moreover, it was also indicated that the use of mobile devices for educational games (Kahoot etc.) and reading activities have significantly reduced post pandemic. However, these behaviours may be due to the challenges in multitasking numerous learning activities, especially in a virtual classroom environment and the rise of video-based instructions as the main medium of instruction. Nevertheless, this also questions how fully online virtual classes integrate engaging active learning strategies besides educational games. Additionally, the use of cloud storage, reading, creating and editing documents seem to also reduce. Next, we also observed that while the use of WhatsApp increased post pandemic, it was not as significant compared to the increased use of Telegram as a means for communication with peers (15.80% → 35.03%), lecturers (11.15% → 41.77%) and data sharing (15.37% →41.77%). Besides, other applications such as Facebook and Short Messaging System (SMS) also portrayed a receding trend. As for internet access, personal data plans usage almost doubled (32.39–59.39%) while showing a reduction in the use of public WIFI (19.63–4.11%) and WIFI at the institute (28.46–10.65%).
Next, based on the findings as reflected in Figure 2, the overall intention (BI) to use mobile learning reduced from 81.1% → 77.9% yet the strength of the relationship remains the same at β = 0.443 with ATT. According to Loh et al. [22], HEI students post-pandemic are showing reduced intention in using m-learning due to technostress, exhaustion and other issues faced in facilitating online teaching and learning. Furthermore, it was also observed that pre-pandemic, SN were found not to predict BI, therefore, rejecting H7 and accepting all other hypotheses. However, post-pandemic, this shifted as only PU was found to have a non-significant relationship with BI, thus accepting all other hypotheses and rejecting H6. Therefore, this outcome contradicts the finding of [19], indicating social influence lack influence on m-learning BI post-pandemic. The importance of SN is even more evident post-pandemic as PU’s non-significant effect indicates that m-learning essentialness to facilitate teaching and learning is not only based on the effectiveness but also on how a learning community defines its relevance in the context. However, PU still remains as an important predictor for ATT.
Next, the variance of ATT (72.0% → 67.8%) and PU (68.2% →66.3%) also reduced post-pandemic. Before the pandemic, PU had higher predicting strength on ATT than PE; nevertheless, both had almost similar effects. Based on the IPMA results, pre-pandemic, the importance in predicting BI was defined in order of PE, ATT, and PU, whereas post-pandemic was ATT, PE and MSE. Even though [20] claimed that PE and PU of m-learning are significant factors predicting intention as it suppresses fear in such technology, yet our findings indicate that m-learning self-efficacy could also be an important factor to consider in reducing fear. Eneje [23] added that for engineering education to embrace m-learning based on affordance, the technology must facilitate visualization, high-speed processing, and high control. By so, indicated the importance of PE. Alternatively, Kumar, Rajamanickam, et al. [6] indicated such use are often unsupported, and m-learning for engineering education mostly focuses as a means for non-formal learning activities. Lastly, we conclude that a shift indicates a marginal reduction in intention. Moreover, m-learning is not solely determined by usefulness but also by the learning community defined under the term subjective norm. Nevertheless, usefulness, directly and indirectly, affects ATT by mediating through ease of use. Therefore, while we encourage the need for personalized m-learning platforms, the pandemic has forced us to re-evaluate the importance of a learning community and how it is also vital in encouraging sustainable engineering education.

6. Limitations and Future Studies

This study, without hesitation, has limitations that should be addressed. Firstly, the study only limits to undergraduates from Malaysian polytechnics focusing on electrical engineering therefore, the outcome of this study could not be generalized to other engineering disciplines as the use of m-learning may be field-specific. Consequently, while there is still a lack of studies of m-learning acceptance for engineering education where other factors could also be considered to fulfil technical fit [24], effectiveness, efficiency [19], and usability [10] to ensure robustness with the engineering discipline. Moreover, as post-pandemic m-learning has become an essential instrument for telematic teaching, there is still a lack of applications supporting the need for autonomous m-learning [25]. According to Ong et al. [51], engineering students are often technologically inclined, thus prefer autonomous learning with instructors guidance. They added the need for auto evaluating knowledge that we perceive could be established through other mobile based interventions such as chatbots [52] due to positive acceptance and opportunities to personalize learning [53]. Lastly, it is also crucial to consider a mixed-method approach to provide an in-depth understanding in investigating the occurrence of nonlinear relationships to evaluate the behavior of engineering undergraduates.

Author Contributions

Conceptualization, J.A.K. and R.R.; methodology, J.A.K. and S.O.; formal analysis, J.A.K.; investigation, J.A.K. and R.R.; writing—original draft preparation, J.A.K.; writing—review and editing, J.A.K., S.O. and M.S.; funding acquisition, J.A.K. All authors have read and agreed to the published version of the manuscript.

Funding

This study was funded under the Universiti Sains Malaysia, Short Term Research Grant 304/PMEDIA/6315219.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Informed consent was obtained from the respondents.

Data Availability Statement

Not applicable.

Acknowledgments

The authors would like to thank Dulina Tholibon, Gauri Birasamy, Suzana Ahmad and Caroline Dame Siagian for their assistance in collecting the data for this study.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Criollo-C, S.; Guerrero-Arias, A.; Jaramillo-Alcázar, Á.; Luján-Mora, S. Mobile learning technologies for education: Benefits and pending issues. Appl. Sci. 2021, 11, 4111. [Google Scholar] [CrossRef]
  2. Qiang, T.; Gao, H.; Ma, X. Pro-environmental behavior and smartphone uses of on-campus engineering students in Xi’an, China. PLoS ONE 2021, 16, 1–15. [Google Scholar] [CrossRef] [PubMed]
  3. Samad, M.R.A.; Ihsan, Z.H.; Khalid, F. The use of mobile learning in teaching and learning session during the Covid-19 pandemic in Malaysia. J. Contemp. Soc. Sci. Educ. Stud. 2021, 1, 46–65. [Google Scholar]
  4. Buabeng-Andoh, C. Exploring university students’ intention to use mobile learning: A research model approach. Educ. Inf. Technol. 2021, 26, 241–256. [Google Scholar] [CrossRef]
  5. Almaiah, M.A.; Al Mulhem, A. Analysis of the essential factors affecting of intention to use of mobile learning applications. Educ. Inf. Technol. 2019, 24, 1433–1468. [Google Scholar] [CrossRef]
  6. Kumar, J.A.; Rajamanickam, S.; Osman, S. Exploring the use of mobile apps for learning: A case study on final year engineering undergraduates in Malaysia. ASM Sci. J. 2020, 13, 63–67. [Google Scholar]
  7. Gezgin, D.M.; Adnan, M.; Acar Guvendir, M. Mobile learning according to students of computer engineering and computer education: A comparison of attitudes. Turk. Online J. Distance Educ. 2018, 19, 4–17. [Google Scholar] [CrossRef] [Green Version]
  8. Alturki, U.; Aldraiweesh, A. Students’ perceptions of the actual use of mobile learning during COVID-19 pandemic in higher education. Sustainability 2022, 14, 1125. [Google Scholar] [CrossRef]
  9. Güler, Ç. Use of WhatsApp in higher education what’s up with assessing peers anonymously? J. Educ. Comput. Res. 2017, 55, 272–289. [Google Scholar] [CrossRef]
  10. Kumar, J.A.; Bervell, B.; Annamalai, N.; Osman, S. Behavioral intention to use mobile learning: Evaluating the role of self-efficacy, subjective norm, and WhatsApp use habit. IEEE Access 2020, 8, 208058–208074. [Google Scholar] [CrossRef]
  11. Viberg, O.; Grönlund, Å. Understanding students’ learning practices: Challenges for design and integration of mobile technology into distance education. Learn. Media Technol. 2017, 42, 357–377. [Google Scholar] [CrossRef]
  12. Al-Emran, M.; Elsherif, H.M.; Shaalan, K. Investigating attitudes towards the use of mobile learning in higher education. Comput. Human Behav. 2016, 56, 93–102. [Google Scholar] [CrossRef]
  13. Almaiah, M.A.; Alamri, M.M.; Al-Rahmi, W. Applying the UTAUT model to explain the students’ acceptance of mobile learning system in higher education. IEEE Access 2019, 7, 174673–174686. [Google Scholar] [CrossRef]
  14. Mallya, K.R.; Srinivasan, B. Effect of cloud based mobile learning on engineering education. Int. J. Mech. Eng. Technol. 2019, 10, 614–621. [Google Scholar]
  15. Annamalai, N.; Kumar, J.A. Understanding smartphone use behavior among distance education students in completing their coursework in English: A mixed-method approach. Ref. Libr. 2020, 61, 199–215. [Google Scholar] [CrossRef]
  16. Garcia, A.; Vidal, E. Mobile-learning experience as support for improving the capabilities of the English area for engineering students. In Proceedings of the International Conference on Virtual Reality and Visualization (ICVRV), Hong Kong, China, 18–19 November 2019; pp. 202–204. [Google Scholar]
  17. Rafiq, K.R.M.; Hashim, H.; Yunus, M.M. Sustaining education with mobile learning for English for specific purposes (ESP): A systematic review (2012–2021). Sustainability 2021, 13, 9768. [Google Scholar] [CrossRef]
  18. Saikat, S.; Dhillon, J.S.; Ahmad, W.F.W.; Jamaluddin, R.A. A systematic review of the benefits and challenges of mobile learning during the covid-19 pandemic. Educ. Sci. 2021, 11, 459. [Google Scholar] [CrossRef]
  19. Almaiah, M.A.; Almomani, O.; Al-Khasawneh, A.; Althunibat, A. Predicting the Acceptance of Mobile Learning Applications During COVID-19 Using Machine Learning Prediction Algorithms. In Emerging Technologies During the Era of COVID-19 Pandemic. Studies in Systems, Decision and Control; Arpaci, I., Al-Emran, M., A. Al-Sharafi, M., Marques, G., Eds.; Springer: Cham, Switzerland, 2021; Volume 348, pp. 319–332. [Google Scholar]
  20. Al-Hamad, M.Q.; Mbaidin, H.O.; Alhamad, A.Q.M.; Alshurideh, M.T.; Al Kurdi, B.H.; Al-Hamad, N.Q. Investigating students’ behavioral intention to use mobile learning in higher education in UAE during Coronavirus-19 pandemic. Int. J. Data Netw. Sci. 2021, 5, 321–330. [Google Scholar] [CrossRef]
  21. Khlaif, Z.N.; Sanmugam, M.; Ayyoub, A. Impact of technostress on continuance intentions to use mobile technology. Asia Pac. Educ. Res. 2022, 1–12, Online ahead of print. [Google Scholar] [CrossRef]
  22. Loh, X.-K.; Lee, V.-H.; Loh, X.-M.; Tan, G.W.-H.; Ooi, K.-B.; Dwivedi, Y.K. The dark side of mobile learning via social media: How bad can it get? Inf. Syst. Front. 2021, 1–18. [Google Scholar] [CrossRef]
  23. Eneje, S. Real-world applications of mobile learning tools in engineering: Prospects, hindrances and accessibility in conjunction with scholastic views. In Proceedings of the IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), London, ON, Canada, 30 August–2 September 2020; Volume 2020, pp. 1–8. [Google Scholar]
  24. Rumreich, L.E.; Kecskemety, K.M. First-year engineering student perceptions and use of iPad technologies: A quantitative investigation of mobile learning. In Proceedings of the 2019 IEEE Frontiers in Education Conference (FIE), Covington, KY, USA, 16–19 October 2019; pp. 1–5. [Google Scholar] [CrossRef] [Green Version]
  25. Corral Abad, E.; Gómez García, M.J.; Diez-Jimenez, E.; Moreno-Marcos, P.M.; Castejón Sisamon, C. Improving the learning of engineering students with interactive teaching applications. Comput. Appl. Eng. Educ. 2021, 29, 1665–1674. [Google Scholar] [CrossRef]
  26. Arain, A.A.; Hussain, Z.; Rizvi, W.H.; Vighio, M.S. Extending UTAUT2 toward acceptance of mobile learning in the context of higher education. Univers. Access Inf. Soc. 2019, 18, 659–673. [Google Scholar] [CrossRef]
  27. Cheon, J.; Lee, S.; Crooks, S.M.; Song, J. An investigation of mobile learning readiness in higher education based on the theory of planned behavior. Comput. Educ. 2012, 59, 1054–1064. [Google Scholar] [CrossRef]
  28. Gómez-Ramirez, I.; Valencia-Arias, A.; Duque, L. Approach to M-learning acceptance among university students. Int. Rev. Res. Open Distrib. Learn. 2019, 20, 141–164. [Google Scholar] [CrossRef]
  29. Taylor, S.; Todd, P.A. Understanding information technology usage: A test of competing models. Inf. Syst. Res. 1995, 6, 144–176. [Google Scholar] [CrossRef]
  30. Lai, C.L. Trends of mobile learning: A review of the top 100 highly cited papers. Br. J. Educ. Technol. 2020, 51, 721–742. [Google Scholar] [CrossRef]
  31. Oke, A.; Fernandes, F.A.P. Innovations in teaching and learning: Exploring the perceptions of the education sector on the 4th industrial revolution (4IR). J. Open Innov. Technol. Mark. Complex. 2020, 6, 31. [Google Scholar] [CrossRef]
  32. Yang, H.; Su, C. Learner behaviour in a MOOC practice-oriented course: In empirical study integrating TAM and TPB. Int. Rev. Res. Open Distrib. Learn. 2017, 18, 35–63. [Google Scholar] [CrossRef] [Green Version]
  33. Al-Mamary, Y.H.; Al-nashmi, M.; Hassan, Y.A.G.; Shamsuddin, A. A critical review of models and theories in field of individual acceptance of technology. Int. J. Hybrid Inf. Technol. 2016, 9, 143–158. [Google Scholar] [CrossRef]
  34. Morchid, N. The current state of technology acceptance: A comparative study. IOSR J. Bus. Manag. 2020, 22, 1–16. [Google Scholar] [CrossRef]
  35. Han, S.; Yi, Y.J. How does the smartphone usage of college students affect academic performance? J. Comput. Assist. Learn. 2019, 35, 13–22. [Google Scholar] [CrossRef] [Green Version]
  36. Moorthy, K.; Tzu Yee, T.; Chun T’ing, L.; Vija Kumaran, V. Habit and hedonic motivation are the strongest influences in mobile learning behaviours among higher education students in Malaysia. Australas. J. Educ. Technol. 2019, 35, 174–191. [Google Scholar] [CrossRef] [Green Version]
  37. Alasmari, T.; Zhang, K. Mobile learning technology acceptance in Saudi Arabian higher education: An extended framework and A mixed-method study. Educ. Inf. Technol. 2019, 24, 2127–2144. [Google Scholar] [CrossRef]
  38. Briz-Ponce, L.; Pereira, A.; Carvalho, L.; Juanes-Méndez, J.A.; García-Peñalvo, F.J. Learning with mobile technologies—Students’ behavior. Comput. Hum. Behav. 2017, 72, 612–620. [Google Scholar] [CrossRef]
  39. Park, S.Y. An analysis of the technology acceptance model in understanding university students’ behavioral intention to use e-learning research hypotheses. Educ. Technol. Soc. 2009, 12, 150–162. [Google Scholar]
  40. Cain, M.K.; Zhang, Z.; Yuan, K.-H. Univariate and multivariate skewness and kurtosis for measuring nonnormality: Prevalence, influence and estimation. Behav. Res. Methods 2016, 49, 1716–1735. [Google Scholar] [CrossRef]
  41. Zhang, Z.; Yuan, K.-H. Practical Statistical Power Analysis Using Webpower and R; ISDSA Press: Granger, IN, USA, 2018. [Google Scholar]
  42. Hair, J.F.; Risher, J.J.; Sarstedt, M.; Ringle, C.M. When to use and how to report the results of PLS-SEM. Eur. Bus. Rev. 2019, 31, 2–24. [Google Scholar] [CrossRef]
  43. Ramayah, T.; Yeap, J.A.L.J.; Ahmad, N.N.H.; Abdul-Halim, H.; Rahman, S.A.; Halim, H. Testing a confirmatory model of Facebook usage in SmartPLS using consistent PLS. Int. J. Bus. Innov. 2017, 3, 1–14. [Google Scholar] [CrossRef] [Green Version]
  44. Hair, J.F.; Ringle, C.M.; Sarstedt, M. Partial least squares structural equation modeling: Rigorous applications, better results and higher acceptance. Long Range Plan. 2013, 46, 1–12. [Google Scholar] [CrossRef]
  45. Dijkstra, T.K.; Henseler, J. Consistent and asymptotically normal PLS estimators for linear structural equations. Comput. Stat. Data Anal. 2015, 81, 10–23. [Google Scholar] [CrossRef] [Green Version]
  46. Henseler, J.; Ringle, C.M.; Sarstedt, M. A new criterion for assessing discriminant validity in variance-based structural equation modeling. J. Acad. Mark. Sci. 2015, 43, 115–135. [Google Scholar] [CrossRef] [Green Version]
  47. Henseler, J.; Hubona, G.; Ray, P.A. Using PLS path modeling in new technology research: Updated guidelines. Ind. Manag. Data Syst. 2016, 116, 2–20. [Google Scholar] [CrossRef]
  48. Hair, J.F.; Howard, M.C.; Nitzl, C. Assessing measurement model quality in PLS-SEM using confirmatory composite analysis. J. Bus. Res. 2019, 109, 101–110. [Google Scholar] [CrossRef]
  49. Shmueli, G.; Sarstedt, M.; Hair, J.F.; Cheah, J.H.; Ting, H.; Vaithilingam, S.; Ringle, C.M. Predictive model assessment in PLS-SEM: Guidelines for using PLSpredict. Eur. J. Mark. 2019, 53, 2322–2347. [Google Scholar] [CrossRef]
  50. Ringle, C.M.; Sarstedt, M. Gain more insight from your PLS-SEM results the importance-performance map analysis. Ind. Manag. Data Syst. 2016, 116, 1865–1886. [Google Scholar] [CrossRef]
  51. Ong, A.K.S.; Prasetyo, Y.T.; Young, M.N.; Diaz, J.F.T.; Chuenyindee, T.; Kusonwattana, P.; Yuduang, N.; Nadlifatin, R.; Redi, A.A.N.P. Students’ preference analysis on online learning attributes in industrial engineering education during the covid-19 pandemic: A conjoint analysis approach for sustainable industrial engineers. Sustainability 2021, 13, 8339. [Google Scholar] [CrossRef]
  52. Kumar, J.A. Educational chatbots for project-based learning: Investigating learning outcomes for a team-based design course. Int. J. Educ. Technol. High. Educ. 2021, 18, 65. [Google Scholar] [CrossRef]
  53. Kumar, J.A.; Silva, P.A. Work-in-progress: A preliminary study on students’ acceptance of chatbots for studio-based learning. In Proceedings of the IEEE Global Engineering Education Conference (EDUCON), Porto, Portugal, 27–30 April 2020; pp. 1627–1631. [Google Scholar]
Sustainability 14 03197 g001 550
Figure 1. Hypothesized model based on C-TAM-TPB.
Figure 1. Hypothesized model based on C-TAM-TPB.
Sustainability 14 03197 g001
Sustainability 14 03197 g002 550
Figure 2. Parameter estimates for Pre and Post pandemic for mobile learning acceptance.
Figure 2. Parameter estimates for Pre and Post pandemic for mobile learning acceptance.
Sustainability 14 03197 g002
Table
Table 1. Kolmogorov-Smirnov test of normality pre and post-COVID.
Table 1. Kolmogorov-Smirnov test of normality pre and post-COVID.
Study1Study2
StatisticdfSig.StatisticdfSig.
Perceive ease of use (PE)0.1713260.0000.1073490.000
Perceive usefulness (PU)0.1273260.0000.1143490.000
Attitude (ATT)0.1533260.0000.1423490.000
Behavioural intention (BI)0.1663260.0000.1473490.000
Mobile learning self-efficacy (MSE)0.1703260.0000.1433490.000
Subjective norm (SN)0.1723260.0000.1653490.000
Table
Table 2. Demographic profile m-learning preference of respondents pre and post pandemic.
Table 2. Demographic profile m-learning preference of respondents pre and post pandemic.
ItemDescriptionStudy1Study2
n%n%
1.GenderMale19559.822363.9
Female13140.212636.1
2. Mobile devices commonly used for learningNetbook/Laptop11326.4626243.74
Smart Phone30370.9631552.59
Tab112.58223.67
Others00.0000.00
Total427 599
3. Mobile applications most commonly used to communicate with learning peerWhatsApp29634.2634546.12
Facebook13315.39324.28
Telegram13715.8626235.03
Short Messaging System (SMS)13515.63152.01
Google Hangout839.61283.74
IMO video chat778.91385.08
Others (WeChat, Instagram, Snapchat, MS Teams, Discord)30.35283.74
Total864 748
4. Mobile applications most commonly used to communicate with lecturersWhatsApp29347.3331150.16
Facebook589.3750.81
Telegram6911.1525941.77
Short Messaging System (SMS)8914.3820.32
Google Hangout558.89274.35
IMO video chat345.4950.81
Others (WeChat, MS Teams)213.39111.77
Total619 620
5. Mobile applications most commonly used to share information/notes with learning peers?WhatsApp29940.6834051.83
Facebook11515.65101.52
Telegram11315.3727441.77
Short Messaging System (SMS)7410.0740.61
Google Hangout7910.75162.44
IMO video chat506.8050.76
Others (WeChat, MS Teams)50.6871.07
Total735 656
6. Mobile applications used for learningEmail (e.g., Gmail, Yahoo Mail etc.)25814.6021416.58
Calendar (e.g., Google Calendar)1397.87393.02
Cloud storage (e.g., Dropbox, Google Drive etc.)18010.191189.14
Creating and editing documents (e.g., Google docs, Mobile Office 365 etc.)23213.1315912.32
Educational games (e.g., Kahoot)19110.81896.89
Online learning platforms (e.g., Moodle, MOOC, CIDOS, Google Classroom)25614.4927221.07
Internet surfing for learning Contents (e.g., Chrome, Firefox, YouTube)27915.7920716.03
Reading (e.g., Adobe, Epub reader)23213.13564.34
Video conferencing (WebEx, Zoom, etc.)00.0013710.61
Total1767 1291
7. Access to internetPersonal data plan29732.3931959.63
WIFI at the institution26128.465710.65
WIFI at home17919.5213725.61
Public WIFI18019.63224.11
Total917 535
Note. Study 1 refers to pre-COVID and Study 2 to post COVID.
Table
Table 3. Results of indicator reliability, composite reliability, and convergent validity analysis.
Table 3. Results of indicator reliability, composite reliability, and convergent validity analysis.
VARItemStudy1Study2
LoadingsIRCRαrho_AAVELoadingsIRCRαrho_AAVE
ATTATT10.8570.9260.9040.8410.8420.7590.8850.9410.8860.8060.820.886
ATT20.8650.9300.7740.880
ATT30.8910.9440.8850.941
BIBI10.8960.9470.9350.8950.8960.8270.8890.9430.9240.8770.8770.924
BI20.9180.9580.9070.952
BI30.9130.9560.8900.943
MSEMSE10.8810.9390.9290.8840.8850.8120.9360.9670.9470.9170.9210.947
MSE20.9200.9590.9230.961
MSE30.9020.9500.9180.958
PEPE10.8160.9030.9080.8650.8660.7120.8560.9250.8950.8430.8470.895
PE20.8370.9150.7690.877
PE30.8740.9350.8640.930
PE40.8470.9200.8100.900
PUPU10.8320.9120.8910.8360.8440.6720.8350.9140.8900.8340.8410.890
PU20.7880.8880.7920.890
PU30.8930.9450.8800.938
PU40.7590.8710.7600.872
SNSN10.9230.9610.9360.8970.9040.8290.9220.9600.9330.8920.8950.933
SN20.9110.9540.9230.961
SN30.8980.9480.8760.936
Note. VAR: Variable; IR: Indicator Reliability; CR: Composite Reliability; ATT: Attitude; BI: Behavioral intention; MSE: Mobile learning self-efficacy; PE: Perceive ease of use; PU: Perceive usefulness; SN: Social norm.
Table
Table 4. HTMT ratio of the studies.
Table 4. HTMT ratio of the studies.
VariablesATTBIMSEPEPUSN
Study1
Attitude (ATT)-
Behavioural intention (BI)0.989-
Mobile self-efficacy (MSE)0.9020.935-
Perceive ease of use (PE)0.9260.8730.807-
Perceive usefulness (PU)0.9860.9110.8620.966-
Subjective norm (SN)0.8360.8140.860.7820.84-
Study2
Attitude (ATT)-
Behavioural intention (BI)0.984-
Mobile self-efficacy (MSE)0.8630.894-
Perceive ease of use (PE)0.9490.8580.785-
Perceive usefulness (PU)0.9520.8700.8560.964-
Subjective norm (SN)0.8560.8510.8300.8190.854-
Table
Table 5. Model fit of the study.
Table 5. Model fit of the study.
ModelStudy 1Study 2
Saturated ModelHI95ConclusionSaturated ModelHI95Conclusion
SRMR0.0290.062Supported0.0520.084Supported
d_ULS0.1740.817Supported0.5591.484Supported
d_G0.2230.324Supported0.4500.526Supported
Table
Table 6. Results for path analyses, corresponding t-value, VIF, confidence intervals and f2.
Table 6. Results for path analyses, corresponding t-value, VIF, confidence intervals and f2.
RelationshipStudy 1Study 2
βStdevtpCI
2.5%		CI 95%f2VIFβStdevtpCI
2.5%		CI 95%f2VIF
H1:
ATT → BI		0.4430.0666.7020.0000.3110.5710.2623.9610.4430.0646.8980.0000.3130.5620.2793.177
H2:
MSE → BI		0.3700.0705.2550.0000.2360.5120.2173.3200.3290.0595.6080.0000.2200.4450.1603.082
H3:
PE → ATT		0.3400.0655.2580.0000.2100.4700.1313.1400.4440.0617.2920.0000.3230.5590.2062.968
H4:
PE → PU		0.8260.02041.3680.0000.7840.8632.1401.0000.8140.01941.9860.0000.7730.8511.9681.000
H5:
PU →ATT		0.5460.0628.7510.0000.4200.6680.3383.1400.4200.0656.4570.0000.2950.5490.1842.968
H6:
PU →BI		0.1220.0602.0430.0410.0060.2410.0223.6360.0330.0580.5660.571-0.0810.1470.0013.297
H7:
SN →BI		0.0380.0650.5880.557-0.0890.1670.0032.9030.1610.0572.8180.0050.0520.2800.0412.860
Note. ATT: Attitude; BI: Behavioural intention; MSE: Mobile learning self-efficacy; PE: Perceive ease of use; PU: Perceive usefulness; SN: Social norm.
Table
Table 7. Coefficient determination (R2) of the model.
Table 7. Coefficient determination (R2) of the model.
VariablesStudy 1Study 2
R2R2 AdjustedR2R2 Adjusted
ATT0.7200.7180.6780.676
BI0.8110.8090.7790.777
PU0.6820.6810.6630.662
Note. ATT: Attitude; BI: Behavioural intention; PU: Perceive usefulness.
Table
Table 8. Predictive Relevance (Q2) of the model.
Table 8. Predictive Relevance (Q2) of the model.
VariableStudy 1Study 2
SSOSSEQ² (1-SSE/SSO)SSOSSEQ² (1-SSE/SSO)
ATT978.000451.1950.5391047.000542.6010.482
BI978.000330.3390.6621047.000401.7460.616
MSE978.000978.0000.0001047.0001047.0000.000
PE1304.0001304.0000.0001396.0001396.0000.000
PU1304.000716.5540.4501396.000782.6150.439
Note. ATT: Attitude; BI: Behavioural intention; MSE: Mobile learning self-efficacy; PE: Perceive ease of use; PU: Perceive usefulness; SN: Social norm.
Table
Table 9. Out-of-sample predictive power (Q2predict) of the model.
Table 9. Out-of-sample predictive power (Q2predict) of the model.
VariableIndicatorStudy 1Study 2
PLSLMa-bPLSLMa–b
RMSE (a)Q2
predict		RMSE
(b)		Q2
predict		 RMSE
(a)		Q2
predict		RMSE
(b)		Q2
predict
ATTATT10.5040.5570.4990.5650.0050.6410.5360.6130.5750.028
ATT20.7100.3840.6430.4950.0670.6640.3030.6180.3950.046
ATT30.5910.4650.5540.5310.0370.6430.4830.6220.5170.021
BIBI10.5360.5780.5430.567−0.0070.5660.520.5680.517−0.002
BI20.5060.6060.5070.604−0.0010.5750.5390.5790.532−0.004
BI30.4980.6380.4910.6480.0070.5620.6380.5410.6660.021
PUPU10.6890.3570.6980.340−0.0090.7130.4850.6580.5600.055
PU20.5680.5240.5340.5790.0340.7210.3980.7020.4300.019
PU30.4700.5770.4630.5890.0070.6150.5420.5740.6010.041
PU40.6680.3540.6430.4010.0250.7450.3340.7220.3750.023
Note: RMSE: root mean squared error; PLS: partial least squares; LM: linear model; Q2: predictive relevance; ATT: Attitude; BI: Behavioural intention; PU: Perceive usefulness.
Table
Table 10. Performance Index Values and Total Effects for BI.
Table 10. Performance Index Values and Total Effects for BI.
VariableStudy 1Study 2
Importance-
Total Effects		Performance-
Index Values		Importance- Total EffectsPerformance- Index Values
Attitude (ATT)0.46380.5720.45874.574
Mobile self-efficacy (MSE)0.37273.0960.30064.756
Perceive ease of use (PE)0.51479.470.37370.055
Perceive usefulness (PU)0.41177.140.22169.492
Subjective norm (SN)0.03972.9340.15267.345
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Kumar, J.A.; Osman, S.; Sanmugam, M.; Rasappan, R. Mobile Learning Acceptance Post Pandemic: A Behavioural Shift among Engineering Undergraduates. Sustainability 2022, 14, 3197. https://doi.org/10.3390/su14063197

AMA Style

Kumar JA, Osman S, Sanmugam M, Rasappan R. Mobile Learning Acceptance Post Pandemic: A Behavioural Shift among Engineering Undergraduates. Sustainability. 2022; 14(6):3197. https://doi.org/10.3390/su14063197

Chicago/Turabian Style

Kumar, Jeya Amantha, Sharifah Osman, Mageswaran Sanmugam, and Rasammal Rasappan. 2022. "Mobile Learning Acceptance Post Pandemic: A Behavioural Shift among Engineering Undergraduates" Sustainability 14, no. 6: 3197. https://doi.org/10.3390/su14063197

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop