1 Introduction

Online educational technologies have long been regarded as supplemental resources that promote traditional classroom education. However, with the COVID-19 pandemic, people witnessed a boom of online learning trend, mainly due to the partial or complete school closures across the world (Adnan & Anwar, 2020). During this period, all educational institutions including universities have sought ways to immediately adapt to the emergency remote teaching process by implementing distance or hybrid programs. Researchers from all over the world have paid mounting attention to understanding the efficacy of these remote teaching and learning processes.

As in all other countries, the pandemic has been a milestone for online education in the Turkish education system. As a response to the need of emergency remote teaching Turkish Council of Higher Education (CoHE) and universities have accepted digitalization of learning systems as an important component of strategic policy in higher education. This approach was embraced by the Turkish government after the pandemic and has triggered to capability enhancement of academic and administrative staff, and students in terms of digitalization of learning environment in all higher education institutions (HEIs) especially including the ones located in disadvantageous regions in Türkiye. Thanks to various projects, thousands of staff and millions of students have taken digital literacy courses. Also, a Massive Open Online Course (MOOC), named “Teaching and Learning at Higher Education in Digital Age” was developed both for academics and students. Furthermore, CoHE bolsters academic studies and practices that promote openness at enhancing the digital literacy levels of learners and faculty members (Bozkurt et al., 2022). Türkiye, home to 208 HE institutions and almost 7 million students, has the highest number of university student population in Europe. Turkish higher education authorities have made significant efforts to ensure uniformity in students’ ability to take the advantage of modern digital technologies for learning purposes in online education. Educational institutions and the government have invested in developing digital infrastructure, such as better internet connectivity and providing students with the necessary electronic devices. The Republic of Türkiye offers all university students 10 GB of free wireless internet every month and provides financial support to buy new tablets or computers (Official Gazette of the Republic of Türkiye, 2023) which makes it easily available for even students with limited opportunities to access internet resources and online materials. In addition, CoHE established an online educational commission that developed an e-learning portal to provide essential tutoring support for faculty members and learners in adapting to online educational modalities. Moreover, CoHE created a free and open-access platform through compiling educational materials provided by HEIs (AA, 2021). Also, Turkish HEIs have integrated purpose-specific digital instruments and resources to optimize and enhance the capability and success of their potential. Several institutions including the six university in this study also invested in institutional open courseware projects that offer a non-profit and open-access online resources for academics, self-learners across the world and share the lesson materials created for all learners through an innovative online learning portal (TÜBA, 2023). In addition, MOOCS developed by universities and other institutions in Türkiye (e.g. Akadema by Anadolu University, AtademiX by Atatürk University, Ninova by Istanbul University, BTKAkademi by Information Technology and Communications Authority in Türkiye, YOK Dersleri by CoHE etc.) provide all university students with access to online educational materials regarding their needs. Similarly, The Ministry of Education projects fulfills ambitious goals to improve the hardware and software infrastructure in the national education system such as providing VPN-broadband and high-speed access; installing interactive whiteboards; wired/wireless internet access; education information network applications including a digital material sharing platform, cloud accounts for digital learning and individualized learning materials to every student. On the other hand, like any system, online education in Türkiye faces challenges, such as ensuring equal access for all students, maintaining the quality of education, and addressing non-cognitive aspects of learning that is often missing in online environments.

Students’ online learning engagement referring to learners’ active participation and persistent endeavors within an online learning setting to accomplish intended learning goals (Jung & Lee, 2018) is regarded as a crucial indicator of effective online education (e.g., Dumford & Miller, 2018; Martin & Bolliger, 2018). Accumulated evidence in the literature has persistently shown that students’ learning engagement plays a significant role in their achievement (e.g., García-Martinez et al., 2021). It is considered a constant predictor of students’ performance (e.g., Miltiadous et al., 2020). Learning engagement has critical importance in terms of the realization and permanence of learning (Chen et al., 2008). Banna et al. (2015) underline that learning engagement is a fundamental solution to the problem of student isolation, low retention, and dropout rates in online education. Therefore, individuals need to have some responsibilities and actively involve in learning activities to get maximum benefit from online education. Research conducted to reveal the variables which impact online learning engagement have created a growing interest not only in online learning readiness but also in academic resilience, both of which are crucial drivers of students’ learning engagement (e.g., Martin & Marsh, 2009).

Online learning readiness, a multifaceted variable that includes individuals’ different competency domains related to online learning such as computer, internet and online communication self-efficacy, self-directed learning, learner control in an online context and motivation level that influence success in online education (Hung et al., 2010) supports individuals to boost their active learning effectiveness and academic achievement in online education (e.g., Ergün & Kurnaz Adıbatmaz, 2019; Wei & Chou, 2020). It is also regarded as a prerequisite for effective learning and active participation. So, if individuals’ readiness is at an unsatisfactory level, the process may turn into a waste of time, resources, budgets, and efforts (Moftakhari, 2013). Likewise, academic resilience the capacity of individuals to successfully manage obstacles and challenges in educational context to attain academic achievement (Cassidy, 2016; Jowkar et al., 2014; Martin, 2013) is considered another crucial variable in online learning. Findings in favor of resilient students in terms of behaviors that are important in the direct learning process, such as attending class on time and being prepared, making the necessary effort to complete school tasks or assignments, and participating in class, highlight the decisive role of academic resilience in their learning engagement (Finn & Rock, 1997).

Studies on online education have remarkably expanded during the pandemic and post-pandemic periods, resulting in a vast body of knowledge. However, there is still a gap in this growing literature with regards to understanding the student-related variables that impact the quality, effectiveness, and permanence of online education. Also, insufficient quantitative evidence on the impact of which student-related variables effect on students’ online learning engagement levels, coupled with the mechanisms of the possible interactions among those variables points out a need for research. In this study, the direct and indirect effects of individuals’ online learning readiness on their online learning engagement are investigated, with evidence from higher education institutions across Türkiye. Thus, it fills a crucial gap in the extant literature: the interplay between learners’ readiness in online learning and their engagement levels, with a specific focus on the mediating role of academic resilience. This approach offers a novel perspective by correlating these variables with the effectiveness and quality of online learning, providing valuable insights for educators, policymakers, and researchers in Türkiye and beyond. This study is predicated upon the following research questions:

  1. 1.

    To what extent is learners’ online learning readiness related to their online learning engagement?

  2. 2.

    How does academic resilience mediate the effects of learners’ online learning readiness on their online learning engagement?

2 Conceptual framework

This section presents theoretical explanations of the constructs and provides rationales for the study’s hypotheses based upon a conceptual framework and empirical relationships among the selected variables in the study. In outlining the study’s conceptual framework, first theoretical rationale for selecting the independent variables is explained. Then, online learning engagement, the main dependent variable is discussed. Next, academic resilience and online learning readiness are addressed. Finally, the focus is placed on the theoretical associations among the aforementioned variables. Following this section, the conceptual model guiding the study design is presented.

2.1 Theoretical rationale

The rationale for choice of the independent variables in the research is based on a robust theoretical framework which discuss online education as a complicated, dynamic and adjustable system. This viewpoint calls for a thorough investigation into the factors that significantly affect the success of online education. First, learning readiness is considered a critical determinant in studies on online education, because it epitomizes the readiness of individuals to engage with online learning activities. Learning readiness comprises factors such as technological skills, digital literacy, capacity for self-directed learning, and flexibility in adapting to online teaching and learning approaches, techniques and methodologies (Hung et al., 2010). The integration of learning readiness is in line with the Technology Acceptance Model (Davis, 1989) emphasizing readiness as a vital element in the adoption of new technologies including online education and related aspects. Likewise, the choice of academic resilience variable is grounded on the theory of psychological resilience (Masten et al., 1990) that claims the capacity to thrive in the face of adversity or tough conditions is crucial for academic achievement. The distinctive challenges of online education such as coping with distractions, isolation, lack of peer interaction or technological barriers underscore the need for academic resilience as a pivotal factor for comprehending learners’ perseverance, success and performance (Hartley, 2012). Third, engagement in online education is a multi-faceted component that comprises of behavioral, emotional, and cognitive aspects (Fredricks et al., 2004). The choice of this variable is grounded on the Self-Determination Theory (Deci & Ryan, 2000) which proposes individuals’ participation is prerequisite for permanent learning and academic success, particularly in settings that need higher levels of self-regulation and in online education.

2.2 Online learning engagement

Individuals’ active participation and sustained effort throughout the learning process to achieve intended learning outcomes in educational settings refer to their engagement in online learning (Jung & Lee, 2018). Robinson and Hullinger (2008) adopted five dimensions defined by the National Survey of Student Engagement to estimate student engagement in online courses. They referred to five benchmarks in their study: academic rigor, a nurturing campus atmosphere, enriching learning experiences, faculty-student engagement, and interactive and cooperative learning. Addressing learning engagement in online contexts, Huang (2013) claims that engagement refers to students’ choice to absorb learning material as a consequence of their interaction in an online learning environment. It also serves as a signifcant indicator of student outcome and academic performance (e.g., Chen et al., 2010). Prior research has proven that evidence that engagement in online learning is significantly linked to quality learning outcomes and academic performance (e.g., Axelson & Flick, 2010). Studies have also demonstrated the significance of engagement in long-term learning success by illustrating a strong relationship between learning engagement and retention (e.g., Xiong et al., 2015). Besides, research have indicated that increased learning engagement could reduce sense of isolation (Banna et al., 2015) and enhance motivation and satisfaction levels (Dunne & Owen, 2013).

Although a handful of research has studied learning engagement as a single variable (e.g., Hui et al., 2019), many quantitative and qualitative research findings support its multi-faceted construct. Fredricks et al. (2004) categorized learning engagement into behavioral, cognitive, and emotional factors. On the other hand, Kahu (2013) identified four distinct approaches to theoretical frameworks on engagement: behavioral, psychological, socio-cultural, and holistic. Handelsman et al. (2005) offered four distinct factors to assess student engagement in a traditional educational environment. These are labeled as skills engagement composed of individuals’ general learning strategies; emotional engagement referring to the emotional attachment with learning content; participation to interactive and cooperative learning; and performance engagement.

This research is guided by Luan et al.’s (2020) conceptualization of online learning engagement due to its inclusive theoretical background. Researchers built the framework on four dimensions. The first dimension, cognitive engagement points out the cognitive endeavors that learners make to obtain sophisticated knowledge or develop specific abilities through online learning. The second dimension, emotional engagement, refers to individuals’ feelings about online lectures. Social engagement relates to students’ social experiences with their environment and enthusiasm for developing and sustaining networks around the instructional context. Finally, behavioral engagement, which involves learning activities such as taking part in online conversations or sharing ideas about a task, is considered an important factor that influences both cognitive and emotional engagement (Li & Lerner, 2013).

2.3 Academic resilience

Resilience, a distinct trait of human characteristics, is the ability of human beings to deal with adversity and continue to function normally in the face of stressors (Masten & Tellegen, 2012). It indicates the capability to recover and overcome difficulties (Cassidy, 2016). Academic resilience is characterized as learners’ potential for academic achievement regardless of individual difficulties, challenges, or vulnerabilities associated with environmental circumstances (Colp & Nordstokke, 2014). Academically resilient individuals are described as those who have remarkable capabilities to bounce back from academic challenges or failures and achieve success, while others continue to underperform (Aydın & Erdem, 2022; Martin & Marsh, 2009).

Although the concept of academic resilience was previously considered primarily as a psychological construct that was studied in the context of students from disadvantaged groups in earlier studies (e.g., Finn & Rock, 1997; Lindstroem, 2001), current literature has revealed that academic resilience is a valid and important factor for all students. Martin and Marsh (2009) reported that academic resilience facilitates and supports to improve learners’ competencies in dealing with setbacks, challenges, and stress in the educational context. They regarded academic resilience as the most powerful indicator of joy at school, classroom engagement, and self-esteem. They also found that five factors related to motivation and engagement (self-efficacy, control, planning, low anxiety, and persistence) are important indicators of academic resilience. Some empirical research (e.g., McLafferty et al., 2012) clearly report that increased resilience enables students to grapple with adversity in academic settings. In addition, many research findings (e.g., Karabıyık, 2020; Shaw, 2008) reveal the interplay among academic resilience and academic achievement or effective instructional practices.

In this study, Martin and Marsh’s (2009) instrument, used as a basis for developing the others, was employed. In the scale development process, the researchers examined the educational correlates of academic resilience. The researchers included scale items that measure learners’ ability to manage difficulties, problems, and stress in academic settings which are specifically related to mental resilience, bouncing back from low grades, coping with school pressures and stress, and obstacles like poor academic performance and negative feedback.

2.4 Online learning readiness

Warner et al. (1998), the first to characterize online learning readiness,, underlined the importance of students’ perceptions of instructional preferences, their confidence and competence levels with novel instructional technologies, and their perceptions of their abilities as autonomous learners. Lopes (2007) explained online learning readiness as the capacity of learners to utilize the advantages of digital learning opportunities. Tang et al. (2021) explicated the notion based on definitions proposed in different studies in the literature as individuals’ perception in educational delivery, self-confidence on online educational platforms, and learners’ autonomy in learning engagement.

Learners with lower readiness for online learning might possibly have negative experiences and biases towards e-learning (Guglielmino & Guglielmino, 2003). Prior research have proven the effectof online learning readiness on students’ motivation and satisfaction levels (e.g., Tang et al., 2021). Akaslan and Law (2011) underlined that students’ learning readiness level is a strong indicator of students’ learning performance in online education. ’In addition, earlier research has demonstrated a clear and distinctive link between online learning readiness and academic achievement (e.g., Wei & Chou, 2020).

Upon examining the existing models and instruments, this study chose to utilize the structural framework built by Hung et al. (2010), adopting a more inclusive and multi-faceted approach compared to previous onesfor a total evaluation of students’ online learning readiness levels in this study. It includes five dimensions: self-directed learning (SL), motivation for learning (ML), learner control (LC), computer and internet self-efficacy (CIS) and online communication self-efficacy (OCS). SL, as Zimmerman (2002) defines the concept, encourages students to take responsibilities of their learning This includes setting learning objectives based on their needs, actively selecting resources, planning, and applying effective learning methods and measuring, monitoring, and evaluating their learning outcomes. Secondly, ML, which is related to several positive student outcomes (e.g., Kaplan et al., 2003), considerably promotes individuals’ attempts to improve their learning, retention, and retrieval. The third dimension, LC refers to the extent to which a learner may influence learning experience through changing their learning environment, particularly in computer-based education (Friend & Cole, 1990). CIS includes two interrelated variables: computer self-efficacy and internet self-efficacy. CIS is directly linked to students’ perceptions of their technological competency associated with knowledge and skills of how to use the internet and learning software (Torkzadeh et al., 2006). Finally, OCS is associated with how individuals perceive their comprehension of the language and culture specific to online learning settings and their capability to communicate effectively (Yurdugül & Alsancak Sırakaya, 2013).

2.5 The relationships between online learning engagement, online learning readiness and academic resilience

Martin (2002), who developed a comprehensive psychological and behavioral model of engagement called “Student Motivation and the Wheel of Engagement”, proposed that engagement is one of the underlying dimensions of academic resilience, along with the joy of school and overall self-esteem (Martin & Marsh, 2009). Researchers also highlighted the link between engagement and resilience. They hypothesized that learners incapable of coping efficiently with such challenges are also less likely to be actively engaged in courses. Kaur and Abas (2004) drew attention to the positive correlation between online learning readiness and individuals’ participation in the online learning process. Ergün and Kurnaz Adıbatmaz (2020), who investigated learners’ level of readiness to predict their engagement, revealed that SL and CIS, two distinct factors of readiness, are potent indicators of students’ behavioral engagement. Additionally, they noted that while LC and SL, as well as ML and OCS, are the strongest predictors of emotional engagement, LC and SL are the key determinants of cognitive engagement. Recent empirical studies reveal significant positive correlation between learning readiness and learning engagement, supporting this relationship (e.g. Ravandpour, 2022; Yıldız Durak, 2018). In addition, Jiang et al. (2021) not only explained a direct cause-and-effect link between readiness and engagement but also went beyond this and discussed the dimensions of these variables that could shape the relationships. Besides, a recent study (Ramadhana et al., 2021) illuminated that learning readiness is regarded as a key underpinning element in academic resilience. Accordingly, readiness for online learning was an important indicator of learners’ academic resilience through ML and SL. In addition, resilience is positively linked with learning engagement, and is a mediator in the interplay between families’ socio-economic levels and learning engagement (Chen et al., 2021).

A holistic view of the existing research relationships suggests that academic resilience is likely to play a key mediating role between the two constructs. In this context, academic resilience is expected to mediate between online learning readiness and engagement. In other words, individuals’ learning readiness can predict their learning engagement in online education through academic resilience. Their academic resilience levels could explain online learning readiness’ role on learning engagement. The conceptual model (see Fig. 1) guiding this research is based on the theoretical foundation outlined above, supported by the earlier empirical research findings. Based on the existing literature explained in the previous section, this study proposes that learning readiness may have direct and indirect effects on learning engagement in online education, and academic resilience may act as a potential mediating variable. More specifically, the current study hypothesizes that: (H1) online learning readiness positively affects learning engagement; (H2) online learning readiness positively affects academic resilience; (H3) academic resilience positively affects learning engagement; and (H4) academic resilience mediates the effects of online learning readiness on learning engagement.

Fig. 1
figure 1

Conceptual framework of the study

Full size image

The model hypothesizes that readiness affects engagement both directly and indirectly through resilience. This is in line with the Interactive Model of Resilience (Masten & Reed, 2002), which suggests that individual characteristics (like readiness) interact with environmental factors (such as the online learning context) to influence resilience, which in turn impacts engagement. In conclusion, the selection of readiness, resilience, and engagement as the variables of this study is deeply rooted in established theoretical frameworks relevant to online education literature. These variables collectively present a holistic insight into the dynamics in online learning environments, thus addressing a crucial gap in the related literature.

The first hypothesis that online learning readiness positively influences learning engagement is based on the self-determination theory (Deci & Ryan, 2000), emphasizing the role of individual competence in fostering engagement. Online learning readiness, encompassing aspects like self-directed learning and technological proficiency, contributes to a learner’s perceived competence in online environments (Bandura, 1997). Warner et al. (1998), propose that online learning readiness enables learners with core skills and confidence to engage efficiently with online learning resources, thus enhance their engagement. Empirical studies (e.g. Hung et al., 2010), supports this link, underlining significant correlations between readiness components like self-efficacy and online engagement.The second hypothesis of the study builds on the notion of academic resilience, claiming that is anchored in the concept of academic resilience which is rooted in positive psychology. The notion is that students who are well-prepared and confident students in their online learning skills are more likely to exhibit academic resilience when confronted academic challenges. The self-efficacy theory by Bandura (1997) also supports this idea, plays a significant role.suggesting that The theory suggests that a learner’s trust in their ability to plan and act in online learning scenarios can contributelead to greater resilience. The third hypothesis, the positive impact of academic resilience on learning engagement is supported by the resilience theory (Masten, 2001), which posits that resilience can enhance a student’s ability to remain engaged in challenging contexts, such as online learning. Resilient students are likely to exhibit persistence, a key component of engagement, even when faced with difficulties (Masten & Reed, 2002). Empirical evidence (Azila-Gbettor, 2023; Martin & Marsh, 2009; Pidgeon et al., 2014; Tudor & Spray, 2017; Versteeg et al., 2022) demonstrates a significant relationship between resilience and engagement, where students who show higher resilience tend to maintain better engagement in their academic activities. The mediation role of academic resilience between online learning readiness and engagement is proposed based on the interactionist perspective (Lazarus & Folkman, 1984), which suggests that individuals’ reactions (engagement) to environmental demands (online learning) are mediated by personal factors (resilience). This hypothesis is further reinforced by the ecological systems theory (Bronfenbrenner, 1979), which underlines that personal growth is molded by the interaction between personal attributes and environmental factors.

3 Method

A cross-sectional quantitative design was employed to explore the effects of language learners’ online learning readiness on their online learning engagement levels through academic resilience as the mediating variable in the present study. In this section, the study sample and data collection procedure, variables and measures, and analytical strategy were described.

3.1 Study sample and data collection

In order to obtain a nationally representative sample, using the convenience sampling technique 6 universities were selected from different cities in the eastern, central and western regions of Türkiye according to the convenience sampling technique. Then, the ethics committee document, the purpose, method, measurement tools to be used, and an extended summary of the importance of the research were sent to the rectorates of these universities. After the approval of the rectorates, the researcher contacted faculty administrators and heads of departments through e-mail. They were given detailed information about the research by organizing an online meeting and department heads shared infographics prepared by the researcher giving brief information about the research through the communication channels they communicate with their students who had successfully completed the upper intermediate level of English preparatory class (e-mail, Facebook, instant messaging groups, etc.) and shared a WhatsApp group link created for this research for volunteer students to involve in the study. To ensure that participants give informed consent by understanding the aim of current research and how their data will be used, the researcher shared all the information about the research again clearly and answered the participants’ questions about the research. Then a link to an online data collection form with informative instructions created by the researcher was shared with the participants. Students who wished to voluntarily participate in the research could answer the validated and reliable scales (Hung et al., 2010; Luan et al., 2020; Martin and Marsh, 2009). The user-friendly online data collection platform, which pays attention to the security and confidentiality of personal data, has implemented features to prevent duplicate responses. In addition to the advantages of digitalization of data collection such as accessing data from a wide range of participants with different profiles in a small amount of time, reduction of paper usage, transportation costs, and the use of space for storing papers (Fallaize et al., 2014), by enabling individuals to take part in online data collection, participants are more likely to share information they would otherwise be reluctant to disclose (Bonini-Campos et al., 2011). As a result, this can lead to increased accuracy in reporting as opposed to socially desirable responses, commonly seen during face-to-face data collection (Aust et al., 2013). Even though 4020 students voluntarily joined the research group, only 3620 of them responded to the questionnaires. Of these, 70 were discarded from the dataset because of a great deal of missing data. Consequently, a total of 3550 response were utilized in data analysis process. Of the participants, 30.15% (n = 1070) of the students were female, and 69.85% (n = 2480) were male.

Table 1 Demographic details about the participants
Full size table

3.2 Variables and measures

Three fundamental variables used in this research were online learning readiness, online learning engagement, and academic resilience. The original versions of the instruments were employed in this research, since the study sample consisted of learners successfully completed the upper intermediate level of English preparatory class.

Online learning engagement was used as the main endogenous variable, and the Online English Learning Engagement (OELE) scale by Luan et al. (2020) was used in this study. OELE comprises a total of 21 items grouped under four factors: cognitive engagement (CE) (6 items), behavioral engagement (BE) (5 items), emotional engagement (EE) (4 items), and social engagement (SE) (6 items). The scale uses a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Academic resilience was used as the mediating variable between online learning readiness and online language learning engagement. ARS-30 developed by Martin and Marsh (2009) was composed of six items loaded on one strong factor structure. Scale items were individually rated on a 7-point Likert scale with anchors ranging from 1 (strongly disagree) to 7 (strongly agree).

The main exogenous variable was online learning readiness. Online Learning Readiness Scale (OLRS) originally developed by Hung et al. (2010) was composed of 18 items, which were grouped into five factors: self-directed learning (SL) (5 items), motivation for learning (ML) (4 items), computer/internet self-efficacy (CIS) (3 items), learner control (LC) (3 items), and online communication self-efficacy (OCS) (3 items). Scale items were individually rated on a Likert scale with anchors ranging from 1 (strongly disagree) to 5 (strongly agree).

3.3 Analytic strategy

The data analysis process was performed through sequential steps. First of all, descriptive statistics were conducted, and Cronbach’s alpha coefficients and composite reliability values were calculated using SPSS and Microsoft Excel. Second, the convergent, divergent, and construct validity analyses were performed. Third, Variance Inflation Factor (VIF), Tolerance Value (TV) and Condition Index (CI) were computed to test for multicollinearity. Fourth, Pearson correlations among the main variables and the factors were obtained to determine the relationships among variables. Structural Equation Modelling (SEM) is regarded as an effective multivariate statistical method to examine the interplay among various components to reveal the level of fit for the conceptual model and to determine the direct, indirect, and total effects of the endogenous variable(s) on the exogenous variable(s) through mediator(s). Thus, SEM was performed to analyze structural relations in the proposed model (see Fig. 1) with AMOS 26. In addition, a bootstrap method with 5,000 samples was performed through The Hayes’ Process Macro for SPSS, which provides more robust findings than Sobel test, to assess the effect size and to evaluate the confidence intervals (CI) and significance levels for the paths in the model as proposed by Preacher and Hayes (2008).

4 Findings

Before conducting SEM, the normality assumption was checked by means of skewness and kurtosis coefficients. The values in Table 1 were between − 1 and + 1 which proved that the data yielded a normal distribution (Tabachnick & Fidell, 2019). The mean values ranging between 2.79 and 3.36 display that while students’ online learning readiness (M = 3.18) and online learning engagement levels (M = 3.04) are moderate on the 5-point Likert-type scales, their academic resilience levels are relatively lower (M = 4.06) on the 7-point Likert-type scale (See Table 2).

Table 2 Descriptive statistics
Full size table

Multicollinearity is an important issue in the process of SEM, which might challenge the results of the analysis so, multicollinearity was tested before SEM. The obtained VIF values (ranging from 1.54 to 3.14), TV values (ranging from 0.32 to 0.65), and CI values (ranging from 1 to 24.45), which are indicators for identifying multicollinearity, illustrate that there was no multicollinearity problem among variables (Weisberg, 2005). Also, the results of correlation analyses were conducted to examine the interrelationships among the variables (see Table 3) demonstrated no potential multicollinearity problem among the scales and their factors. Besides, the correlation between academic resilience and online learning engagement (r = .49) was larger than those between academic resilience and online learning readiness (r = .37) and the correlation between online learning engagement and online learning readiness (r = .34). However, all of the correlation values among the scales and factors were positive and significant (p < .001).

Table 3 Correlations among variables and measures
Full size table

In order to test the reliability, CR and α were calculated. Also, AVE and square root of the AVE values of the scales and factors were computed for the convergent and discriminant validity. Furthermore, CFA was performed to explore the measures’ construct validity. The reliability scores (see Table 4) acquired through CA and CR of all the scales and factors were high and met the desired standard (α = 0.70). Moreover, AVE coefficients were greater than 0.50 and CR was higher than AVE which points out that convergent validity was ensured (Zait & Bertea, 2011). Also, the discriminant validity was tested through Fornell-Lacker criterion through the comparison of the square root of AVE values with the correlations among the latent constructs (Hair et al., 2022). The results in Table 3 showed that square roots of the AVE coefficients are higher than the correlation coefficients that proved the discriminant validity of the constructs (Fornell & Larcker, 1981).

Table 4 Reliability & validity analyses
Full size table

Construct validity of the instruments was tested with CFA through the AMOS 26 software. Table 5 demonstrates the CFA results. The results proved that the 5-dimensional and 18-item structure of the online learning readiness scale, the one-dimensional and 6-item structure of the academic resilience scale, and the 4-dimensional and 21-item structure of the online language learning scales were confirmed. As presented in Table 5, the χ2/df ratios of the three measures were below the optimal cut-off point 3 and the RMSEA values were under 0.06, pointing to a good fit (Steiger, 2007). Similarly, all of the absolute and incremental fit indices were at good or acceptable levels (Jöreskog & Sörbom, 1993; Kline, 2011).

Table 5 CFA Results
Full size table

The partial mediation model fit indices obtained from the SEM analysis (see Table 6; Fig. 2) demonstrated an overall good fit (χ2/df = 2,79, RMSEA = 0.071, SRMR = 0.071, TLI = 0.935, CFI = 0.951, GFI = 0.929). The findings indicated that academic resilience mediated the impact of online learning readiness on online learning engagement. The findings also revealed that there were positive and statistically significant relationships among the variables in the conceptual model proposed in Fig. 1. Firstly, online learning readiness had a moderate and statistically significant direct effect on online learning engagement (β = 0.332, p < .001). The direction of the relationship was positive: the greater the readiness of students in online learning, the higher their engagement in learning activities.

Fig. 2
figure 2

Mediation model of the research

Full size image
Table 6 Bootstrapping results for the conceptual model
Full size table

Secondly, the findings obtained from SEM analysis also indicated a moderate but positive and statistically significant effect of online learning readiness on academic resilience (β = 0.370, p < .01). It shows that students with a higher readiness for online learning typically have increased levels of academic resilience. This finding upholds the idea that learners’ readiness level shapes their academic resilience levels directly. Thirdly, the results produced a relatively higher, positive and statistically significant relationship between academic resilience and online learning engagement (β = 0.414, p < .01). It confirms that resilient students tend to participate more actively in online learning activities.

Finally, the SEM model signifies a moderate, positive and statistically significant effect of online learning readiness on online learning engagement, mediated by academic resilience (β = 0.153, p < .01). This means that the increase in students’ online learning readiness levels brings about a moderate increase in their academic resilience levels, leading to a moderate improvement in their online learning engagement. The total effect of OLRS on OLLE is positive and statistically significant and moderate in size (β = 0.332, p < .001). The direct effects of OLRS on OLLE explained 14% of the total variance in AR (R2 = 0.142), while the total effects of OLRS on OLLE accounted for 38% of the total variance in OLLE (R2 = 0.383). To sum up, the findings confirmed H1, H2, H3 and H4.

5 Discussion and conclusion

This research contributes to the burgeoning field of online education through elucidating the complex relationships among learning readiness, academic resilience, and learning engagement. The findings transcend mere statistical representation, providing deeper insights into the dynamics of digital learning environments. Even though there were limited number of descriptive and correlational studies examining the effects of academic resilience or learning readiness on online learning engagement (e.g. Prihastiwi et al., 2021; Shao & Kang, 2022), there was no research testing the complex relationships between these concepts. This section discusses the hypotheses tested grounded on the conceptual framework that explains relationships among these three variables and findings from research closely related to the topic.

The findings of this research have validated the hypotheses about the expected relationships among the variables in the conceptual model. First of all, this study, utilizingbringing together evidence from six different Turkish HEIs,higher education institutions in Türkiye illustrates a moderate yet a significant correlation between suggest that students’ online learning readiness and their engagement in online education is a significant predictor with a moderate effect size in identifying how actively they are engaged in online education. This finding echoes through global educational trends emphasizing the importance of digital literacy and self-regulation for quality online learning outcomes. Empirical evidence consistently highlights students’ readiness for online learning, including technological proficiency and self-directed learning, is a strong predictor to determine engagement levels. Prihastiwi et al. (2021), for instance, found that online learning readiness accounted for 26% of variance of engagement and reported that self-directed learning, a dimension of online learning readiness, has the highest correlation and contribution to the explained variance. Confirming this link, Kim et al., (2019) revealed a strong, positive and direct relationship between students’ digital readiness levels and their academic engagement. Besides, Ergün and Kurnaz Adıbatmaz (2020) reported that higher readiness levels predicted increased engagement in online learning. They explained that while self-directed learning and computer internet proficiency are significant predictors of students’ behavioral engagement; learner control, self-directed learning, motivation to learn, and online communication self-efficacy are effective predictors of emotional engagement. They also added that students’ cognitive engagement is predicted by their self-directed learning and learner control levels. Apart from these above studies, research in similar online contexts (e.g., Jiang et al., 2021; Ravandpour, 2022; Yıldız Durak, 2018) underline significant positive correlations between the variables, supporting this study. Therefore, the above findings align with the previous research even with various methods and the effect sizes or correlation levels among the variables, suggesting that readiness for online learning is a key driver of engagement in online education. Results point out that students demonstrate higher levels of engagement when they are good at self-directed learning and learning control to actively manage their performance and possess motivation and confidence in technical skills required for in online education. Furthermore, various research (e.g., Axelson & Flick, 2010; Martin & Bolliger, 2018; Miltiadous et al., 2020) have claimed that learners participating in online education exhibit higher levels of academic achievement. Thus, it is crucial to understand how readiness shapes students’ online learning experiences as readiness significantly predicts engagement. This finding also indicates a paradigmatic shift in educational landscape of digital learning platforms as learning readiness plays a vital role on learning engagement. Unlike traditional approach where only cognitive variables or external components such as instructional quality or classroom dynamics are crucial, this research reveals the central role of learning readiness in online education. This finding that resonates with Barbour and Reeves’s (2009) research emphasizing the importance of self-regulation and technological competence in virtual education. The moderate effect size in the study points out the importance of the readiness factors, referring a need for reconsideration the design and implementation of educational strategies to foster student readiness in online education (Hart, 2012). The effect size is not just a numerical value, but it represents the reflection of the significant effect of students’ self-regulation, technological proficiency, and adaptability to the online learning environments. Thus, this finding signifies critical implications for educational policy and practice, pointing out that equipping students with the necessary skills and mindset for online learning is as crucial as the investing and developing technological infrastructure for online education.

Second, the findings also demonstrate that online learning readiness has a positive and direct effect on academic resilience. This finding expands the conventional understanding of student engagement by revealing resilience as an important determinant in online education. Furthermore, this crucial link between readiness and academic resilience indicates that well-designed online programs have the potential to positively foster resilience particularly in non-Western contexts like Türkiye where integration of such non-cognitive factors could bolster online learning outcomes. This result matches what Ramadhana et al. (2021) found in their study, which is the only one to explore the link between these two factors. Researchers have reported a moderate yet significant connection between learning readiness and academic resilience in online education. They have also indicated that learning readiness is a key predictor of resilience, particularly through the factors of learning motivation and self-direction. Positive and significant impact of online learning readiness on academic resilience reveals the transformative role of preparedness in cultivating resilience. In this context, resilience transcends academic boundaries, becoming a vital skill for navigating online learning challenges (Martin & Marsh, 2009). This study’s identification of motivation and self-directed learning as significant resilience predictors aligns with the findings of Zimmerman (2002), who highlights the importance of self-regulation in academic success. Martin and Marsh (2009) identified that five factors labeled as self-efficacy, planning, persistence, anxiety (negatively), and uncertain control (negatively) are predictors of academic resilience. Cassidy (2016) also indicated that academic self-efficacy is correlated with, and a significant predictor of academic resilience. Suryani and Wulandari (2018) reported that self-esteem is a major predictor of academic resilience. Alva (1991) described students with academic resilience as the ones who stay motivated and perform well in school despite encountering difficulties that could lead to academic failure or dropping out. Thus, the fact that students with higher levels of readiness for online education have higher motivation, stronger beliefs, and perceptions that they have the competencies needed in this process, and that they adopt self-directed learning and have control over the learning process leads them to be more academically resilient. The positive effect correlation between learning readiness and academic resilience goes beyond a simple statistical connection as it illustrates that readiness bolsters resilience in online education. This is important in everchanging landscape of online education, particularly during the COVID-19 era since academic resilience could serve as a lifeguard throughout the uncertain and challenges times. Similarly, this study findings of motivation and self-directed learning as strong predictors of online education align with the Zimmerman’s (2002) that highlights the role of self-regulation in academic success.

Thirdly, the existence of positive relationships between students’ academic resilience and their learning engagement in online education means that students who can cope with academic challenges, stress and overcome adversities are more engaged and put more effort into developing their knowledge and competencies based on learning objectives in online education. Considering the impact of learning engagement on students’ effective learning and academic achievement as evidenced by research findings presented in the theoretical framework, the importance of the relatively higher, positive, and statistically significant relationship between these two variables is noteworthy. It is also in accordance with previous studies (e.g., McKeering et al., 2021; Romano et al., 2021) indicating that academic resilience has a positive effect on students’ learning engagement. For example, McKeering et al. (2021) reported a significant correlation between early-adolescent students’ emotional engagement and resilience. Previous research has also shown that academically resilient students are more engaged in the learning process and report higher academic performance than their peers because they are more protected from serious forms of maladaptation (e.g., Romano et al., 2021). Positive correlations between academic resilience and online learning engagement elucidate a critical pathway for educational success in virtual settings. This relationship speaks to the ability of resilient students to not only endure but thrive amidst the challenges of online education. It underscores resilience as a key driver for active and meaningful engagement, a prerequisite for deep learning and academic achievement. Hence, the findings advocate for an educational approach that is not just reactive in addressing challenges but proactive in cultivating resilience as a core competency for 21st-century learners.

Finally, the mediation model proposed in this study (see Fig. 1) was accepted. The findings revealed that students’ academic resilience emerged as a mediating factor altering and influencing the extent to which learning readiness impacted student engagement in online education. It suggests that resilience, often overlooked in educational discourse, is a critical component that can enhance students’ engagement levels, especially in online settings where self-motivation and persistence are key.This finding also contributes to the academic debate about the important variables on students’ performance and academic achievement because, as explained in the theoretical framework, the current literature comprises several studies confirming that both individuals’ engagement and academic resilience have a direct and indirect effect on effective learning and academic success in online education. Regarding the mediating role of academic resilience, a systematic literature review by Kurniadi et al. (2022) highlights the crucial role of academic resilience in online education. They underline the importance of resilience in helping students overcome challenges in online courses, improving their academic performance, and increasing their engagement. This partial mediation model in the current study that provides an improvement in students’ online learning readiness leads to a moderate increase in their level of academic resilience. This, in turn, results in a moderate enhancement in learning engagement in online education. The proposed mediation model in the current research provides a clear understanding of how academic resilience mediates the associations between learning readiness and engagement. The model offers that building up readiness could indirectly enhance engagement through increased resilience, a framework drawn by Tinto (1993) that discusses for the need to create holistic educational environments. This finding has critical and practical implications with regards to educational policy and practice, especially in non-Western contexts like Türkiye, in which the recognition and integration of these factors in online education could be instrumental to optimize teaching and learning process that overestimate non-cognitive components in online education. The validation of the mediation model in the current research is not simply a statistical representation but a conceptual framework for understanding and enhancing academic performance in online education. The elucidation that academic resilience mediates the effect of learning readiness on online learning engagement reflects the complicated and multifaceted nature of learning processes in online education. This model provides educators and policymakers with a more holistic understanding of the levers that can be adjusted to optimize learning outcomes. It also opens up new avenues for research, suggesting that exploring other potential mediators could further unravel the intricate web of factors influencing online educational success. It also points out that there could be other mediators to be explored in this structure. In the same vein, not only other individual-based factors (e.g., motivation, satisfaction, psychological capital) but also other variables related their teachers, friends or families (e.g., teacher attitudes, collaboration, parental support etc.), could be investigated to understand the relationship better.

Finally, this research enriches the field of online education research by highlighting the integral roles of readiness and resilience in enhancing student engagement. The findings extend the existing body of knowledge by exploring the nuanced interplay between online learning readiness, academic resilience, and online learning engagement. The results complement Kuh et al.’s (2006) study underscoring the importance of understanding the multi-faceted nature of learner engagement. The findings of the current research present a pathway to transform online educational practices, emphasizing the need for a more learner-centered strategy recognizing and nurturing the inherent capabilities of students to adjust, participate, and excel in the ever-shifting structure of online education. The research reveals detailed aspects of the operational dynamics and the multifaceted nature of online education in higher education contexts. The results emphasize the requirement of an educational paradigm shift which attach more importance student-related variables such as learning readiness and academic resilience as indispensable to learning engagement in online education, corresponding with the current literature on effective learning mechanism in the 21st century called digital age (Siemens, 2005). Moreover, the results indicate that educational policy makers, strategists, curriculum and instruction and educational technology specialists, and other stakeholders need to take account of such variables in the process of curriculum development, online learning activities design to ensuring a more holistic and comprehensive approach for learner development and create an effective learning atmosphere in online education.

6 Limitations and future directions

This study research has some limitations to keep in mind in the interpretation of the results. First of all, it is critical to consider that the scales utilized in the research were grounded on learners’ self-assessments, which may probably cause a few biases. In other words, the research instruments possibly might not be objective enough to represent the students’ real viewpoints truthfully. Particularly, learners could overrate the extent to which they are prepared, resilient, or actively engaged in online education, which may potentially affect the objectivity of the research data. Hence, further in-depth qualitative research through semi-structured interviews, focus group interviews or classroom observations, can strengthen the understanding of the interrelationships among the variables. Furthermore, as the current research was designed as a cross-sectional study, it was unable to establish causal relationships or dependencies among the variables. Therefore, future research utilizing mixed-methods, longitudinal, and experimental designs is necessary to establish a causal relationship between variables and uncover more in-depth information. Third, this is the first research to analyze the interplay among online learning readiness, academic resilience, and online learning engagement. Also, it should be acknowledged that the primary focal point of this research was individual-level factors. Hence, further investigations are required to understand the findings of this research in the scope of family, peer, instructor, and school-related effects. Moreover, additional studies examining different psychological, cognitive, and environmental variables may provide a better understanding of the relationships among these variables. This study makes a valuable contribution to the international literature on effective variables of online education by providing empirical evidence from a non-Western context where studies are relatively limited, and where heavily prioritize cognitive aspects of learning and academic performance based on the scores of standardized tests (e.g. Altan, 2021yBüköztürk, 2016) and non-cognitive variables such as resilience, emotional engagement or motivation in online education (e.g. Temircan, 2023) have been frequently ignored by practitioners or policymakers. However, these elements are crucial for a holistic understanding of online education’s efficacy and student outcomes but have not been adequately explored or utilized by researchers and policymakers in the region of the study. Nevertheless, as the analyses are based on a dataset collected through convenience sampling, the findings might not be generalizable to the whole population. Hence, further research in Türkiye and in other developing countries could be beneficial in delving into the effects of the same or related variables using nationally representative sample groups.

The study’s results emphasize the significance of online learning readiness in determining their engagement in online education. Students who are proficient in self-directed learning and possess strong learning management, motivation, and technical abilities tend to participate learning activities. Therefore, it is essential for educational institutions to extend support, resources and provide education to assist and encourage learners develop the skills and confidence required for successful online learning. The findings can provide the basis for professional development programs aimed at training educators to recognize and promote the factors that are components of online learning readiness. Furthermore, the study found that academic resilience plays a significant mediating role in the relationship between students’ online learning readiness and engagement. Thus, it is crucial for all stakeholders to assist students in fostering academic resilience by providing opportunities to enhance their self-efficacy, planning, persistence, and self-esteem. Given the positive correlation between academic resilience and engagement, a revision of student support services should be realized. Expanding these services to include resilience-building programs, informed by the study’s findings, could help students navigate the complexities of online learning more effectively. Also, the data suggest the need for online course to promote students’ online learning readiness and its dimensions named self-directed learning, computer and internet self-efficacy, motivation for learning, online communication self-efficacy and learner control for online learning and resilience. Integrating such features into online learning platforms could result in more engaging and effective learning experiences, as indicated by the strong relationship between online learning readiness, resilience, and engagement in the study. The results also indicate that online education can be an effective opportunity to improve academic achievement if students are well-equipped and supported. Next, this study, situated in a non-Western context, Türkiye, offers unique insights relevant to similar educational settings. Educational stakeholders in non-Western countries could consider these findings to customize their online education strategies, acknowledging cultural and contextual nuances in learning readiness and resilience. Moreover, the study lays the groundwork for future longitudinal research to investigate how changes in online learning readiness and resilience over time impact student engagement and success. This implication emerges from our study’s snapshot of these relationships, suggesting the value of exploring their evolution over longer periods.