Skip to main content
NCBI home page
Springer Nature - PMC COVID-19 Collection logoLink to Springer Nature - PMC COVID-19 Collection
. 2023 Mar 30:1–23. Online ahead of print. doi: 10.1007/s10639-023-11749-x

An extended hedonic motivation adoption model of TikTok in higher education

Xinjie Deng 1, Zhonggen Yu 1,✉
PMCID: PMC10061356  PMID: 37361837

Abstract

As information technologies develop, social networking services have gradually gained attention from both researchers and practitioners. However, little is known about the technology adoption of social networking from the perspective of hedonic motivation. For this purpose, this study applied the hedonic motivation system adoption model (HMSAM) to TikTok and incorporated two innovative factors, i.e., perceived boredom and personal innovativeness. Via structural equation modeling (SEM), this study used SmartPLS 4.0.8 to analyze 246 valid responses from Chinese university students via an online survey. The results showed that the research model was adequate for the adoption of TikTok. Curiosity and perceived boredom significantly mediated the positive relationships between perceived ease of use and behavioral intention. Additionally, the educational level moderated the relationship between joy and focused immersion. The results of this study provided insights for future researchers and innovative teaching.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10639-023-11749-x.

Keywords: Hedonic motivation system adoption model (HMSAM), TikTok, Perceived ease of use, Perceived boredom, Personal innovativeness

Introduction

The advancement of information technologies promoted the development of social network services in the twenty-first century. TikTok (Fig. 1) as a short video platform has experienced growing popularity across the world since its launch in 2016. As of September 27th 2021, TikTok has reached over one billion monthly active users worldwide (TikTok, 2021). TikTok, also known as Douyin in China, is characterized by music videos between 15 and 60 s long (Zeng & Kaye, 2022) on various topics ranging from fashion to sports. Since the mission of TikTok is to “inspire creativity and bring joy” (TikTok, 2021), people from different countries use TikTok to learn, laugh, and connect themselves to the outside world. TikTok users can interact with the application by sharing, liking, commenting on, and even imitating original videos (Montag et al., 2021).

Fig. 1.

Fig. 1

Open in a new tab

The screenshot of TikTok (https://tiktokapp.cc/; www.tiktok.com)

Although TikTok is usually a platform for entertainment, it can also serve as a pedagogical tool in different disciplines and learning contexts. TikTok can offer a place and space for higher dance education and shed light on post-humanist thinking in artistic practices (Heyang & Martin, 2022). Similarly, through music and movement, TikTok can meet the requirements of sports science courses, thus promoting students’ motivation, creativity, and curiosity (Escamilla-Fajardo et al., 2021). In chemistry education, TikTok videos can improve undergraduate students’ learning interest and engagement (Hayes et al., 2020). The technological features of TikTok, i.e., instantaneity, interactivity, mobility, and multimodality, also contributes to online informal language learning (Lee, 2022).

While previous studies explored the educational use of TikTok, few of them further examined the role of intrinsic motivation in TikTok use. Intrinsic motivation could be defined as the doing of an activity because of inner satisfaction rather than external inducement (Ryan & Deci, 2000). In the context of technology adoption, intrinsic motivation could prompt students to perceive the potential benefits of technology (Sun & Gao, 2020). Intrinsic motivation was also a significant factor to influence college students’ continuance intention in blended learning (Yang et al., 2022). Therefore, this study aimed to identify an extended TikTok adoption model based on the hedonic-motivation system adoption model (HMSAM) by including personal innovativeness and perceived boredom.

Literature Review

Theoretical framework

The hedonic system acceptance model (HSAM) (van der Heijden, 2004) was essentially a variation of the technology acceptance model (TAM). The TAM (Davis, 1989) was used to predict and explain users’ acceptance of information systems. Two salient constructs in the model were perceived ease of use (PEOU) and perceived usefulness (PU). Since Venkatesh (2000) added perceived enjoyment (i.e., joy) which was a type of intrinsic motivation to TAM, many studies have paid attention to the role of intrinsic motivation in technology acceptance research (e.g., van der Heijden, 2004). HSAM involved PEOU and joy to represent intrinsic motivation, and included PU to measure extrinsic motivation. PU and joy were two significant mediators in the relationship between PEOU and behavioral intention to use cinema websites (van der Heijden, 2004).

Based on van der Heijden’s (2004) model, Lowry et al. (2013) further proposed the hedonic-motivation system adoption model (HMSAM). The latter considered both intrinsic motivation and cognitive absorption. Flow theory, as the conceptual basis of cognitive absorption (Lowry et al., 2013), well explained involvement and concentration in the context involving intrinsic motivation. This theory believed that if individuals were internally motivated, they would logically experience immersion and be absorbed in their ongoing activities (Guo & Poole, 2009; Lee, 2010). Flow theory was thus frequently used to extend existing models, such as the technology acceptance model (TAM) (Lee, 2010) and the unified theory of acceptance and use of technology (UTAUT) (Zhou, 2011). The final HMSAM consisted of seven constructs, i.e., perceived ease of use (PEOU), perceived usefulness (PU), curiosity, joy, control, focused immersion (FI), and behavioral intention to use (BIU).

The HMSAM was designed and applicable for hedonic motivation systems, including video games, online shopping, virtual worlds, and social networking (Lowry et al., 2013; van der Heijden, 2004). Among previous studies, Oluwajana et al. (2019) used HMSAM to investigate students’ acceptance of gamified learning environments. Likewise, HMSAM was also applied to Kahoot!, which was a game-based student response system (Palos-Sanchez et al., 2022). Drawing upon HMSAM, researchers focused on both American and Chinese consumers’ mobile shopping intentions (Ertz et al., 2022). Furthermore, perceived enjoyment and subjective well-being were integrated with the HMSAM to explain consumers’ continued intention to use virtual reality for tourism (Kim & Hall, 2019). However, little was known about users’ hedonic motivation and adoption of social networking. This study thus focused on TikTok use for learning to further understand the acceptance of social networks among students in higher education institutions.

Research model and hypotheses

Perceived ease of use (PEOU) and perceived usefulness (PU) were two essential elements in technology acceptance research and models. PEOU was defined as the extent to which users believed that interacting with a certain system would require little effort, and PU referred to the degree to which individuals thought that using a particular system would improve their learning efficiency or job performance (Davis, 1989). However, hedonic systems were not designed for productivity goals. PU in the HMSAM emphasized hedonic aspects instead of a utilitarian function, thus focusing on the usefulness of pursuing pleasure (Gu et al., 2010; Lowry et al., 2013). Many previous studies have identified the positive relationship between PEOU and PU in different research models from both utilitarian (e.g., Alyoussef, 2021; Yu, 2020) and hedonic (e.g., Oluwajana et al., 2019; Palos-Sanchez et al., 2022) perspectives. Therefore, the authors proposed the following hypothesis.

  • H1. Perceived ease of use positively and significantly affects perceived usefulness in TikTok use at the 0.05 level.

Curiosity, a subconstruct of cognitive absorption, indicated the extent to which users’ experience with a specific system aroused their sensory and cognitive curiosity (Agarwal & Karahanna, 2000). Curiosity was derived from flow theory (Csikszentmihalyi, 1990) and was deemed as an important feature of intrinsic motivation. PEOU of instant messaging technology significantly influenced Kuwait university students’ curiosity (Rouibah & Hamdy, 2009). Other studies on game-based systems also verified the positive relationship between PEOU and curiosity (Oluwajana et al., 2019; Palos-Sanchez et al., 2022). The authors thus presented the following hypothesis.

  • H2. Perceived ease of use positively and significantly affects curiosity in TikTok use at the 0.05 level.

Control measured the extent to which users perceived that they were in charge of the interaction with systems (Agarwal & Karahanna, 2000). Specifically, individuals with a high level of control could skillfully either adapt the interaction to their needs or interrupt the experience (Lowry et al., 2013). It was found that if a system was easy to manipulate, users would be in control of the interaction (Oluwajana et al., 2019). When people used technology tools with ease, their confidence and self-efficacy increased. Consequently, users were likely to control their interactions (Lowry et al., 2013). Therefore, the authors formulated the following hypothesis.

  • H3. Perceived ease of use positively and significantly affects control in TikTok use at the 0.05 level.

Joy, also termed as heightened enjoyment in the HMSAM, referred to “the pleasurable aspects of the interaction” with systems (Agarwal & Karahanna, 2000, p. 673). It was usually described as being enjoyable and fun. PEOU positively predicted joy when students used gamification tools such as Kahoot! (Palos-Sanchez et al., 2022) and storyboards (Lowry et al., 2013). In the mobile shopping context, PEOU was a significant antecedent to perceived enjoyment, contributing to the enjoyable experience (Chang & Chen, 2021; Ertz et al., 2022). Thus, the following hypothesis was proposed.

  • H4. Perceived ease of use positively and significantly affects joy in TikTok use at the 0.05 level.

Boredom was a negative deactivating emotion frequently occurring with enjoyment (Dewaele & Li, 2021). It was defined as an unpleasant psychological state where people were in low physical and cognitive activation and thus wanted to escape from ongoing activities (Li, 2021). The under-stimulation theory (Larson & Richards, 1991) believed that repetitive stimuli resulted in the lack of arousal and involvement. According to the control-value theory, boredom may result from a low level of value and control over tasks (Pekrun, 2006). The integration of perceived boredom (PB) and the HMSAM followed Lowry et al.’s (2013) suggestions that future research could examine the effect of emotions on hedonic motivation system acceptance. Based on these considerations, the authors proposed the following hypothesis.

  • H5. Perceived ease of use negatively and significantly affects perceived boredom in TikTok use at the 0.05 level.

Behavioral intention to use (BIU) was conceptualized as the degree to which people consciously established plans to perform or not to perform particular future behavior (Warshaw & Davis, 1985). Many acceptance models have investigated the antecedents of BIU. Among them, PU was the most studied and common one (Palos-Sanchez et al., 2022; Ruangkanjanases et al., 2020). University students’ hedonic motivation significantly influenced their continued use intention of travel applications (Zhou et al., 2022). In the HMSAM, curiosity and joy were another two positive predictors of BIU (Ertz et al., 2022; Lowry et al., 2013). This study attempted to introduce an additional predictor, i.e., perceived boredom. Therefore, the following hypotheses were determined.

  • H6. Perceived usefulness positively and significantly affects behavioral intention to use TikTok at the 0.05 level.

  • H7. Curiosity positively and significantly affects behavioral intention to use TikTok at the 0.05 level.

  • H8. Joy positively and significantly affects behavioral intention to use TikTok at the 0.05 level.

  • H9. Perceived boredom negatively and significantly affects behavioral intention to use TikTok at the 0.05 level.

Focused immersion (FI) referred to a state of flow where individuals were totally engaged in their tasks and ignored other attentional demands (Agarwal & Karahanna, 2000). During the COVID-19 pandemic period, customers’ hedonic motivation positively influenced their flow when they used live-streaming shopping applications (Zhao & Bacao, 2021). In virtual reality tourism, perceived enjoyment was an important antecedent of both visitors’ and non-visitors’ flow state (Kim & Hall, 2019). The HMSAM also confirmed that curiosity, control, and joy were three determinants of FI (Lowry et al., 2013; Oluwajana et al., 2019). Moreover, individuals with a high level of boredom showed a deficit in sustained attention (Malkovsky et al., 2012). Thus, the authors presented the following hypotheses.

  • H10. Curiosity positively and significantly affects focused immersion in TikTok use at the 0.05 level.

  • H11. Control positively and significantly affects focused immersion in TikTok use at the 0.05 level.

  • H12. Joy positively and significantly affects focused immersion in TikTok use at the 0.05 level.

  • H13. Perceived boredom negatively and significantly affects focused immersion in TikTok use at the 0.05 level.

Personal innovativeness (PI) was defined as the willingness of a person to try out any new system or information technology (Agarwal & Prasad, 1998). The diffusion of innovation theory (Rogers, 1995) posited that people held different attitudes toward new things and did not adopt an innovation simultaneously. PI, when integrated into the TAM, significantly exerted a positive influence on PEOU in students’ mobile learning (Joo et al., 2014). Similarly, teachers’ PI contributed to the prediction of their PEOU and behavioral intention to adopt emerging technologies (Akar, 2019). PI was also incorporated into the UTAUT model to explore Indonesian students’ mobile learning (Sidik & Syafar, 2020). Although PI was a critical construct for understanding the acceptance of technology innovations, it remained underexplored (Agarwal & Prasad, 1998) and was not included in the HMSAM. The authors thus formulated the following hypothesis. Based on previous studies and hypotheses development, the final research model was proposed in Fig. 2.

  • H14. Personal innovativeness positively and significantly affects perceived ease of use in TikTok at the 0.05 level.

Fig. 2.

Fig. 2

Open in a new tab

The proposed research model

Methods

Participants and procedures

The participants of this study were students from higher education institutions in China with the experience of using TikTok. The researchers obtained ethical approval from their organizations before data collection. A convenience sampling method was used both online and offline at the first author’s university. The participants were informed about the confidentiality policy and instructions to fill in the survey.

Data were collected using an online survey from October 9, 2022 to November 6, 2022. The researchers distributed the survey to their students through an online platform named Questionnaire Star during the 10-min break. The first author also posted messages in mainstream social software to recruit participants, including WeChat, QQ, Weibo, TikTok, and Xiaohongshu (an Instagram-like lifestyle app in China). Once respondents finished the survey, they would receive either a participation reward or a corresponding gift. A total of 277 students were willing to participate in this study and submitted their informed consent. The researchers excluded 31 invalid questionnaires and analyzed the remaining 246 responses. Therefore, the valid response rate was 88.9%. Table 1 provides the profile of participants.

Table 1.

The profile of participants

Item Category Frequency Percentage (%)
Gender Male 105 42.68
Female 141 57.32
Educational level Bachelor 143 58.13
Master 92 37.40
Doctor 11 4.47
Frequency of using TikTok Always 63 25.61
Often 68 27.64
Sometimes 49 19.92
Seldom 66 26.83
Total 246 100.00
Open in a new tab

Instruments

The researchers used the back-translation method and designed the questionnaire in both English and Chinese to avoid misunderstanding. They also invited 10 TikTok users and 3 professors to complete the survey before data collection. Based on their suggestions, the researchers optimized and improved the questionnaire regarding wording, format, numbering, and sequence of the items.

The final version of the questionnaire consisted of two parts. The first part included the willingness of participating in the survey, gender, educational level, and frequency of using TikTok. The second part was made up of 34 items (see Appendix in supplementary files) to measure the nine constructs included in the proposed research model: PEOU (4 items), PU (5 items), curiosity (3 items), control (3 items), joy (4 items), perceived boredom (4 items), PI (4 items), BIU (3 items), and FI (4 items). All items in the second part were measured using five-point Likert scales adapted from previous studies (Agarwal & Karahanna, 2000; Bieleke et al., 2021; Lowry et al., 2013; Venkatesh, 2000).

Data analysis

This study adopted partial least squares structural equation modeling (PLS-SEM) to test the research model. PLS-SEM was suitable for theory development and highly complex models with many constructs (Hair et al., 2012). This study included nine constructs and mediating effects, aiming to develop a theoretically comprehensive model by introducing boredom theory. PLS-SEM was also especially useful for investigating relatively new phenomena (Chin & Newsted, 1999), such as learning in TikTok. Thus, the researchers employed SmartPLS 4.0.8 (Ringle et al., 2022) to analyze measurement and structural models. The analysis of PLS-SEM results involved two stages (Sarstedt et al., 2014). The first stage examined the measurement model by reliability and validity. The second stage evaluated the structural model by predictive relevance, significance of path coefficients, indirect effects, and moderating effects.

Results

Measurement model assessment

To examine the reflective measurement model, the researchers performed an ordinary PLS-SEM algorithm to obtain indicator loadings first (Table 2). Loadings greater than 0.70 revealed that the construct could explain over 50% of the indicator’s variance (Sarstedt et al., 2014). Since the loadings of FI1 and PU4 were far below 0.70, the researchers decided to delete the two items. After deletion, all factor loadings were higher than 0.70 and significant at the 0.05 level, indicating acceptable item reliability (Chin, 1998). Following Hair et al.’s guidelines (2019), the researchers also assessed internal consistency reliability using Cronbach’s alpha and composite reliability (CR). The Cronbach’s alpha and CR for each construct exceeded the recommended threshold, i.e., 0.70. The third step was to evaluate the convergent validity. The average variance extracted (AVE) of all constructs were greater than the minimum acceptable value of 0.50, suggesting a satisfactory level of convergent validity.

Table 2.

Reliability and convergent validity of the measurement model

Construct Item Factor loading Cronbach’s alpha Composite reliability (CR) Average variance extracted (AVE)
Behavioral intention to use (BIU) BIU1 0.947* 0.931 0.931 0.879
BIU2 0.939*
BIU3 0.926*
Control (CTL) CTL1 0.891* 0.782 0.907 0.681
CTL2 0.791*
CTL3 0.791*
Curiosity (CUR) CUR1 0.876* 0.872 0.876 0.796
CUR2 0.928*
CUR3 0.871*
Focused immersion (FI) FI1 (deleted) 0.216 0.716 0.746 0.635
FI2 0.734*
FI3 0.868*
FI4 0.783*
Joy (JOY) JOY1 0.897* 0.910 0.924 0.790
JOY2 0.940*
JOY3 0.920*
JOY4 0.789*
Perceived boredom (PB) PB1 0.822* 0.788 0.829 0.608
PB2 0.866*
PB3 0.708*
PB4 0.709*
Perceived ease of use (PEOU) PEOU1 0.772* 0.791 0.821 0.611
PEOU2 0.702*
PEOU3 0.799*
PEOU4 0.847*
Personal innovativeness (PI) PI1 0.874* 0.894 0.898 0.760
PI2 0.908*
PI3 0.826*
PI4 0.876*
Perceived usefulness (PU) PU1 0.875* 0.848 0.872 0.685
PU2 0.832*
PU3 0.851*
PU4 (deleted) 0.605*
PU5 0.746*
Open in a new tab

*p < 0.05

The fourth step was to measure discriminant validity using three statistical criteria. First, cross-loadings required that factor loadings of each component item belonging to its corresponding construct should be higher than those of other constructs (Urbach & Ahlemann, 2010). Second, as shown in Table 3, the discriminant validity met the Fornell-Larcker criterion (Fornell & Larcker, 1981) that square root of each construct’s AVE was larger than inter-construct correlation between the same construct and all the others. However, this metric received criticism because variance-based SEM methods may overestimate factor loadings and underestimate structural model relationships (Henseler et al., 2015). The third criterion was thus proposed.

Table 3.

Fornell-Larcker test of discriminant validity

Construct Correlation of the constructs
BIU Control Curiosity FI Joy PB PEOU PI PU
BIU 0.937
Control 0.157 0.825
Curiosity 0.590 0.035 0.892
FI 0.408 0.024 0.562 0.797
Joy 0.506 0.244 0.676 0.560 0.889
PB -0.512 -0.005 -0.566 -0.610 -0.630 0.780
PEOU 0.229 0.229 0.350 0.371 0.428 -0.296 0.782
PI 0.155 0.092 0.251 0.107 0.125 -0.109 0.164 0.871
PU 0.453 0.189 0.594 0.559 0.801 -0.553 0.525 0.172 0.828
Open in a new tab

The diagonal values in bold are square roots of AVEs. BIU: Behavioral intention to use; FI: Focused immersion; PB: Perceived boredom; PEOU: Perceived ease of use; PI: Personal innovativeness; PU: Perceived usefulness

The heterotrait-monotrait ratio (HTMT) was an estimate of the correlations. It was the average of the item correlations across constructs relative to those within the same construct (Henseler et al., 2015). The threshold value would be set at 0.90 if constructs were conceptually similar, and a threshold value of 0.85 was suggested when constructs were distinct. Since perceived usefulness (PU) focused on hedonic aspects, it was reasonable that PU was slightly similar to joy. Therefore, the HTMT ratio (Table 4), ranging from 0.120 to 0.909, generally validated the discriminant validity of this study.

Table 4.

Heterotrait-monotrait ratio (HTMT) of discriminant validity

BIU Control Curiosity FI Joy PB PEOU PI PU
BIU
Control 0.175
Curiosity 0.654 0.058
FI 0.493 0.130 0.691
Joy 0.548 0.283 0.756 0.673
PB 0.576 0.144 0.655 0.762 0.715
PEOU 0.247 0.261 0.410 0.465 0.474 0.335
PI 0.166 0.120 0.285 0.138 0.141 0.134 0.194
PU 0.511 0.210 0.692 0.685 0.909 0.652 0.582 0.201
Open in a new tab

BIU: Behavioral intention to use; FI: Focused immersion; PB: Perceived boredom; PEOU: Perceived ease of use; PI: Personal innovativeness; PU: Perceived usefulness

Structural model analysis

Having confirmed the reliability and validity of the measurement model, the researchers first assessed the coefficient of determination (R2). R2 indicated the relationship of a construct’s variance that could be explained to its total variance, ranging from 0 to 1. The calculation of R2 was sum of squares due to regression (SSR) divided by total sum of squares (SST). Thus, R2 could reflect the explanatory or predictive power of a model (Shmueli & Koppius, 2011). The R2 value of 0.67, 0.33, and 0.19 could be considered substantial, moderate, and lower weak (Chin, 1998). Figure 3 presents that the research model explained a minimum of 45.4% of focused immersion and 40.0% of behavioral intention to use TikTok, suggesting moderate explanatory power.

Fig. 3.

Fig. 3

Open in a new tab

The research model with path coefficients. ***p < 0.001, **p < 0.01, *p < 0.05

The blindfolding technique was removed in SmartPLS 4, and PLSpredict was recommended for the assessment of predictive power. Following Hair et al.’s (2019) suggestions, the researchers chose to compare the root mean squared error (RMSE), rather than the mean absolute error (MAE), with the benchmark, i.e., the linear regression model (LM) values. After PLSpredict executed tenfold cross-validation 10 times, the results (Table 5) showed that none of indicators yielded greater RMSE values compared to the LM benchmark, which indicated a high predictive power (Shmueli et al., 2019). All the positive Q2 values also confirmed the predictive relevance of the model.

Table 5.

The results of PLSpredict

Construct Item Q2 predict PLS-SEM analysis LM benchmark
RMSE MAE RMSE MAE
Behavioral intention to use (BIU) BIU1 0.019 1.057 0.878 1.070 0.893
BIU2 1.101 0.908 1.107 0.904
BIU3 1.168 0.977 1.180 0.967
Control (CTL) CTL1 0.005 0.993 0.834 1.007 0.842
CTL2 0.981 0.825 0.995 0.838
CTL3 1.101 0.883 1.119 0.901
Curiosity (CUR) CUR1 0.052 0.854 0.604 0.864 0.609
CUR2 0.946 0.750 0.954 0.760
CUR3 1.025 0.839 1.029 0.837
Focused immersion (FI) FI2 0.009 0.865 0.627 0.880 0.636
FI3 0.882 0.604 0.896 0.614
FI4 1.087 0.946 1.097 0.947
Joy (JOY) JOY1 0.010 0.94 0.678 0.954 0.690
JOY2 0.961 0.679 0.971 0.695
JOY3 1.009 0.760 1.020 0.765
JOY4 1.094 0.931 1.100 0.930
Perceived boredom (PB) PB1 0.007 1.086 0.926 1.098 0.928
PB2 1.102 0.911 1.119 0.918
PB3 1.151 0.956 1.156 0.950
PB4 1.153 0.990 1.168 1.011
Perceived ease of use (PEOU) PEOU1 0.015 0.635 0.464 0.641 0.470
PEOU2 0.829 0.571 0.840 0.604
PEOU3 0.878 0.606 0.893 0.626
PEOU4 0.672 0.560 0.680 0.565
Perceived usefulness (PU) PU1 0.018 0.998 0.769 1.012 0.781
PU2 0.806 0.576 0.817 0.575
PU3 1.099 0.878 1.110 0.879
PU5 1.075 0.883 1.086 0.888
Open in a new tab

The researchers also verified the hypothesized relationships using the PLS-SEM algorithm and bootstrapping techniques. The standardized root mean square residual (SRMR) reflected the approximate model fit. The value of SRMR was 0.065 and lower than 0.08 (Henseler et al., 2016a, 2016b), indicating that the model fit was adequate. Path coefficients explained the strength of the relationship between two constructs, which should exceed 0.10 (Urbach & Ahlemann, 2010). The resampling technique such as bootstrapping was run to determine the significance, and the number of bootstrap subsamples was set to 5000.

As presented in Table 6, 11 out of 14 hypothesized paths were statistically significant. Perceived ease of use positively affected perceived usefulness (β = 0.525, p < 0.001), curiosity (β = 0.350, p < 0.001), control (β = 0.229, p < 0.01), and joy (β = 0.428, p < 0.001), which supported H1, H2, H3, and H4. In contrast, perceived ease of use negatively predicted perceived boredom (β = -0.296, p < 0.001), supporting H5. Perceived usefulness (β = 0.045, p > 0.05) and joy (β = 0.063, p > 0.05) did not significantly affect behavioral intention, which rejected H6 and H8. However, curiosity (β = 0.395, p < 0.001) and perceived boredom (β = -0.224, p < 0.01) were significant predictors of behavioral intention, supporting H7 and H9. Likewise, curiosity (β = 0.236, p < 0.01), joy (β = 0.178, p < 0.05), and perceived boredom (β = -0.364, p < 0.001) also significantly predicted focused immersion, which validated H10, H12, and H13. Nevertheless, control failed to significantly affect focused immersion (β = -0.029, p > 0.05), thus rejecting H11. Personal innovativeness positively influenced perceived ease of use (β = 0.164, p < 0.05), which supported H14.

Table 6.

The results of hypothesis testing

Hypothesis Path Path coefficient T statistics p value 95% confidence interval Results
Lower Upper
H1 PEOU → PU 0.525 8.900 0.000 0.408 0.637 Accepted
H2 PEOU → Curiosity 0.350 4.390 0.000 0.184 0.495 Accepted
H3 PEOU → Control 0.229 3.403 0.001 0.103 0.361 Accepted
H4 PEOU → Joy 0.428 6.265 0.000 0.287 0.553 Accepted
H5 PEOU → PB -0.296 4.321 0.000 -0.425 -0.157 Accepted
H6 PU → BIU 0.045 0.494 0.622 -0.122 0.231 Rejected
H7 Curiosity → BIU 0.395 5.062 0.000 0.228 0.533 Accepted
H8 Joy → BIU 0.063 0.637 0.524 -0.134 0.254 Rejected
H9 PB → BIU -0.224 2.938 0.003 -0.376 -0.075 Accepted
H10 Curiosity → FI 0.236 3.112 0.002 0.088 0.384 Accepted
H11 Control → FI -0.029 0.487 0.626 -0.148 0.092 Rejected
H12 Joy → FI 0.178 2.058 0.040 0.006 0.344 Accepted
H13 PB → FI -0.364 5.217 0.000 -0.503 -0.233 Accepted
H14 PI → PEOU 0.164 2.575 0.010 0.064 0.295 Accepted
Open in a new tab

BIU: Behavioral intention to use; FI: Focused immersion; PB: Perceived boredom; PEOU: Perceived ease of use; PI: Personal innovativeness; PU: Perceived usefulness

Mediating analysis

The researchers performed the mediation analysis with a 95% confidence interval using the bias-corrected and accelerated bootstrap method. Among the mediating relations in the research model, 10 mediation paths were significant (Table 7). Curiosity (B = 0.125, 95% CI [0.060, 0.216]) and perceived boredom (B = 0.066, 95% CI [0.022, 0.128]) were two significant mediators of the relationship between PEOU and BIU. They also significantly mediated the effect of PEOU on FI (B = 0.075, 95% CI [0.022, 0.160]; B = 0.108, 95% CI [0.051, 0.184]). The results also confirmed the mediating role of curiosity between PI and BIU (B = 0.079, 95% CI [0.027, 0.151]), and between PI and FI (B = 0.048, 95% CI [0.011, 0.104]). The effect of PI on hedonic aspects (i.e., PU, joy, and curiosity) was significantly mediated by PEOU.

Table 7.

Significant mediating effects of the research model

Path Indirect effect Standard deviation T statistics p value 95% confidence interval
Lower Upper
PEOU → Curiosity → BIU 0.125 0.039 3.194 0.001 0.060 0.216
PEOU → PB → BIU 0.066 0.027 2.485 0.013 0.022 0.128
PEOU → Curiosity → FI 0.075 0.035 2.144 0.032 0.022 0.160
PEOU → PB → FI 0.108 0.034 3.176 0.002 0.051 0.184
PI → Curiosity → BIU 0.079 0.031 2.572 0.010 0.027 0.151
PI → Curiosity → FI 0.048 0.024 2.011 0.044 0.011 0.104
PI → PEOU → PU 0.085 0.038 2.241 0.025 0.013 0.158
PI → PEOU → Joy 0.069 0.032 2.163 0.031 0.013 0.135
PI → PEOU → PB -0.048 0.022 2.138 0.033 -0.097 -0.011
PI → PEOU → Curiosity 0.051 0.026 1.996 0.046 0.010 0.108
Open in a new tab

BIU: Behavioral intention to use; FI: Focused immersion; PB: Perceived boredom; PEOU: Perceived ease of use; PI: Personal innovativeness; PU: Perceived usefulness

Multigroup analysis

The researchers also performed permutation multigroup analysis to investigate moderating roles of gender and educational level. All participants in this study were divided into male and female in terms of gender. Regarding educational level, master and doctor were merged into one category because of the small sample size of the doctor category. Before multigroup analysis, measurement invariance of composite models (MICOM) should be examined in three steps, i.e., configural invariance, compositional invariance, and the equality of composite mean values and variances (Henseler et al., 2016a, 2016b). In step 1, configural invariance was established (Table 8). In step 2, correlations of all constructs in both groups of gender or educational level were not significantly lower than one, indicating that compositional invariance was established (Henseler et al., 2016a, 2016b). In step 3, the 95% confidence interval of differences in mean values and variances generally included the original difference, which suggested that the composite mean values and variances were equal.

Table 8.

The results of the three-step MICOM procedure

Moderating variable Construct Step 1 Step 2 Step 3a Step 3b
Configural invariance Original correlation p value Compositional invariance Original difference 95% CI p value Equal means Original difference 95% CI p value Equal variances
Gender (Male vs. Female) BIU Yes 1.000 0.923 Yes -0.064 [-0.259, 0.250] 0.624 Yes 0.339 [-0.319, 0.304] 0.034 No
CTL Yes 0.999 0.978 Yes -0.071 [-0.258, 0.245] 0.579 Yes 0.229 [-0.338, 0.308] 0.163 Yes
CUR Yes 0.999 0.166 Yes -0.138 [-0.257, 0.251] 0.287 Yes 0.508 [-0.421, 0.394] 0.017 No
FI Yes 0.998 0.724 Yes 0.060 [-0.253, 0.253] 0.647 Yes 0.313 [-0.436, 0.414] 0.148 Yes
JOY Yes 1.000 0.252 Yes -0.193 [-0.262, 0.256] 0.142 Yes 0.362 [-0.422, 0.421] 0.098 Yes
PB Yes 0.999 0.657 Yes 0.002 [-0.254, 0.250] 0.985 Yes 0.252 [-0.351, 0.334] 0.145 Yes
PEOU Yes 0.994 0.362 Yes -0.012 [-0.248, 0.247] 0.936 Yes 0.456 [-0.565, 0.569] 0.142 Yes
PI Yes 0.996 0.362 Yes 0.062 [-0.257, 0.249] 0.623 Yes -0.043 [-0.311, 0.296] 0.788 Yes
PU Yes 0.999 0.571 Yes -0.083 [-0.257, 0.253] 0.519 Yes 0.258 [-0.496, 0.494] 0.312 Yes
Educational level (Bachelor vs. Master and doctor) BIU Yes 1.000 0.549 Yes -0.210 [-0.253, 0.248] 0.102 Yes 0.17 [-0.294, 0.318] 0.265 Yes
CTL Yes 0.889 0.068 Yes 0.023 [-0.251, 0.255] 0.852 Yes -0.227 [-0.305, 0.332] 0.165 Yes
CUR Yes 1.000 0.557 Yes -0.217 [-0.255, 0.255] 0.096 Yes -0.023 [-0.398, 0.412] 0.914 Yes
FI Yes 0.996 0.453 Yes -0.121 [-0.253, 0.239] 0.344 Yes 0.089 [-0.416, 0.419] 0.678 Yes
JOY Yes 0.999 0.159 Yes -0.087 [-0.248, 0.252] 0.498 Yes 0.145 [-0.399, 0.431] 0.502 Yes
PB Yes 0.998 0.601 Yes 0.098 [-0.255, 0.248] 0.447 Yes 0.106 [-0.330, 0.342] 0.543 Yes
PEOU Yes 0.988 0.145 Yes -0.228 [-0.256, 0.253] 0.079 Yes 0.545 [-0.580, 0.549] 0.061 Yes
PI Yes 0.999 0.796 Yes 0.139 [-0.254, 0.253] 0.282 Yes -0.179 [-0.306, 0.305] 0.251 Yes
PU Yes 0.999 0.705 Yes -0.169 [-0.255, 0.255] 0.200 Yes 0.169 [-0.469, 0.511] 0.496 Yes
Open in a new tab

MICOM: Measurement invariance of composite models; BIU: Behavioral intention to use; CTL: Control; CUR: Curiosity; FI: Focused immersion; JOY: Joy; PB: Perceived boredom; PEOU: Perceived ease of use; PI: Personal innovativeness; PU: Perceived usefulness

The permutation multigroup analysis showed the absence of the moderating effect of gender. Relationships in the research model did not vary with gender. However, the relationship between joy and focused immersion was moderated by the educational level (original difference = 0.378, p < 0.05). It indicated that the extent to which the interaction with TikTok was enjoyable differently influenced users’ ability to be engrossed in videos according to the educational level. Undergraduate students were more likely to concentrate on their learning because of joy than postgraduate students.

Discussion

This study aimed to establish an acceptance model of TikTok system based on the HMSAM and examine students’ behavioral intention and focused immersion. The results revealed that the HMSAM was applicable and suitable for social networking, and 11 of the 14 hypotheses were finally supported. Curiosity and perceived boredom significantly mediated the relationship between perceived ease of use and two dependent variables. Gender could not moderate any relationships in the model, while the educational level moderated the relationship between joy and focused immersion.

Perceived ease of use (PEOU) significantly influenced perceived usefulness (PU), curiosity, control, joy, and perceived boredom. The results were consistent with previous studies (e.g., Chang & Chen, 2021; Oluwajana et al., 2019; Palos-Sanchez et al., 2022), which implied that the ease of use of TikTok increased learners’ hedonic benefits. Without complex procedures and instructions to cope with, users could focus on their experience and feelings, e.g., joy and perceived boredom. Users’ attention could also be free to explore novel interactions (Lowry et al., 2013), thus stimulating users’ curiosity. In contrast, if TikTok was difficult to use, students had to take much time or energy to get familiar with the system and struggle to control their interactions. Gradually, they may feel frustrated and bored to overcome the initial barrier.

Personal innovativeness (PI) was a significant antecedent of PEOU, supporting the findings of Joo et al. (2014). Innovativeness, a positive attitude toward technology, contributed to an increase in technology readiness (Chang & Chen, 2021). This personal characteristic encouraged users to actively try out new technology and experiment with the newly-launched system although they may encounter some technical and operational issues. For that reason, individuals with a higher level of innovativeness were more risk-taking (Sidik & Syafar, 2020) and easier to interact with TikTok than those who refused emerging applications. It was reasonable that innovative learners perceived TikTok as easy to use.

The main predictors of behavioral intention to use (BIU) TikTok were curiosity and perceived boredom. This suggested that either the presence of curiosity or the absence of perceived boredom was a motivation for students to adopt TikTok-assisted learning. However, the results did not completely confirm the influence of PU and joy, unlike other studies (Lowry et al., 2013; Palos-Sanchez et al., 2022). The possible reason was that they focused on the game-based response system (Palos-Sanchez et al., 2022) or actual games (Lowry et al., 2013). The game per se was enjoyable and interesting, while not all educational videos were entertaining. Furthermore, since PU in the HMSAM placed emphasis on pleasure, a similar situation occurred with PU.

Similarly, the main predictors of focused immersion (FI) were curiosity, joy, and perceived boredom, which was in line with Palos-Sanchez et al. (2022). Students would bury themselves in videos if TikTok provoked curiosity, improved enjoyment, and prevented boredom. Educational videos on TikTok were usually accompanied by detailed explanations and background music, which might be more interesting and diverting than words. Students did not need to burn much energy to understand, contributing to absorption and immersion. Nevertheless, the findings also showed that control did not significantly predict FI. It meant that whether users could control their interactions with TikTok, they might still be deeply intrigued by educational videos. This was caused mainly because TikTok could automatically play the next video even if users did nothing, facilitating sustained attention.

The effect of PEOU on BIU and FI was fully mediated by curiosity and perceived boredom, corroborated by Oluwajana et al. (2019). TikTok boasted the algorithm of personalized recommendations based on previous videos users watched. Users were likely to receive the video that suited their tastes. However, they would never guess what exactly the next video was about until the next video played, stimulating their curiosity. The ease of use allowed different kinds of interactions with TikTok, e.g., giving a like, commenting, following, and posting, encouraging to avoid boredom that resulted from simply browsing similar videos. Users thus concentrated on educational videos and wanted to continue using TikTok application for learning.

The results shed light on the role of TikTok in improving higher-order thinking skills. Technology could potentially develop students’ higher-order thinking skills (Nussbaum et al., 2021) such as problem-solving skills, critical thinking skills, and creative thinking skills (Priyaadharshini & Sundaram, 2018). The aim of TikTok was to stimulate creativity, which was a vital element in the problem-solving process (Nussbaum et al., 2021) and the enhancement of students’ creative thinking. Curiosity could support information seeking, question asking, and deeper learning, which were three mechanisms for the development of scientific thinking (Jirout, 2020). Since deep learning approaches directly influenced higher-order thinking skills (Lee & Choi, 2017), students should be encouraged to do critical analyses rather than repetition after watching the same series of Tik Tok videos.

This study also examined the moderating effect of both gender and educational level. Gender was not a significant moderator, indicating that relationships among variables were not moderated by gender. The reason might be that most educational videos on TikTok were designed and available for both males and females. The educational level moderated the relationship between joy and FI. The possible explanation might come from different learning styles and approaches. The learning styles of most undergraduate students were auditory-reading and auditory-kinesthetic types. However, compared with undergraduate students, graduate students were more likely to be unimodal learners and adopt deep and strategic learning approaches (Samarakoon et al., 2013). They systematically acquired professional knowledge by reading books and journal articles, attending academic lectures, and writing papers. They were probably not accustomed to enjoying short videos with fragmented information and failed to gain pleasure, thus not being involved in TikTok videos.

Conclusion

Major findings

This study developed an adoption model of TikTok system by incorporating two constructs into the HMSAM, i.e., perceived boredom and personal innovativeness. The results showed that the proposed research model was adequate for the adoption of TikTok. Perceived innovativeness significantly and positively influenced perceived ease of use. Additionally, curiosity and perceived boredom mediated the positive relationships between perceived ease of use and behavioral intention and between perceived ease of use and focused immersion. Multigroup analysis revealed that the educational level moderated the relationship between joy and focused immersion.

Limitations of this study

Although this research shed light on TikTok adoption, there were still some limitations in this study. First, this study was conducted among Chinese university students. The results should be generalized to other countries with caution. Second, this study used cross-sectional data to validate the research model. However, students’ perceptions and feelings may change over a period of time with the advancement of technology. Third, the research model only explained 40.0% of the total variance of behavioral intention to use TikTok. Other potential factors may also influence behavioral intention and focused immersion.

Implications for future research

This research provided some valuable theoretical implications. First, this study validated the theoretical framework, i.e., the HMSAM, in the context of TikTok-assisted learning in the university. Future research could investigate the adoption of other social media platforms, e.g., Facebook and Twitter. Learners in other phases such as high school students should also be considered. Second, perceived usefulness in the HMSAM was pleasure-oriented, which did not focus on students’ performance. Researchers in the future are thus expected to compare pleasure-focused and performance-focused usefulness in one model. Third, researchers could enrich and expand the HMSAM by introducing other influencing factors or theories, such as the expectation confirmation theory. Future studies could also examine the effect of other moderating variables, e.g., age, frequency of using the application, and learning content, on behavioral intention and focused immersion. Researchers in the future are also expected to further explore other differences between undergraduate and graduate students in the use of TikTok.

This study helped practitioners better understand the adoption of social network tools such as TikTok. Video makers are advised to determine a clear topic and include interesting elements to ensure usefulness and kindle viewers’ curiosity. Both users and the platform should provide different music to suit the content. Developers should attach importance to improving the ease of use of emerging technologies and simplify the interaction process (Chang & Chen, 2021), e.g., registering an account. The results of this study also provided insights into innovative learning and teaching. Teachers could consider learners’ intrinsic motivation and properly integrate TikTok into activities to help students enjoy learning and develop higher-order thinking skills. The critical role of personal innovativeness reminded teachers to consciously encourage students to be positive about new technologies. Moreover, the moderating role of the educational level implied that teachers in higher education institutions should adopt different teaching strategies. For example, teachers could enhance undergraduate students’ joy of learning to help them keep focused.

In addition, social media companies tend to attract users’ attention through highly immersive services (Montag et al., 2019), e.g., endless and personalized content. However, it was the immersive design that may result in problematic use, i.e., TikTok addiction (Montag et al., 2021). Teachers should control the extent to which students adopt TikTok to complete tasks, guide them to use it in moderation, and request parents’ cooperation when necessary. Future research could explore the problematic use of social media tools and propose appropriate solutions.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (XLSX 42 KB) (41.6KB, xlsx)
Supplementary file2 (DOCX 30 KB) (30KB, docx)

Authors’ contributions

Xinjie Deng: Methodology, Investigation, Editing, and Writing-Original Draft.

Zhonggen Yu: Conceptualization and Funding acquisition.

Funding

This work is supported by 2019 MOOC of Beijing Language and Culture University (MOOC201902) (Important) “Introduction to Linguistics”; "Introduction to Linguistics" of online and offline mixed courses in Beijing Language and Culture University in 2020; Special fund of Beijing Co-construction Project-Research and reform of the "Undergraduate Teaching Reform and Innovation Project" of Beijing higher education in 2020-innovative "multilingual + " excellent talent training system (202010032003); The research project of Graduate Students of Beijing Language and Culture University "Xi Jinping: The Governance of China" (SJTS202108); The Fundamental Research Funds for the Central Universities, and the Research Funds of Beijing Language and Culture University (22YCX017); and the Research Project of the 8th Education and Teaching Reform in Dali University (2022JGY08-129).

Data availability

We make sure that all data and materials support our published claims and comply with field standards.

Declarations

Conflict of interest/Competing interests

We have no conflicts of interest to declare that are relevant to the content of this article.

Footnotes

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Xinjie Deng, Email: 1255811529@qq.com, Email: 202121198711@stu.blcu.edu.cn.

Zhonggen Yu, Email: yuzhonggen@blcu.edu.cn, Email: 401373742@qq.com.

References

  1. Agarwal R, Karahanna E. Time flies when you're having fun: Cognitive absorption and beliefs about information technology usage. Mis Quarterly. 2000;24(4):665–694. doi: 10.2307/3250951. [DOI] [Google Scholar]
  2. Agarwal R, Prasad J. A conceptual and operational definition of personal innovativeness in the domain of information technology. Information Systems Research. 1998;9(2):204–215. doi: 10.1287/isre.9.2.204. [DOI] [Google Scholar]
  3. Akar SGM. Does it matter being innovative: Teachers' technology acceptance. Education and Information Technologies. 2019;24(6):3415–3432. doi: 10.1007/s10639-019-09933-z. [DOI] [Google Scholar]
  4. Alyoussef IY. Factors influencing students' acceptance of m-learning in higher education: An application and extension of the UTAUT model. Electronics. 2021;10(24):3171. doi: 10.3390/electronics10243171. [DOI] [Google Scholar]
  5. Bieleke M, Gogol K, Goetz T, Daniels L, Pekrun R. The AEQ-S: A short version of the achievement emotions questionnaire. Contemporary Educational Psychology. 2021;65:101940. doi: 10.1016/j.cedpsych.2020.101940. [DOI] [Google Scholar]
  6. Chang YW, Chen JH. What motivates customers to shop in smart shops? The impacts of smart technology and technology readiness. Journal of Retailing and Consumer Services. 2021;58:102325. doi: 10.1016/j.jretconser.2020.102325. [DOI] [Google Scholar]
  7. Chin WW. The partial least squares approach to structural equation modeling. In: Marcoulides GA, editor. Modern methods for business research. Lawrence Erlbaum Associates; 1998. pp. 295–336. [Google Scholar]
  8. Chin WW, Newsted PR. Structural equation modeling analysis with small samples using partial least squares. In: Hoyle R, editor. Statistical Strategies for Small Sample Research. Sage Publications; 1999. pp. 307–341. [Google Scholar]
  9. Csikszentmihalyi M. Flow: The psychology of optimal experience. Harper and Row; 1990. [Google Scholar]
  10. Davis FD. Perceived usefulness, perceived ease of use, and user acceptance of information technology. Mis Quarterly. 1989;13(3):319–340. doi: 10.2307/249008. [DOI] [Google Scholar]
  11. Dewaele JM, Li CC. Teacher enthusiasm and students' social-behavioral learning engagement: The mediating role of student enjoyment and boredom in Chinese EFL classes. Language Teaching Research. 2021;25(6):922–945. doi: 10.1177/13621688211014538. [DOI] [Google Scholar]
  12. Ertz M, Jo MS, Kong Y, Sarigollu E. Predicting m-shopping in the two largest m-commerce markets: The United States and China. International Journal of Market Research. 2022;64(2):249–268. doi: 10.1177/14707853211023036. [DOI] [Google Scholar]
  13. Escamilla-Fajardo P, Alguacil M, Lopez-Carril S. Incorporating TikTok in higher education: Pedagogical perspectives from a corporal expression sport sciences course. Journal of Hospitality Leisure Sport & Tourism Education. 2021;28:100302. doi: 10.1016/j.jhlste.2021.100302. [DOI] [Google Scholar]
  14. Fornell C, Larcker DF. Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research. 1981;18(1):39–50. doi: 10.2307/3151312. [DOI] [Google Scholar]
  15. Gu JC, Fan L, Suh YH, Lee SC. Comparing utilitarian and hedonic usefulness to user intention in multipurpose information systems. Cyberpsychology Behavior and Social Networking. 2010;13(3):287–297. doi: 10.1089/cyber.2009.0167. [DOI] [PubMed] [Google Scholar]
  16. Guo YM, Poole MS. Antecedents of flow in online shopping: A test of alternative models. Information Systems Journal. 2009;19(4):369–390. doi: 10.1111/j.1365-2575.2007.00292.x. [DOI] [Google Scholar]
  17. Hair JF, Risher JJ, Sarstedt M, Ringle CM. When to use and how to report the results of PLS-SEM. European Business Review. 2019;31(1):2–24. doi: 10.1108/EBR-11-2018-0203. [DOI] [Google Scholar]
  18. Hair JF, Sarstedt M, Ringle CM, Mena JA. An assessment of the use of partial least squares structural equation modeling in marketing research. Journal of the Academy of Marketing Science. 2012;40(3):414–433. doi: 10.1007/s11747-011-0261-6. [DOI] [Google Scholar]
  19. Hayes C, Stott K, Lamb KJ, Hurst GA. "Making every second count": Utilizing TikTok and systems thinking to facilitate scientific public engagement and contextualization of Chemistry at home. Journal of Chemical Education. 2020;97(10):3858–3866. doi: 10.1021/acs.jchemed.0c00511. [DOI] [Google Scholar]
  20. Henseler J, Hubona G, Ray PA. Using PLS path modeling in new technology research: Updated guidelines. Industrial Management & Data Systems. 2016;116(1):2–20. doi: 10.1108/IMDS-09-2015-0382. [DOI] [Google Scholar]
  21. Henseler J, Ringle CM, Sarstedt M. A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science. 2015;43(1):115–135. doi: 10.1007/s11747-014-0403-8. [DOI] [Google Scholar]
  22. Henseler J, Ringle CM, Sarstedt M. Testing measurement invariance of composites using partial least squares. International Marketing Review. 2016;33(3):405–431. doi: 10.1108/IMR-09-2014-0304. [DOI] [Google Scholar]
  23. Heyang T, Martin R. Teaching through TikTok: A duoethnographic exploration of pedagogical approaches using TikTok in higher dance education in China and Norway during a global pandemic. Research in Dance Education. 2022 doi: 10.1080/14647893.2022.2114446. [DOI] [Google Scholar]
  24. Jirout JJ. Supporting early scientific thinking through curiosity. Frontiers in Psychology. 2020;11:1717. doi: 10.3389/fpsyg.2020.01717. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Joo YJ, Lee HW, Ham Y. Integrating user interface and personal innovativeness into the TAM for mobile learning in Cyber University. Journal of Computing in Higher Education. 2014;26(2):143–158. doi: 10.1007/s12528-014-9081-2. [DOI] [Google Scholar]
  26. Kim MJ, Hall CM. A hedonic motivation model in virtual reality tourism: Comparing visitors and non-visitors. International Journal of Information Management. 2019;46:236–249. doi: 10.1016/j.ijinfomgt.2018.11.016. [DOI] [Google Scholar]
  27. Larson RW, Richards MH. Boredom in the middle school years: Blaming schools versus blaming students. American Journal of Education. 1991;99(4):418–443. doi: 10.1086/443992. [DOI] [Google Scholar]
  28. Lee MC. Explaining and predicting users' continuance intention toward e-learning: An extension of the expectation-confirmation model. Computers & Education. 2010;54(2):506–516. doi: 10.1016/j.compedu.2009.09.002. [DOI] [Google Scholar]
  29. Lee YJ. Language learning affordances of Instagram and TikTok. Innovation in Language Learning and Teaching. 2022 doi: 10.1080/17501229.2022.2051517. [DOI] [Google Scholar]
  30. Lee J, Choi H. What affects learner’s higher-order thinking in technology-enhanced learning environments: The effects of learner factors. Computers & Education. 2017;115:143–152. doi: 10.1016/j.compedu.2017.06.015. [DOI] [Google Scholar]
  31. Li CC. A control-value theory approach to boredom in English classes among university students in China. Modern Language Journal. 2021;105(1):317–334. doi: 10.1111/modl.12693. [DOI] [Google Scholar]
  32. Lowry PB, Gaskin JE, Twyman NW, Hammer B, Roberts TL. Taking "fun and games" seriously: Proposing the hedonic-motivation system adoption model (HMSAM) Journal of the Association for Information Systems. 2013;14(11):617–671. doi: 10.17705/1jais.00347. [DOI] [Google Scholar]
  33. Malkovsky E, Merrifield C, Goldberg Y, Danckert J. Exploring the relationship between boredom and sustained attention. Experimental Brain Research. 2012;221(1):59–67. doi: 10.1007/s00221-012-3147-z. [DOI] [PubMed] [Google Scholar]
  34. Montag C, Lachmann B, Herrlich M, Zweig K. Addictive features of social media/messenger platforms and freemium games against the background of psychological and economic theories. International Journal of Environmental Research and Public Health. 2019;16(14):2612. doi: 10.3390/ijerph16142612. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Montag C, Yang HB, Elhai JD. On the psychology of TikTok use: A first glimpse from empirical findings. Frontiers in Public Health. 2021;9:641673. doi: 10.3389/fpubh.2021.641673. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Nussbaum M, Barahona C, Rodriguez F, Guentulle V, Lopez F, Vazquez-Uscanga E, Cabezas V. Taking critical thinking, creativity and grit online. Educational Technology Research and Development. 2021;69(1):201–206. doi: 10.1007/s11423-020-09867-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Oluwajana D, Idowu A, Nat M, Vanduhe V, Fadiya S. The adoption of students' hedonic motivation system model to gamified learning environment. Journal of Theoretical and Applied Electronic Commerce Research. 2019;14(3):156–167. doi: 10.4067/S0718-18762019000300109. [DOI] [Google Scholar]
  38. Palos-Sanchez PR, Saura JR, Velicia-Martin F. A case study on a hedonic-motivation system adoption model in a game-based student response system. International Journal of Human-Computer Interaction. 2022 doi: 10.1080/10447318.2022.2121801. [DOI] [Google Scholar]
  39. Pekrun R. The control-value theory of achievement emotions: Assumptions, corollaries, and implications for educational research and practice. Educational Psychology Review. 2006;18(4):315–341. doi: 10.1007/s10648-006-9029-9. [DOI] [Google Scholar]
  40. Priyaadharshini M, Sundaram BV. Evaluation of higher-order thinking skills using learning style in an undergraduate engineering in flipped classroom. Computer Applications in Engineering Education. 2018;26(6):2237–2254. doi: 10.1002/cae.22035. [DOI] [Google Scholar]
  41. Ringle, C. M., Wende, S., & Becker, J.-M. (2022). SmartPLS 4. Boenningstedt: SmartPLS. https://www.smartpls.com.  Accessed 4 Nov 2022.
  42. Rogers EM. Diffusion of innovation. Free Press; 1995. [Google Scholar]
  43. Rouibah K, Hamdy H. Factors affecting information communication technologies usage and satisfaction: Perspective from instant messaging in Kuwait. Journal of Global Information Management. 2009;17(2):1–29. doi: 10.4018/jgim.2009040101. [DOI] [Google Scholar]
  44. Ruangkanjanases A, Hsu SL, Wu YJ, Chen SC, Chang JY. What drives continuance intention towards social media: Social influence and identity perspectives. Sustainability. 2020;12(17):7081. doi: 10.3390/su12177081. [DOI] [Google Scholar]
  45. Ryan RM, Deci EL. Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology. 2000;25(1):54–67. doi: 10.1006/ceps.1999.1020. [DOI] [PubMed] [Google Scholar]
  46. Samarakoon L, Fernando T, Rodrigo C, Rajapakse S. Learning styles and approaches to learning among medical undergraduates and postgraduates. BMC Medical Education. 2013;13:42. doi: 10.1186/1472-6920-13-42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Sarstedt M, Ringle CM, Smith D, Reams R, Hair JF. Partial least squares structural equation modeling (PLS-SEM): A useful tool for family business researchers. Journal of Family Business Strategy. 2014;5(1):105–115. doi: 10.1016/j.jfbs.2014.01.002. [DOI] [Google Scholar]
  48. Shmueli G, Koppius OR. Predictive analytics in information systems research. Mis Quarterly. 2011;35(3):553–572. doi: 10.2139/ssrn.1606674. [DOI] [Google Scholar]
  49. Shmueli G, Sarstedt M, Hair JF, Cheah JH, Ting H, Vaithilingam S, Ringle CM. Predictive model assessment in PLS-SEM: Guidelines for using PLSpredict. European Journal of Marketing. 2019;53(11):2322–2347. doi: 10.1108/EJM-02-2019-0189. [DOI] [Google Scholar]
  50. Sidik D, Syafar F. Exploring the factors influencing student's intention to use mobile learning in Indonesia higher education. Education and Information Technologies. 2020;25(6):4781–4796. doi: 10.1007/s10639-019-10018-0. [DOI] [Google Scholar]
  51. Sun YY, Gao F. An investigation of the influence of intrinsic motivation on students' intention to use mobile devices in language learning. Educational Technology Research and Development. 2020;68(3):1181–1198. doi: 10.1007/s11423-019-09733-9. [DOI] [Google Scholar]
  52. TikTok. (2021). Thanks a billion! TikTok Newsroom. https://newsroom.tiktok.com/en-us/1-billion-people-on-tiktok. Accessed 1 Nov 2022.
  53. Urbach N, Ahlemann F. Structural equation modeling in information systems research using partial least squares. Journal of Information Technology Theory and Application. 2010;11(2):2. [Google Scholar]
  54. van der Heijden H. User acceptance of hedonic information systems. Mis Quarterly. 2004;28(4):695–704. doi: 10.2307/25148660. [DOI] [Google Scholar]
  55. Venkatesh V. Determinants of perceived ease of use: Integrating control, intrinsic motivation, and emotion into the technology acceptance model. Information Systems Research. 2000;11(4):342–365. doi: 10.1287/isre.11.4.342.11872. [DOI] [Google Scholar]
  56. Warshaw PR, Davis FD. Disentangling behavioral intention and behavioral expectation. Journal of Experimental Social Psychology. 1985;21(3):213–228. doi: 10.1016/0022-1031(85)90017-4. [DOI] [Google Scholar]
  57. Yang H, Cai J, Yang HH, Wang XC. Examining key factors of beginner's continuance intention in blended learning in higher education. Journal of Computing in Higher Education. 2022 doi: 10.1007/s12528-022-09322-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Yu ZG. Extending the learning technology acceptance model of WeChat by adding new psychological constructs. Journal of Educational Computing Research. 2020;58(6):1121–1143. doi: 10.1177/0735633120923772. [DOI] [Google Scholar]
  59. Zeng J, Kaye DBV. From content moderation to visibility moderation: A case study of platform governance on TikTok. Policy and Internet. 2022;14(1):79–95. doi: 10.1002/poi3.287. [DOI] [Google Scholar]
  60. Zhao YY, Bacao F. How does gender moderate customer intention of shopping via live-streaming apps during the COVID-19 pandemic lockdown period? International Journal of Environmental Research and Public Health. 2021;18(24):13004. doi: 10.3390/ijerph182413004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Zhou T. Understanding mobile Internet continuance usage from the perspectives of UTAUT and flow. Information Development. 2011;27(3):207–218. doi: 10.1177/0266666911414596. [DOI] [Google Scholar]
  62. Zhou T, Song Y, Zhou P. Continued use intention of travel apps: From the perspective of control and motivation. Technology Analysis & Strategic Management. 2022;34(6):703–716. doi: 10.1080/09537325.2021.1916457. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary file1 (XLSX 42 KB) (41.6KB, xlsx)
Supplementary file2 (DOCX 30 KB) (30KB, docx)

Data Availability Statement

We make sure that all data and materials support our published claims and comply with field standards.

Articles from Education and Information Technologies are provided here courtesy of Nature Publishing Group

RESOURCES