Exploring Motivations for TikTok Usage and Impact Factors of TikTokers’ Continuance Intention

2.1 Technology acceptance model

In 1985, Davis developed TAM to explain technology adoption behavior and was deemed by Chuttur [6] to be the leading model that explains and predicts system use. According to Davis [4], TAM focuses on two predictors of PEOU and PU that help to understand technology usage behaviors. A variety of researchers have often employed TAM as a means to gain insight into how people perceive their use of different technologies. Scherer et al. [7] report behavioral intentions and technology usage as outcome variables. In relation to the outcomes cited above, PEOU and PU of the system by individuals are key independent variables [8, 9]. Using TAM, it is attempted to develop an understanding of the reasons behind people’s adoption or exclusion of a certain technology [4, 10].

It is important to note, however, that TAM does have two limitations. Due to its generality and simplicity, the initial model does not identify antecedents that may have influenced PEOU and PU [11, 12]. Moreover, as Park [12] pointed out, while the model can be used to identify factors influencing individuals’ adoption and usage of certain technologies, it cannot adequately explain why people do so. The TAM must, furthermore, include factors related to target technology, context, and users, according to Shyu and Huang [13].

Considering the weakness of TAM, this paper introduces UGT, a commonly applied approach in communication research. As TAM focuses on the technology itself, whereas UGT looks at the motivations and experiences that are generated by the technology, it is crucial to integrate both theories into a broader conceptualization [14].

2.2 Uses and gratifications theory

UGT is an established theory in mass communications that explains how rational individuals choose particular media to fulfill certain socio-psychological needs [15]. In many studies, UGT has been used to examine how individuals utilize different media and the motives behind them. It is possible for people to use identical media for different reasons, as well as to experience different levels of gratifications. With no predefined set of constructs, UGT has been constantly revised and updated as research progresses.

This is not a new area of research to apply UGT to study social media. Using the framework of UGT, a popular line of studies on social media’s continued usage emphasizes the impact of distinct gratifications on continuation intentions as seen in Facebook [16, 17], microblogging [18], online games [19], WeChat [20], Instagram [21] and KakaoTalk [22]. The intent to continue TikTok’s usage has, however, never been addressed. UGT has been revived and rejuvenated by new forms of media, according to Rubin [23] and Ruggiero [24]. It was suggested by Bucknell Bossen and Kottasz [25] that new media platforms with distinctive features, such as TikTok, could be studied using this theoretical framework.

Unlike TAM, which focuses on external factors, UGT focuses on internal factors. Accordingly, we utilize UGT to identify internal factors of individuals, which is a weakness of TAM. Two reasons justify combining UGT with TAM in this study. Firstly, the combination incorporates user motivations while maintaining TAM’s parsimonious nature. Furthermore, this integration enables us to predict the continuation of TikTok usage.

4.1 Data collection and sample

The authors obtained ethical approval for this study from University Malaysia Sabah’s Research and Ethics Committee. All participants gave their informed consent to participate in this study. Researchers complied with ethical guidelines when gathering data and ensured that all information provided by respondents was used only for research purposes.

Owing to practicality and expense restrictions, it is unrealistic and impossible to research the whole population. Even if it was possible to study the entire population, it is prohibitive in terms of time, costs, and other human resources [42]. To gather data for this research, non-probability sampling methods were used. This is due to the fact that not every unit in the population, namely TikTokers, can be assured to be included in the sample since this is a voluntary survey and only interested respondents are asked to participate. The appropriateness of the sample size may be generally evaluated using the following criteria, according to studies of Tabachnick and Fidell [43] and Comrey and Lee [44]: 100 is considered poor; 200 is considered fair; 300 is considered good; 500 is considered very good; and 1000 or more is considered excellent. As a result, a sample size of over 300 is considered good.

As part of this study, data were collected from July 3 through August 29, 2022 using Tencent’s questionnaire (https://wj.qq.com/). Convenience sampling was utilized to distribute the online survey link. Only one submission was allowed per IP address. There was no name associated with any of the respondents and all information was voluntarily disclosed. A description of the research scheme and its goals were presented in the questionnaire at the outset. As a means of minimizing meaning errors and guaranteeing accuracy, back translation [45] was employed. The survey, written in English initially, was then translated into mandarin by a linguistically proficient third party before being translated back into English. There is an appendix listing the items measured for the study. Pre-testing was conducted on 44 individuals prior to the survey to make sure every question was well-structured, thorough, and reliable. The pilot study feedback led to the revision of several perplexing questions. There were a total of 505 questionnaires collected. Those questionnaires scoring the same or taking less than two minutes to complete were eliminated from the analysis. In total, 421 valid samples were processed subsequently for analysis. A demographic breakdown of the sampled participants is shown in Table 1. In this sample, 44.2% of the participants are between the ages of 21-25; 39.4% are under 20. As for gender distribution, the majority of samples are female, making up 68.2%. In terms of education level, there are relatively more “College/University” participants with a proportion of 62.7% in the sample. 26.6% of TikTok users have used the platform for less than half a year. For TikTok usage frequency, “Less than once every week” accounts for the highest proportion of 35.2%. There are relatively more people who spend less than 15 minutes on TikTok on a daily basis, accounting for 36.8% of the sample.

4.2 Construct measurement

According to our research model, there are ten constructs, including CI, ATT, PEOU, PU, ESC, ENT, FAS, SS, MM and IS. Measures of each construct were based on multiple item scales. All measurement instruments derived from existing research were then adjusted in order to fit the TikTok research scenario. Bhattacherjee [26] was the source of items for CI. The list of ATT items was based on Taylor and Todd [46]. Several items from Joo and Sang [32] were modified to serve as PEOU items. There were items for PU that came from Camilleri and Falzon [34]. For ESC, items were drawn from Meng and Leung [41] and Lee et al. [47]. In this study, we modified the items of ENT from Meng and Leung [41] & Tao and Zhu [48]. FAS items were sourced from Meng and Leung [41]. This article borrowed SS items from Meng and Leung [41] and Lee et al. [47]. Modifications to items for MM have been made based on Meng and Leung [41]. A selection of items for IS was taken from Meng and Leung [41]. All measuring items were rated from strong disagreement to strong agreement using a 7-point Likert scale.

4.3 A brief overview of analysis

Data were analyzed via confirmatory factor analysis (CFA) and SEM. In CFA, the underlying latent variable structure is already known, such as in this study, and the validity of the measurement items was evaluated against the proposed factor structures. In the current study, hypotheses were tested using SEM. For estimating relationships among different independent and dependent variables, SEM is commonly used in empirical research. An examination of the factors that influence people’s CI when it comes to TikTok was conducted using the two-stage analytical technique developed by Anderson and Gerbing [49].

Reliability and validity tests were undertaken first. To form model fit indices, the ratio of chi-square to degrees of freedom (χ²/df), the goodness-of-fit index (GFI), the adjusted goodness-of-fit index (AGFI), the incremental fit index (IFI), the comparative fit index (CFI) and the Tucker-Lewis index (TLI), the standardized root mean squared residual (SRMR), the root mean square error of approximation (RMSEA) were combined [50-52]. Cronbach’s Alpha (α) and composite reliability (CR) [53] were utilized to test the internal consistency of all constructs in the conceptual model. It was recommended that CR be at least 0.70 to be adequate [54]. The average variance extracted (AVE), the square root of AVE, and the correlation values [55] served to validate our measurement model’s convergent and discriminant validity. To assess the convergent and discriminant validity, we checked whether the AVEs of each construct are greater than their correlation with other constructs, as well as whether each item’s loading on its assigned construct is higher than that on other constructs [55, 56].

Secondly, this research utilized structural equation analysis for testing the hypotheses, including investigating whether the hypotheses hold and an analysis of the relationships among the different variables under study. Statistical analyses were performed with SPSS 23.0, SPSSAU and AMOS 24.0.

6.1 Discussion

As per survey results, TikTok is used primarily for ENT and IS motivations, followed by SS and ESC motivations, similar to that of Meng and Leung [41]. Based on the data, contrary to what Lu and Lu [59] and Meng and Leung [41] claimed, it is evident that most people don’t use TikTok for FAS or MM reasons. The participants believe that using TikTok helps them relax or rest, as well as relieve their stress, in line with studies on UGT, in which scholars conclude that media technologies are often viewed as forms of ENT [60-62]. To satisfy their information needs, most respondents use TikTok to keep up with recent events and to find interesting information, which is also consistent with previous research [10, 63]. Contrary to Mull and Lee [64] and Vaterlaus and Winter [65], the current research finds most people do not use TikTok for FAS or MM motivations. As for FAS, the finding of this research is supported by Curras-Perez et al. [66], who claimed that people use social networks to form social links with friends, family, and acquaintances, and to have fun and entertainment, rather than being fashionable or up-to-date. According to Meng and Leung [41], MM is an additional motivation for TikTokers to use the platform to earn money through livestreaming and product promotion. Despite the fact that our research findings do not support it at present, TikTok's commercial and money-making advantages may become increasingly important as more users adopt the app and continue to use it [65].

This study shows that people generally perceive TikTok as an easy-to-use and useful application, in agreement with Vaterlaus and Winter [65]’s finding. Participants find it easy to learn how to operate TikTok, with most reporting no difficulty using the app skillfully. Furthermore, they have an easy time interacting with TikTok thanks to its simple interface, similar to what Vaterlaus and Winter [65] found. With TikTok, people are able to access content faster than with traditional video services, providing them with a useful alternative, consistent with Lu and Lu [59]. Additionally, TikTok offers a better experience when it comes to watching informative or entertaining videos, the same as Wang [67] mentioned. Besides that, TikTok also facilitates people’s studies or work, which Scherr and Wang [68] also noted. Most people have a positive ATT toward TikTok since they find it enjoyable and satisfying to use the service and consider using it a good idea. Consequently, much of the research participants are committed to using TikTok and to doing so regularly in the future.

Based on SEM analysis, PU and ATT are significant positive antecedents of people’s intention to continue using TikTok, whereas PEOU is a significant negative predictor, contrary to what many researchers have reported PEOU to be a positive predictor, such as Kim and Wang [69], Dhir et al. [70] and Dalvi-Esfahani et al. [71]. When the usefulness of TikTok is perceived and a positive ATT towards the app is established, people are more likely to continue using the service. Yet, the easier the software is to operate, the more likely users are to discontinue using it. Possibly, due to the simplicity of the platform, it may be easy for people to become addicted to it, thereby wanting to get rid of it. As Vaterlaus and Winter [65], Chakraborty et al. [72] and Lu et al. [73] have pointed out, people avoid TikTok due to concerns about low-quality content and vulgarity, as well as potential time-wasting. While PEOU significantly influences PU in a positive way, both PEOU and PU have significant positive impacts on ATT. This means people are more likely to perceive TikTok’s usefulness if they perceive its ease of use. Additionally, when people perceive TikTok as easy to use and useful, they will develop a more favorable ATT towards it. PEOU of TikTok is significantly influenced by ENT and IS in a positive way, by FAS in a negative way, while ESC, SS and MM have only limited influences. In other words, people who use TikTok for ENT and IS purposes are more likely to find the app easy to use. Nevertheless, the more people use TikTok for FAS motivation, the more people will find the software hard to use. It could be that TikTok is only a popular social media that can’t meet people’s FAS needs. Moreover, ENT, SS and IS stand to have significant positive effects on PU of TikTok, in contrast to ESC, FAS and MM, which have negligible effects. This means that the more people use TikTok for ENT, SS, and IS motivations, the more useful it will be to them.

Social media such as TikTok have seen a surge in use following the unexpected spread of Coronavirus epidemic and its resulting social distancing measures. Paid streaming services also saw a spike in subscriptions [74]. A positive ATT and a perception of ease of use and usefulness of media platforms like TikTok would encourage consumers to continue entertaining themselves and searching for interesting information on the software.

6.2 Theoretical implications

Theoretical contributions of this research are as follows. To begin with, by integrating UGT and TAM in a single theoretical framework, the current study advances the existing literature. To understand how information and communications technologies are adopted, UGT and TAM are both well-known theories. UGT is concerned with intrinsic factors of information system adoption, whereas TAM focuses on extrinsic factors, so this study utilized UGT to shed light on individual users’ intrinsic motivations, a weakness of TAM. In this study, we successfully developed a model that would be able to explain and predict the CI for TikTok users. This study also explores the antecedents of PEOU and PU. The research model can assist in addressing not only the underlying research questions, but also to examine why people continue to engage in other popular social media similar to TikTok. To some extent, the current work is a novel attempt combining UGT and TAM to examine people’s motivations for using newer social media, such as TikTok, and their CI towards them.

Furthermore, as TikTok continues to grow, this research serves as a springboard for further academic exploration of topics that have yet to be explored. Based on our findings, TikTok users have different motives than those who use other social media platforms. As far as TikTok use motives are concerned, most users prioritize ENT and IS. Due to social media’s predominant role as a “social” tool, previous social media UGT studies have focused on the importance of social interaction [75]. On TikTok, users primarily seek ENT and information, while the majority consumes content, whereas the minority builds communities and interacts with others.

6.3 Practical implications

This study provides valuable insights for social media marketers seeking to connect and engage with Chinese consumers on TikTok. As a start, marketers who are interested in TikTok influencer marketing need to take into account the motivations of its users. TikTok users mainly consume time for ENT and IS, so practitioners must develop entertaining and informative contents that meet their primary needs. Aside from that, social media marketers should understand consumers’ motivations, as well as their CI on the platform in order to collaborate with appropriate influencers to accomplish a variety of influencer marketing goals.

TikTok platform designers can also benefit from our findings. TikTok users tend to use the platform more for ENT and IS motivations than for FAS or MM. To satisfy people’s need for ENT, platform designers should include features that help them express their enjoyment during consumption, such as avatars and emojis [76]. Also, more convenient search methods will help people find relevant information more easily, such as adding hashtag search function to satisfy their need for IS by making it easy to find videos they’ve watched recently. SEM results show that PU and ATT are significant positive antecedents of people’s intention to continue using TikTok, whereas PEOU is a significant negative predictor, which could be explained by its simplicity, allowing people to become addicted easily, and thus want to get rid of it. Consequently, TikTok designers should strive to make users feel that the platform is useful and maintain their positive ATT towards it. As well as that, the designers should develop a mechanism that will prevent addiction, so that users won’t stop using the software completely because they feel too addicted to it.

6.4 Research limitations and future directions

A few research limitations should be noted, along with directions for future research. As a result of the COVID-19 epidemic, a nonprobability sampling was utilized to draw the sample for this research, and therefore, the sample for this study was neither representative nor diversified. Therefore, when interpreting the findings, one should keep in mind that the research may not be generalizable. Moreover, our survey had a much higher percentage of female participants than male participants. An appropriate sample of Chinese TikTok users should be recruited with a balance of male and female respondents in future research. A further limitation of this study could be that the six motivations examined here are inadequate to account for the time users spend on TikTok. Future TikTok research can explore other motivations, as well as the relationship between motivations and users’ CI, which has not been explored thus far.