Human papillomavirus (HPV) is the most common viral infection, responsible for approximately 630,000 new cancer cases worldwide each year, including cervical and other types of cancer (de Martel et al., 2017). Recognizing its high prevalence and severity, many national immunization programs have been implemented to promote HPV vaccination, which helps protect against HPV transmission and cancer development (Bruni et al., 2016). Despite the efforts, the HPV vaccination rate remains low globally, particularly in less developed regions (Bruni et al., 2021). Recent studies have identified social media as a promising tool for strengthening HPV vaccine confidence and increasing vaccination rates (Chen et al., 2021; Ortiz et al., 2019). As a significant source of health information, social media now plays an instrumental role in educating the public about HPV, shaping attitudes, and influencing behaviors toward HPV vaccination. Research on the impact of social media on HPV vaccination has emerged across various platforms, including Facebook (Loft et al., 2020), Twitter (Allen et al., 2020), and YouTube (Briones et al., 2012).

Mobile short video (MSV) applications are an emerging form of social media that has skyrocketed in popularity over the past few years (Zhang et al., 2019). These platforms enable users to create and share entertaining videos that last for a few minutes, or even a few seconds, with simplified editing functions. TikTok, one of the most popular MSV applications, boasts approximately 1.1 billion active users worldwide who record short videos about singing, cooking, and other daily activities, and upload them with hashtags to join various topics and trends (Geyser, 2021). TikTok is also notable for its uniquely addictive algorithm, which has driven extensive use among teenagers and young adults (Zhang et al., 2019). The explosive growth and attractive features of MSVs have made them effective tools for disseminating health-related information (Basch et al., 2020; Song et al., 2021b). Compared to other social media platforms, the originality and interactivity of MSVs elicit a stronger sense of social presence and immersion (Zhang et al., 2019), which may further motivate changes in individuals’ health attitudes and behaviors, especially among the young generation. Previous studies of MSVs have mostly focused on identifying the factors that influence user experience and intention to use (Song et al., 2021c; Wang, 2020; Zhang et al., 2019), but have yet to examine how to enhance their persuasive impacts on health communication. Investigating MSVs and health communication is important, as it will contribute to more effective strategies for health message design and media intervention implementation.

In HPV-related health research, although relevant findings have provided an overview of how health interventions impact HPV vaccination knowledge and attitudes, most have focused on traditional formats such as texts or images (Mavundza et al., 2021). Little research has devoted attention to the short video format emerging on social media, and exploration of its specific impact on persuading people to adopt HPV vaccination remains lacking. Considering the scarcity of relevant literature in this new research area, this study employs an experimental approach to examine the potential factors of MSVs that may influence people’s attitudes and intentions toward HPV vaccination, through the theoretical lens of the elaboration likelihood model (ELM) and media richness theory. The findings can extend the literature by assessing the informational and technical stimuli that contribute to the persuasive effect.

Theoretical background

Persuasion and health messages

Persuasion is an important concept in mass communication, referring to the intentional effort to change audiences’ attitudes and behaviors by disseminating relevant information (Perloff, 2008). The existing literature offers a sound understanding of persuasive messages that lead to health-related behavior changes, including quitting smoking (So, 2022), undergoing mammography screening (Kemp et al., 2024), adopting a healthy diet (Friedman et al., 2022), and promoting home-based physical activity (Carfora and Catellani, 2021). Substantial research has also examined the health messages effective in changing attitudes and intentions toward HPV vaccination. For example, findings often suggest that framing strategies and narrative types of health messages are important factors in predicting individuals’ perception of and intention toward HPV vaccination (Gainforth et al., 2012; Nan et al., 2015), while graphics also play a role in promoting HPV vaccination among young adults (Teoh et al., 2019). Persuasive strategies used in video-based media such as YouTube to promote HPV vaccines were further examined (Kim et al., 2021), but their actual effects on attitudinal and behavioral changes warrant further investigation.

Elaboration likelihood model

First introduced by Petty and Cacioppo (1986), the elaboration likelihood model (ELM) is the most popular persuasion theory in social psychology and behavioral science. It seeks to understand the underlying processes of persuasion and describes two basic routes that explain how humans process stimuli and the subsequent changes in attitude and behavior (Petty and Cacioppo, 1986). The “central route” requires a high level of message elaboration and, consequently, a high level of audience motivation. In this scenario, attitude change occurs through a cognitive (or rational) process based on the detailed information presented in the message, such as the quality of the central argument, personal relevance, and other constructs related to the technological medium (Liang and Lin, 2018; Petty and Cacioppo, 1986; Tang et al., 2012). The persuasive effect of the central route is often enduring and consistent. Conversely, under the “peripheral route,” a person lacks motivation or ability and thus makes simple inferences backed by superficial cues or heuristics, such as the credibility or attractiveness of the source, or the production quality of the message (Petty et al., 2009; Teng and Khong, 2015). The results of attitude change through the peripheral route are often short-term and unpredictable.

ELM is a highly adaptable and inclusive model that considers all elements of a message and different ways of processing information. It aids in constructing and delivering persuasive messages to target audiences by considering their motivations and abilities (SanJosé-Cabezudo et al., 2009). ELM has been applied to investigate the multiple roles of variables in influencing health communication efficacy (e.g., Guo et al., 2020; Yada and Head, 2019) and to test various media message strategies used in health promotion (e.g., Flynn et al., 2011; Thompson et al., 2007). More importantly, unlike other classical theories, ELM accommodates the integration of new and different variables associated with novel technologies (Shahab et al., 2021). Consequently, ELM is well-suited to understanding the new technology of MSVs and exploring their persuasive impact on HPV vaccination in the current study.

Media richness theory

Apart from the content and source of information, it is also essential to consider the medium that transmits information. Media richness theory, introduced by Daft and Lengel (1986), describes and evaluates the ability of a communication medium to reproduce the information sent over it. The degree of this ability is known as “richness,” indicating the number of multiple cues in audio, visual, or textual formats. The use of richer media allows for more information to be conveyed and reduces ambiguity during the transmission process (Dennis and Kinney, 1998). With the emergence of new media, researchers have identified the varying richness among different forms of social media (Kaplan and Haenlein, 2010). For instance, the richness of textual media like blogs is considered low, while that of video sites such as YouTube is high due to the inclusion of visual and auditory cues. Based on these classifications, visual- or audio-based forms of social media are assumed to be more effective in communication than text-based forms.

Furthermore, findings from existing studies indicated that the use of pictures and videos in social media is more effective in triggering users’ emotions, capturing attention, and promoting behavior changes (Bhattacharya et al., 2017; Guidry et al., 2020). Nevertheless, some studies have reported controversial results, showing non-significant impacts of rich media (Chen et al., 2020; Kim and Yang, 2017). This may be because visual and audio cues can distract the audience, particularly when the primary objective is to convey a straightforward message. Robert and Dennis (2005) further argued that the use of rich media could increase motivation but decrease the ability to process information. In health promotion, an important target is to examine the impacts of media interventions on people’s attitudes and behavior toward a health issue. However, research on the association between media richness and health perception (and behavior) is largely limited. Therefore, the current study applies media richness theory to further explore the effects of medium type on attitude and behavioral intention in the context of MSV interventions.

Research hypotheses

Drawing on the theoretical frameworks of both ELM and media richness theory, the current study aims to explore the persuasive impacts of information quality, source credibility, and media richness of MSV on promoting HPV vaccination intent, via the mediation effect of attitude toward HPV vaccine information.

Attitude toward health information

Despite the long-standing recognition of the important role of attitude on behavioral intention in the psychology literature (Danes, 1981; Kim and Nan, 2012), the relationship between attitude toward health information and the intention to adopt protective health behaviors is not well articulated. For information to influence behavioral intention, individuals must believe and process the information, and then develop an intent to adopt the message’s recommendation (Somera et al., 2016). The attitude toward health information significantly affects people’s responses to health messages. Some scholars have examined the predictive effects of attitude toward health information (e.g., information trust) on the intention to seek and use health information (Deng et al., 2015). Additionally, recent evidence shows a positive relationship between health information-seeking and health-promoting behaviors (Lee et al., 2015; Liu et al., 2024). The extent and nature of gathered health information can change people’s beliefs and attitudes regarding specific health behaviors, thus promoting the decision-making process for these anticipated health behaviors (Lalazaryan and Zare-Farashbandi, 2014). Taking all this together, we posit the following hypothesis in the context of HPV vaccination:

H1: Attitude toward HPV vaccine information has a positive impact on HPV vaccination intention.

Information quality

Under the central route of ELM, people tend to engage in high elaboration when evaluating the content of persuasive information they receive. Consequently, the strength of information arguments determines recipients’ attitude formation and change (Teng and Khong, 2015). If the received information is regarded as high quality, strong, and convincing, individuals are likely to develop a positive attitude towards it and adopt corresponding decisions (O’Keefe, 1990). Information quality is primarily related to the accuracy, relevancy, and sufficiency of a communication topic (Zhou, 2012). In ELM studies, information quality serves as an important central cue in a person’s decision-making process (Zha et al., 2018; Zhou, 2012). Research from e-commerce has indicated that information quality positively affects users’ attitudes toward the information, including perceived information usefulness (Cho et al., 2019) and the tendency to adopt information (Jiang et al., 2021). Lin and colleagues (2017) further reported that the information quality of health messages had a positive impact on recipients’ attitudes toward the messages. In the field of information management, the influence of information quality on users’ behavioral intentions was also supported by prior studies (Jeong and Lambert, 2001; Muslichah, 2018), with this relationship being partially mediated by attitude toward the information (Lu et al., 2010). Therefore, it is reasonable to assume similar influencing paths in the context of MSVs and HPV vaccination information. The following hypotheses are posited:

H2a: Information quality of MSVs has a positive impact on attitude toward HPV vaccine information.

H2b: Information quality of MSVs has a positive impact on HPV vaccination intent.

H2c: The effect of information quality of MSVs on HPV vaccination intent is mediated by attitude toward HPV vaccine information.

Source credibility

When individuals are not highly motivated or lack the ability to engage deeply with the content, they are more likely to process information through the peripheral route, relying on source credibility to form their attitudes. Source credibility is understood as the degree to which a person or agency generating information is perceived to be trustworthy, knowledgeable, and believable (Zha et al., 2018). Generally, messages obtained from professional and credible sources (e.g., experts) are often more persuasive and have a significant impact on receivers’ attitudes (López and Sicilia, 2014). In the context of social media, source credibility is the most salient cue in the peripheral route of ELM, as other online peripheral cues are difficult to obtain (Teng and Khong, 2015). An extensive body of previous research in information studies has supported the impacts of source credibility on users’ perceptions and attitudes (Lee et al., 2022; Weber et al., 2023). The impact of source credibility has also been extended to behavioral intentions for recommended actions (Warkentin and Johnston, 2010). Evidence from Farhadpoor & Dezfuli (Farhadpoor and Dezfuli, 2021) confirmed the significant indirect effect of source credibility on shopping behavior through the attitude towards the source.

In the health promotion process, experimental studies (Hu and Sundar, 2009; Niu et al., 2020) have demonstrated that source credibility of online health information influences users’ attitudes toward the information and encourages corresponding behavioral intentions. Barnes et al. (2018) further identified the credibility of HPV vaccine information sources as a major theme that shaped parents’ attitudes toward the HPV vaccine. In a study of online health communities, findings from Xu et al. (2021) indicated that perceived message credibility indirectly affected HPV vaccination intention via information selection. However, the mediation role of attitude remains unexplored in this particular health context. Therefore, three hypotheses are therefore proposed related to MSVs and HPV:

H3a: Source credibility of MSVs has a positive impact on attitude toward HPV vaccine information.

H3b: Source credibility of MSVs has a positive impact on HPV vaccination intent.

H3c: The effect of source credibility of MSVs on HPV vaccination intent is mediated by attitude toward HPV vaccine information.

Media richness

When interacting with rich media that include visual and audio cues, users often demonstrate high engagement with the message and are therefore more likely to process the information (Robert and Dennis, 2005). As the level of engagement increases, the effort and intention devoted to the communication process also increase. In other words, the high richness of media leads to a high level of attention and motivation among users. Additionally, richer media are more capable of facilitating users’ understanding of ambiguous or equivocal messages (Daft and Lengel, 1986), enhancing their ability to process the message. As mentioned above, ELM proposes that the motivation and ability to process information determine the extent of elaboration, which in turn affects attitude and behavior (Petty and Cacioppo, 1986). Therefore, as some scholars argue (e.g., Yen, 2018), media richness might act as an important cue in the central route of ELM, influencing perceptual and behavioral changes. Nevertheless, other studies consider visual and audio factors to play a peripheral role in message processing (Lazard and Atkinson, 2014; Shin et al., 2020; Song et al., 2021a). Whether via the central or peripheral route, it is reasonable to assume that the integration of visuals and audio in MSVs influences attitudes and decisions regarding behaviors.

In terms of delivering health messages, recent research suggested that websites with multimedia elements, such as videos, were more persuasive in changing attitudes (2009) and promoting risk-protective behaviors (Perrault and Silk, 2014). Besides the significant main effect, media richness has been found to have an indirect effect on health behavioral intentions through the mediation of attitude (Lu et al., 2014). To explore the relationships on the emerging MSV platform and determine the persuasive impact on HPV vaccination intent at different levels of media richness, the following hypotheses are posited:

H4a: Media richness of MSVs has a positive impact on attitude toward HPV vaccine information.

H4b: Media richness of MSVs has a positive impact on HPV vaccination intent.

H4c: The effect of media richness of MSVs on HPV vaccination intent is mediated by attitude toward HPV vaccine information.

A conceptual model is proposed based on the above hypotheses, which is shown in Fig. 1.

Fig. 1
figure 1

Proposed conceptual model in this study.

Full size image

Furthermore, it is essential to explore the potential interactions among the three factors in terms of persuasive outcomes. Studying the moderating effects of media richness within the cognitive process of information is of theoretical interest (Gosain et al., 2019). Although ELM and media richness theory are derived from different theoretical perspectives, they both help explain individuals’ motivation and ability to process information. Therefore, it is possible that media, based on their levels of richness, will produce different influences—either supporting or hindering—the persuasive effects of central or peripheral cues. For example, Gosain et al. (2019) examined whether media richness could strengthen the relationship between information quality and customer engagement behaviors, but no significant moderation impact was detected. Another experimental study (Hu and Sundar, 2009) indicated that the persuasive effects of different sources (i.e., expert or layperson) on information attitude and behavior intentions varied on specific forms of social media. Their findings further confirmed that messages with different health contents produced different interactions between source and media type, as the health issues contained in them caused differential levels of heuristic processing. Moreover, the interplay of message quality and source credibility was examined in the context of information validation, with untrustworthy sources found to disrupt the processing of plausible texts (Wertgen and Richter, 2022). Given the scarcity of relevant evidence in the context of MSVs and HPV vaccination, an exploratory research question is thus posed in the current study:

RQ: Are there any interaction effects among information quality, source credibility, and media richness of MSVs on HPV vaccination intent?

Methods

This study has employed an experiment with a 2 (information quality: high vs. low) × 2 (source credibility: high vs. low) × 2 (media richness: high vs. low) between-subject factorial design to investigate the impact of MSV on HPV vaccination intent. The study has gained ethical approval from the university’s institutional review board.

Participants

Considering the high incidence and mortality rates of cervical cancer in China, the Chinese government has established a national target to improve HPV vaccine coverage among young adolescent females (National Health Commission of the People’s Republic of China, 2023). Given the specific socio-political context, this study excluded male participants but only recruited female adolescents and young adults (aged below 25 years). They are recruited from local high schools and universities in a southern province of China, between March and April 2021. Participants would be excluded from the experiment if they: (a) were majored in a medical background (to reduce the effect of existing knowledge); (b) got the HPV vaccine before; or (c) planned to get the HPV vaccine within the coming months. An a priori power analysis was conducted using G*Power (Faul et al., 2007) to determine the minimum sample size required to test the study hypotheses. To achieve 80% power for detecting a small-to-medium effect (d = 0.3) at a significance criterion of 0.05, the required sample size was 90 for main effects and 175 for interaction effects, respectively. The study initially recruited 266 participants, but 18 of them were then excluded because of conflicting schedules or previous history of HPV vaccination. Thus, the final sample size who completed the whole experiment is 248, which is adequate to test all the hypotheses.

Procedure

The participants were randomly assigned to one of the eight experimental conditions, in which they received customized short videos about the HPV vaccine once per day, for a total of two days. Rather than a laboratory experiment, this study employed the field experiment design which occurred in the natural environment (i.e., everyday life) of the participants. There were no restrictions on when and where they watched these short videos—participants only needed to follow their own habits for smartphone usage. The natural setting is able to increase the external validity of the experiment, and the results can be generalized to other contexts. A post-study survey was issued to all the participants after the two-day intervention, to collect their information attitudes, and behavioral intentions toward the HPV vaccine. The actual exposure to the videos was further asked in the survey, in which all participants self-reported that they had completed watching all the videos in the two days. Figure 2 illustrates the flowchart of the experiment.

Fig. 2
figure 2

Study flowchart and details of interventions.

Full size image

Intervention

A total of 16 customized short videos (2 videos per condition) were used in this experiment. They were designed based on the characteristics of the TikTok platform, and the length of the videos was about one minute. The short videos in different conditions varied from each other in terms of information quality, source credibility, and media richness. Details of the short videos in eight conditions were also shown in Fig. 2. Detailed descriptions and screenshots of all the experimental conditions were demonstrated in Appendix 1. The manipulation of the three factors was further described below:

  1. a.

    Information quality. The central route of ELM is logic-driven and often uses data and facts to convince the audience. In the design of information quality in MSVs, we adopted the theoretical framework of the health belief model (HBM; Rosenstock, 1974) to frame the persuasion of HPV vaccination, which includes perceived susceptibility, perceived severity, perceived benefits, perceived barriers, and cues to action. Specifically, short videos with high information quality include detailed figures and scientific explanations when discussing susceptibility and severity of cervical cancer, benefits, and cues to action toward the HPV vaccine. In contrast, those with low information quality cover these topics only with ambiguous wording and emotional arguments.

  2. b.

    Source credibility. Given the context of MSV, the presence of source credibility is mainly manipulated by the account details and appearance of the “vlogger” (i.e., persuader) in the short videos. More specifically, short videos with high source credibility link to an authorized account with a medical-related account name (e.g., “Doctor Niu in Gynecology”). The account details were shown right below the videos, and the audience could easily notice them while watching the videos. Meanwhile, the vlogger in these videos acted as a health professional, with a professional dress code such as wearing a white coat and a facial mask. For the short videos with low source credibility, they link to an unauthorized account with a random account name. The person talking in these videos appeared to be a layperson, wearing casual and normal clothes just like ordinary people.

  3. c.

    Media richness. Following the practice from previous studies (Dennis and Kinney, 1998; Kaplan and Haenlein, 2010), media richness was mainly manipulated by the number of ways in which HPV vaccine information can be communicated, i.e., visual and audio cues in the MSV such as animation, subtitle, and music. For example, short videos with high media richness have vivid animation and/or visual effects when educating the audience on HPV illness and vaccines. Clear subtitles and relevant background music were also added to aid the persuasion process. However, all these elements were absent in those short videos with low media richness, leaving only the oral presentation of the persuader.

Manipulation check

Before the actual experiment, a manipulation check was first performed to ensure that three factors (i.e., information quality, source credibility, and attractiveness) had been effectively manipulated across the experimental conditions. An extra group of 92 participants with similar backgrounds (but didn’t participate in the final experiment) were invited to watch all 16 short videos. They were then asked to finish a survey that evaluated each video in terms of information quality, source credibility, and attractiveness. The measurement of information quality was a seven-item scale (e.g., “The information presented in the video is scientific”) adapted from Pee (2012). Source credibility was assessed by a five-item scale (e.g., “This account has a good reputation”) adapted from Ohanian (1990). A three-item scale (e.g., “This video has various ways for expression”) from Yang and Chen (2020) was used to measure media richness. The sum of the item scores was used to perform the statistical test in each of the three variables. Results from independent t-test supported the effective manipulation of experiment conditions, by indicating a significant difference between videos with high and low information quality (t = −2.528, p = 0.013), source credibility (t = −2.766, p = 0.007), and media richness (t = −2.766, p < 0.001). Table 1 indicates the details results of the manipulation check.

Table 1 Results of manipulation check (n = 92).
Full size table

Measurement

The main outcomes measured in this experiment were attitude toward HPV vaccine information and HPV vaccination intent. Four items of attitude toward HPV vaccine information were originally from Bhattacherjee and Sanford (2006) but modified to fit the current context. It assessed whether participants perceived the HPV vaccine information as objective, safe, comfortable, and reliable for decision-making. Six items of HPV vaccination intent were developed based on scales from Gerend and Shepherd (2007) and Fu (2015), to measure the likelihood that participants would try to get more information about and/or consider getting the HPV vaccine once it is available. A seven-Likert scale ranging from strongly disagree (1) to strongly agree (7) was used to measure all the question items. Both attitudes toward HPV vaccine information and HPV vaccination intent resulted in good reliability in the current study, with a Cronbach’s α of 0.833 and 0.892, respectively. Good convergent validity and discriminant validity were further reported among these two outcome measurements (see Appendix 2).

Results

Table 2 demonstrates the descriptive statistics of outcome measures in all eight experiment conditions. Path analysis was applied to examine the proposed model and test the persuasive effects of the three factors, i.e., information quality, source credibility, and media richness. Path analysis is a form of multiple regression statistical analysis that is commonly used to evaluate causal models. Different from those in cross-sectional studies, the model in the current study involved experimentally manipulated independent variables. Thus, we followed the guidelines from Bollen (1989) and MacCallum and Austin (2000), to model the relationships of the categorical exogenous (predictor) variables to mediators and outcomes. The path analysis was conducted in Mplus version 6.0.

Table 2 Outcome measures for all the experiment conditions (n = 248).
Full size table

Hypotheses testing

Before the model examination, goodness-of-fit was first assessed to determine how the proposed model fits the actual data. Similar to previous studies (Li et al., 2020; Xu et al., 2016), multiple fit statistics were used including the Chi-square test, Comparative Fit Index (CFI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR). The recommended values of good fit in these indices were adopted from prior literature (Bentler, 1990; Hu and Bentler, 1990; Steiger and Lind, 1980), and demonstrated in Table 3. Based on the recommended criteria, the proposed model has resulted in a good fit for data (refer to Table 3). Figure 3 shows the estimations of the path coefficient and hypotheses testing results. Particularly, regarding the central route, significant positive effects of information quality in MSV were observed on attitude toward HPV vaccine information (β = 0.242, p = 0.004), and HPV vaccination intent (β = 0.191, p = 0.007), thereby supporting H2a and H2b. Regarding the peripheral route, the positive effects of source credibility and media richness of MSV on attitude toward HPV vaccine information were also significant, with β = 0.188, p = 0.008, and β = 0.193, p = 0.021, respectively. Similar persuasive effects of source credibility and media richness were also found on HPV vaccination intent, with β = 0.219, p = 0.002, and β = 0.181, p = 0.040, respectively. Thus H3a, H3b, H4a, and H4b were supported. Lastly, H1 was strongly supported by a significant path from the attitude toward HPV vaccine information to HPV vaccination intent (β = 0.464, p < 0.001).

Table 3 Goodness-of-fit indices and model fit.
Full size table
Fig. 3: Results of path analysis.
figure 3

Note. Path coefficients are standardized. The solid line indicates a significant path with *p < 0.05, **p < 0.01, ***p < 0.001.

Full size image

The mediation role of attitude toward HPV vaccine information was further examined on Mplus with Andrew Hayes’ PROCESS analysis, followed by the guideline from Stride et al. (2015). Findings from the mediation tests indicated a strong indirect effect of information quality on HPV vaccination intent via medication of attitude toward HPV vaccine information (β = 0.452, p < 0.001). Similar results were also found on the mediation effects of source credibility (β = 0.458, p < 0.001) and media richness (β = 0.427, p < 0.001). Consequently, H2c, H3c, and H4c were supported.

A significantly large R2 of 0.435 in HPV vaccination intent was reported in the model. It demonstrated that 43.5% of the unique variance of HPV vaccination intent was explained by the independent and mediating variables mentioned above. Generally, the findings have revealed significant persuasive impacts of three factors on people’s decision-making process toward HPV vaccine, indicating that MSV with high information quality, source credibility, and media richness would lead to a more positive attitude toward HPV vaccine information among the audience, and further increase their likelihood to go for HPV vaccination.

Interaction effects

A series of multiple-way analyses of variance (ANOVA) was conducted in IBM SPSS, to examine the RQ of interaction effects among the three factors. Findings from the three-way ANOVA did not indicate a significant three-way interaction effect among three factors on HPV vaccination intent (p = 0.067). Nevertheless, a significant two-way interaction effect was observed between information quality and media richness (F(1239) = 68.793, p < 0.001, η2 = 0.224) on HPV vaccination intent. These results suggested that the persuasive effect of media richness varied across different levels of information quality, such that MSV with richer visual and audio cues tended to be even more persuasive in promoting HPV vaccination if they communicated high information quality when compared to those with limited cues. The two-way interaction between source credibility and media richness is approaching significant (F(1239) = 3.779, p = 0.053, η2 = 0.016), releasing a similar conclusion on HPV vaccination intent. However, this significant two-way interaction effect was not found between information quality and source credibility (p = 0.065). The results of ANOVAs are further shown in Table 4.

Table 4 Results of multiple-way ANOVAs on HPV vaccination intent.
Full size table

Discussion

This study seeks to investigate if and how HPV vaccine information delivered via mobile short videos (MSVs) can persuade individuals to decide to get the HPV vaccine. By varying the form of information quality, source credibility, and media richness, it provides valuable insights into how health professionals can best adopt emerging social media to maximize their persuasive effects in health education and promotion. Additionally, this research applies the theoretical lenses of the elaboration likelihood model (ELM) and media richness theory to understand the persuasive mechanisms of short-form videos in this health context. Primary results indicate that the quality of health content communicated in MSVs has a direct impact on individuals’ protective behavioral intentions and an indirect impact through changing their attitudes toward health information. This aligns with the central route of the ELM (Petty and Cacioppo, 1986) and confirms the role of information quality as a major determinant of health behavioral intentions on the new medium of MSVs. Health messages with strong arguments and clear evidence have been reported to be more persuasive in previous research (Goh and Chi, 2017; Lin et al., 2017). Our study further demonstrates that health messages with high argument quality of HBM constructs are more persuasive than those with low argument quality. The findings can inspire health professionals to develop better educational materials on social media to encourage HPV vaccination or other similar protective health behaviors.

Consistent with prior findings (Lin et al., 2017; Niu et al., 2020), short videos featuring health professionals were more persuasive than those featuring laypersons in affecting attitudes toward HPV vaccine information and HPV vaccination intent. In the case of health messages related to a major cancer risk, the presence of a credible and authorized communicator might be considered evidence of argument quality (Petty and Cacioppo, 1984), leading to increasingly positive attitudes toward the messages and the subsequent health behaviors they advocate. Additionally, source credibility was reported to be the most critical predictor of HPV vaccination intent. This finding represents an important step forward in revealing the role of source credibility in the new form of video-based media. Unlike traditional online videos, the success of MSVs relies on the core concept of “fast entertainment.” With limited video length and bitrate, users are not expected to spend much time and cognitive effort processing the content. Instead, they may rely more on peripheral cues, such as source credibility, to judge an argument within a limited time period without deep cognitive elaboration. This reliance on peripheral cues highlights the significant impact of source credibility in influencing viewers’ attitudes and behavioral intentions.

The findings also shed light on how media use influences individuals’ perceptions and decisions regarding health behaviors. Consistent with previous studies (Mohamad, 2020; Perrault and Silk, 2014; Yen, 2018), MSVs with high media richness resulted in more positive attitudes toward HPV vaccine information and increased intent for HPV vaccination. The heightened attention and motivation elicited by multiple visual and audio cues may explain the persuasive impact on attitudes and behavioral intentions. However, this explanation warrants further validation in future studies through proper measurement of attention and motivation. Furthermore, in predicting the persuasive impact, it is crucial to point out a (approaching) significant interaction between source credibility and media richness on HPV vaccination intent. These findings have integrated media richness into the extended ELM framework: For MSVs with multiple cues, connecting to a credible source has proven more effective in enhancing health persuasion for HPV vaccination. One possible explanation is that peripheral cues are more effectively conveyed through media with high cue multiplicity (Gold-Nöteberg, 2004). High media richness facilitates the evocation of peripheral cue processes, thereby enhancing message persuasiveness. Robert and Dennis (2005) argued that individuals presented with complex messages in high-richness environments tend to rely on peripheral cues and avoid cognitive evaluation. While much remains to be explored in this emerging field, this study has illuminated an important mechanism linking media richness and peripheral cues.

Additionally, this study identified a significant interaction between central cues and media richness, indicating that media richness enhances the persuasive effect of promoting HPV vaccination when the presented vaccine information is of high quality. Previous research has primarily focused on distinguishing the influences of information quality and media richness on psychological outcomes (Chen and Chang, 2018; Zhang et al., 2021), with few exploring their interaction effects within an integrated framework. This study addresses this gap by demonstrating how these two distinct characteristics collaborate to shape decision-making processes. In the context of MSVs, while high media richness enhances communication efficiency, it also presents the challenge of managing a large volume of HPV vaccine information within a short period. The abundance of novel and complex information conveyed through high-richness media increases the likelihood of information overload among recipients (Schneider, 1987). However, information overload can be mitigated through the delivery of accurate and clear information (Kugbey et al., 2019). Thus, high information quality may amplify the persuasive impact of media richness by reducing information overload. This finding holds significant implications for public health campaigns aiming to effectively communicate high-quality health messages using rich media with multiple cues.

Limitations and implications

Several limitations were identified in this study. The primary limitation is the relatively short intervention duration, which was primarily due to constraints in time and manpower. Future research could enhance validity by allowing for longer intervention periods to determine the optimal frequency of MSVs for health persuasion. Additionally, although the study interventions aimed to simulate real-world contexts, they lacked an important element of MSV—interactivity. Incorporating features such as commenting, liking, or retweeting in experiments could provide insights into the effects of MSVs in a more natural environment. Furthermore, the persuasive effects of MSVs can be significantly influenced by technological features (e.g., screen size, resolution, mobility) of the mobile medium through which they are presented, which were not controlled in this experiment. Future research could improve by systematically testing the effects of these technological variables. Lastly, the study’s use of a sample consisting solely of female students may limit the generalizability of its findings. Future studies should aim to include a broader population encompassing diverse socio-economic backgrounds.

Overall, these findings underscore the importance of information quality, source credibility, and media richness when communicating HPV vaccine information via MSVs. They offer both theoretical insights and practical implications for health persuasion in the evolving digital landscape. From a theoretical perspective, this study establishes a framework integrating ELM and media richness theory to explore their application in promoting HPV vaccination. However, in light of contemporary digital communication advancements, concerns have been raised regarding the adequacy of ELM alone (Kitchen et al., 2014). The study’s identification of two-way interaction effects contributes to the existing literature by confirming media richness’s moderating role in both central and peripheral routes to health persuasion. Moreover, the proposed framework extends beyond psychological outcomes to include behavioral intentions. Given the high correlation between intention and actual behavior, it serves as a more reliable indicator of the persuasive impact of health messages. The results also support the mediating role of information attitude in the persuasion process, highlighting how individuals’ beliefs about health information from various sources influence their adoption of healthy behaviors.

Regarding practical implications, the results highlight that MSV is a viable medium for promoting health messages, particularly those related to HPV vaccination. However, different methods of message delivery within short videos significantly influence their behavioral effects. Health professionals and practitioners should therefore prioritize information quality and media richness when developing strategies to disseminate health knowledge through MSVs. Given the brief duration of short videos, effectively presenting comprehensive content with a clear layout becomes crucial for audience persuasion. Theoretical frameworks such as HBM can provide guidance for designing messages. Additionally, Chinese health professionals can leverage emerging short video platforms to create HPV educational materials that utilize visual and audio cues, such as animations or scenario-based presentations, to achieve optimal persuasive effects. Furthermore, while credible and authorized sources are perceived as more persuasive than lay sources, it would be beneficial to post videos using official accounts or indicating affiliation with reputable organizations. Although not fully explored in this study, a combination of these strategies holds promise for maximizing the persuasive impact of MSVs in health promotion.

Conclusion

In conclusion, this study has delved into the effectiveness of using MSV to promote HPV vaccination, focusing on the interaction of information quality, source credibility, and media richness. By applying the ELM and media richness theory, we have uncovered significant insights into the persuasive mechanisms at play in this emerging digital health communication context. Our findings underscored that the quality of health information conveyed through short videos directly influences individuals’ protective behavioral intentions towards HPV vaccination. Moreover, the presence of credible sources, particularly health professionals, enhanced positive attitudes and intentions toward vaccination. Furthermore, the role of media richness emerged prominently, with higher levels of visual and auditory cues in MSVs correlating positively with increased attitude change and behavioral intent. The integration of peripheral cues, facilitated by media richness, was found to augment the persuasive impact of health messages, particularly when paired with high-quality health information. Overall, this research contributes to both theoretical advancements in digital health communication and practical guidance for leveraging emerging media formats to promote public health initiatives effectively.