The stimulus–organism–response (S-O-R) model proposed by Mehrabian and Russell (1974), has been widely adopted to explain how consumers process stimuli in digital environments, including tourism contexts. In this framework, external stimuli (such as visual or narrative content from an influencer) evoke internal responses in the organism (emotions, attitudes, and perceptions), which subsequently lead to behavioral responses, such as the intention to travel or share information online.

Liu et al. (2023) applied the S-O-R model to examine how destination marketing using short videos influences travel intentions. In their study, audiovisual stimuli activated internal mechanisms, such as telepresence, perceived enjoyment, and perceived value, which mediated the relationship between the content and the intention to visit the destination. These results empirically validate the role of internal states in tourists’ behavioral responses. Zhang et al. (2021) similarly utilized the S-O-R model in the context of influencer marketing via live streaming, identifying stimuli such as source attractiveness, content utility, perceived social norms, and virtual enjoyment. These stimuli affect the user’s organism, measured by attitudes toward the influencer and impulsive buying tendencies, ultimately determining offline purchase intentions. Their findings demonstrated that intermediate affective and cognitive mechanisms are crucial for understanding the impact of influencers.

Complementarily, Matiza and Slabbert (2024) employed the S-O-R model to explain how celebrity-generated social media content influences factors such as trust, safety perception, and country image evaluation. These factors, which represent the organism, mediate the effect of stimuli (attributes of the influencer) on travel intentions. Collectively, these studies underscore that the S-O-R model provides a robust explanatory framework for analyzing how influencer-generated digital content influences travel decisions, enabling the sequential mapping of stimulus impacts on internal user states and behavioral responses.

The Travel Influencer Construct (TIC) proposed by Manthiou et al. (2024) offers a multidimensional and context-sensitive framework that integrates five perceptual dimensions—attractiveness, expertise, trustworthiness, authenticity, and inspiration—to explain how users form attitudes toward tourism content creators. This model addresses conceptual limitations in previous literature, which tended to rely on fragmented constructs such as source credibility or parasocial interaction without fully capturing the experiential and affective nature of tourism marketing on social media (Djafarova and Rushworth, 2017; Lou and Yuan, 2019). Unlike earlier models, the TIC was empirically validated using both qualitative and Bayesian quantitative methods in highly visual and interactive tourism contexts, making it especially relevant for platforms such as TikTok (Manthiou et al., 2024).

Moreover, the TIC incorporates dimensions, such as authenticity and inspiration, that have proven particularly salient in digital tourism environments shaped by emotional engagement and visual storytelling. For example, inspiration operates as an aspirational cue that allows users to mentally project themselves into narrated travel experiences and motivates future behavior (Pop et al., 2022; Saini et al., 2023). Authenticity, in turn, reinforces identification and credibility, especially when influencers convey consistency and alignment with their audiences’ values (Matiza and Slabbert, 2024). While these attributes have been discussed in prior studies, they have seldom been systematically integrated into formal conceptualizations of influencer impact.

Given the growing relevance of short-form, algorithmically curated platforms in shaping travel perceptions, especially among younger users, there is increasing interest in frameworks that reflect the emotional, symbolic, and social dynamics of such environments (Cao et al., 2021; Liu et al., 2023). In this context, the TIC offers a more comprehensive and theoretically coherent approach for understanding influencer effectiveness in the tourism domain.

While this study adopts the TIC framework proposed by Manthiou et al. (2024), its primary contribution lies in applying this model to a distinct sociocultural and platform-specific context. Rather than focusing on scale validation, the study examines the TIC’s conceptual utility within TikTok’s short-form, algorithm-driven video ecosystem and among young Chilean users. This approach enables an exploration of how each dimension of the TIC operates in an environment characterized by high visual intensity, rapid content consumption, and platform-specific engagement logics, thereby contributing to the broader understanding of the model’s adaptability and relevance across digital contexts marked by diverse technological and cultural dynamics. In light of this, each of the five dimensions is discussed below.

The attitude toward a travel influencer represents the consumer’s internal response resulting from exposure to stimuli such as the presenter’s attractiveness, the enjoyment derived from the content, and its perceived usefulness (Manthiou et al., 2024). This response manifests in behaviors such as purchase intention (Zhang et al., 2021). This evaluation comprises cognitive components, such as the perceived usefulness of the content, affective elements, such as emotional connection with the influencer, and behavioral aspects, such as the predisposition to follow their suggestions or share their content. Zhang et al. (2021) demonstrate that in the context of live streaming, attitude toward the influencer emerges as a key internal response influenced by source attractiveness, perceived usefulness, and virtual enjoyment, which in turn predicts offline purchase intention. This approach, based on the S-O-R model, posits attitude as the organism between digital environmental stimuli and consumer behavioral responses.

In tourism, Najar et al. (2024) provided empirical evidence that a favorable attitude toward an influencer increases destination trust, visit intention, and electronic word-of-mouth (eWOM). In their study, influencer attributes (such as authenticity, trustworthiness, and expertise) directly predicted users’ attitudes. Additionally, Pop et al. (2022) and Rodriguez-Hidalgo et al. (2023) emphasized that such attitudes are not formed in isolation; rather, they are shaped by perceived congruence between the influencer’s values and those of the user as well as thematic consistency across content. These conditions strengthen user identification with the source, fostering a more stable and positive attitude toward the content. Thus, attitude toward the travel influencer emerges as a central construct in explaining influencer marketing effectiveness, as it reflects the psychological link between the emitted content and the recipient’s decisions in digital tourism environments.

Behavioral intention regarding a travel influencer refers to users’ willingness to act based on the influencer’s recommendations, encompassing behaviors such as seeking additional information, sharing content, or planning a trip (Purwandari et al., 2022). This variable is critical in assessing the actual impact of influencer marketing in the digital tourism context as it allows the estimation of the extent to which content exposure translates into concrete decisions. Multiple studies have shown that the perceived attributes of an influencer significantly influence users’ attitudes, which, in turn, predict behavioral intentions. For example, Purwandari et al. (2022) found that affective engagement generated through interactions with the influencer, along with the perceived informational and emotional value of the content, positively affects the intention to follow travel recommendations. Similarly, Sokolova and Kefi (2020) demonstrated that influencer credibility and parasocial interaction directly affect purchase intention, a finding that can be extrapolated to tourism when specific destinations are promoted via social media. In a similar manner, Pop et al. (2022) highlighted that trust in the influencer, derived from perceived similarity and authenticity, enhances the intention to visit the promoted destinations and share recommendations through eWOM. In this case, behavioral intention arises not solely as a rational response to content, but also as a result of a symbolic relationship connecting the user with both the message and the source.

Theoretically, the S-O-R model offers a robust framework for understanding this phenomenon. Zhang et al. (2021) applied this model and confirmed that influencer characteristics (stimuli), such as attractiveness or content usefulness, generate a positive attitude toward the influencer (organism), which then directly influences purchase intention (response). This framework has been validated in tourism contexts, reinforcing the role of attitude in the relationship between perceived influencer attributes and behavioral intentions regarding recommendations.

Behavioral intention to visit the recommended destination refers to the user’s conscious willingness to visit a given location in the future, influenced by perceptions formed through exposure to digital content, particularly that generated by social media influencers. This variable has been widely validated as a behavioral outcome of influencer marketing in tourism, with a strong predictive value for actual travel decisions (Priyanga and Ashokkumar, 2023; Zhang and Xu, 2023). Several studies show that visit intention is significantly strengthened when information provided by influencers perceived as credible, attractive, or aligned with the recipient’s values (Meng Monroe et al., 2025; Pop et al., 2022). For example, followers have been shown to develop stronger travel intentions when they identify value, background, or attitudinal similarities with the influencer, thereby enhancing the perceived credibility and usefulness of the content (Pop et al., 2022).

In addition, the study by Meng Monroe et al. (2025) shows that congruence between the type of influencer (human or virtual) and the type of destination (natural or cultural) significantly affects visit intention. This effect is explained by differential perceptions of credibility and activation of the self-referencing mechanism, which mediates the relationship between influencer type and willingness to visit the destination. In summary, behavioral intention toward a destination is shaped not only by attributes of the place itself, but also by a complex network of symbolic stimuli, identity congruence, and perceived source credibility. Understanding these factors allows for the optimization of tourism promotion strategies through the strategic use of digital influencers.

Electronic word-of-mouth (eWOM) has become an essential component of the dissemination of tourism content in digital environments. In the context of influencer marketing, eWOM refers to the voluntary transmission of opinions, experiences, and recommendations generated by users after interacting with the content posted by travel influencers. This form of communication has proven to be highly persuasive, particularly when it originates from sources that are perceived as trustworthy and relatable (Saini et al., 2023). Pop et al. (2022) emphasized that influencers who generate trust tend to foster active participation from their followers, who not only adopt recommendations, but also retransmit them through posts, comments, or interactions on social media. This dynamic transforms the consumer into a “prosumer” (both producer and consumer of content), thereby expanding the reach of tourism campaigns through horizontal social diffusion. Additionally, Rahman and Mia (2025) empirically confirms that perceived source credibility directly influences eWOM effectiveness, as users are more likely to trust recommendations from figures they perceive as authentic, competent, and emotionally connected to their audience. In this sense, the impact of eWOM lies not only in the content itself but also in how the recipient perceives the influencer as a source. Rodriguez-Hidalgo et al. (2023) highlighted that the perceived congruence between the influencer’s values and the user’s interests strengthens the latter’s willingness to share content, creating a cascading effect that amplifies the reach of the original message. Consequently, eWOM emerging from influencers does not merely entail direct recommendations, but also activates social and emotional processes that support the co-creation of tourism content on social media.

This quantitative cross-sectional study analyzed how social media users, particularly TikTok, modify their information-seeking behaviors related to tourism. The target population focused on young Chilean users, primarily Generation Z and Millennials, who represent the most active TikTok users in the country (Statista, 2024a). According to The Social Shepherd (2025), TikTok was the most downloaded app globally in 2024, with 773 million downloads. Additionally, Backlinko (2025) reported that 55% of the platform’s weekly active users were aged between 18 and 34, and adults spent, on average, 53.8 min daily on the app. However, regional data suggest much higher engagement levels: Chile ranks fourth worldwide in average monthly TikTok usage, with users spending approximately 45 h and 30 min per month, or about 1.5 h per day, on the platform (Exploding Topics, 2025). This aligns with the usage profile of our sample, in which over 80% of respondents reported using TikTok for more than 1 h per day, and reinforces the study’s focus on intensively engaged users who are more likely to be exposed to and influenced by travel content on the platform.

The sample was selected using non-probabilistic convenience sampling, due to accessibility and alignment with the objectives of the study. Prior to the face-to-face data collection conducted by professional interviewers between August and November 2024, a pre-test was administered to 30 participants. A total of 237 valid responses were obtained, of which 66.2% corresponded to Generation Z and 33.8% to Generation Y. Regarding gender, 54.9% were identified as female, 44.3% as male, and 0.8% as other. The age distribution was concentrated within the 18–27 years age range (54.8% of the total), with an average age of 27 years. Geographically, 41.4% resided in the northern region, 28.7% in the central region, and 29.9% in the southern region of the country. In relation to TikTok usage, 30.4% indicated using the platform for 2 h per day, while 57.4% reported using it between one and 3 h daily. Regarding tourism mobility, 31.2% reported traveling for leisure once per year, followed by 26.6% who traveled less than once per year and 22.4% who traveled twice annually. The demographic data are summarized in Table 1.

The measurement instrument consisted of a structured, self-administered questionnaire designed to assess the factors influencing the use of TikTok as a search engine for planning tourism-related travel. The scales employed were adapted from validated empirical literature and semantically adjusted to the Chilean context, with a specific focus on Generation Z and Generation Y.

The Travel Influencer Construct (TIC) was operationalized as a second-order variable composed of five dimensions: inspiration, authenticity, attractiveness, content expertise, and trustworthiness. These dimensions were measured using 17 items, each evaluated on a seven-point Likert scale (1 = “strongly disagree”; 7 = “strongly agree”), based on the model developed and validated by Manthiou et al. (2024). Sample items included statements such as: “Travel influencers inspire me to travel,” “Travel influencers are visually appealing,” “Travel influencers know what they are talking about,” and “I trust travel influencers.”

User behavioral intentions were incorporated into the model as dependent variables, following the framework proposed by Manthiou et al. (2024), who identified three key responses: (1) intention to follow the influencer’s recommendations, (2) intention to visit the promoted destination, and (3) intention to share content through social media. These variables were measured using seven-point Likert scales. Before the final implementation, the questionnaire was reviewed by experts in digital marketing and tourism to ensure content validity and contextual appropriateness. Subsequently, a pre-test was conducted with 30 participants to verify the semantic clarity, internal consistency, and overall functionality of the instrument.

Data analysis was conducted using partial least squares structural equation modeling (PLS-SEM), a variance-based technique chosen for its flexibility and suitability in estimating complex models involving multiple latent constructs and observed variables. This methodology is particularly appropriate for explanatory and predictive research, where the modeling of composite relationships and assessment of explained variance are prioritized (Henseler, 2018; Sarstedt et al., 2017). Unlike covariance-based approaches, PLS-SEM allows for analyses with smaller sample sizes, non-normal data distributions, and both formative and reflective model structures (Henseler et al., 2016). Its application has been widely validated in studies on marketing, tourism, and consumer behavior.

Following methodological recommendations, the analysis was conducted in two stages: first, the evaluation of the measurement model, focusing on reliability, convergent validity, and discriminant validity, and second, the evaluation of the structural model, which examined path coefficients and the coefficient of determination (R²; Gudergan et al., 2008; Henseler et al., 2016). This analytical strategy enables an empirical assessment of the measurement quality and the robustness of the proposed theoretical model. The analysis was performed using the SmartPLS4 software (Ringle et al., 2022).

To examine potential differences between generational cohorts, a multi-group analysis (PLS-MGA) was conducted, comparing Generation Z (n = 157) and Generation Y (n = 80). This analysis followed the non-parametric procedure proposed by Henseler et al. (2016) and used 5,000 bootstrap resamples to test for significant differences in path coefficients across groups. Prior to conducting the multi-group comparison, measurement invariance was evaluated using the Measurement Invariance of Composite Models (MICOM) procedure, confirming configural and compositional invariance. The MGA results allowed us to assess whether the structural relationships varied significantly between generations, offering additional insight into audience heterogeneity.

Measurement models were evaluated to verify the reliability and validity of the model’s latent constructs. Regarding internal reliability, all constructs exhibited satisfactory values for Cronbach’s alpha (CA), composite reliability (Rho_C), and Dijkstra–Henseler’s reliability (Rho_A), exceeding the recommended threshold of 0.70 established in the literature (J. F. Hair et al., 2022). In addition, the individual item loadings were above 0.70, indicating adequate internal consistency. Convergent validity was confirmed through Average Variance Extracted (AVE), with all values exceeding the minimum recommended threshold of 0.50 (Fornell and Larcker, 1981). This indicates that the items explained a substantial proportion of the variance in their respective constructs. Table 2 presents the results.

The structural model was assessed through the analysis of path coefficients, coefficients of determination (R²), and model fit index SRMR. SRMR yielded a value of 0.04, which is below the threshold of 0.08, indicating an acceptable model fit (Henseler et al., 2016; Table 3).

The model explained 75.1% of the variance in attitude toward travel influencers (R² = 0.751). Among the dimensions of the TIC construct, attractiveness had the strongest impact on attitude (β = 0.373, p < 0.001), followed by content expertise (β = 0.282, p < 0.001), and trustworthiness (β = 0.254, p < 0.001). Authenticity (β = −0.06, p = 0.364) and inspiration (β = 0.101, p = 0.094) were not significant in this context.

Attitude toward influencers showed a significant direct impact on all three dependent variables. It explained 70.4% of behavioral intention toward influencers (R² = 0.704; β = 0.839, p < 0.001), 63.8% of behavioral intention toward the destination (R² = 0.638; β = 0.799, p < 0.001), and 39.9% of electronic word-of-mouth (eWOM; R² = 0.399; β = 0.632, p < 0.001). These results underscore the importance of producing attractive, trustworthy, and professional content to enhance the influence of content creators on tourism-related decisions. The results are shown in Figure 2.

The results of the hypothesis testing (Table 4) allowed for the evaluation of the structural relationships proposed in the model. Of the eight formulated hypotheses, six were statistically significant (p < 0.001), while two did not meet the required threshold (p > 0.05). First, H1, which posited a positive effect of the inspiration dimension on attitude toward travel influencers, was not supported (β = 0.101; p = 0.094). This suggests that, in this context, influencers’ capacity to evoke travel desires does not constitute a decisive factor in users’ attitudinal evaluations. Similarly, H2 was rejected, as influencer authenticity did not have a significant impact on attitude (β = −0.060; p = 0.364), contradicting prior literature and highlighting a potential loss of centrality for this attribute within certain digital environments. Conversely, H3 was confirmed, establishing influencer attractiveness as exerting the strongest effect on attitude (β = 0.373; p < 0.001), followed by content expertise (H4, β = 0.282; p < 0.001) and trustworthiness (H5, β = 0.254; p < 0.001), all of which demonstrated positive and significant effects. These three dimensions of the TIC construct jointly explained 75.1% of the variance in attitude toward influencers (R² = 0.751), evidencing their strong explanatory power.

Regarding behavioral outcomes, attitude toward influencers was positively associated with behavioral intention toward the influencers themselves (H6, β = 0.839; R² = 0.704; p < 0.001), intention to visit promoted destinations (H7, β = 0.799; R² = 0.638; p < 0.001), and willingness to share content via electronic word-of-mouth (H8, β = 0.632; R² = 0.399; p < 0.001). Collectively, these findings underscore the relevance of producing visually attractive, professionally crafted, and trustworthy content to enhance the influence of content creators on users’ travel decisions.

Measurement invariance was assessed using the Measurement Invariance of Composite Models (MICOM) procedure, as described by Hair et al. (2017). In Step 1 (configural invariance), all constructs shared identical model specifications, indicators, and estimation settings across generational groups. In Step 2 (compositional invariance), all constructs showed non-significant permutation p-values (p > 0.05), confirming full compositional invariance. In Step 3 (equality of means and variances), most constructs demonstrated invariance; however, Authenticity and Electronic Word-of-Mouth (eWOM) presented significant differences (p < 0.05), indicating partial measurement invariance. The establishment of partial measurement invariance (i.e., configural and compositional invariance) is sufficient to proceed with multi-group comparisons of path coefficients using PLS-MGA (Hair et al., 2017). Therefore, the multi-group analysis was conducted, interpreting findings related to Authenticity and eWOM with caution.

To examine potential generational differences in the structural relationships of the model, a multi-group analysis (PLS-MGA) was conducted comparing participants from Generation Z (n = 157) and Generation Y (n = 80). The analysis aimed to test whether the path coefficients between constructs differed significantly across these two groups.

The results, presented in Table 5, indicate that no statistically significant differences were observed in the relationships between the five dimensions of the Travel Influencer Construct (TIC), inspiration, authenticity, attractiveness, content expertise, and trustworthiness, and attitude toward travel influencers (H1 to H5). This suggests that both generations evaluate the influencer attributes in a similar manner when forming their attitudes.

However, significant generational differences emerged in the paths from attitude toward travel influencers to the three behavioral outcome variables. Specifically, the path from attitude to behavioral intention regarding the influencer (H6) was significantly stronger for Generation Y (β = 0.929) than for Generation Z (β = 0.751), with a PLS-MGA p-value of 0.000. Likewise, the effect of attitude on intention to visit the promoted destination (H7) was higher among Generation Y participants (β = 0.894) compared to Generation Z (β = 0.703), with a p-value of 0.001. A similar pattern was observed in the path from attitude to electronic word-of-mouth (eWOM; H8), where Generation Y exhibited a stronger effect (β = 0.760) than Generation Z (β = 0.517), with a p-value of 0.003. These findings suggest that although both generations rely on similar criteria when forming attitudes toward travel influencers, the influence of those attitudes on behavioral intentions is significantly stronger among Millennials than among Gen Z participants.