The proliferation of digital platforms has propelled influencer marketing into a highly effective and cost-efficient tool, particularly for products targeting young, digitally native consumers. The beauty industry, specifically, has seen a rise in "beauty vloggers" who create online content reviewing and demonstrating cosmetic products. This study addresses the gap in existing research by holistically examining the source credibility model, encompassing trustworthiness, expertise, and attractiveness, to understand their impact on consumer trust and purchase intention. Existing literature often treats these aspects of credibility in isolation. The study also explores the role of parasocial interaction (PSI), the illusionary interpersonal relationship between audience and influencer, in shaping consumer behavior, a factor under-researched in emerging markets like India. The research questions are: 1) Do credible attributes of beauty vloggers influence consumer attitudes? 2) Does trust mediate the relationship between credible attributes and purchase intention? 3) Do repeated parasocial interactions affect purchase intention? This study uses Ohanian's (1991) source credibility model and Horton and Wohl's (1956) parasocial interaction theory to build a comprehensive framework illustrating how beauty vloggers influence purchasing behavior, particularly in the Indian context, where influencer marketing is a rapidly expanding phenomenon with a market value expected to reach 28 billion rupees by 2026 (Statista, 2023). The study's significance lies in providing actionable strategies for cosmetic marketers to enhance their influencer marketing campaigns by focusing on building consumer trust and leveraging the power of parasocial interaction.

The theoretical foundation of this research lies in the source credibility model, initially proposed by Hovland and Weiss (1951), which highlights trustworthiness and expertise as key dimensions. McGuire (1985) later expanded the model to include attractiveness. This study utilizes all three dimensions to assess beauty vlogger credibility, recognizing their significance in shaping consumer perceptions and purchase decisions. Numerous studies (e.g., Lee & Watkins, 2016; Sokolova & Kefi, 2019; Manchanda et al., 2022) have explored various aspects of influencer credibility, but often in a fragmented manner. This research aims to provide a more holistic understanding by incorporating all three dimensions. Furthermore, the role of trust is highlighted, considering it crucial for effective communication and positive attitudes towards sponsored messages (Jiang et al., 2015; Wang et al., 2017). The concept of parasocial interaction (PSI), introduced by Horton and Wohl (1956), is also central to this study. PSI describes the one-sided relationship between media personalities and their audience, creating an illusion of intimacy and connection. Prior research (e.g., Choi & Lee, 2019; Lee & Lee, 2022) has explored PSI's impact on purchase intention, but further investigation is needed in emerging market contexts like India. The review thus lays the groundwork for hypothesizing that source credibility positively influences both trust and purchase intention, trust mediates the relationship between credibility and purchase intention, and parasocial interactions positively influence purchase intention.

This quantitative study employed a survey methodology using a questionnaire adapted from existing literature to measure trustworthiness, expertise, attractiveness (all components of source credibility), trust, parasocial interaction, and purchase intention. A 5-point Likert scale was used for responses. Data was collected from 357 North Indian respondents (aged 16-45) using a non-probability convenience sampling technique. The sample was drawn from individuals who actively engage with beauty vloggers' content daily for 1-3 hours on social media. The sample comprises 56.3% male and 43.7% female, primarily from Punjab, Haryana, Chandigarh, and Delhi NCR, using predominantly YouTube, Instagram, and Facebook. The data collection period spanned from June 2022 to March 2023, with ethical clearance secured from the institutional review board. Partial Least Squares Structural Equation Modeling (PLS-SEM) using SmartPLS 4.0 was employed to analyze the data. PLS-SEM was chosen for its suitability in handling complex path models with latent variables and its ability to manage non-parametric data without requiring assumptions of multivariate normality. The model was assessed for reliability and validity using established criteria including Cronbach's alpha, composite reliability, average variance extracted (AVE), and the heterotrait-monotrait ratio of correlations (HTMT). The structural model was evaluated for multicollinearity and the hypotheses were tested using bootstrapping. PLS-predict was used to evaluate the predictive relevance of the model. The gender effect was also analyzed as a control variable for direct and indirect effects.

The results of the study support all five hypotheses. Confirmatory factor analysis confirmed the reliability and validity of the measurement model for both lower-order and higher-order constructs. All items exhibited factor loadings exceeding the acceptable threshold of 0.708. Cronbach's alpha values fell within the acceptable range (0.70–0.95), and AVE values surpassed the threshold of 0.50, confirming convergent validity. Discriminant validity was established via HTMT values, all of which remained below 0.85. The structural model assessment confirmed the absence of multicollinearity, with VIF values below 3. The analysis revealed significant positive relationships between variables: H1: Source credibility significantly and positively influences trust (β = 0.628, p < 0.001, f² = 0.650, strong effect); H2: Source credibility positively influences purchase intention (β = 0.226, p < 0.001, f² = 0.049, weak effect); H3: Trust positively influences purchase intention (β = 0.416, p < 0.001, f² = 0.202, moderate effect); and H5: Parasocial interaction positively influences purchase intention (β = 0.179, p < 0.001, f² = 0.040, weak effect). Mediation analysis (H4) showed that trust partially mediates the relationship between source credibility and purchase intention (β = 0.261, p < 0.001). The R-square values indicated that the model explains 39.2% of the variance in trust and 48.6% of the variance in purchase intention. PLS-predict analysis confirmed the model's predictive relevance, with Q² values for both trust and purchase intention being greater than 0, and RMSE values indicating higher predictive power compared to a linear regression model. Gender had no significant effect on either the direct or indirect relationships. The effect size analysis indicated that source credibility has a stronger effect on trust than on purchase intention directly; trust has a moderate effect on purchase intention; and parasocial interaction exhibits a weak effect on purchase intention.

The findings strongly support the influence of beauty vloggers on consumer purchase intention through a mechanism involving credibility, trust, and parasocial interaction. The mediating role of trust highlights its crucial role in translating perceived credibility into actual purchasing behavior. Consumers are more likely to trust and consequently purchase products endorsed by beauty vloggers they perceive as trustworthy, expert, and attractive. The positive impact of parasocial interaction, although weaker than the influence of credibility, underscores the importance of building genuine connections with the audience. The study's findings align with previous research (Kim & Kim, 2021; Lee & Watkins, 2016; Lee & Lee, 2022; Sokolova & Kefi, 2019; Manchanda et al., 2022; Lou & Yuan, 2019) while contributing to a more nuanced understanding of the Indian context, where consumer skepticism and cultural factors might influence the relative strength of credibility and parasocial interaction's impact. The weaker effect size of parasocial interaction in the Indian context compared to findings in other studies could be attributed to factors such as collectivist culture, high-context communication style, and respect for authority figures. Unlike some western contexts where parasocial connections might be stronger due to different cultural dynamics, this study suggests Indian consumers may place a higher initial value on established credibility before forming parasocial bonds.

This study makes significant theoretical contributions by integrating source credibility and parasocial interaction theories to explain consumer purchase intention in influencer marketing. It emphasizes the critical role of trust as a mediator and highlights the importance of all three aspects of source credibility (trustworthiness, expertise, and attractiveness) in influencer selection. Practically, the findings offer cosmetic companies valuable insights for developing effective influencer marketing strategies, particularly in emerging markets. Future research could explore the influence of age and regional variations across India, investigate the role of other moderating variables such as relationship strength, and compare the effectiveness of beauty vloggers versus virtual influencers.

The study's reliance on convenience sampling limits the generalizability of the findings beyond the specific North Indian sample. Future research should employ more representative sampling techniques to broaden the scope of the study's conclusions. The focus on a specific age range (16-45) also restricts the generalizability to other age groups. Additionally, the study focuses on the Indian context; cultural differences may affect the results in other regions. Finally, future research could explore additional factors that influence purchase intentions, such as the impact of audience characteristics.