The theoretical underpinnings of this study rely on key frameworks that explain how celebrity personality traits influence perceived product quality and purchasing intentions, and why these effects vary across social media platforms.

Source Credibility Theory posits that an endorser's persuasiveness depends on expertise, trustworthiness, and likability (26–28). In this study, likability, expertise, credibility, and physical attractiveness represent facets of source credibility shaping consumer evaluations. A celebrity with high expertise enhances perceived product quality, increasing purchase intentions. Similarly, likable and credible celebrities foster positive affect, supporting product-related decisions.

Elaboration Likelihood Model (ELM) explains how consumers process persuasive messages via central or peripheral routes, depending on involvement and cognitive capacity (28–30). On social media, peripheral cues such as likability, attractiveness, and charisma often dominate, as millennials frequently consume content quickly and impressionistically. This lens supports the hypothesis that traits like likability and attractiveness influence both perceived product quality and purchase intentions, especially when consumers do not deeply scrutinize product details.

Cognitive Response Model suggests that consumers actively process endorsement messages, generating cognitive and affective responses that shape attitudes and behavior (31, 32). Perceived product quality is a key cognitive response: when a celebrity demonstrates expertise or authenticity, consumers interpret the product as reliable and valuable, influencing purchase intentions. This model provides the rationale for including perceived product quality as a mediator between celebrity traits and purchasing decisions.

Attitude-Toward-the-Ad Model emphasizes that favorable attitudes toward advertisements strengthen the link between endorsements and purchase intentions (33). In social media, engaging and authentic content amplifies the persuasive impact of celebrity traits. For instance, emotional engagement with a likable celebrity fosters positive attitudes toward the product, enhancing the effects of credibility and attractiveness on consumer behavior.

Platform-specific mechanisms further clarify these dynamics. On visually rich platforms like Instagram, peripheral cues such as likability and attractiveness dominate, enhancing perceived product quality and purchase intentions. Twitter's text-oriented environment emphasizes central route processing, where expertise and credibility drive cognitive evaluations. Snapchat's informal and ephemeral content facilitates quick trust-building and engagement, leveraging a combination of likability and credibility.

Taken together, these frameworks provide a coherent rationale for hypothesizing how celebrity personality traits influence perceived product quality and purchasing intentions across Instagram, Snapchat, and Twitter. Source Credibility Theory explains why traits such as likability, expertise, credibility, and attractiveness enhance message acceptance, elevating perceived product quality—a key precursor to behavioral intention. ELM shows that millennials often rely on peripheral cues, making personality traits especially impactful in shaping perceived quality. The Cognitive Response Model clarifies that consumers generate cognitive evaluations based on affective and cognitive impressions of the celebrity. Finally, the Attitude-Toward-the-Ad Model supports the role of perceived product quality as a mediator, as consumers' evaluations of the message and the celebrity shape beliefs about the product, influencing purchase decisions.

These theories also guide interpretation of qualitative data, providing a basis for understanding how consumers describe experiences with celebrity endorsements, including perceptions of trust, expertise, and appeal. By examining how traits such as likability and credibility affect millennial purchasing intentions, this study uncovers nuanced mechanisms driving successful endorsements on Instagram, Snapchat, and Twitter. The mediating role of perceived product quality underscores the importance of aligning endorsements with consumer expectations to optimize marketing outcomes.

Likability plays a pivotal role in the success of celebrity endorsements, enabling endorsers to forge positive associations with the products they promote. Consumers often perceive products endorsed by likable celebrities as higher quality due to the endorser's personal appeal (17). This is especially relevant on visually driven platforms like Instagram, where relatability and personal engagement are paramount. Millennials are drawn to sports celebrities who project warmth and authenticity on social media, creating a sense of trust and personal connection that can translate to the endorsed product (20, 34). For instance, a sports celebrity sharing glimpses of their personal life and interacting with fans is more likely to enhance the perceived quality and desirability of the products they endorse (35). Likable celebrities not only foster trust but also elevate the attractiveness and credibility of the associated products, shaping millennial purchasing behavior (36–38). According to ELM, Instagram's visually rich and peripheral cue–dominated environment amplifies the impact of likability, whereas Snapchat leverages informal and spontaneous interactions to reinforce trust. Twitter relies less on peripheral cues, emphasizing quick impressions rather than detailed engagement. Accordingly, based on Source Credibility Theory and ELM, likability serves as a persuasive cue that increases message acceptance and influences the cognitive evaluation of the product, thereby strengthening both perceived quality and purchasing intentions.

H1: Sports celebrity likability positively influences millennial purchasing intentions on social media platforms (a. Instagram, b. Snapchat, c. Twitter).

This study adopted a mixed-methods sequential explanatory design (88), in which quantitative data collection and analysis were followed by qualitative interviews to provide deeper insight into survey findings. Quantitative results guided the development and focus of the interviews, allowing for rich integration of participants' perceptions of sports celebrity endorsements with the Structural Equation Modeling (SEM) results.

A cross-sectional survey was conducted among Egyptian millennials aged 23–41 who actively use Instagram, Snapchat, and Twitter. Because participants were recruited via online platforms (Google Forms) and social media networks, a true probability sampling frame was not feasible. A stratified quota sampling approach was adopted to approximate representativeness across platforms. Strata were defined by participants' primary social media platform usage. Recruitment involved sharing the survey link through social media posts, university networks, and snowballing. Participants were not randomly selected from a master list; rather, quota targets ensured adequate numbers for each platform. Out of the final 228 respondents, 80 were from Instagram, 75 from Snapchat, and 73 from Twitter. Participants could use multiple platforms, with 60% reporting multi-platform usage.

Inclusion criteria required participants to be aged 23–41, reside in Egypt, and actively use at least one of the target platforms. Exclusion criteria were non-users or participants outside the age range. The minimum sample size was determined using the formula suggested by Nyalungwe et al. (62):n=(z2×p×(1−p))/e2=((1.96)2×0.5×(1−0.5))/(0.1)2≈96.04⇒n=97<228The achieved sample of 228 exceeded the minimum required for analysis. Self-selection bias may have influenced responses. A conventional response rate cannot be calculated due to the open distribution method, which is discussed as a limitation in Section 7.

A structured questionnaire was developed using validated scales from prior literature, with items adapted to the context of sports celebrity endorsements on social media. The questionnaire comprised four sections. The first collected demographic information, including age, gender, and social media usage patterns. The second assessed sports celebrity personality traits—likability, expertise, credibility, and physical attractiveness—using items adapted from McCormick (63). Items were adjusted to refer specifically to social media posts.

The third section measured perceived product quality, based on Akoglu and Özbek (64), with items reflecting consumer perceptions after exposure to social media content shared by sports celebrities. The fourth section focused on purchasing intentions, adapted from Priyankara et al. (65), capturing the likelihood of purchase following social media exposure. All measurement items, response anchors (1 = strongly disagree to 5 = strongly agree), and their sources are provided in Appendix A, with modifications clearly documented.

The quantitative survey was distributed via Google Forms over a one-month period using social media posts, university networks, and snowballing. Procedural remedies reduced common method bias, including anonymous responses, mixing item order, and separating predictors from outcomes.

Data were analyzed using SPSS for descriptive statistics and Partial Least Squares Structural Equation Modeling (PLS-SEM) to examine relationships among latent constructs. PLS-SEM was chosen over CB-SEM due to the moderate sample size (N = 228) and model complexity, which included multiple latent constructs, mediating paths, and multigroup comparisons. Reliability and validity were assessed using Cronbach's Alpha, Composite Reliability (CR), Average Variance Extracted (AVE), and factor loadings. Discriminant validity was evaluated with both Fornell-Larcker and HTMT methods. Multigroup analysis was conducted based on participants' primary platform (Instagram, Snapchat, Twitter), with the Welch–Satterthwaite test complementing PLS-MGA to identify significant differences in path coefficients across groups.

Semi-structured interviews were conducted with 12 participants purposively selected from the survey sample to ensure variation in gender, engagement level, and primary platform (four participants per platform). Participants' ages ranged from 25 to 32 years, reflecting the millennial target group. Data saturation was reached when no new themes emerged.

Interviews explored participants' perceptions of celebrity traits, platform-specific content, emotional reactions, and the influence of endorsements on perceived product quality and purchase intention. Questions covered both general impressions and platform-specific perceptions, including engagement with content, emotional responses, and judgments of authenticity, expertise, credibility, and attractiveness. Main topics and example prompts guided the discussion to ensure coverage of areas relevant to the hypotheses (see Appendix A2 for the full interview guide).

Transcripts were analyzed using thematic analysis (89). Open coding identified meaningful data segments, axial coding grouped codes into categories, and selective coding identified overarching themes. Coding reliability was ensured through careful review and discussion by the research team, with examples of how raw statements were transformed into codes, categories, and themes provided in Appendix A3 for methodological transparency.

Qualitative findings were explicitly integrated with SEM results to confirm and nuance specific relationships, providing richer context for platform-specific variations. Integration strengthened the mixed-methods interpretation.

Participant data were handled anonymously to ensure confidentiality. All participants received explanations of the study objectives, procedures, and voluntary participation requirements. Informed consent was obtained, and ethical approval was granted by the Institutional Review Board (IRB) at Badr University in Cairo (BUC).

To complement the quantitative analysis, qualitative data were gathered through semi-structured interviews with 12 millennial participants selected from the original survey pool. These interviews were designed to deepen understanding of how sports celebrity traits influence perceptions of product quality and purchase intention across Instagram, Snapchat, and Twitter. All interviews were transcribed verbatim and analyzed thematically, resulting in four key themes that support and extend the statistical findings.

This study advances theory-building by demonstrating that the influence of sports celebrity traits is platform-dependent rather than universal, thereby challenging earlier assumptions that endorsement strategies function similarly across social media environments. A key theoretical contribution lies in showing that traits such as likability (26, 27), expertise (82, 83), credibility (28, 84), and physical attractiveness (85, 86) exhibit varying levels of effectiveness depending on the platform through which endorsements are communicated. The qualitative findings reinforce this interpretation by revealing that Instagram's visual modality amplifies likability, Snapchat's informal and ephemeral environment heightens the importance of authenticity and relatability, and Twitter's text-driven format foregrounds expertise and informational value.

These findings highlight the need for future theoretical models of celebrity endorsement to explicitly incorporate platform context, including content modality, interaction norms, and platform-specific affordances. Rather than treating social media as a homogeneous communication environment, the results suggest that consumer evaluations are shaped by the functional purpose and interaction style of each platform, with users relying on different evaluative cues accordingly. This insight extends prior endorsement research by conceptualizing consumer engagement as a multi-platform and context-sensitive process.

In addition, this research extends existing frameworks such as the Cognitive Response Model (31, 87) by identifying perceived product quality as a central mechanism linking celebrity traits to purchasing intentions. While earlier studies acknowledged cognitive evaluations in endorsement effectiveness, they often paid limited attention to how product quality perceptions emerge from trait-based cues. The present findings, supported by both SEM results and qualitative insights, indicate that perceived quality integrates emotional responses (e.g., likability), trust-based judgments (e.g., credibility), and competence-based assessments (e.g., expertise), thereby offering a more comprehensive explanation of endorsement influence.

Finally, the study addresses a key limitation in prior literature that frequently treated social media platforms as interchangeable endorsement contexts. The findings demonstrate that millennials engage with platforms in distinct ways, seeking inspirational and lifestyle-oriented content on Instagram, spontaneous and authentic moments on Snapchat, and informational or expertise-driven cues on Twitter. These patterns underscore the importance of incorporating platform-specific user expectations and evaluative pathways into future theoretical development, particularly with respect to the balance between emotional and cognitive processing in digital persuasion.

From a managerial perspective, this study offers clear, evidence-based guidance for designing more effective sports celebrity endorsement strategies across social media platforms. A central practical contribution lies in demonstrating that endorsement success depends not only on selecting the “right” celebrity, but on aligning specific celebrity traits with the functional logic of each platform.

Likability emerged as a consistent driver of perceived product quality and purchasing intentions, particularly on visually oriented platforms. Brands targeting millennial consumers should therefore prioritize endorsers who convey warmth, relatability, and authenticity rather than relying solely on fame or performance credentials. Content strategies that showcase personal narratives, humor, and everyday moments are especially effective on Instagram and Snapchat, where emotional connection plays a dominant role. Interview insights confirm that behind-the-scenes and lifestyle-oriented posts increase trust and perceived realism, thereby strengthening endorsement impact.

Credibility represents another critical managerial consideration, particularly in environments where consumers actively assess authenticity. The findings indicate that frequent brand switching or inconsistent endorsements erode consumer trust, suggesting that long-term partnerships and visible product usage are more effective than short-term promotional campaigns. Snapchat's real-time and informal nature appears especially effective for signaling authenticity, making it a valuable channel for reinforcing credibility through unscripted or spontaneous content.

Expertise was shown to be particularly influential in shaping perceptions of product quality and purchase confidence, especially on Twitter. For technical, performance-related, or high-involvement products, brands should collaborate with athletes whose experience and knowledge are clearly relevant to the offering. The results suggest that detailed explanations, personal usage accounts, and informational content outperform image-based appeals in text-driven environments, reinforcing the need for message–platform congruence.

Physical attractiveness, while still relevant, was found to be insufficient as a standalone strategy. Its influence was strongest when combined with perceived authenticity, relevance, and demonstrated product fit, particularly in categories such as fitness, health, and wellness. This finding cautions managers against overreliance on aesthetic appeal and highlights the importance of holistic endorser–brand alignment.

Finally, a major practical contribution of this research is the demonstration that each platform requires a distinct endorsement strategy rather than content replication across channels. Instagram is best suited for emotionally engaging, visually rich storytelling; Twitter is optimal for expertise-driven, informational messaging; and Snapchat facilitates rapid trust-building through informal and authentic interactions. Marketers who adapt endorsement content, tone, and format to these platform-specific affordances are more likely to enhance perceived product quality and purchasing intentions among millennial consumers.