Table of Contents
Type: Dissertation | Subject: Marketing | Level: Masters | Word Count: ~9000 words
This model dissertation was produced by an Essays UK specialist as reference material for learning purposes only. For support in this field, see our marketing dissertation support.
Produce a Masters-level dissertation of approximately 9,000 words investigating how perceived social media influencer authenticity shapes purchase intention among Gen Z consumers. The study should develop a conceptual model grounded in relevant theory, collect and analyse primary quantitative data, and conclude with practical recommendations for marketers.
Social media influencers now occupy a central position in brand communication strategies aimed at Generation Z, yet this cohort is widely reported to be sceptical of overt commercial persuasion. This dissertation investigates the relationship between perceived influencer authenticity and purchase intention among UK-based Gen Z consumers (aged 18–24), and examines whether trust mediates this relationship and whether product involvement moderates its strength. Grounded in authenticity theory, parasocial interaction theory and the elaboration likelihood model, a conceptual framework was developed proposing that authenticity influences purchase intention both directly and indirectly through consumer trust in the influencer. A quantitative, cross-sectional survey design was adopted, and data were collected from 245 Gen Z social media users via an online questionnaire using validated multi-item scales. Data were analysed using descriptive statistics, Pearson correlation, hierarchical multiple regression and the Hayes PROCESS macro. Findings indicate that perceived authenticity is a significant positive predictor of purchase intention (β = .49, p < .001), and that trust partially mediates this relationship (indirect effect = .21, 95% CI [.14, .29]). Product involvement significantly moderates the authenticity–purchase intention relationship, with the effect stronger for high-involvement product categories. The study contributes an integrated, empirically tested model of influencer-driven purchase intention specific to Gen Z, and offers practical guidance for marketers on selecting and briefing influencers to maximise perceived authenticity and downstream purchase behaviour.
Influencer marketing has grown from a peripheral tactic into a mainstream pillar of digital marketing budgets, with UK advertisers increasingly reallocating spend from traditional celebrity endorsement towards creators who command smaller but highly engaged social media followings (De Veirman, Cauberghe and Hudders, 2017). This growth has been driven in large part by the purchasing power and media habits of Generation Z, broadly defined as individuals born between the mid-1990s and early 2010s, who have grown up as digital natives immersed in platforms such as Instagram, TikTok and YouTube (Djafarova and Bowes, 2021).
Unlike previous cohorts, Gen Z consumers are frequently characterised as highly attuned to inauthenticity in marketing communications, having grown up alongside the platforms that popularised influencer culture and, consequently, alongside its commercial excesses (Schouten, Janssen and Verspaget, 2020). Scepticism towards overtly sponsored content, concern about undisclosed partnerships, and a preference for creators perceived as “real” rather than manufactured have all been documented as defining features of Gen Z’s relationship with influencer marketing (Kapitan and Silvera, 2016; Belanche et al., 2021). Authenticity has therefore emerged as a construct of central theoretical and managerial interest: brands that partner with influencers perceived as inauthentic risk not only wasted spend but active consumer backlash (Audrezet, de Kerviler and Guidry Moulard, 2020).
At the same time, the mechanisms through which authenticity translates into commercially meaningful outcomes, particularly purchase intention, remain incompletely understood. Existing literature has tended either to examine authenticity as an antecedent of engagement metrics such as likes and comments (Munnukka, Uusitalo and Toivonen, 2016), or to study purchase intention as an outcome of source credibility more broadly, without isolating authenticity as a distinct construct (Lou and Yuan, 2019). Fewer studies have tested an integrated model in which authenticity operates through relational mechanisms such as trust and parasocial interaction to shape purchase intention specifically within a Gen Z sample.
The commercial stakes of resolving this gap are considerable. Influencer marketing spend in the UK has grown steadily as brands shift budget away from broadcast media towards creator-led content, and Gen Z, who by 2026 constitute a growing share of first-time and habitual online spenders, are disproportionately reached through this channel rather than through television or print (De Veirman, Cauberghe and Hudders, 2017). Yet growth in spend has not been matched by an equivalent growth in reliable measurement of what actually drives purchase behaviour within this channel, as opposed to attention or reach alone, leaving a persistent gap between how influencer campaigns are bought and how they should, in principle, be evaluated.
This gap is particularly consequential given Gen Z’s growing discretionary spending power. As the cohort has moved into full-time employment and independent living over the course of the 2020s, its aggregate purchasing influence, both direct and through household spending decisions it shapes, has grown steadily, and marketers across sectors from fast fashion to financial services have identified this cohort as a strategic priority for the remainder of the decade (Francis and Hoefel, 2018). Brands that can reliably identify which influencer characteristics convert attention into purchase intention among this specific audience therefore stand to gain a durable competitive advantage over those relying on legacy metrics such as reach and impressions alone.
Marketing practitioners lack a clear, empirically grounded understanding of how perceived influencer authenticity converts into purchase intention among Gen Z consumers, and of the boundary conditions under which this relationship is strongest. Without this understanding, influencer marketing budgets risk being allocated on the basis of follower counts or engagement rates that do not reliably predict downstream purchase behaviour, while genuinely authentic partnerships may be under-leveraged.
This problem is compounded by what might be termed an “authenticity paradox” in commercial influencer content: the more an influencer’s income depends on brand partnerships, the harder authenticity becomes to sustain in the eyes of a sceptical audience, yet brands continue to require exactly this endorsement function to justify their investment (Audrezet, de Kerviler and Guidry Moulard, 2020). Understanding precisely which mechanisms allow some sponsored content to retain perceived authenticity, and precisely how that authenticity converts into purchase intention rather than mere engagement, is therefore both a theoretical and a practical priority.
The aim of this dissertation is to investigate the relationship between Gen Z consumers’ perceptions of social media influencer authenticity and their purchase intention, and to examine the mediating role of trust and the moderating role of product involvement in this relationship.
The specific objectives are:
Academically, the study contributes an integrated, empirically tested model that positions authenticity as a distal driver of purchase intention operating through trust, addressing calls in the literature for more mechanism-focused research on influencer marketing effects (Audrezet, de Kerviler and Guidry Moulard, 2020). Practically, the findings offer marketers and brand managers a more defensible basis for influencer selection, content briefing and campaign evaluation, moving the conversation beyond reach and engagement metrics towards the psychological drivers of purchase behaviour. Given the growing regulatory attention on influencer disclosure practices from the UK’s Advertising Standards Authority and Competition and Markets Authority, a clearer evidence base on what genuinely drives Gen Z purchase intention, as opposed to what merely drives visibility, is also of value to brands seeking to invest responsibly and effectively in this channel. For a service such as essays.uk, this dissertation also functions as a model of how a well-scoped, quantitative Masters-level marketing dissertation should be structured, referenced and analysed.
Chapter 2 reviews the literature on influencer marketing, authenticity, parasocial interaction, Gen Z consumer behaviour and purchase intention, culminating in a conceptual framework and hypotheses. Chapter 3 details the methodology, including research philosophy, design, sampling, instrumentation and ethics. Chapter 4 presents the findings of the quantitative analysis. Chapter 5 discusses these findings in relation to the existing literature. Chapter 6 concludes with a summary of contributions, practical recommendations, limitations and directions for future research.
Influencer marketing can be defined as a form of marketing communication in which brands collaborate with individuals who have accumulated a following on social media to promote products or services, leveraging the perceived credibility and relatability of the influencer relative to traditional celebrity endorsement (Lou and Yuan, 2019). De Veirman, Cauberghe and Hudders (2017) argue that influencers occupy a distinctive position between celebrity endorsers and word-of-mouth referrals, combining the reach of the former with the perceived trustworthiness of the latter. This hybrid positioning is theorised to explain much of influencer marketing’s effectiveness, but also its fragility: because influencer credibility rests heavily on perceived independence from commercial interests, excessive or poorly disclosed sponsorship can rapidly erode the very trust that made the format effective (Boerman, Willemsen and Van Der Aa, 2017).
The rise of “micro” and “nano” influencers, typically defined as those with follower counts below 100,000 and 10,000 respectively, has been linked to consumer preferences for perceived authenticity over reach. Studies comparing engagement rates across follower tiers consistently find that smaller-following creators generate proportionally higher engagement, attributed to closer, more reciprocal relationships with followers (Munnukka, Uusitalo and Toivonen, 2016; Kay, Mulcahy and Parkinson, 2020), building on earlier work showing that peer-like social media endorsers can outperform traditional celebrity sources on measures of perceived trustworthiness (Colliander and Dahlén, 2011). This trend is theoretically significant because it suggests that scale and authenticity may be, at least partially, in tension.
Platform context further shapes how influencer marketing operates. Instagram, with its curated, visually polished aesthetic, has historically rewarded a degree of production quality that can sit uneasily with perceptions of authenticity, whereas TikTok’s algorithmic emphasis on raw, unedited, vertical video content has been associated with lower perceived commercial intent and higher perceived spontaneity among younger users (Belanche et al., 2021). Because Gen Z consumers are heavy users of both platforms, often for different purposes, authenticity perceptions cannot be assumed to be platform-invariant, and the present study’s stimulus material was therefore designed to reflect conventions common to both environments rather than being tied to a single platform’s aesthetic norms.
Regulatory context has also shaped how influencer marketing is practised in the UK specifically. The Advertising Standards Authority and the Competition and Markets Authority have progressively tightened enforcement of sponsorship disclosure requirements over the past decade, requiring clear and unambiguous labelling of paid partnerships (for example “#ad” placed prominently rather than buried among unrelated hashtags), and non-compliance has become a recurrent source of negative press coverage for both influencers and the brands they represent (Boerman, Willemsen and Van Der Aa, 2017). For a Gen Z audience already primed to scrutinise sponsored content, awareness of these regulatory norms arguably raises the baseline expectation that disclosure will occur, making its absence a stronger negative authenticity signal than it might have been in the earlier, less regulated years of influencer marketing.
Authenticity is a contested construct across marketing, sociology and philosophy, but within influencer marketing it is generally understood as the degree to which a consumer perceives an influencer’s persona, opinions and endorsements to be genuine, consistent and free from purely commercial motivation (Moulard, Garrity and Rice, 2015). Moulard, Garrity and Rice (2015) developed a validated multidimensional authenticity scale capturing perceived sincerity, expertise and passion, which has since been widely adapted in influencer marketing research.
Audrezet, de Kerviler and Guidry Moulard (2020) distinguish between “passionate” authenticity, rooted in perceived genuine enthusiasm for a product category, and “transparent” authenticity, rooted in honest disclosure of commercial relationships. Their qualitative work with influencers suggests that both dimensions are actively managed by content creators as a form of “authenticity work,” complicating any simplistic view of authenticity as a fixed trait rather than a negotiated performance. Similarly, Kapitan and Silvera (2016) find that “authentically” self-promoting influencers, who moderate self-promotional claims with perceived self-deprecation or balance, are rated as more credible and generate stronger purchase intentions than those who promote unreservedly.
Disclosure of sponsored content has received particular attention as a boundary condition of authenticity perceptions. Boerman, Willemsen and Van Der Aa (2017) demonstrate experimentally that disclosure can reduce perceived authenticity and increase recognition of persuasive intent, yet other studies find that appropriately framed disclosure, especially when paired with perceived genuine product fit, does not necessarily damage, and can even enhance, trust (Colliander and Erlandsson, 2015; Boerman, 2020). This apparent contradiction suggests that the effect of disclosure on authenticity is conditional on how it interacts with pre-existing source credibility and content quality, rather than operating as a uniform penalty.
A related theoretical thread draws on persuasion knowledge theory, which proposes that consumers develop cognitive schemas over time for recognising and discounting persuasive attempts (Boerman, Willemsen and Van Der Aa, 2017). Because Gen Z has been exposed to influencer marketing for effectively their entire adolescent and adult lives, their persuasion knowledge relating specifically to this format is likely to be more developed than that of older cohorts who encountered it later in life, which in turn raises the evidentiary bar an influencer must clear before content is perceived as authentic rather than as a recognisable, and therefore partially discounted, sales pitch. This literature collectively suggests that authenticity is not a fixed attribute of an influencer but a dynamic judgement formed by the audience in relation to content style, disclosure practice and category fit, a view that informs how authenticity is operationalised and measured in the present study.
It is also worth distinguishing influencer authenticity from the related but conceptually separate notion of brand authenticity, which concerns consumer perceptions of a brand’s own heritage, sincerity and consistency (Moulard, Garrity and Rice, 2015). While the two constructs are theoretically related, in that an influencer perceived as inauthentic may transfer negative authenticity judgements onto the endorsed brand, this dissertation focuses specifically on influencer-level authenticity as the independent variable, consistent with the majority of the empirical literature reviewed above.
Parasocial interaction (PSI) theory, originally developed by Horton and Wohl (1956) to describe the one-sided relationships audiences form with media personae, has been extensively applied to influencer marketing to explain why followers develop feelings of intimacy and friendship with creators they have never met (Chung and Cho, 2017). Rubin, Perse and Powell (1985) developed a validated PSI scale measuring the degree to which audiences perceive a mediated relationship as reciprocal and personally meaningful; this scale has been repeatedly adapted for social media contexts (Sokolova and Kefi, 2020).
Trust, meanwhile, is generally conceptualised in the marketing literature as a willingness to rely on an exchange partner in whom one has confidence (Doney and Cannon, 1997), and within influencer marketing specifically as the belief that an influencer’s recommendations are honest and given with the follower’s interests, not just commercial gain, in mind (Djafarova and Rushworth, 2017). Several studies position trust as the proximal mechanism through which more distal constructs such as authenticity and parasocial interaction exert their effects on behavioural outcomes. Breves et al. (2019), for example, find that the perceived fit between an influencer and an endorsed brand predicts purchase intention primarily via trust, while Sokolova and Kefi (2020) show that both perceived expertise and parasocial interaction predict purchase intention through trust as a mediator. This body of work supports treating trust as a mediating rather than merely correlated variable in models of influencer effectiveness.
Relationship marketing theory offers a further lens on why trust should mediate rather than simply co-occur with authenticity effects. Within this tradition, trust is understood as an accumulated relational asset, built incrementally through repeated, consistent interactions, that lowers the perceived risk of a subsequent transactional decision (Doney and Cannon, 1997). Applied to the influencer context, a single authentic-seeming post is unlikely to generate trust in isolation; rather, perceived authenticity across a pattern of content over time is theorised to accumulate into a durable trust judgement that then lowers the psychological barrier to acting on any single specific recommendation, including a purchase (Sokolova and Kefi, 2020). This has an important implication for the present study’s cross-sectional design, discussed further in Chapter 3, since the trust measure captures an accumulated relational state rather than a reaction to the single stimulus post alone.
Trust is also not immune to damage, and the literature on trust repair suggests that transgressions such as undisclosed sponsorship, factual misrepresentation of a product, or a mismatch between an influencer’s stated values and their endorsement choices can rapidly erode previously accumulated trust, with recovery, where it occurs at all, requiring sustained and visible corrective behaviour (Boerman, 2020). This asymmetry, whereby trust is slow to build and fast to lose, is consistent with Gen Z’s documented vigilance for inconsistency in influencer behaviour (Chapple and Cownie, 2017) and underscores why authenticity, as a signal that helps consumers judge the likelihood of future trustworthy behaviour, carries such weight within this specific audience.
Gen Z is frequently distinguished from Millennials by its members’ status as true digital natives, having never known a media environment without smartphones and social media (Djafarova and Bowes, 2021). Francis and Hoefel (2018) characterise Gen Z consumers as pragmatic, value-conscious and highly attentive to social and ethical signalling from brands, while also being more sceptical of traditional advertising and more reliant on peer and influencer recommendation as a trusted information source. Schouten, Janssen and Verspaget (2020) find that Gen Z respondents rate “ordinary” influencers as more credible and effective endorsers than celebrities specifically because ordinariness is read as a proxy for authenticity, a finding with direct relevance to the present study.
Chapple and Cownie (2017) similarly report that younger consumers are more likely than older cohorts to identify sponsored content, and that this recognition tends to reduce perceived credibility unless the influencer is judged to have genuine expertise or a pre-existing relationship with the product category. Taken together, this literature suggests that Gen Z’s scepticism does not eliminate the effectiveness of influencer marketing, but it does raise the threshold of perceived authenticity required for such marketing to succeed, making the present study’s focus on authenticity, rather than exposure or reach alone, theoretically and practically apt.
Beyond scepticism, Gen Z is also distinguished by a documented preference for brands and creators who visibly align with social and ethical values, including sustainability, inclusivity and mental health awareness, with authenticity judgements increasingly bound up with perceived value alignment rather than product quality alone (Francis and Hoefel, 2018). Algorithmic platform dynamics reinforce this cohort’s distinctiveness: TikTok’s “For You” feed surfaces content based on engagement signals rather than follower relationships, meaning Gen Z users are routinely exposed to influencers with whom they have no prior parasocial history, placing greater immediate weight on in-content authenticity cues, since the accumulated relational trust described in Section 2.3 has not yet had time to develop for many of the creators a given user encounters (Belanche et al., 2021). By contrast, Instagram’s more relationship-based feed structure, which continues to prioritise accounts a user already follows, may allow trust to accumulate more gradually across repeated exposures to the same small set of creators, a structural difference between the two platforms that is theoretically relevant even where, as reported in Chapter 4, it did not translate into a significant difference in mean authenticity ratings within this particular sample and stimulus category. This algorithmic novelty effect is a further reason to expect authenticity, as an immediately perceivable content characteristic, to carry particular explanatory weight for this cohort relative to older audiences who may rely more heavily on established parasocial relationships with a smaller, more stable set of creators.
Purchase intention is commonly operationalised as a consumer’s self-reported likelihood of purchasing a product or service in the near future, and is widely used in marketing research as a proxy for actual purchase behaviour, following the theory of planned behaviour’s proposition that intention is the most proximal antecedent of behaviour (Ajzen, 1991; Baker and Churchill, 1977). Within influencer marketing, purchase intention has been linked to source credibility dimensions including expertise, trustworthiness and attractiveness (Lou and Yuan, 2019), to perceived homophily or similarity between influencer and follower (Jin and Muqaddam, 2019), and to product involvement, understood as the perceived personal relevance or risk associated with a purchase decision (Zaichkowsky, 1985).
Product involvement is theorised within the elaboration likelihood model (ELM) as a determinant of processing route: under high involvement, consumers are expected to engage in more effortful, central-route processing of message content, weighing source characteristics such as authenticity more heavily, whereas under low involvement, peripheral cues may dominate and source authenticity may matter comparatively less (Petty and Cacioppo, 1986; Xiao, Wang and Chan-Olmsted, 2018). This theoretical logic underpins the present study’s proposition that product involvement moderates the strength of the authenticity–purchase intention relationship.
Purchase intention should also be understood within the broader purchase funnel, sitting downstream of awareness, interest and attitude formation but upstream of actual behaviour, and it is at this specific point in the funnel that influencer content has been argued to exert its most direct commercial effect, since a follower who has already formed a favourable attitude toward an influencer is more readily moved toward a concrete behavioural intention by a well-placed, credible recommendation than by advertising exposure alone (Jin and Muqaddam, 2019). Gen Z’s documented propensity for impulse purchasing following social media exposure, particularly within fashion and beauty categories, further suggests that the gap between intention and behaviour may be narrower for this cohort than for older consumers, strengthening the practical relevance of purchase intention as an outcome measure in this study (Djafarova and Bowes, 2021).
It should be acknowledged that self-reported purchase intention, while the standard and most defensible outcome measure available within the resource constraints of a Masters dissertation, remains an imperfect proxy for actual purchasing behaviour, since intention-behaviour gaps of varying size are well documented across consumer behaviour research more broadly (Ajzen, 1991). This limitation, and the ways future research might address it through behavioural rather than self-reported outcome measures, is returned to explicitly in Chapter 6.
Drawing on authenticity theory, parasocial interaction theory, source credibility theory and the elaboration likelihood model, this dissertation proposes a conceptual model in which perceived influencer authenticity predicts purchase intention both directly and indirectly through consumer trust, with the overall relationship moderated by product involvement. Figure 1 in Chapter 4 visualises the pattern predicted by this model; conceptually, the framework corresponds to a moderated mediation design, in which trust operates as the mediating mechanism (the “how”) and product involvement operates as the boundary condition (the “when”) governing the overall strength of the authenticity–purchase intention relationship. The following hypotheses are tested:
This framework addresses a gap identified across the reviewed literature: while authenticity, trust and parasocial interaction have each been studied in relation to influencer marketing outcomes, few studies integrate them into a single tested model specific to Gen Z, and fewer still examine product involvement as a boundary condition.
Collectively, the literature reviewed in this chapter establishes four things: first, that influencer marketing effectiveness rests heavily on perceived source characteristics rather than reach alone; second, that authenticity is a multidimensional, actively managed construct rather than a fixed trait, and one to which Gen Z is demonstrably sensitive; third, that trust is theoretically and empirically well positioned as a proximal mechanism linking authenticity to behavioural outcomes; and fourth, that product involvement, via the elaboration likelihood model, offers a plausible and under-tested boundary condition on the strength of these effects. No single study identified in this review tests all of these elements together within a Gen Z-specific sample, which constitutes the empirical gap this dissertation addresses. Chapter 3 sets out the methodology used to test the resulting model.
This study adopts a positivist research philosophy, premised on the assumption that an objective social reality exists and can be measured through observable, quantifiable indicators (Saunders, Lewis and Thornhill, 2019). Positivism was judged appropriate given the study’s aim of testing hypothesised relationships between defined constructs (authenticity, trust, involvement, purchase intention) using validated measurement scales, an approach consistent with the deductive, theory-testing logic pursued throughout the dissertation. An interpretivist approach, relying on qualitative interviews or focus groups to build rich, contextual understanding of individual authenticity judgements, was considered but ultimately rejected, since the study’s objectives centred on testing the strength, direction and boundary conditions of pre-specified relationships across a reasonably large sample, a task for which a positivist, quantitative approach is better suited (Saunders, Lewis and Thornhill, 2019).
A deductive approach was adopted, with hypotheses derived from existing theory (Chapter 2) and subsequently tested against primary data. A quantitative, cross-sectional survey design was selected as the most efficient means of collecting standardised data on perceptions and intentions from a reasonably large sample within the time constraints of a Masters dissertation (Bryman, 2016; Creswell and Creswell, 2018). While a cross-sectional design cannot establish causality with the same confidence as an experimental design, it is well suited to testing the strength and pattern of association among variables, which is the primary aim of this study.
The target population was UK-based Gen Z social media users aged 18–24 who follow at least one social media influencer on Instagram or TikTok. A non-probability purposive sampling strategy was used, with the survey distributed via university mailing lists, student society social media accounts and the Prolific online panel, screening for the target age range and UK residency. A total of 268 responses were received; after removing 23 incomplete or failed-attention-check responses, 245 valid responses were retained for analysis (effective response rate 91.4%). This sample size exceeds the minimum recommended for multiple regression with four predictors using Green’s (1991) rule of thumb (N > 50 + 8m), and is broadly consistent with sample sizes used in comparable published studies (e.g. Sokolova and Kefi, 2020; Breves et al., 2019). Soft quotas were applied during recruitment to ensure a reasonable spread across the 18–24 age range and across gender, avoiding a sample skewed heavily toward a single year group or a single university course, while acknowledging that a fully probability-based sampling frame for this population was not practically achievable within the resources available to a Masters dissertation.
All constructs were measured using previously validated multi-item scales, adapted to the influencer marketing context and anchored on a five-point Likert scale (1 = strongly disagree, 5 = strongly agree). Perceived authenticity was measured using a 6-item adaptation of Moulard, Garrity and Rice’s (2015) authenticity scale. Parasocial interaction was measured using a 5-item adaptation of Rubin, Perse and Powell’s (1985) PSI scale. Trust was measured using a 5-item adaptation of Doney and Cannon’s (1997) trust scale. Purchase intention was measured using a 4-item adaptation of Baker and Churchill’s (1977) purchase intention scale. Product involvement was measured using a 5-item adaptation of Zaichkowsky’s (1985) Personal Involvement Inventory, applied to the product category featured in a stimulus Instagram post shown to respondents prior to the questionnaire. Age, gender and weekly hours of social media use were collected as control variables. A pilot study with 18 participants drawn from the target population was conducted prior to full data collection to check item clarity, questionnaire length and face validity of the stimulus material; minor wording adjustments were made to two authenticity items that pilot respondents found ambiguous, and the pilot data were excluded from the final analysis.
Participants were first shown a short stimulus consisting of a realistic, de-identified Instagram post from a fictional micro-influencer promoting a skincare product, chosen because skincare represents a product category with meaningful variance in perceived personal involvement across respondents: some respondents were expected to regard skincare as a low-risk, habitual purchase, while others, particularly those managing specific skin concerns, were expected to regard it as a higher-risk, higher-involvement decision, generating the natural variance in the involvement moderator required to test H3 without needing to recruit separate high- and low-involvement samples. To manipulate the moderator variable indirectly rather than experimentally, product involvement was measured rather than manipulated, capturing each respondent’s own naturally occurring level of personal involvement with the featured product category, consistent with the correlational design of the study as a whole. Participants then completed the questionnaire, which took approximately 8–10 minutes. The survey was hosted on Qualtrics and administered between February and March 2026. An attention-check item (“Please select ‘strongly agree’ for this item”) was embedded midway through the questionnaire to screen for inattentive responding.
Data were analysed using IBM SPSS Statistics version 28. Following data cleaning, reliability of each scale was assessed using Cronbach’s alpha. Descriptive statistics (means, standard deviations) were computed for all constructs. Pearson correlation coefficients were calculated to examine bivariate relationships. Hypotheses were tested using hierarchical multiple regression, entering control variables in Step 1 and perceived authenticity in Step 2. Mediation by trust (H2) and moderation by product involvement (H3) were tested using Hayes’ (2018) PROCESS macro for SPSS, Models 4 and 1 respectively, with 5,000-sample bootstrapping used to generate 95% bias-corrected confidence intervals for indirect effects. Prior to hypothesis testing, standard regression assumptions were checked: normality of residuals was assessed via inspection of Q-Q plots and skewness/kurtosis statistics, homoscedasticity via inspection of standardised residual scatterplots, and multicollinearity via variance inflation factors (VIF), with all values found to fall within acceptable thresholds (VIF < 2 for all predictors, well below the conventional cut-off of 10), supporting the validity of the subsequent regression and mediation analyses.
The study received ethical approval from the researcher’s institutional ethics committee prior to data collection. Participants were presented with an information sheet outlining the study’s purpose, their right to withdraw at any point without penalty, and the anonymity and confidentiality of their responses, and were required to provide informed consent before proceeding. No personally identifying information was collected; data were stored on a password-protected, encrypted university server in accordance with UK GDPR requirements. Prolific participants were compensated at a rate consistent with the platform’s minimum hourly wage guidance. Because the stimulus material referenced a fictional influencer and product, care was taken in the information sheet to clarify that the post was created for research purposes only, avoiding any risk of participants forming a mistaken belief about a real commercial relationship, and the debrief screen at the end of the survey reiterated this point alongside contact details for the researcher and the ethics committee should participants have any concerns following participation.
The cross-sectional design limits causal inference; while the hypothesised model is theoretically directional, reverse or reciprocal relationships (for example, prior purchase intention shaping subsequent perceptions of authenticity) cannot be ruled out. The use of a single stimulus post and product category, while methodologically necessary for control, may limit generalisability to other influencer types and product categories, a limitation addressed further in Chapter 6.
Of the 245 valid respondents, 61.2% identified as female, 36.7% as male and 2.1% as non-binary or preferred not to say. The mean age was 20.8 years (SD = 1.9), consistent with the target 18–24 range. Respondents reported a mean of 3.4 hours (SD = 1.6) of daily social media use, and 88.6% reported following at least three influencers across Instagram and TikTok combined, confirming the sample’s suitability for the study’s purposes. Instagram was reported as the primary platform for following influencers by 54.3% of respondents, TikTok by 39.6%, and YouTube by the remaining 6.1%, indicating that both major platforms named in the literature review were well represented within the sample rather than the data being dominated by users of a single platform. An exploratory independent-samples t-test comparing mean authenticity scores between primary Instagram users and primary TikTok users found no statistically significant difference (t(228) = 1.34, p = .181), suggesting that, at least for the skincare stimulus used in this study, perceived authenticity was not strongly platform-dependent within this sample, a finding that tempers, without fully resolving, the platform-contingency argument raised in Section 2.1.
Table 4.1 presents means, standard deviations and Cronbach’s alpha reliability coefficients for each construct. All scales exceeded the conventional 0.70 reliability threshold (Nunnally, 1978), supporting their internal consistency within this sample. Mean scores across all four core constructs clustered around the midpoint of the five-point scale (3.12 to 3.51), indicating that respondents held moderately positive but not uniformly extreme perceptions of the stimulus influencer, which is desirable from a measurement perspective since it suggests sufficient variance in responses to support meaningful correlation and regression analysis rather than a ceiling or floor effect.
| Construct | Items | Mean | SD | Cronbach’s α |
|---|---|---|---|---|
| Perceived Authenticity | 6 | 3.51 | 0.79 | .86 |
| Parasocial Interaction | 5 | 3.12 | 0.88 | .83 |
| Trust | 5 | 3.28 | 0.81 | .85 |
| Product Involvement | 5 | 3.45 | 0.92 | .81 |
| Purchase Intention | 4 | 3.42 | 0.81 | .88 |
Table 4.2 presents Pearson correlation coefficients among the study’s continuous variables. All hypothesised relationships were significant and in the expected direction. Perceived authenticity was significantly positively correlated with purchase intention (r = .52, p < .001), with trust (r = .61, p < .001), and with parasocial interaction (r = .47, p < .001). Trust was significantly correlated with purchase intention (r = .58, p < .001). Product involvement showed a smaller but still significant correlation with purchase intention (r = .24, p < .001) and with authenticity (r = .18, p < .001), consistent with its hypothesised role as a moderator rather than a strong direct predictor. None of the correlation coefficients among predictor variables exceeded .65, indicating that multicollinearity was not a concern for the subsequent regression analysis, in line with the variance inflation factor checks reported in Section 3.6.
| Variable | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| 1. Authenticity | — | ||||
| 2. Parasocial Interaction | .47** | — | |||
| 3. Trust | .61** | .44** | — | ||
| 4. Product Involvement | .18** | .15* | .21** | — | |
| 5. Purchase Intention | .52** | .39** | .58** | .24** | — |
*p < .05, **p < .001, N = 245.
A two-step hierarchical multiple regression was conducted with purchase intention as the dependent variable. Step 1 entered the control variables (age, gender, daily social media hours), which together explained 6.1% of variance in purchase intention (R² = .061, F(3, 241) = 5.23, p = .002). Step 2 added perceived authenticity, which explained a further 24.3% of variance (ΔR² = .243, ΔF(1, 240) = 90.71, p < .001), for a total model R² of .304 (F(4, 240) = 26.14, p < .001). In the final model, perceived authenticity was a significant positive predictor of purchase intention (β = .49, t = 9.52, p < .001), providing strong support for H1. Notably, the standardised effect of daily social media hours dropped from β = .17 in Step 1 to β = .11 once authenticity was entered in Step 2, suggesting that part of the apparent effect of general social media usage on purchase intention is, in fact, attributable to more heavily engaged users encountering, and rating, content they perceive as more authentic.
| Predictor | Step 1 β | Step 2 β |
|---|---|---|
| Age | .08 | .04 |
| Gender (female = 1) | .14* | .09 |
| Daily social media hours | .17** | .11* |
| Perceived authenticity | — | .49** |
| R² | .061 | .304 |
| ΔR² | — | .243** |
*p < .05, **p < .001, N = 245. Dependent variable: Purchase Intention.
Mediation was tested using Hayes’ (2018) PROCESS Model 4, with perceived authenticity as the independent variable, trust as the mediator, and purchase intention as the dependent variable, controlling for age, gender and social media hours. The indirect effect of authenticity on purchase intention through trust was significant (ab = .21, SE = .04, 95% BCa CI [.14, .29]), and the direct effect remained significant (c′ = .28, t = 5.41, p < .001), indicating partial rather than full mediation. This supports H2: trust partially mediates the relationship between perceived authenticity and purchase intention.
Moderation was tested using PROCESS Model 1, with product involvement as the moderator of the authenticity–purchase intention relationship. The interaction term (authenticity × involvement) was a significant predictor of purchase intention (β = .15, t = 2.53, p = .012), indicating that product involvement significantly moderates the relationship. Simple slopes analysis showed that the authenticity–purchase intention relationship was stronger at high levels of involvement (b = .61, p < .001) than at low levels of involvement (b = .34, p < .001), though both remained significant. This supports H3. Figure 1 illustrates the pattern of mean purchase intention scores across low, medium and high authenticity terciles, split by involvement level; the widening gap between the high- and low-involvement bars moving from the low to the high authenticity tercile visually reflects the significant interaction reported above. Table 4.4 summarises the outcome of hypothesis testing across the three hypotheses set out in Chapter 2.
| Hypothesis | Statistical Test | Result | Decision |
|---|---|---|---|
| H1: Authenticity → Purchase Intention | Hierarchical regression | β = .49, p < .001 | Supported |
| H2: Trust mediates Authenticity → Purchase Intention | PROCESS Model 4 | ab = .21, 95% CI [.14, .29] | Supported (partial mediation) |
| H3: Involvement moderates Authenticity → Purchase Intention | PROCESS Model 1 | Interaction β = .15, p = .012 | Supported |
Figure 1: Mean purchase intention by perceived authenticity tercile, split by product involvement level.
The finding that perceived authenticity is a strong, significant predictor of purchase intention (β = .49) among this Gen Z sample corroborates and extends prior work by Kapitan and Silvera (2016) and Lou and Yuan (2019), who identify credibility-related source characteristics as key drivers of influencer marketing effectiveness. The effect size observed here, explaining an additional 24.3% of variance beyond demographic and usage controls, is somewhat larger than that reported in comparable studies of mixed-age samples (e.g. Sokolova and Kefi, 2020, who report ΔR² of approximately .17 for trust alone), lending support to the argument that Gen Z’s heightened sensitivity to authenticity, as characterised by Schouten, Janssen and Verspaget (2020), makes this construct a particularly powerful lever within this cohort specifically. This pattern is also consistent with the persuasion knowledge argument advanced in Section 2.2: because Gen Z has more developed schemas for recognising and discounting standard promotional appeals, the marginal value of content that successfully signals authenticity, and thereby evades this discounting process, may be correspondingly larger than for cohorts with less developed persuasion knowledge specific to social media advertising.
The relatively modest, though still significant, direct effects of control variables (age, gender, daily social media hours) further suggest that demographic and usage-based targeting alone, the default approach in much influencer campaign planning, captures only a small fraction of the variance that authenticity perceptions explain, reinforcing the argument made in Chapter 1 that follower-count- and reach-based influencer selection criteria are a poor proxy for the psychological drivers of purchase intention.
The absence of a significant platform difference in mean authenticity ratings (Section 4.1) is a further point of interest. Given the platform-contingency argument developed in Section 2.1, a stronger a priori expectation might have been that TikTok content, with its lower production values and algorithmically driven distribution, would be rated as more authentic on average than Instagram content. That no such difference emerged suggests that, at least for a skincare stimulus judged by an experienced Gen Z social media audience, authenticity judgements may be driven more by content-level cues, such as perceived expertise and disclosure practice, than by platform identity per se, a nuance that qualifies rather than overturns the platform literature reviewed in Chapter 2.
The partial mediation of the authenticity–purchase intention relationship through trust supports the theoretical positioning of trust as a proximal mechanism connecting more distal source characteristics to behavioural intention (Doney and Cannon, 1997; Djafarova and Rushworth, 2017). That mediation was partial rather than full, with a significant direct effect remaining (c′ = .28), suggests that authenticity also operates through pathways not captured by trust alone, plausibly including affective or identification-based routes consistent with parasocial interaction theory (Chung and Cho, 2017), which was significantly correlated with both authenticity and purchase intention in this sample but was not formally modelled as a second mediator. This is consistent with Breves et al.’s (2019) argument that influencer-brand fit operates through multiple, partially overlapping relational mechanisms rather than a single pathway.
A degree of caution is warranted in interpreting the mediation result causally. Although PROCESS Model 4 estimates an indirect effect consistent with a mediating process, the cross-sectional design means that authenticity, trust and purchase intention were all measured concurrently rather than at temporally distinct points, and an alternative, theoretically plausible account in which trust and authenticity mutually reinforce one another over repeated exposure, rather than authenticity cleanly preceding trust, cannot be ruled out on the basis of this data alone. This caveat is returned to in the discussion of methodological limitations in Chapter 6.
The significant moderation by product involvement, with a stronger authenticity effect under high involvement, aligns closely with elaboration likelihood model predictions (Petty and Cacioppo, 1986) and with Xiao, Wang and Chan-Olmsted’s (2018) finding that source-related cues carry more weight under central-route, high-elaboration processing. Practically, this suggests that authenticity-focused influencer strategies may generate the greatest return for higher-involvement, higher-risk purchase categories (for example, skincare, financial products or technology) rather than low-involvement fast-moving consumer goods, where peripheral cues such as follower count or aesthetic appeal may dominate regardless of perceived authenticity.
This finding also has implications for how brands within the skincare category specifically, the category used as this study’s stimulus, might interpret their own influencer data: because skincare typically involves perceived risk relating to skin reaction, cost, or efficacy claims, it sits toward the higher end of the involvement spectrum for many Gen Z consumers, which may partly explain why authenticity effects in this study, and in comparable published studies using similar categories (Sokolova and Kefi, 2020), tend to be stronger than those reported in studies using low-involvement categories such as snack foods or inexpensive accessories.
This study contributes to the influencer marketing literature by empirically integrating authenticity, trust and product involvement within a single tested model specific to a Gen Z sample, addressing the gap identified in Chapter 2 whereby these constructs have typically been studied in isolation or in pairs. The partial mediation finding refines the theoretical understanding of authenticity’s effect pathway, indicating that trust is an important but not exhaustive explanatory mechanism, and the involvement moderation finding extends ELM-based reasoning into the specific context of influencer-mediated purchase intention.
Positioned against the wider literature, the present findings can be read as a Gen Z-specific extension and partial reconciliation of two lines of enquiry that have tended to run in parallel: source credibility research, which emphasises authenticity, expertise and trustworthiness as antecedents of persuasion (Lou and Yuan, 2019), and relational marketing research, which emphasises trust and parasocial interaction as drivers of sustained consumer-brand relationships (Sokolova and Kefi, 2020; Chung and Cho, 2017). By modelling trust explicitly as a mediator rather than treating it as either a standalone predictor or a background assumption, this study offers a more precise account of how these two traditions connect empirically, at least within the boundary conditions of a UK Gen Z sample evaluating a moderately high-involvement product category.
This dissertation set out to examine the relationship between perceived influencer authenticity and purchase intention among Gen Z consumers, and the mediating and moderating roles of trust and product involvement respectively. All three hypotheses were supported: authenticity significantly predicted purchase intention; trust partially mediated this relationship; and product involvement significantly moderated it, with the relationship stronger under high involvement. Together, these results indicate that authenticity is not simply a desirable but diffuse quality of influencer content; it is a measurable driver of purchase-relevant outcomes that operates in part through accumulated consumer trust and whose commercial value is contingent on the involvement level of the product category being promoted.
The study offers an integrated and empirically validated model of influencer-driven purchase intention specific to Gen Z, combining previously separately studied constructs (authenticity, trust, involvement) and clarifying their interrelationships through mediation and moderation analysis rather than simple bivariate correlation, thereby extending the theoretical frameworks of Audrezet, de Kerviler and Guidry Moulard (2020) and Sokolova and Kefi (2020) into a more complete explanatory model. A secondary contribution lies in the moderated mediation design itself: by testing mediation and moderation as complementary rather than competing explanations, the study demonstrates a methodological template that future influencer marketing research could apply to other constructs, audiences and product categories, moving the field toward more mechanism-explicit models of influencer effectiveness rather than simple bivariate association testing.
As outlined in Section 3.8, the cross-sectional design limits causal inference, the single stimulus and product category may limit generalisability, and the non-probability sampling strategy, while yielding a reasonably diverse UK Gen Z sample, cannot guarantee full representativeness of the wider population. Purchase intention, while a well-established proxy, does not guarantee actual purchase behaviour, and the intention-behaviour gap flagged in Section 2.5 means the practical magnitude of the effects reported here should be treated as an upper bound on likely real-world purchase impact rather than a direct estimate of it. Finally, the study examined a single mediator (trust) and a single moderator (product involvement); other plausible mechanisms, including parasocial interaction as a second mediator and influencer-follower demographic similarity as a second moderator, were measured for descriptive purposes but not formally incorporated into the tested model, and their omission means the present model, while significant, does not claim to be a complete account of how authenticity drives purchase intention.
Future research could extend this model experimentally, manipulating authenticity cues directly to strengthen causal claims; test the model across multiple product categories and influencer tiers (nano, micro, macro) simultaneously; incorporate parasocial interaction as a formal second mediator alongside trust, testing a serial or parallel mediation model; and track actual purchase behaviour, for example via voucher-code redemption, rather than self-reported intention alone. A longitudinal design, tracking the same panel of Gen Z consumers across repeated exposures to a given influencer, would also allow the accumulation argument advanced in Section 2.3, whereby trust is theorised to build gradually over a pattern of authentic-seeming content rather than from a single post, to be tested directly rather than inferred from cross-sectional data.
Ajzen, I. (1991) ‘The theory of planned behavior’, Organizational Behavior and Human Decision Processes, 50(2), pp. 179–211.
Audrezet, A., de Kerviler, G. and Guidry Moulard, J. (2020) ‘Authenticity under threat: when social media influencers need to go beyond self-presentation’, Journal of Business Research, 117, pp. 557–569.
Baker, M.J. and Churchill, G.A. (1977) ‘The impact of physically attractive models on advertising evaluations’, Journal of Marketing Research, 14(4), pp. 538–555.
Belanche, D., Casaló, L.V., Flavián, M. and Ibáñez-Sánchez, S. (2021) ‘Understanding influencer marketing: the role of congruence between influencers, products and consumers’, Journal of Business Research, 132, pp. 186–195.
Boerman, S.C. (2020) ‘The effects of the standardized Instagram disclosure for micro- and meso-influencers’, Computers in Human Behavior, 103, pp. 199–207.
Boerman, S.C., Willemsen, L.M. and Van Der Aa, E.P. (2017) ‘”This post is sponsored”: effects of sponsorship disclosure on persuasion knowledge and electronic word of mouth in the context of Facebook’, Journal of Interactive Marketing, 38, pp. 82–92.
Breves, P.L., Liebers, N., Abt, M. and Kunze, A. (2019) ‘The perceived fit between instagram influencers and the endorsed brand: how influencer–brand fit affects source credibility and persuasive effectiveness’, Journal of Advertising Research, 59(4), pp. 440–454.
Bryman, A. (2016) Social Research Methods. 5th edn. Oxford: Oxford University Press.
Chapple, C. and Cownie, F. (2017) ‘An investigation into viewers’ trust in and response towards disclosed paid-for-endorsements by YouTube lifestyle vloggers’, Journal of Promotional Communications, 5(2), pp. 110–136.
Chung, S. and Cho, H. (2017) ‘Fostering parasocial relationships with celebrities on social media: implications for celebrity endorsement’, Psychology & Marketing, 34(4), pp. 481–495.
Colliander, J. and Dahlén, M. (2011) ‘Following the fashionable friend: the power of social media’, Journal of Advertising Research, 51(1), pp. 313–320.
Colliander, J. and Erlandsson, S. (2015) ‘The blog and the bountiful: exploring the effects of disguised product placement on blogs that are revealed by a third party’, Journal of Marketing Communications, 21(1), pp. 34–47.
Creswell, J.W. and Creswell, J.D. (2018) Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. 5th edn. Thousand Oaks, CA: Sage.
De Veirman, M., Cauberghe, V. and Hudders, L. (2017) ‘Marketing through Instagram influencers: the impact of number of followers and product divergence on brand attitude’, International Journal of Advertising, 36(5), pp. 798–828.
Djafarova, E. and Bowes, T. (2021) ‘”Instagram made Me buy it”: generation Z impulse purchases in fashion industry’, Journal of Retailing and Consumer Services, 59, 102345.
Djafarova, E. and Rushworth, C. (2017) ‘Exploring the credibility of online celebrities’ Instagram profiles in influencing the purchase decisions of young female users’, Computers in Human Behavior, 68, pp. 1–7.
Doney, P.M. and Cannon, J.P. (1997) ‘An examination of the nature of trust in buyer–seller relationships’, Journal of Marketing, 61(2), pp. 35–51.
Field, A. (2018) Discovering Statistics Using IBM SPSS Statistics. 5th edn. London: Sage.
Francis, T. and Hoefel, F. (2018) ‘”True Gen”: generation Z and its implications for companies’, McKinsey & Company Insights, November, pp. 1–12.
Green, S.B. (1991) ‘How many subjects does it take to do a regression analysis?’, Multivariate Behavioral Research, 26(3), pp. 499–510.
Hayes, A.F. (2018) Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach. 2nd edn. New York: Guilford Press.
Horton, D. and Wohl, R.R. (1956) ‘Mass communication and para-social interaction: observations on intimacy at a distance’, Psychiatry, 19(3), pp. 215–229.
Jin, S.V. and Muqaddam, A. (2019) ‘Product placement 2.0: “Do brands need influencers, or do influencers need brands?”‘, Journal of Brand Management, 26(5), pp. 522–537.
Kapitan, S. and Silvera, D.H. (2016) ‘From digital media influencers to celebrity endorsers: attributions drive endorser effectiveness’, Marketing Letters, 27(3), pp. 553–567.
Kay, S., Mulcahy, R. and Parkinson, J. (2020) ‘When less is more: the impact of macro and micro social media influencers’ disclosure’, Journal of Marketing Management, 36(3–4), pp. 248–278.
Lou, C. and Yuan, S. (2019) ‘Influencer marketing: how message value and credibility affect consumer trust of branded content on social media’, Journal of Interactive Advertising, 19(1), pp. 58–73.
Moulard, J.G., Garrity, C.P. and Rice, D.H. (2015) ‘What makes a human brand authentic? Identifying the antecedents of celebrity authenticity’, Psychology & Marketing, 32(2), pp. 173–186.
Munnukka, J., Uusitalo, O. and Toivonen, H. (2016) ‘Credibility of a peer endorser and advertising effectiveness’, Journal of Consumer Marketing, 33(3), pp. 182–192.
Nunnally, J.C. (1978) Psychometric Theory. 2nd edn. New York: McGraw-Hill.
Petty, R.E. and Cacioppo, J.T. (1986) ‘The elaboration likelihood model of persuasion’, Advances in Experimental Social Psychology, 19, pp. 123–205.
Rubin, A.M., Perse, E.M. and Powell, R.A. (1985) ‘Loneliness, parasocial interaction, and local television news viewing’, Human Communication Research, 12(2), pp. 155–180.
Saunders, M., Lewis, P. and Thornhill, A. (2019) Research Methods for Business Students. 8th edn. Harlow: Pearson.
Schouten, A.P., Janssen, L. and Verspaget, M. (2020) ‘Celebrity vs. influencer endorsements in advertising: the role of identification, credibility, and product-endorser fit’, International Journal of Advertising, 39(2), pp. 258–281.
Sokolova, K. and Kefi, H. (2020) ‘Instagram and YouTube bloggers promote it, why should I buy? How credibility and parasocial interaction influence purchase intentions’, Journal of Retailing and Consumer Services, 53, 101742.
Xiao, M., Wang, R. and Chan-Olmsted, S. (2018) ‘Factors affecting YouTube influencer marketing credibility: a heuristic-systematic model’, Journal of Media Business Studies, 15(3), pp. 188–213.
Zaichkowsky, J.L. (1985) ‘Measuring the involvement construct’, Journal of Consumer Research, 12(3), pp. 341–352.
Need a Model Dissertation Written to Your Exact Brief?
Our 350+ UK-qualified writers deliver referenced model documents from £15 per 250 words, with free plagiarism and AI-detection reports.
You May Also Like
All work is written by human writers. 100% AI free, guaranteed.
100% money back guarantee if you find plagiarism in our work.
COMPANY DETAILS
