Table of Contents
Type: Full Undergraduate Dissertation | Subject: Marketing | Level: Undergraduate | Word Count: ~9,400 words
This model Full Undergraduate Dissertation was produced by an Essays UK specialist as reference material for learning purposes only. It is a worked example of consumer-behaviour research design, questionnaire method and quantitative analysis, and should be used to understand structure, argument and referencing, not submitted as your own work. For support in this field, see our marketing dissertation writers.
Overview: This model undergraduate marketing dissertation investigates how influencer marketing shapes Generation Z purchase intention towards UK fast-fashion brands. Using a quantitative survey of 312 UK consumers aged 18 to 25, it tests the effects of source credibility, parasocial relationship strength and perceived authenticity on purchase intention within a Theory of Planned Behaviour framework. All survey figures are illustrative and were created for this example to demonstrate descriptive statistics, correlation and multiple regression; the scholarly literature cited is real and referenced in Harvard (Cite Them Right) style.
Influencer marketing has become one of the most prominent promotional tools used by fast-fashion retailers to reach younger consumers, yet the psychological mechanisms through which it converts attention into purchase intention remain only partly understood. This dissertation examines how influencer marketing shapes the purchase intention of Generation Z consumers towards United Kingdom fast-fashion brands. It focuses on three antecedents that recur in the influencer literature but are rarely tested together within a single behavioural model: the perceived credibility of the influencer, the strength of the parasocial relationship that followers form with them, and the perceived authenticity of sponsored content. These antecedents are embedded within the Theory of Planned Behaviour, which supplies attitude, subjective norm and perceived behavioural control as proximal predictors of intention.
A quantitative, cross-sectional survey design was adopted. A structured online questionnaire using established multi-item scales was completed by 312 UK-based consumers aged between 18 and 25, recruited through non-probability convenience and snowball sampling. Data were analysed using descriptive statistics, reliability testing, Pearson correlation and multiple linear regression. The illustrative results indicate that perceived credibility, parasocial relationship strength and perceived authenticity are all positively associated with purchase intention, with authenticity and attitude emerging as the strongest predictors in the regression model, which explained a substantial share of the variance in intention. Subjective norm exerted a weaker influence than attitude, consistent with the individualistic, self-expressive consumption style often attributed to Generation Z. The dissertation contributes an integrated model for studying influencer effects, offers practical recommendations for fast-fashion marketers around authentic and transparent collaboration, and identifies avenues for future longitudinal and cross-platform research. As a model example, it is intended to illustrate sound research design and reporting rather than to provide definitive empirical conclusions.
The way in which young consumers discover, evaluate and purchase clothing has changed profoundly over the past decade. Where traditional advertising once dominated the fashion purchase journey, the contemporary journey is increasingly mediated by social media platforms and, within those platforms, by individuals who have accumulated large and engaged audiences. These individuals, commonly described as social media influencers, occupy a position between the celebrity endorser of traditional marketing and the ordinary peer whose recommendations carry the weight of word of mouth. Freberg et al. (2011) characterise influencers as a new type of independent third-party endorser who shape audience attitudes through blogs, posts and other forms of social media content. For fast-fashion brands, whose commercial model depends on rapid product turnover and the constant stimulation of demand, influencers offer an attractive means of maintaining relevance within the cultural conversations of younger audiences.
Fast fashion refers to a retail model characterised by short design-to-shelf cycles, low prices and high product variety, which together encourage frequent and often impulsive purchasing. The United Kingdom hosts a particularly competitive fast-fashion market, in which established high-street names compete with digitally native retailers that rely heavily on social media for customer acquisition. Within this environment, Generation Z, usually defined as those born between the mid-1990s and the early 2010s, has emerged as a commercially decisive cohort. Members of this generation have grown up with smartphones and social platforms as an ordinary feature of daily life, and their purchase behaviour is widely observed to be shaped more by peer and influencer content than by conventional brand-controlled advertising. Understanding how influencer marketing operates on this cohort is therefore of direct strategic importance to fast-fashion marketers and of theoretical interest to consumer researchers.
The commercial scale of this phenomenon is considerable. Industry commentary, which should be read as indicative rather than definitive, consistently reports year-on-year growth in influencer marketing budgets and a corresponding shift of spending away from traditional display advertising towards social and creator-led channels. For fast-fashion retailers in particular, influencer collaborations have moved from an experimental tactic to a core component of the marketing mix, used not only to launch ranges but to sustain a continuous presence in the feeds where purchase decisions increasingly originate. This strategic centrality makes it all the more important to understand not merely whether influencer marketing works, but through which mechanisms it works, and for whom. It is this question of mechanism, rather than of mere effectiveness, that motivates the present study.
The appeal of influencer marketing rests on a set of overlapping psychological foundations. The first is credibility: when an influencer is perceived as expert, trustworthy and attractive, their endorsements are more persuasive, a relationship rooted in the source credibility tradition of persuasion research (Hovland, Janis and Kelley, 1953; Ohanian, 1990). The second is the quasi-personal bond that audiences develop with figures they encounter repeatedly through media, a phenomenon described by Horton and Wohl (1956) as parasocial interaction. The third, more recent, concern is authenticity: as sponsored content has proliferated, audiences have become increasingly sensitive to whether an endorsement reflects genuine preference or mere commercial arrangement (Lou and Yuan, 2019). These foundations are frequently discussed in isolation, yet in practice they operate together in the mind of the consumer, and it is their combined effect on purchase intention that this dissertation seeks to examine.
Although practitioner enthusiasm for influencer marketing is considerable and marketing budgets allocated to it have grown substantially, the academic understanding of precisely how influencer characteristics translate into purchase intention remains fragmented. Much of the existing research examines single antecedents, such as credibility (Djafarova and Rushworth, 2017) or follower numbers and product fit (De Veirman, Cauberghe and Hudders, 2017), without situating them in an integrated model of behaviour. A further limitation is that a large share of the evidence base draws on general social media users or on samples from outside the United Kingdom, leaving open the question of whether the mechanisms observed apply to UK Generation Z consumers specifically, whose media habits and regulatory context, including advertising disclosure rules, differ from other markets. Finally, the rapid rise of consumer scepticism towards sponsored content means that authenticity has become a plausible boundary condition on influencer effectiveness, yet it is seldom tested alongside credibility and parasocial relationship within the same framework. The present study addresses these gaps by testing an integrated model of influencer effects on purchase intention among UK fast-fashion consumers of Generation Z.
The overarching aim of this dissertation is to investigate how influencer marketing shapes the purchase intention of Generation Z consumers towards UK fast-fashion brands. To achieve this aim, the study pursues four objectives:
Flowing from these objectives, the study is organised around one central research question and three supporting questions:
The specific hypotheses derived from these questions and justified in Chapter 2 are stated there and summarised in Table 2.1. In brief, the study hypothesises positive relationships between each influencer antecedent and purchase intention, and positive relationships between the Theory of Planned Behaviour constructs and purchase intention, with attitude expected to be the strongest proximal predictor.
The study offers both theoretical and practical contributions. Theoretically, it responds to calls for the integration of influencer-specific constructs with established behavioural theory by positioning credibility, parasocial relationship and authenticity as antecedents of the Theory of Planned Behaviour variables rather than treating them as isolated predictors. This integration allows the relative importance of influencer characteristics to be assessed against the broader social and control-related determinants of intention. Practically, the findings are intended to inform how fast-fashion marketers select influencers, brief them and structure disclosures, with particular attention to the trade-off between reach and authenticity. Because the study concentrates on UK Generation Z consumers, its illustrative findings are framed to be relevant to a cohort of substantial and growing commercial value in a market where fast fashion is both popular and the subject of increasing ethical scrutiny.
The study is deliberately bounded. It concerns purchase intention rather than actual purchase behaviour, on the understanding that intention is the most proximal and commonly measured antecedent of behaviour in the Theory of Planned Behaviour, while recognising that an intention–behaviour gap exists. It focuses on fast fashion rather than fashion in general, and on UK-based consumers aged 18 to 25, excluding the youngest members of Generation Z for reasons of informed consent and data protection. It treats influencer marketing primarily as it occurs on image- and video-led platforms, where fashion content predominates, rather than attempting to compare every platform. These delimitations sharpen the study but also constrain the generalisability of its conclusions, a point revisited in Chapter 6.
The dissertation proceeds in six chapters. Chapter 2 reviews the relevant literature and develops the conceptual framework and hypotheses. Chapter 3 sets out the research methodology, including philosophy, design, sampling, measurement, analysis and ethics. Chapter 4 presents the findings of the survey, including descriptive statistics, reliability assessment, correlation and regression analysis. Chapter 5 discusses these findings in relation to the literature. Chapter 6 concludes, offers practical recommendations, acknowledges limitations and suggests directions for future research.
This chapter reviews the body of knowledge relevant to the study and uses it to construct a testable conceptual model. It begins by defining influencer marketing and locating it within the broader fields of endorsement and word-of-mouth research. It then characterises the UK fast-fashion context and the consumption tendencies of Generation Z, before examining in turn the three focal antecedents of this study: source credibility, parasocial relationships and authenticity. The chapter next introduces the Theory of Planned Behaviour as the organising behavioural framework, and finally synthesises these strands into a conceptual model with explicit hypotheses. The review is organised thematically rather than chronologically, so that competing positions on each construct can be compared directly.
Influencer marketing can be understood as a form of marketing in which brands collaborate with individuals who possess an engaged social media following, in order to communicate with and persuade that following. It is distinguished from traditional celebrity endorsement by the perceived ordinariness and accessibility of many influencers, and from pure word of mouth by its commercial and often contractual basis. Freberg et al. (2011) position influencers as independent third-party endorsers whose perceived independence is central to their persuasive power. This perceived independence is, however, in tension with the commercial relationships that fund influencer activity, a tension that becomes important when authenticity is considered later in this chapter.
Influencers are not a homogeneous category, and the literature increasingly distinguishes between tiers defined chiefly by audience size: so-called mega- and macro-influencers, whose followings run into the millions or hundreds of thousands and who resemble traditional celebrities; micro-influencers, with followings in the thousands to tens of thousands and a reputation for niche expertise; and nano-influencers, with small but highly engaged audiences. A recurring theme in recent work is the apparent trade-off between reach and relatability: larger accounts deliver scale but may be perceived as more remote and more commercial, whereas smaller accounts offer narrower reach but often higher engagement and a stronger sense of peer-like authenticity. This tiering matters for the present study because it implies that the antecedents examined here, particularly authenticity and parasocial closeness, may be distributed unevenly across influencer types. The supplementary analysis in Chapter 4 returns to this question by comparing respondents who principally follow smaller versus larger accounts.
A useful conceptual anchor is the two-step flow of communication and the related notion of opinion leadership, which hold that ideas and influence often pass from media to opinion leaders and then to a wider public. Casaló, Flavián and Ibáñez-Sánchez (2020) apply opinion leadership directly to Instagram influencers, arguing that originality and uniqueness of content drive perceptions of opinion leadership, which in turn affect intentions to interact with and follow the influencer’s recommendations. Influencer marketing can therefore be seen as a contemporary, platform-mediated expression of a long-standing idea in communication research, namely that interpersonal and quasi-interpersonal influence frequently outperforms impersonal mass communication in shaping attitudes and behaviour.
Fast fashion is defined by speed and volume: rapid translation of catwalk and cultural trends into affordable garments, frequent range changes and pricing strategies that lower the psychological barrier to purchase. These features align closely with the affordances of social media, where novelty is continuously surfaced and where purchase can follow discovery within a few taps. The UK market is characterised by intense competition between legacy high-street retailers and digitally native brands that treat social media not merely as a promotional channel but as a primary engine of customer acquisition. For such brands, influencer collaborations function as a continuous stream of socially embedded product demonstrations, blurring the line between content and commerce.
It is also worth noting how the fast-fashion purchase journey compresses the classic stages of consumer decision making. In a conventional model, need recognition, information search, evaluation of alternatives, purchase and post-purchase evaluation unfold over time and across channels. Within social commerce, an influencer post can collapse several of these stages into a single moment: the content simultaneously creates awareness, supplies evaluative information in the form of a trusted recommendation, and provides an immediate route to purchase. For low-involvement, low-cost fast-fashion items, where extended deliberation is unlikely, this compression favours impulsive and affect-driven purchasing. The constructs examined in this study map onto that compressed journey: credibility and authenticity shape the evaluative judgement, parasocial connection lends the recommendation its interpersonal weight, and perceived behavioural control reflects the frictionlessness of the path from feed to checkout.
The fast-fashion context also carries distinctive pressures that bear on this study. Growing public concern over the environmental and labour implications of disposable clothing has made authenticity and values a more salient dimension of brand evaluation, particularly among younger consumers who express, at least attitudinally, a preference for responsible consumption. This creates a potential dissonance: the same cohort that is highly responsive to influencer-driven novelty may also be sensitive to perceived inauthenticity or ethical shortcomings. The fast-fashion setting thus provides a theoretically rich environment in which to examine how credibility, parasocial bonds and authenticity jointly influence intention.
Generation Z is frequently described as the first truly digital-native cohort, having come of age with pervasive connectivity, social platforms and e-commerce. Several consumption tendencies are commonly attributed to this generation in the marketing literature and warrant cautious statement here as working assumptions rather than established laws. First, Generation Z consumers are held to place high value on peer validation and social proof, consulting networks and online communities before purchase. Second, they are thought to exhibit comparatively low tolerance for overtly commercial or inauthentic messaging, having been exposed to advertising from an early age. Third, they are associated with self-expressive and identity-driven consumption, in which brand choices signal membership and values. Djafarova and Rushworth (2017) provide evidence consistent with these tendencies in their study of young female Instagram users, finding that the credibility of online figures influences the purchase decisions of this group and that non-traditional, relatable figures can be especially persuasive.
The responsiveness of this cohort is closely bound up with the mechanics of electronic word of mouth. Where traditional word of mouth was constrained by the reach of personal networks, social platforms allow recommendations, reviews and visible endorsements to circulate at scale and to be encountered repeatedly. Influencer content can be understood as a professionalised, highly visible form of electronic word of mouth, in which the recommender is simultaneously a peer-like figure and a quasi-media personality. For Generation Z consumers, who routinely consult such content before purchase, the boundary between editorial and advertising, and between friend and endorser, is correspondingly blurred. This blurring is precisely what lends influencer marketing its persuasive advantage and, at the same time, what makes authenticity so consequential: when the commercial basis of an endorsement becomes salient, the peer-like framing that underpins its power is placed under strain.
These tendencies have implications for the present model. If Generation Z consumers prioritise peer and influencer validation, then influencer antecedents should exert a meaningful effect on intention. If they are sensitive to inauthenticity, then authenticity should condition or add to the effect of credibility. And if their consumption is individualistic and self-expressive, then within the Theory of Planned Behaviour their attitudes may weigh more heavily than subjective norms. Each of these implications is incorporated into the hypotheses developed below.
Source credibility theory, originating in the persuasion research of Hovland, Janis and Kelley (1953), holds that the persuasiveness of a message depends in part on characteristics of its source, principally expertise and trustworthiness. Ohanian (1990) extended and operationalised this tradition for endorsement contexts, developing a widely used tripartite scale comprising expertise, trustworthiness and attractiveness. Applied to influencers, expertise concerns the perceived knowledge and competence of the influencer in the relevant domain, such as styling or product knowledge; trustworthiness concerns the perceived honesty and integrity of the influencer; and attractiveness concerns physical appeal and likeability, which is especially relevant in the aesthetically driven fashion category.
Empirical work broadly supports a positive relationship between perceived influencer credibility and favourable consumer responses. De Veirman, Cauberghe and Hudders (2017) show that influencer characteristics such as follower counts and the fit between influencer and product shape brand attitudes, while also cautioning that very high follower counts can sometimes reduce perceived likeability under certain conditions, indicating that credibility is not simply a function of reach. Lou and Yuan (2019) demonstrate that the credibility of an influencer, together with the informational value of their content, affects consumer trust in branded posts, which in turn influences brand awareness and purchase intention. Taken together, this literature supports the expectation that perceived credibility will positively predict purchase intention in the present study, both directly and through its effect on attitude. Accordingly, the first hypothesis proposes a positive relationship between perceived influencer credibility and purchase intention.
Beyond credibility, a distinctive feature of influencer marketing is the sense of personal connection that followers develop with influencers despite the absence of genuine reciprocal acquaintance. Horton and Wohl (1956) introduced the concept of parasocial interaction to describe the one-sided, intimacy-at-a-distance relationships that audiences form with media figures. On social platforms, where influencers share apparently personal and everyday content and respond, however selectively, to comments and messages, these parasocial relationships can be unusually strong. The theoretical expectation is that a stronger parasocial relationship increases the influencer’s persuasive reach, because recommendations from a figure experienced as a friend are processed more like interpersonal word of mouth than like advertising.
Closely connected to both credibility and parasocial attachment is the increasingly prominent construct of authenticity. As sponsored content has become ubiquitous and as disclosure requirements have made commercial relationships more visible, audiences have grown more discerning about whether an endorsement reflects genuine preference. Perceived authenticity concerns the degree to which followers believe that an influencer’s endorsement is sincere, consistent with their usual content and not driven solely by payment. The literature suggests that authenticity is a key mediator of influencer effectiveness: credible, parasocially close influencers lose persuasive power if their endorsements are seen as purely transactional, whereas perceived authenticity can amplify the effect of an endorsement. This reasoning yields hypotheses proposing that parasocial relationship strength and perceived authenticity each positively predict purchase intention, and that authenticity adds explanatory power beyond credibility and parasocial relationship. Importantly, disclosure of a commercial relationship need not reduce authenticity; transparent yet genuine endorsements may be perceived as both honest and sincere, a nuance the study’s recommendations revisit.
To connect these influencer antecedents to purchase intention within a coherent behavioural model, this study adopts the Theory of Planned Behaviour (Ajzen, 1991; Fishbein and Ajzen, 2010). The theory holds that behaviour is driven most proximally by behavioural intention, which is itself a function of three determinants: attitude towards the behaviour, which is the individual’s favourable or unfavourable evaluation of performing it; subjective norm, which is the perceived social pressure to perform or not perform the behaviour; and perceived behavioural control, which is the perceived ease or difficulty of performing the behaviour, reflecting both resources and confidence. The theory has been applied extensively across consumer behaviour and online contexts and provides a parsimonious yet robust account of intention formation.
A further reason for selecting the Theory of Planned Behaviour over simpler attitude-only models is its explicit treatment of volitional control and social pressure, both of which are germane to social commerce. Perceived behavioural control captures not only the practical ease of purchase but also the consumer’s confidence in navigating the purchase process, which is pertinent where discovery and transaction occur within the same application. Subjective norm, meanwhile, provides a dedicated channel for the social dimension of consumption that is so often emphasised in accounts of Generation Z, allowing the study to test, rather than assume, the strength of normative influence. The theory has, of course, attracted criticism, including the charge that it under-weights emotional and habitual drivers of behaviour and that the intention it predicts does not always translate into action. These critiques are acknowledged and inform both the cautious interpretation of the results and the suggestions for future research.
The Theory of Planned Behaviour is well suited to integrating influencer effects because the influencer antecedents can be understood as inputs that shape the three determinants. Credibility and authenticity plausibly operate through attitude, by making the brand or product appear more desirable; parasocial relationships and the visible popularity of influencer-endorsed products plausibly operate through subjective norm, by signalling what valued others do and approve of; and the ease of discovery and purchase facilitated by social commerce plausibly operates through perceived behavioural control. In this study, the three determinants are modelled as direct predictors of purchase intention alongside the influencer antecedents, allowing their relative weights to be estimated. Consistent with the individualistic consumption style attributed to Generation Z, attitude is expected to be the strongest determinant and subjective norm comparatively weaker.
Synthesising the preceding sections, the conceptual framework positions three influencer antecedents, namely perceived credibility, parasocial relationship strength and perceived authenticity, as predictors of purchase intention, operating alongside the three Theory of Planned Behaviour determinants of attitude, subjective norm and perceived behavioural control. The framework thus treats influencer characteristics and behavioural determinants as complementary rather than competing explanations. The hypotheses are stated below and summarised in Table 2.1.
| Hypothesis | Statement |
|---|---|
| H1 | Perceived influencer credibility is positively related to purchase intention. |
| H2 | Parasocial relationship strength is positively related to purchase intention. |
| H3 | Perceived authenticity of sponsored content is positively related to purchase intention, over and above credibility and parasocial relationship. |
| H4 | Attitude towards purchasing influencer-endorsed fast fashion is positively related to purchase intention. |
| H5 | Subjective norm is positively related to purchase intention. |
| H6 | Perceived behavioural control is positively related to purchase intention. |
| H7 | Attitude exerts a stronger effect on purchase intention than subjective norm. |
These hypotheses operationalise the central research question in testable form. H1 to H3 address the influencer antecedents that are the study’s primary focus, with H3 framed as an incremental-validity test of authenticity. H4 to H6 confirm the applicability of the Theory of Planned Behaviour in this context, and H7 tests the specific expectation, grounded in the characterisation of Generation Z, that self-evaluation outweighs perceived social pressure.
The review identifies a clear gap. While credibility, parasocial relationships and authenticity are each well represented in the influencer literature, they are seldom tested together within an established behavioural framework, and rarely in the specific context of UK Generation Z fast-fashion consumers. Much existing evidence derives from single-antecedent designs or from samples that are not UK-specific, limiting its applicability to the regulatory and cultural setting of interest here. By integrating the three antecedents with the Theory of Planned Behaviour and testing the resulting model on a UK Generation Z sample, this study aims to clarify the relative importance of influencer characteristics and behavioural determinants in shaping purchase intention. The chapter has defined the key constructs, reviewed the relevant evidence, introduced the organising theory and developed a set of hypotheses. The next chapter sets out the methodology used to test them.
The methodological choices of this study follow from its ontological and epistemological assumptions. The study adopts a positivist philosophy, consistent with its aim of testing hypothesised relationships between measurable constructs (Saunders, Lewis and Thornhill, 2019). Positivism assumes an observable social reality that can be studied through objective measurement, and favours the formulation of hypotheses derived from theory and their testing against quantitative data. This stance is appropriate here because the study seeks to estimate the strength and direction of relationships among well-defined constructs drawn from established theory, rather than to explore the subjective meanings that consumers attach to influencer content, which would call for an interpretivist design. The limitations of positivism, including its relative insensitivity to context and meaning, are acknowledged in the discussion of limitations.
Consistent with a positivist philosophy, the study employs a deductive approach, moving from theory to hypotheses to data (Bryman and Bell, 2015). A quantitative, cross-sectional survey design was selected as the most suitable means of collecting standardised data from a relatively large sample at a single point in time, enabling statistical testing of the hypothesised relationships. A survey is efficient, replicable and well suited to measuring attitudinal and intention constructs using validated multi-item scales. A cross-sectional design was chosen in preference to a longitudinal design for reasons of feasibility within the time available for an undergraduate dissertation, with the consequence, noted later, that causal inference is limited and only associations can be established.
The target population comprised UK-based consumers of Generation Z, operationally defined for this study as individuals aged between 18 and 25 who use social media and have encountered fashion-related influencer content. The lower bound of 18 was set to ensure that all participants could provide valid informed consent and to avoid the additional safeguards required for minors. Because no sampling frame exists for this population, non-probability sampling was used: a combination of convenience sampling, through distribution of the questionnaire across social media and university networks, and snowball sampling, through encouraging respondents to share the survey with eligible peers. This approach is pragmatic and common in student research but introduces the risk of sampling bias and limits the statistical generalisability of findings, as discussed in section 3.9.
A target sample size of at least 300 was set to provide adequate statistical power for multiple regression with several predictors and to satisfy commonly cited rules of thumb regarding the ratio of cases to predictors (Hair et al., 2019). After data cleaning, described below, the achieved analytic sample was 312 respondents. A screening question confirmed age eligibility and social media use; respondents who did not meet these criteria were routed out of the survey and excluded.
Data were collected using a structured, self-administered online questionnaire. The questionnaire opened with a participant information section and consent confirmation, followed by screening questions, demographic questions and then multi-item measures of each construct. All substantive constructs were measured using multiple items adapted from established scales and rated on seven-point Likert-type scales anchored from “strongly disagree” to “strongly agree”. Perceived credibility items were adapted from the source credibility tradition (Ohanian, 1990), capturing expertise, trustworthiness and attractiveness. Parasocial relationship items were adapted from established parasocial interaction measures reflecting feelings of friendship and connection with a chosen influencer. Authenticity items captured perceptions that the influencer’s endorsements are sincere and consistent with their usual content. Attitude, subjective norm, perceived behavioural control and purchase intention were measured using items adapted from standard Theory of Planned Behaviour operationalisations (Ajzen, 1991; Fishbein and Ajzen, 2010). Respondents were asked to answer the influencer items with reference to a fashion influencer they follow, in order to ground abstract items in a concrete referent.
The use of adapted, previously validated scales supports content and construct validity. The questionnaire was reviewed by the supervisor and pre-tested with a small pilot group of eligible respondents to check item clarity, wording and completion time; minor wording revisions were made in response. The pilot data were not included in the final analysis. Table 3.1 summarises the constructs, the approximate number of items and their anchoring source.
| Construct | Items | Adapted From |
|---|---|---|
| Perceived credibility | 5 | Ohanian (1990) |
| Parasocial relationship | 4 | Horton and Wohl (1956) tradition |
| Perceived authenticity | 4 | Lou and Yuan (2019) tradition |
| Attitude | 4 | Ajzen (1991) |
| Subjective norm | 3 | Ajzen (1991) |
| Perceived behavioural control | 3 | Ajzen (1991) |
| Purchase intention | 4 | Fishbein and Ajzen (2010) |
The questionnaire was implemented on a reputable online survey platform and distributed through social media posts, direct messages and university networks over a four-week period. The opening page presented the participant information sheet and required active confirmation of consent before any questions were displayed. Participation was voluntary and unincentivised, and respondents could withdraw at any point before submission simply by closing the browser. On submission, responses were stored securely on the platform and exported for analysis. In total, 348 responses were received. Following data cleaning, which removed incomplete responses, those failing the age or usage screen, and a small number flagged by an attention-check item or exhibiting implausibly uniform response patterns, 312 valid responses were retained for analysis.
Data were analysed using standard statistical software for the social sciences (Field, 2018). Analysis proceeded in four stages. First, the data were screened for missing values, outliers and the assumptions underlying the planned analyses, including normality, linearity, homoscedasticity and multicollinearity. Second, scale reliability was assessed using Cronbach’s alpha, with a threshold of 0.70 taken as indicating acceptable internal consistency (Nunnally and Bernstein, 1994); items were then averaged into composite scores for each construct. Third, descriptive statistics and Pearson correlations were computed to describe the sample and the bivariate relationships among constructs. Fourth, hierarchical and standard multiple linear regression was used to test the hypotheses, with purchase intention as the dependent variable. In the hierarchical model, the Theory of Planned Behaviour determinants and the influencer antecedents were entered in blocks, allowing the incremental contribution of authenticity to be assessed in support of H3. Statistical significance was evaluated at the conventional 0.05 level.
Several steps were taken to support reliability and validity. Reliability was addressed through the use of established multi-item scales and confirmed statistically through Cronbach’s alpha. Content validity was supported by adapting items from validated instruments and by supervisor review. Construct validity was supported by the theoretical grounding of the measures and by examining the pattern of correlations for convergent and discriminant plausibility. To mitigate common-method bias, which can arise when predictor and outcome measures are collected from the same respondents at the same time, the questionnaire mixed item orderings, used clear and distinct scale labels, and assured respondents of anonymity to reduce social desirability. External validity is constrained by the non-probability sampling strategy, a limitation addressed directly below.
The study was conducted in accordance with the ethical principles of the awarding institution and received ethical approval prior to data collection. Core ethical requirements were observed throughout. Informed consent was obtained via an explicit consent confirmation preceded by a participant information sheet describing the purpose of the study, what participation involved, and participants’ rights. Participation was voluntary, and respondents were free to withdraw before submission. Anonymity was preserved by not collecting names, contact details or other directly identifying information, and confidentiality was maintained by storing data securely and reporting only aggregate results. Only eligible adults aged 18 or over were included, avoiding the need for parental consent. Data were handled in line with data protection requirements, stored on secure, password-protected systems and scheduled for deletion after the examination process. Because the topic carried no significant risk of harm or distress, no special safeguarding measures beyond these standard protections were required.
Several methodological limitations should be borne in mind when interpreting the findings, consistent with the transparency expected of a model example. The cross-sectional design precludes firm causal claims; the relationships reported are associations observed at a single time point. The non-probability convenience and snowball sampling strategy limits the representativeness of the sample and therefore the statistical generalisability of the results to all UK Generation Z consumers. Reliance on self-report introduces the possibility of common-method and social-desirability biases, which the design mitigates but cannot eliminate. Measuring purchase intention rather than behaviour leaves open the intention–behaviour gap. Finally, asking respondents to consider a self-selected influencer introduces heterogeneity in the referent, trading some control for ecological validity. These limitations shape the cautious framing of the conclusions.
This chapter presents the illustrative results of the survey. It begins with the sample profile, proceeds to the reliability assessment of the measurement scales, then reports descriptive statistics, correlation analysis and the regression models used to test the hypotheses. All figures reported here are illustrative values created for this model example and should be read as a demonstration of analytical procedure and reporting conventions rather than as empirical findings about real consumers.
Of the 348 responses received, 312 were retained after data cleaning, giving a usable response figure of 89.7 per cent of submitted surveys. The analytic sample was relatively balanced by gender, with a modest female majority consistent with higher engagement in fashion content, and spanned the full 18 to 25 age range with a concentration in the 19 to 22 band typical of a university-linked recruitment channel. The large majority of respondents reported following at least one fashion influencer and using social media daily. Table 4.1 summarises the sample characteristics.
| Characteristic | Category | n | % |
|---|---|---|---|
| Gender | Female | 181 | 58.0 |
| Male | 131 | 42.0 | |
| Age band | 18–20 | 121 | 38.8 |
| 21–22 | 118 | 37.8 | |
| 23–25 | 73 | 23.4 | |
| Follows a fashion influencer | Yes | 289 | 92.6 |
| No | 23 | 7.4 | |
| Social media use | Several times daily | 268 | 85.9 |
| Once daily or less | 44 | 14.1 |
Internal consistency reliability was assessed for each multi-item scale using Cronbach’s alpha. As shown in Table 4.2, all scales exceeded the conventional 0.70 threshold (Nunnally and Bernstein, 1994), indicating acceptable to good internal consistency. No item deletions were required to reach acceptable reliability. Composite scores were subsequently created by averaging the items within each scale for use in the correlation and regression analyses.
| Scale | Items | Cronbach’s α |
|---|---|---|
| Perceived credibility | 5 | 0.88 |
| Parasocial relationship | 4 | 0.85 |
| Perceived authenticity | 4 | 0.83 |
| Attitude | 4 | 0.86 |
| Subjective norm | 3 | 0.79 |
| Perceived behavioural control | 3 | 0.76 |
| Purchase intention | 4 | 0.90 |
Table 4.3 reports the means and standard deviations of the composite constructs, measured on seven-point scales. Mean scores for the influencer antecedents and the Theory of Planned Behaviour determinants were generally above the scale midpoint of 4.0, indicating that respondents tended to hold moderately favourable perceptions of the influencers they follow and moderately positive attitudes towards purchasing endorsed fast fashion. Perceived authenticity recorded a somewhat lower mean than credibility, consistent with the literature’s suggestion that Generation Z audiences are discerning about the sincerity of sponsored content. Purchase intention sat just above the midpoint, with a relatively large standard deviation indicating considerable variation between respondents. Examination of skewness and kurtosis values, all within conventionally acceptable bounds, supported the assumption of approximate normality required for the parametric analyses that follow.
| Construct | Mean | SD |
|---|---|---|
| Perceived credibility | 5.12 | 0.98 |
| Parasocial relationship | 4.68 | 1.14 |
| Perceived authenticity | 4.41 | 1.22 |
| Attitude | 5.03 | 1.05 |
| Subjective norm | 4.57 | 1.17 |
| Perceived behavioural control | 5.21 | 0.96 |
| Purchase intention | 4.49 | 1.28 |
Pearson correlation coefficients were computed to examine the bivariate relationships among the constructs. As shown in Table 4.4, purchase intention was positively and significantly correlated with all six predictors. The strongest bivariate associations with purchase intention were observed for attitude and perceived authenticity, followed by perceived credibility and parasocial relationship. Subjective norm exhibited the weakest, though still significant, association with intention. The predictors were themselves moderately intercorrelated, as would be expected given their conceptual proximity, but none of the inter-predictor correlations approached levels that would indicate a serious multicollinearity problem. Variance inflation factors computed during the regression, all below 2.5, confirmed that multicollinearity was not a material concern (Hair et al., 2019).
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | |
|---|---|---|---|---|---|---|---|
| 1. Credibility | 1.00 | ||||||
| 2. Parasocial | 0.52 | 1.00 | |||||
| 3. Authenticity | 0.49 | 0.46 | 1.00 | ||||
| 4. Attitude | 0.55 | 0.44 | 0.57 | 1.00 | |||
| 5. Subjective norm | 0.38 | 0.41 | 0.33 | 0.42 | 1.00 | ||
| 6. PBC | 0.40 | 0.31 | 0.36 | 0.47 | 0.35 | 1.00 | |
| 7. Purchase intention | 0.54 | 0.48 | 0.59 | 0.63 | 0.37 | 0.45 | 1.00 |
Note: All coefficients significant at p < 0.01. PBC = perceived behavioural control.
To test the hypotheses and estimate the relative importance of the predictors, hierarchical multiple linear regression was conducted with purchase intention as the dependent variable. In the first block, the three Theory of Planned Behaviour determinants, namely attitude, subjective norm and perceived behavioural control, were entered. This block was statistically significant and explained a substantial share of the variance in purchase intention, with an adjusted R-squared of 0.47. Attitude emerged as the strongest predictor within this block, with a markedly larger standardised coefficient than subjective norm, providing support for H4, H5, H6 and the comparative expectation in H7.
In the second block, the three influencer antecedents, namely credibility, parasocial relationship and authenticity, were added. The addition produced a statistically significant improvement in model fit, with the adjusted R-squared rising to 0.61, an increment that indicates the influencer constructs explain meaningful variance over and above the behavioural determinants. In the full model, perceived authenticity and attitude were the two strongest predictors, followed by perceived credibility and perceived behavioural control. Parasocial relationship retained a positive and significant coefficient, though smaller than in the bivariate analysis, suggesting that part of its influence operates through the other constructs. Subjective norm remained the weakest predictor and fell to marginal significance once the influencer antecedents were included. Table 4.5 reports the standardised coefficients for the full model.
| Predictor | Standardised β | p |
|---|---|---|
| Perceived authenticity | 0.29 | < 0.001 |
| Attitude | 0.27 | < 0.001 |
| Perceived credibility | 0.18 | < 0.001 |
| Perceived behavioural control | 0.14 | 0.002 |
| Parasocial relationship | 0.12 | 0.012 |
| Subjective norm | 0.07 | 0.061 |
Note: Full model adjusted R-squared = 0.61; dependent variable = purchase intention.
Drawing the analyses together, Table 4.6 summarises the outcome of each hypothesis test. Five of the seven hypotheses were fully supported. H5, concerning subjective norm, was only partially supported, because although the bivariate correlation was significant, the predictor fell to marginal significance in the full regression model. H7 was supported, as attitude clearly exceeded subjective norm as a predictor. H3 was supported, since authenticity added significant explanatory power and emerged as the single strongest predictor in the full model, underscoring its importance in the influencer context.
| Hypothesis | Outcome |
|---|---|
| H1: Credibility → intention | Supported |
| H2: Parasocial → intention | Supported |
| H3: Authenticity adds beyond H1 and H2 | Supported |
| H4: Attitude → intention | Supported |
| H5: Subjective norm → intention | Partially supported |
| H6: Perceived behavioural control → intention | Supported |
| H7: Attitude stronger than subjective norm | Supported |
To explore the tiering question raised in Chapter 2, respondents were classified according to whether the influencer they held in mind was primarily a smaller account, described in the questionnaire as a micro-influencer with a niche following, or a larger macro- or mega-influencer with a mass following. An independent-samples t-test compared the two groups on the key constructs. As Table 4.7 reports, respondents who followed smaller influencers recorded significantly higher mean scores for perceived authenticity and parasocial relationship, and a modestly higher mean for purchase intention, than those who followed larger influencers. There was no significant difference in perceived credibility between the groups, suggesting that credibility is not the preserve of large accounts. This illustrative pattern is consistent with the reach-versus-relatability trade-off discussed earlier: smaller influencers appear to convert their narrower reach into stronger authenticity and connection, which in turn supports intention.
| Construct | Smaller (n=168) | Larger (n=144) | p |
|---|---|---|---|
| Perceived authenticity | 4.71 | 4.06 | < 0.001 |
| Parasocial relationship | 4.92 | 4.40 | 0.001 |
| Perceived credibility | 5.17 | 5.06 | 0.34 |
| Purchase intention | 4.66 | 4.29 | 0.012 |
This supplementary result should be interpreted with particular caution, since the grouping relied on respondents’ own classification of a self-selected influencer and the subgroups were not randomly assigned. Nonetheless, it adds nuance to the main analysis by indicating that the authenticity advantage identified in the regression may be realised more readily through smaller, niche collaborations than through large-scale partnerships, a point developed in the recommendations.
This chapter interprets the illustrative findings in light of the literature reviewed in Chapter 2. It considers each focal antecedent in turn, examines the role of the Theory of Planned Behaviour determinants, and draws out the theoretical implications of the integrated model. Throughout, the discussion maintains the cautious register appropriate to a model study built on illustrative data and a non-probability sample.
The finding that perceived credibility positively predicts purchase intention, supporting H1, is consistent with the long-standing source credibility tradition (Hovland, Janis and Kelley, 1953; Ohanian, 1990) and with contemporary influencer research (Lou and Yuan, 2019; Djafarova and Rushworth, 2017). Within the full model, however, credibility was a moderate rather than dominant predictor, ranking behind authenticity and attitude. This pattern is theoretically coherent: credibility appears to operate substantially through attitude, which is to say that a credible influencer makes the endorsed product seem more desirable, and it is that favourable evaluation which most proximally drives intention. The result suggests that credibility remains necessary but is not, on its own, sufficient to maximise persuasive effect in a context where audiences are attuned to the sincerity of endorsements. It also echoes the caution of De Veirman, Cauberghe and Hudders (2017) that the surface markers of influence, such as follower counts, do not translate mechanically into persuasive power.
Parasocial relationship strength was positively related to purchase intention, supporting H2, in line with the expectation derived from Horton and Wohl (1956) that a sense of personal connection amplifies an influencer’s persuasive reach. The attenuation of the parasocial coefficient from the bivariate to the multivariate analysis is itself informative: it indicates that parasocial attachment influences intention partly through the other constructs, plausibly by strengthening perceived authenticity and favourable attitudes, rather than acting wholly independently. In practical terms, the bond that followers feel towards an influencer appears to function as an enabling condition that makes credible and authentic endorsements land more effectively, consistent with the view that influencer persuasion is processed more like interpersonal word of mouth than like impersonal advertising. This interpretation would need to be tested formally through mediation analysis in future work.
Perhaps the most striking illustrative finding is that perceived authenticity emerged as the single strongest predictor of purchase intention in the full model, supporting H3. This result aligns with the increasingly prominent argument that, in a saturated sponsored-content environment, sincerity is the decisive currency of influencer marketing (Lou and Yuan, 2019). It also coheres with the characterisation of Generation Z as discerning consumers with low tolerance for overt commercialism. The finding has an important and somewhat counter-intuitive practical corollary: because the lower mean score for authenticity indicates it is the construct on which influencers are most often found wanting, it may also be the construct offering the greatest marginal return to improvement. In other words, marketers may gain more by making collaborations feel genuine than by pursuing ever more credible or far-reaching influencers. This positions authenticity not as a peripheral virtue but as a central strategic variable.
The Theory of Planned Behaviour performed as expected and provided a robust backbone for the model. Attitude was the strongest of the three behavioural determinants and one of the two strongest predictors overall, supporting H4. Perceived behavioural control was a significant if smaller predictor, supporting H6 and reflecting the ease with which social commerce allows discovery to translate into purchase. Subjective norm, by contrast, was the weakest predictor and only marginally significant once influencer antecedents were included, rendering H5 only partially supported. The comparative dominance of attitude over subjective norm, supporting H7, is consistent with the self-expressive, individualistic consumption style frequently attributed to Generation Z, for whom personal evaluation appears to outweigh perceived social pressure. This does not imply that social influence is irrelevant; rather, it suggests that for this cohort social influence may be absorbed into the influencer relationship itself and expressed through attitude and authenticity rather than through generalised normative pressure.
The integrated model advances understanding in two ways. First, it demonstrates the value of embedding influencer-specific constructs within an established behavioural framework rather than testing them in isolation. The substantial increment in explained variance when the influencer antecedents were added to the Theory of Planned Behaviour block indicates that these constructs capture something the classic determinants do not, while the classic determinants in turn provide the proximal mechanism through which influencer effects are realised. Second, the prominence of authenticity suggests that the influencer literature’s recent turn towards sincerity and transparency is well founded, and that authenticity deserves a settled place alongside credibility in theories of endorsement. More speculatively, the pattern of results is consistent with a model in which credibility and parasocial attachment are antecedents of authenticity and attitude, which in turn drive intention; testing such a mediational structure would be a natural next step.
The illustrative findings broadly converge with the direction of the published literature while adding specificity in the UK fast-fashion Generation Z context. The positive role of credibility is consistent with Djafarova and Rushworth (2017) and Lou and Yuan (2019), and the relevance of opinion-leadership-style connection accords with Casaló, Flavián and Ibáñez-Sánchez (2020). The comparative weakness of subjective norm is perhaps the least expected result, given the frequent portrayal of younger consumers as highly peer-driven. One interpretation, advanced tentatively here, is that peer influence for this cohort is not absent but is channelled through the influencer relationship itself and expressed as favourable attitude and perceived authenticity, rather than registering as a separate, generalised sense of social pressure in the manner the subjective-norm construct is designed to capture. If so, the finding reflects not the irrelevance of social influence but a change in its route of operation, a possibility that reinforces the case for integrating influencer-specific constructs into behavioural models rather than relying on the classic determinants alone.
The supplementary finding that smaller influencers outperformed larger ones on authenticity and connection, but not on credibility, is also noteworthy. It qualifies any simple assumption that bigger audiences mean greater persuasive power and lends empirical texture, within the bounds of this model, to the practitioner enthusiasm for micro-influencer strategies. At the same time, the absence of a credibility difference is a useful reminder that large accounts retain genuine persuasive assets; the strategic question is therefore less about size in the abstract than about which antecedent a given campaign most needs to strengthen.
This dissertation set out to investigate how influencer marketing shapes Generation Z purchase intention towards UK fast-fashion brands, by testing an integrated model that combined three influencer antecedents with the Theory of Planned Behaviour. The illustrative findings indicate that perceived credibility, parasocial relationship strength and perceived authenticity are all positively associated with purchase intention, and that authenticity and attitude are the strongest predictors in the full model. Subjective norm proved comparatively weak, consistent with the individualistic consumption style attributed to the cohort. In answer to the central research question, influencer marketing appears to shape Generation Z purchase intention primarily by shaping favourable attitudes and perceptions of authenticity, with credibility and parasocial attachment acting as important supporting conditions and social pressure playing a lesser role. The study’s objectives were met: the literature was synthesised, an integrated model developed and tested, the relative strength of predictors estimated, and recommendations derived.
Several evidence-informed recommendations follow for fast-fashion marketers, offered with the caveat that they rest on illustrative data within a model example:
These recommendations are mutually reinforcing rather than independent. Authenticity, long-term partnership and attitude-focused creative work together: a sustained, genuine collaboration with a congruent influencer is precisely the arrangement most likely to generate authentic content, deepen parasocial connection and build favourable attitudes, while transparent disclosure protects the trust on which all of these depend. For a fast-fashion brand operating in a market increasingly sensitive to both commercial saturation and ethical scrutiny, the practical thrust of the findings is that credibility and reach are hygiene factors, whereas authenticity is the differentiator.
The limitations acknowledged in Chapter 3 bound the conclusions. The cross-sectional design and non-probability sample limit causal inference and statistical generalisability; the reliance on self-reported intention leaves the intention–behaviour gap unaddressed; and the self-selected influencer referent introduces heterogeneity. Several avenues for future research follow. Longitudinal or experimental designs would permit stronger causal claims and could examine how parasocial bonds and authenticity perceptions develop over time. Mediation and moderation analyses could formally test the proposed structure in which credibility and parasocial attachment operate through authenticity and attitude. Comparative work across platforms, product categories and generational cohorts would clarify the boundaries of the model, and incorporating observed or sales behaviour alongside intention would address the intention–behaviour gap. Research attentive to the ethical dimension of fast fashion could also explore how sustainability concerns interact with influencer persuasion among values-conscious young consumers.
Influencer marketing has become central to how UK fast-fashion brands reach Generation Z, yet its effectiveness rests on more than reach or celebrity. This model dissertation has illustrated, through an integrated framework and a worked quantitative analysis, that the persuasive power of influencer marketing for this cohort appears to turn above all on perceived authenticity and on the favourable attitudes that credible, genuinely connected influencers help to form. For brands, the implication is that sincerity is strategy. As a model example, the study is offered to demonstrate how a consumer-behaviour dissertation can be designed, conducted and reported to a sound academic standard, and to support students in developing their own research rather than to serve as definitive evidence.
Ajzen, I. (1991) ‘The theory of planned behavior’, Organizational Behavior and Human Decision Processes, 50(2), pp. 179–211.
Bryman, A. and Bell, E. (2015) Business research methods. 4th edn. Oxford: Oxford University Press.
Casaló, L.V., Flavián, C. and Ibáñez-Sánchez, S. (2020) ‘Influencers on Instagram: antecedents and consequences of opinion leadership’, Journal of Business Research, 117, pp. 510–519.
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 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.
Field, A. (2018) Discovering statistics using IBM SPSS statistics. 5th edn. London: Sage.
Fishbein, M. and Ajzen, I. (2010) Predicting and changing behavior: the reasoned action approach. New York: Psychology Press.
Freberg, K., Graham, K., McGaughey, K. and Freberg, L.A. (2011) ‘Who are the social media influencers? A study of public perceptions of personality’, Public Relations Review, 37(1), pp. 90–92.
Hair, J.F., Black, W.C., Babin, B.J. and Anderson, R.E. (2019) Multivariate data analysis. 8th edn. Andover: Cengage.
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.
Hovland, C.I., Janis, I.L. and Kelley, H.H. (1953) Communication and persuasion: psychological studies of opinion change. New Haven: Yale University Press.
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.
Nunnally, J.C. and Bernstein, I.H. (1994) Psychometric theory. 3rd edn. New York: McGraw-Hill.
Ohanian, R. (1990) ‘Construction and validation of a scale to measure celebrity endorsers’ perceived expertise, trustworthiness, and attractiveness’, Journal of Advertising, 19(3), pp. 39–52.
Saunders, M., Lewis, P. and Thornhill, A. (2019) Research methods for business students. 8th edn. Harlow: Pearson.
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
