Frontiers in Business, Economics and Management ISSN: 2766-824X | Vol. 15, No. 2, 2024 19 Social Media and the Electronics Market: The Role and Challenges of Influencer Marketing For Brand Trust Jiajia Dong Adam Smith Business School, University of Glasgow, G12 8QQ, Glasgow, UK 2743112D@student.gla.ac.uk Abstract: This study focuses on the use of influencers on social media to build trust-based brand-consumer connections in the context of the digital age, with a special focus on the Chinese e-market. By analysing the impact of influencer marketing on consumer brand trust, using Zhejiang Province as a case study, it aims to update brand trust theory for the digital marketing environment. The paper highlights the popularity of social media in China and the potential for undue competitive behaviour by influencer marketing. Ultimately, this marketing proposal aims to encourage the electronics industry to actively participate in influencer marketing to enhance brand trust. Keywords: Social media, Electronics market, Influencer marketing, Brand trust, Marketing, Brand. 1. Introduction In the context of the digital age, influencers utilise their reputation and extensive network of relationships on social media to provide new ways of building trust-based connections between brands and consumers, particularly in the area of electronics which is particularly dependent on consumer-perceived brand trust (Brown & Hayes, 2008). This is due to the fact that electronic products often involve higher purchase costs and demands for technical specifications. China, the country with the largest social media presence in the world, has reached an impressive 1.02 billion social media users by 2022, a figure that highlights the popularity of social media in China and the foundation it has laid for influencer marketing (Thomala, 2023). In addition, Zhejiang Province is considered the heartland of e-commerce in China, with e- commerce giants such as Alibaba and a large number of social media users, providing a rich source of case studies and data for the study. (Zhang, 2021). Therefore, this marketing proposal chooses Zhejiang Province as the study area to make full use of its advancement and activity in influencer marketing to explore in-depth the impact of influencer marketing on consumer-perceived brand trust. Furthermore, the theoretical significance of this study is that even though brand trust is a core concept in marketing research, most of the existing theories mainly focus on the impact of traditional marketing channels on brand trust (Keller & Lehman, 2010). Influence (Keller & Lehmann, 2006). As influencer marketing has become an important marketing channel, traditional theories have not adequately considered the impact of factors specific to the digital environment, such as influencer credibilit influencer-brand fit, and audience engagement. Therefore, this study contributes to updating the brand trust theory to make it more relevant to the reality of digital marketing. In terms of practical implications, literature suggests that certain influencers in influencer marketing in the Chinese electronics market engage in inappropriate competitive practices, such as spreading inaccurate information to discredit rivals in order to gain a competitive advantage, and that such behaviours have triggered conflicts within the industry (You, 2023). This marketing proposal hopes that the results of the study will inspire the electronics industry to do positive influencer marketing. Therefore, the main purpose of this study is to verify whether influencer marketing is effective in enhancing consumers' trust in brands in the electronics market in Zhejiang Province. The structure of this marketing proposal coversliterature review, methodology, limitations of the research methodology, and shows the direction for future research. 20 2. Literature Review 2.1. Definition of Key Term Figure 1. Definition of Key Terms 2.2. Research Gap In terms of geography, although there is a large body of research on the impact of influencer marketing on brand trust, relatively few studies have focused on specific regions (e.g., Zhejiang Province, China) and specific industries (e.g., the electronics market). Geographic cultures, economic conditions and consumer behaviours vary significantly across regions and industries, and these factors may influence the mechanisms by which audience engagement, influencer credibility and brand fit act on brand trust. In-depth analyses of the impact of these geographic and industry-specific factors are lacking in the existing literature. In addition, while influencer marketing and brand fit have been identified as key factors influencing consumer trust, there is insufficient research on how to accurately assess and optimise fit. Furthermore, existing literature tends to focus on quantitative metrics (e.g., the number of likes, comments, and shares) when exploring audience engagement, while the quality and depth of engagement is under-explored. Finally, the impact of influencer credibility on consumers' perceived brand trust is a complex process involving multiple factors such as individual differences, behavioural consistency of the influencer, and the closeness of the relationship between the consumer and the influencer. Existing studies tend to use simplified models that fail to adequately capture the dynamic relationships and interactions between these variables. 2.3. The Relationship between Influencer Credibility and Consumer Perceived Brand Trust Based on a previous literature review on influencer credibility and consumer-perceived brand trust, several studies have shown that influencer credibility on social media improves consumer-perceived brand trust by communicating their expertise, reliability, and approachability. Further empirical research has revealed a significant relationship between influencer credibility and consumer-perceived brand trust. In Erdogan's study, although the research focussed on celebrity endorsements, its analysis of reliability can be applied to social media influencers (Erdogan, 1999). It was found that celebrity endorsers who are perceived as reliable can significantly increase brand trust, as consumers tend to trust sources of information that they perceive as honest and credible (Erdogan, 1999). In addition, research by Friedman and Friedman suggests that endorsers with domain-specific expertise are more likely to influence consumer brand perception and trust (Friedman, 1980). In the context of social media influencers, when influencers demonstrate in-depth knowledge in their area of expertise, their recommendations of a brand are more likely to be perceived as credible by consumers. What's more, Kahle and Homer's study explored how the attractiveness of celebrity endorsers can affect advertising effectiveness from a social adaptation perspective. Although this study focused on physical attractiveness, its findings also apply to the personal charisma and attractiveness of influencers, factors that can increase consumer attention and acceptance of brand messages, thereby enhancing brand trust (Kahle & Homer, 1985). 2.3.1. Social Media Value Model The social media value model proposed by Lou and Yuan (2019) can demonstrate the existence of a relationship between influencer credibility and consumers' perceived brand trust through the informational value of influencer- generated content affecting fans' trust in the influencer's brand posts and purchase intentions (Liu & Zheng, 2024). Based on the information value dimension of this model, information provided by highly credible influencers is perceived to be 21 more accurate and reliable, which enhances consumers' trust in the content of the information, and thus indirectly enhances consumers' trust in the brand. However, some studies have shown that in a social media environment flooded with information, even information from credible influencers may be less influential due to overload (Eppler & Mengis). At the level of its social value, when consumers observe highly credible influencers interacting positively with a brand, this sense of social identity and trust is transferred to the brand, as people tend to trust brands recommended within their social circles. At the relationship value level, the trust relationship established between consumers and influencers can be extended to brands recommended by influencers, especially when influencers are perceived to be highly credible and the relationship is perceived to be based on authentic interactions. The higher the credibility of the influencer based on the social media value model of analysis, the stronger the consumer's perceived brand trust; therefore, Hypothesis 1 is proposed that there is a positive correlation between the credibility of the influencer and the consumer's perceived brand trust. 2.4. The Relationship between Influencer- Brand Fit and Consumer-Perceived Brand Trust Based on a review of previous literature on influencer- brand fit and consumer-perceived brand trust, it was found that there is a significant relationship between influencer- brand fit and consumer-perceived brand trust. Boerman's research suggests that when influencers clearly disclose their partnership with a brand on Instagram, this clear disclosure can have a positive impact on brand evaluations. Consumers may respect and trust the brand more as a result of this transparency, as it demonstrates the brand's integrity and respect for consumer intelligence (Boerman, 2020). Through further empirical research, disclosure has a less negative impact on brand evaluation when there is a higher match between the influencer and the brand, which implies that a high influencer-brand match enhances consumer trust in the brand (Boerman, 2020). In addition, research by Amos et al. provided an early understanding of the influencer-brand fit concept, whereby endorsers with a high degree of fit are able to blend more naturally with the brand image, making the advertising message more acceptable and memorable to the target audience (Amos et al., 2008). This is because consumers tend to transfer positive attributes of the endorser to the brand, thereby increasing brand awareness and appeal. Therefore, a good match significantly improves the effectiveness of the advert, which implies the potential positive impact of match on enhancing brand trust. 2.4.1. Match-Up Theory The match-Up theory, also known as the match-Up hypothesis, suggests that the degree of fit (match) between the image of the endorser (influencer) in an advert and the image of the product or brand being endorsed has a significant effect on the effectiveness of the advert (Kamins, 1990). According to the match-Up theory, consumers tend to look for brands that match their values (Kamins, 1990). When a high level of fit is shown between the influencer and the brand, this not only validates the consumer's choice but also strengthens their emotional connection to the brand. This emotional connection is based on identification, which further enhances trust in the brand. Furthermore, consumers experience cognitive dissonance when confronted with information that does not match their expectations, which may reduce trust in the brand. Match-Up theory suggests that high fit influencer-brand combinations reduce this cognitive dissonance because they match consumers' expectations and beliefs about the brand, thus indirectly increasing brand trust. Therefore, Hypothesis 2 is proposed that there is a positive relationship between influencer-brand fit and consumers' perceived brand trust. 2.5. The Relationship between Audience Engagement and Consumer-Perceived Brand Trust Based on the previous literature review on audience engagement and consumers' perceived brand trust, it was found that there is an important relationship between audience engagement and consumers' perceived brand trust. Cheung et al. in their study emphasised the influence of social media comments on consumers' purchasing decisions, especially when these comments originate from consumers' their own social networks (Cheung et al., 2014). The number of likes, comments and shares on social media can be used as proof that the influencer has a high level of audience engagement, indicating the public's recognition and trust in the brand. High engagement (e.g., products with more likes) is often perceived as more popular and trustworthy, which increases consumers' willingness to purchase. Therefore, this study also indirectly implies that the higher the audience engagement, the higher the consumer perceived brand trust. 2.5.1. Social Capital Theory Social capital theory originates from the field of sociology is a theory that explains how individuals or organisations gain value and resources through relationships in social networks (Lin, 2017). This theory helps to identify the relationship between audience engagement and consumers' perceived brand trust. It argues that relationships in social networks can generate value and resources, including information, influence, and so on. Firstly, the accumulation of relational capital, also known as likes, comments and shares on content posted by influencers, directly increases consumer trust in the brand. Second, high audience engagement promotes the spread of positive word-of-mouth through positive social interactions. Word-of-mouth serves as a powerful social proof that further strengthens consumer trust in the brand. Finally, the sharing of cognitive capital makes it easier for consumers to emotionally connect with the brand and feel that the brand's philosophy fits with their personal values. This empathy enhances consumers' trust and loyalty to the brand. Therefore, Hypothesis 3 is proposed that there is a positive relationship between audience engagement and consumers' perceived brand trust. 22 3. Hypothesis Framework Figure 2. Hypothesis framework 4. Methodology 4.1. Research philosophy Positivism is the acquisition of knowledge based on observation and measurement, emphasising objectivity and verifiability (Popper, 2005). The purpose of this study was to objectively assess the pervasive impact of influencer marketing factors on brand trust. This quantitative phase of the research focuses on providing quantifiable evidence to support theoretical hypotheses. In contrast, constructivism believes that knowledge is constructed through people's interactions in a particular social and cultural context, emphasising the importance of individual experience and subjectivity (Berger & Luckmann, 2023). This qualitative phase focuses on gaining a deeper understanding and interpretation of the motivations, attitudes and feelings behind quantitative findings as well as exploring how individual consumers subjectively perceive and interpret influencer marketing and brand trust. In this case, the main study focuses on whether there are influences on consumers' perceived brand trust in the Chinese contextual culture. By combining positivist and constructivist approaches, the study can comprehensively analyse consumers' perceptions and feelings of brand trust from both objectivity and subjectivity perspectives. 4.2. Research Approach The research method chosen for this marketing proposal is explanatory sequential design, which, in contrast to convergent parallel design, allows the researcher to go deeper in response to the initial findings through qualitative research conducted after quantitative research, and is particularly suitable for exploring the Zhejiang Province electronics market's specific cultural and economic factors on brand trust (Creswell, 2014). Secondly, in contrast to exploratory sequential design, explanatory sequential design begins with quantitative analyses to provide direction for qualitative research (Creswell, 2014). This sequence is appropriate for first identifying generalised trends and then delving into and explaining the reasons behind these trends, especially when they are related to geographical and industry-specific factors. Compared to embedded design, explanatory sequential design places more emphasis on the interpretation and deepening of quantitative findings from qualitative data, which helps to provide more in-depth and nuanced insights into the electronics market in Zhejiang Province (Creswell, 2014). However, it also has limitations in that interpretive sequential design requires a phased approach to research, which can lead to longer research cycles and higher resource requirements than other designs, such as parallel convergent designs. 4.3. Time Horizon The time horizon will briefly identify the period in which the study took place (Melnikovas, 2018). This study examines the challenges of influencer marketing on consumer- perceived brands and needs to determine whether influencer credibility, influencer-brand fit and audience engagement are associated with consumer-perceived brands in the Chinese electronics market. In addition, a cross-sectional design is preferred for this study. However, this cross-sectional design only allows researchers to record evidence without intervening in the testing environment and helps to compare multiple variables simultaneously. However, the researchers were unable to understand the actual conditions of the study participants before and after the study. 4.4. Research Design 4.4.1. Quantitative Study (1) Research Hypothesis Hypothesis 1: There is a positive relationship between influencer credibility and consumer perceived brand trust. Hypothesis 2: There is a positive relationship between influencer-brand fit and consumer-perceived brand trust. Hypothesis 3: There is a positive relationship between audience engagement and consumer-perceived brand trust. (2) Sampling This study defines the stratified random sampling technique, which reduces sample selection bias and improves estimation accuracy compared to simple random sampling because it takes into account heterogeneity within the population (Lohr, 2021). Compared to other sampling methods, the study chose to sample from a wider range of the population, including different cities, age stages, and genders within Zhejiang Province. The researcher could evaluate different ranges of population samples to make the conclusions of the study more accurate. In recruiting participants, social media can be used to post survey invitations for younger consumers, while older people who do not use social media often need to be invited by distributing questionnaires in offline nursing homes. Krejcie & Morgan's research suggests that the sample size should range from 300- 500 (Krejcie & Morgan, 1970). (3) Research Tool In quantitative studies, the Likert scale, named by Rensis Likert, is a commonly used survey measurement tool to assess file:///C:/Users/Gang/Desktop/文章排版格式模板2024.3.15/4.4.1.1 file:///C:/Users/Gang/Desktop/文章排版格式模板2024.3.15/4.4.1.2 file:///C:/Users/Gang/Desktop/文章排版格式模板2024.3.15/4.4.1.3 23 the extent to which a respondent holds a certain opinion (appendix 1). The Likert scale in this study is divided into five levels, ranging From "strongly agree" to "strongly disagree". Firstly, the Influencer Credibility Scale is based on Ohanian's (1990) study, which constructed and validated a scale, a theoretical framework and scale that has been widely used to assess the credibility of social media influencers (Ohanian, 1990). Secondly, Kamins' Influencer-Brand Match Scale explores the 'matchmaking assumption' in celebrity advertising, i.e., the impact of the fit between the celebrity's image and the image of the product or brand he or she is endorsing on the effectiveness of the advertisement (Kamins, 1990). This concept also applies to the fit between social media influencers and brands. Thirdly, this research questionnaire references Hollebeek et al. who defined consumer brand engagement in the context of social media and developed and validated a corresponding scale. This work provides a theoretical basis for understanding and measuring consumer engagement in social media interactions with brands (Hollebeek et al., 2014). Finally, the Consumer Perceived Brand Trust Scale references the Commitment-Trust Theory proposed by Morgan and Hunt (1994), which emphasises the central role of trust in building lasting relationship marketing (Morgan & Hunt, 1994). (4) Data During the data preparation phase, data collection for this study collected primary data mainly through online questionnaires and offline distribution of questionnaires. This study entailed rigorous data cleaning measures, including dealing with missing values and outliers, and these methods were chosen to reflect conventional practices in social science research (Tabachnick & Fidell, 2011). In addition, data were coded and variables transformed to accommodate subsequent statistical analyses, a process that follows widely accepted practices in the social sciences (Creswell, 2017). This study focused on the statistical processing of performing descriptive statistics, reliability analyses, correlation analyses, and regression analyses using SPSS analytical tools. Descriptive statistical analyses provide an overview of the dataset, including measures of central tendency and dispersion of the sample, which is essential for understanding the characteristics of the sample in this study (Pallant, 2020). Reliability analyses assess the internal consistency of the scale by calculating Cronbach's alpha coefficient (Cronbach, 1951).Correlation and regression analyses are needed to validate the relationship between the independent variables (e.g., influencer trustworthiness, influencer brand fit, and audience engagement) and the dependent variable (consumer perceived brand trust). Hypothesis 2: There is a positive relationship between influencer-brand fit and consumer-perceived brand trust. Hypothesis 3: There is a positive relationship between audience engagement and consumer-perceived brand trust. 4.4.2. Sampling This study defines the sampling technique as purposive sampling, a sampling method that allows researchers to selectively recruit participants based on specific objectives and research needs. The study chose to sample from a specific population including active social media users, consumers of electronic products in different cities, age stages and genders within Zhejiang Province (Bethlehem, 2009). Compared to other sampling methods, researchers can also evaluate samples in purposeful sampling. This is because this sampling method is useful when researchers need to obtain a representative sample from a diverse population. In terms of recruiting participants, this study can be conducted by posting recruitment threads in a number of forums focusing on electronics as well as by setting up recruitment points through offline community centres. In their study, Guest et al. found that data saturation tends to begin to occur after interviews with 12 to 15 participants (Guest et al., 2006). This suggests that this range could provide sufficient depth of data for the study, whilst also taking into account resource and time constraints. (1) Research Tool In qualitative research, this study used semi-structured interviews, a tool proposed by Berelson (appendix2), where the researcher was able to gain insights into how consumers understand and evaluate influencer marketing and its impact on brand trust (Asheim, 1980). This qualitative approach reveals more about participants' personal experiences, emotional responses and complex decision-making processes than quantitative research. (2) Data In the data preparation phase, data collection for this study was mainly through forum posting of recruitment threads as well as primary data collection through offline community centres. Qualitative researchers have identified the problem and selected a targeted sample to gather knowledge to address the problem (Sekaran & Bougie, 2016). However, the data obtained can be unspecific or daunting due to the sheer volume of material that needs to be processed without adequate review before a decision can be made (Merriam, 2009).Braun and Clark propose six steps to help assess the data, create initial codes, scan for themes, identify and name themes, and analyse the data. These steps include assessing the data, creating initial codes, scanning for themes, identifying and naming themes, and writing a report (Creswell, 2013). Figure 3 below illustrates the review process. file:///C:/Users/Gang/Desktop/文章排版格式模板2024.3.15/4.4.2.3 24 Figure 3. Six-step thematic analysis procedure (Carroll, Booth, Leaviss and Rick, 2013) 5. Ethical consideration Sekaran & Bougie emphasise that ethical considerations should be taken into account when accessing raw data, and that researchers should value private consent, ensuring that it is only used for testing purposes and that details are not disclosed to unrelated third parties. Respondents' personal or invasive details should be respected and kept confidential; this is even more crucial (Saunders et al., 2016). The researcher made it clear that participants should not be asked to answer specific questions because their results were uncomfortable. Interviewers must inform participants that they should not express racist views or use abusive language while participating in the interview. On the other hand, applicants are encouraged to participate in the interview by answering truthfully and accurately the questions posed in the interview. 6. Limitation of the Study There are a number of limitations to the scope of this research proposal, including time, geography, and participants. These limitations may have an impact on the generalisability and replication of the research findings. Firstly, due to the cross-sectional design, the study data only reflects the situation at a specific point in time, which does not capture the trends and dynamics over time, limiting the observation and analysis of long-term effects. Secondly, due to limited resources, only respondents from Zhejiang Province in China participated, which may lead to geographical limitations of the study results, restricting the ability to generalise to other regions or countries. Based on these limitations, this study could have taken a number of steps to ensure rigour. Whilst this study utilised a cross- sectional design, the impact of temporal change on the results could be explored by introducing longitudinal data collection through a follow-up study. As far as possible, participants from different backgrounds, ages and genders were recruited for the sample to ensure that the findings are representative and comprehensive. References [1] Amos, C., Holmes, G. and Strutton, D. (2008) ‘Exploring the relationship between celebrity endorser effects and advertising effectiveness’, International Journal of Advertising, 27(2), pp. 209–234. doi:10.1080/02650487.2008.11073052. [2] Asheim, L. (1980) ‘Bernard Berelson (1912-1979)’ , The Library Quarterly, 50(4), pp. 407–409. doi:10.1086/601015. [3] Brown, D. and Hayes, N. (2008) ‘Evaluating influencer marketing’, Influencer Marketing, pp. 129–137. doi:10.1016/b978-0-7506-8600-6.50017-3. [4] Booth, N. and Matic, J.A. (2011) ‘Mapping and leveraging influencers in social media to shape corporate brand perceptions’ , Corporate Communications: An International Journal, 16(3), pp. 184–191. doi:10.1108/13563281111156853. 25 [5] Boerman, S.C. (2020) ‘The effects of the standardized Instagram disclosure for Micro-and Meso-influencers’ Computers in Human Behavior, 103, pp.199–207. doi:10.1016/j.chb.2019.09.015. [6] Berger, P. and Luckmann, T. (2023) ‘The social construction of Reality’, Social Theory Re-Wired, pp. 92–101. doi:10.4324/9781003320609-11. [7] Bethlehem, J. (2009) Applied survey methods [Preprint]. doi:10.1002/9780470494998. [8] Cheung, C.M.K., Xiao, B.S. and Liu, I.L.B. (2014) ‘Do actions speak louder than voices? the signaling role of social information cues in influencing consumer purchase decisions’ , Decision Support Systems, 65, pp. 50–58. doi:10.1016/j.dss.2014.05.002. [9] Cronbach, L.J. (1951) ‘Coefficient alpha and the internal structure of tests’, Psychometrika, 16(3), pp. 297–334. doi:10.1007/bf02310555. [10] Creswell, J. W. (2014) Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. 4th ed. London: SAGE. [11] De Veirman, M. et al. (2017) ‘Consumers’ motivations for lurking and posting in brand communities on social networking sites’, Digital Advertising, pp. 207–221. doi:10.4324/9781315623252-12. [12] Erdogan, B.Z. (1999) ‘Celebrity endorsement: A literature review’, Journal of Marketing Management, 15(4), pp. 291– 314. doi:10.1362/026725799784870379. [13] Eppler, M.J. and Mengis, J. (no date) ‘The concept of information overload-A review of literature from Organization Science, accounting, marketing, MIS, and relateddisciplines (2004)’, Kommunikationsmanagement im Wandel, pp. 271– 305. doi:10.1007/978-3-8349-9772-2_15. [14] Freberg, K. et al. (2011) ‘Who are the social media influencers? A study of public perceptions of personality’, Public Relations Review, 37(1), pp. 90–92. doi:10.1016/j.pub rev.2010.11.001. [15] Friedman, H.H. (1980) Endorser effectiveness as a function of product type. [16] Guest, G., Bunce, A. and Johnson, L. (2006) ‘How many interviews are enough?’ , Field Methods, 18(1), pp. 59–82. doi:10.1177/1525822x05279903. [17] Hollebeek, L.D., Glynn, M.S. and Brodie, R.J. (2014) ‘Consumer brand engagement scale’, PsycTESTS Dataset [Preprint]. doi:10.1037/t62246-000. [18] Hollebeek, L.D., Glynn, M.S. and Brodie, R.J. (2014) ‘Consumer brand engagement in social media: Conceptualization, scale development and validation’, Journal of Interactive Marketing, 28(2), pp. 149–165. doi:10.1016/j.int mar.2013.12.002. [19] Keller, K.L. and Lehmann, D.R. (2006) ‘Brands and branding: Research findings and future priorities’, Marketing Science, 25(6), pp. 740–759. doi:10.1287/mksc.1050.0153. [20] Kahle, L.R. and Homer, P.M. (1985) ‘Physical attractiveness of the celebrity endorser: A Social Adaptation Perspective’, Journal of Consumer Research, 11(4), p. 954. doi:10.1086/209029. [21] Kamins, M.A. (1990) ‘An investigation into the “match-up” hypothesis in celebrity advertising: When Beauty may be only skin deep’, Journal of Advertising, 19(1), pp. 4–13. doi:10.1080/00913367.1990.10673175. [22] Krejcie, R.V. and Morgan, D.W. (1970) ‘Determining sample size for research activities’, Educational and Psychological Measurement, 30(3), pp. 607–610. doi:10.1177/001316447003000308. [23] Liu, X. and Zheng, X. (2024) The persuasive power of social media influencers in brand credibility and purchase intention, Nature News. Available at: https://www.nature.com/articles/s41599-023-02512-1 (Accessed: 11 March 2024). [24] Lin, N. (2017) ‘Building a network theory of social capital’, Social Capital, pp. 3–28. doi:10.4324/9781315129457-1. [25] Lohr, S.L. (2021) ‘Stratified sampling’, Sampling, pp. 79–120. doi:10.1201/9780429298899-3. [26] Morgan, R.M. and Hunt, S.D. (1994) ‘The commitment-trust theory of Relationship Marketing’, Journal of Marketing, 58(3), p. 20. doi:10.2307/1252308. [27] Morgan, R.M. and Hunt, S.D. (1994) ‘The commitment-trust theory of Relationship Marketing’, Journal of Marketing, 58(3), pp. 20–38.doi:10.1177/002224299405800302. [28] Merriam, S. B. (2009) Qualitative Research: A guide to design and implementation. San Francisco: Jossey-Bass, a Wiley Imprint. [29] Melnikovas, A. (2018) 'Towards an Explicit Research Methodology: Adapting Research Onion Model for Futures Studies ', Journal of Futures Studies, 23(2), pp. 29-44. [30] Ohanian, R. (1990) ‘Source-credibility scale’, PsycTESTS Dataset [Preprint]. doi:10.1037/t66414-000. [31] 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. doi:10.1080/00913367.1990.10673191. [32] Popper, K. (2005) The logic of Scientific Discovery [Preprint]. doi:10.4324/9780203994627. [33] Pallant, J. (2020) SPSS survival manual [Preprint]. doi:10.4324/9781003117407. [34] Sekaran, U. & Bougie, R. (2016) Research Methods for Business.7th ed. West Sussex: John Wiley & Sons Ltd. [35] Saunders, M., Lewis, P., & Thornhill, A. (2016) Research Methods For Business Students.7th ed. Essex: Pearson Education Limited. [36] Thomala, L.L. (2023) China: Number of social media users 2027, Statista. Available at: https://www.statista.com/statistics/277586/number-of-social- network-users-in-china/ (Accessed: 10 March 2024). [37] Tabachnick, B.G. and Fidell, L.S. (2011) ‘Multivariate analysis of variance (MANOVA)’, International Encyclopedia of Statistical Science, pp. 902–904. doi:10.1007/978-3-642- 04898-2_394. [38] You, Xixi (2023) Influencer marketing in China, Croud. Available at: https://croud.com/en-gb/resources/influencer- marketing-in-china/ (Accessed: 10 March 2024).Zhang (2021) An overview of Zhejiang Province and its commerce development. Available at: https://zcom.zj.gov.cn/art/2018/2/26/art_1455367_15616953. html (Accessed: 10 March 2024). https://www.nature.com/articles/s41599-023-02512-1 https://www.statista.com/statistics/277586/number-of-social-network-users-in-china/ https://www.statista.com/statistics/277586/number-of-social-network-users-in-china/ https://croud.com/en-gb/resources/influencer-marketing-in-china/ https://croud.com/en-gb/resources/influencer-marketing-in-china/