Research on the Relationship between Characteristics of Video Bloggers and Consumers’ Purchase Intentions

Abstract

Video blogging channels for social media technology continue to innovate and evolve, and brands are beginning to rely on social media channels for product marketing. The factors influencing consumers’ purchase intention for products advertised by video bloggers are numerous and unclear. The main purpose of this study is to explore the mechanism by which the characteristics of video bloggers influence the formation of emotional cognition and then influences consumers’ purchase intention. Based on the SOR model, 518 questionnaires were administered to Chinese video blogging users. The AMOS 24.0 and SPSSAU 24.0 web tools were used to analyze the study data. Empirical results indicate that consumers’ perceived homogeneity, expertise, degree of emotional attachment to video bloggers, and perceived source confidence have a direct or indirect positive influence on purchase intentions. Evident advertising intention weakens the relationship between emotional attachment and purchase intention. This result is strategically important for a brand’s advertising efforts and market share and provides new research insights and frameworks for studying consumers’ advertised product purchase behaviors in the context of video blogging channels for social media.

1. Introduction

As social media technology advances, the video blogging industry continues to evolve. Vloggers create, record, and upload video logs to social media platforms [1]. For example, on social media platforms such as YouTube, many video bloggers have attracted a large number of followers by sharing their self-generated content [2]. This rapid growth in vlogging has led to the emergence and growth of video blog advertisements. Consumers view blogs as independent and reliable sources of information, which sets them apart from traditional advertising platforms [3,4]. Therefore, in this context, brand marketing communication strategies are often implemented using blogs. Brands and companies are increasingly abandoning traditional celebrity endorsements, and the marketing focus is evolving to shift to people with some influence on social media [5]. More and more companies are adopting this new form of online advertising [1]. Advertising is incentivized by sponsors or brands, providing some benefits to video bloggers [6]. Video bloggers have become bridges between consumers and product sellers by promoting products through brand placement, product reviews, in-depth video experiences, and other forms of marketing [7]. For product marketers and brands, video bloggers with a certain fan base play an important marketing role, creating a “celebrity effect” [8]. Mainolfi and Vergura [3] claim that a blogger’s recommendation of a company’s products can attract and convince consumers, resulting in a great advertising effect.

The concept of video blog advertising has been around for a long time, and scholars [7,9,10] have been slowly investing in this direction in light of the trends in video blog advertising. One scholar found that bloggers’ social media interactions mediate the credibility of such product recommendations, which has a positive impact on brand trustworthiness, thereby enhancing consumers’ purchase intentions [11]. Meanwhile, homophily and expertise have a significant direct effect on social media interactions. Most previous studies [10,12,13,14] have supported that source attractiveness and source expertise are two critical factors affecting the effectiveness of endorsement as well as the strength of social media interactions in the endorsement process; homophily is closely related to source attractiveness and source expertise, so two celebrity characteristics, homophily and expertise, are important predictors of blogger endorsement [15]. Given previous research, we used homogeneity and expertise as our key independent variables. In addition, Munnukka et al. [16] found that video bloggers’ credibility affects product impact and consumer purchase intent. The perceived usefulness and credibility of bloggers’ recommendations significantly affect blog viewers’ online shopping attitudes and intentions [17]. In the beauty product industry, research has demonstrated the significant impact of attitudes and values, which are key elements of homophily, in shaping emotional attachment and the popularity of video bloggers. This popularity greatly influences viewers’ purchasing decisions regarding recommended products, while expertise surprisingly holds no sway over their preferences for these bloggers [2]. Finally, previous studies in the field of video blogging have mainly considered the characteristics of video bloggers, mainly their cognitive variables. Rana [13] discusses the effect of homophily and expertise on purchase intention; Chiu and Ho [18] found that source credibility and endorser-product match affected Gen Z’s purchase intention of endorsing products, there is an indirect relationship between attitude homophily, physical attractiveness and social attractiveness with the purchase intention [19], but lack the exploration of advertising intention. In a study of traditional text blogs, Uribe et al. [4] discussed the role of advertising intention, but there is little research in the field of video blogs. The discussion of advertising intention is a necessary condition for discussing the purchase intention for advertising products. At the same time, when defining the factors of decision-making processes, it is necessary to take into account emotional and affective states in order to avoid the application of limited and biased objective models; the study of consumers’ purchase intention can not only involve the rational cognition of consumers because people are not always rational, there will be emotional behavior, and consumer behavior is the combination of rational thinking and emotional influence [20]. Therefore, based on the SOR model, this study introduced two consumer emotion variables to study the impact of consumers’ rational cognition of vloggers’ characteristics on their emotional perception formation, forming a complete research mechanism different from previous relevant studies.

Earlier research has indicated that academics employ various methodologies to evaluate or assess consumers’ willingness to purchase blog-advertised products. The classical Aristotelian persuasion theory [21], the theory of parasocial relationship [19], the theory of electronic word-of-mouth (eWOM) [13], and the theory of technology acceptance modeling (TAM) [22] are among the frameworks that have been widely used. In reviewing the related literature, we evaluated the most extensive and frequently studied variables and selected the variables for this study based on their relevance and expected validity for research purposes. The SOR model was chosen as the framework for this study because the SOR theory analyzes behavior in which the environment stimulates an individual to produce internal changes that lead to behavior [23,24]. Specifically, the SOR model is a theoretical model that interprets individual behavior as a learning response to an external stimulus, consisting of three components: stimulus, organism, and response [25]. Stimulus refers to external environmental factors. An organism is a person’s emotional and cognitive state, including the entire process of intervening in response to a stimulus, transforming that stimulus into meaningful information that further assists them in making decisions [24]. Among them, affective states reveal the psychological emotions induced by the stimulus and the feelings and emotions expressed under the stimulus. The cognitive state refers to all the information acquired, processed, retained, and retrieved in people’s consciousness, such as experience, feeling, and thinking activities. The response is a behavioral response that represents an individual’s ultimate response to a particular stimulus [26,27], which is consistent with the research scenario, research variables, and research framework of this study. This study examines how homophily and expertise affect consumers’ emotional attachment to vloggers and perceived source credibility (organism), which leads to consumers’ purchasing behavior (response). In a study of traditional blogs, Uribe et al. [4] found that blogs’ sponsorship information may have a certain negative impact on stimulating consumers’ willingness to buy, and sponsorship information includes more positive information about products or services. Consumers are usually skeptical and critical of messages with commercial intent; therefore, sponsored messages may have a negative impact on persuading them to buy [4]. Therefore, this paper introduces the variable of advertising intention into the research in the field of video blogging and combines it with previous related variables and studies as a key moderator in the study of video blogger advertisement product purchase intention.

Based on the limited research on the product purchase intention of video blog advertisements, this study adds key factors targeting advertisement behavior and cascades important influencing factors from previous related studies, including homophily, expertise, perceived source credibility, and emotional attachment. Structural equation modeling was established to comprehensively investigate the determinants shaping consumers’ purchase intentions toward products advertised on video blogs. The mature scale of the research questionnaire developed by previous scholars was utilized to collect relevant data. Subsequently, the hypotheses affecting video bloggers’ willingness to purchase advertising products are tested through modeling and quantitative analysis, and conclusions are drawn.

2. Literature Review and Research Hypotheses

2.1. Homophily and Consumer Purchase Intention

Homophily refers to the extent to which individuals who interact with one another share similar social statuses, beliefs, and so on [28]. This study utilized McCroskey’s conceptual framework, viewing homophily as a four-dimensional structure that includes morality, appearance, attitudes, and background [29]. We drew here on one of the most utilized and influential dimensions, attitudes, for our study. The tendency to seek advice on social networking sites and consumer empathy is affected by homophily. Homophily, reflecting the cognitive and emotional connection of consumers to the celebrity, is the most likely predictor of consumer–celebrity parasocial interaction [15]. It can explain how consumers reflect on user-generated content, content from influencers, or content from celebrities (e.g., video bloggers). Similarly, similarity inspires trust and attachment [2]. When consumers perceive a connection between themselves and endorsers, such as shared interests, values, or traits, they are more likely to embrace endorsers’ attitudes, behaviors, and beliefs [5]. In other words, consumers tend to mirror the consumption choices of online celebrities who share similar values by purchasing products they endorse and recommend. Scholars have found that homophily sources play a vital role in influencing consumers’ purchase intentions and their willingness to adopt product information [2,13,30]. Therefore, homophily with a video blogger directly influences consumers’ willingness to make a purchase. Meanwhile, one study found that quasi-social interactions had a significant positive correlation with attitude homophily [10]. The degree of perceived alignment among consumer name culture, values, and other indicators strongly predicts quasi-social relationships [30]. Additionally, the perception of similarity influences how online shoppers view trustworthiness, social presence, ease of use, usefulness, level of interactive enjoyment, and intention to reuse [2]. In accordance with the concept of social comparison theory, individuals have a tendency to compare their skills and attitudes with those of others. When individuals notice similarities between themselves and others, they tend to assume that they have similar needs and preferences [31]. Bu et al. [30] found that consumers are more prone to encountering increased levels of interaction, trust, and comprehension with individuals who share similarities with them. Consumers tend to prioritize information from individuals with similar views, experiences, and preferences as they find it more useful, and recipients of information find similar sources more persuasive than those who do not appear similar [32]. Therefore, consumers find recommendations from video bloggers who share similar interests, attitudes, and preferences to be more influential as they align with their own [14]. Similar arguments have been made that consumers’ purchase intentions can be stimulated through homophily, which deepens their sense of emotional trust and enhances their perception of emotional value [22]. Therefore, this study proposes the following hypothesis:

Hypothesis H1a. 

Homophily in video bloggers is positively associated with consumers’ purchase intentions.

2.2. Expertise and Consumer Purchase

Expertise is based on how consumers perceive a source’s knowledge, abilities, or specialized knowledge in a specific field [33]. The perception of an endorser’s expertise is more important than their actual knowledge and skills [34]. Previously, consumers were accustomed to being motivated by the appearance, attractiveness, and charisma of celebrities, but today, consumers prefer expertise, competence, and knowledge of celebrities. Therefore, companies are widely utilizing celebrity expert endorsements to direct consumers’ attention toward impulse buying [35]. Expertise in electronic word-of-mouth sources increases their persuasive effects. One scholar summarized 20 other articles in his study, and the results showed that source expertise has a significant effect on the perceived usefulness and credibility of eWOM, purchase intention, and information adoption [13]. The importance of celebrity expertise is emphasized by the celebrity endorsement model, which asserts that consumers are motivated to make purchases based on the perceived expertise of the celebrity, leading to more thorough buying behavior. This suggests that video bloggers’ expertise positively affects customers’ purchase intentions [36]. In recent research, this conclusion has been reaffirmed by scholars: specialized knowledge plays an important role in predicting purchase intentions [12,37,38]. In addition, the literature suggests that celebrities are more persuasive when they possess a higher level of expertise in convincing customers to make decisions [39]. The expertise of the influencer will result in trust in the relationship [12].In the eyes of consumers, the expertise of celebrity endorsers has been shown to have an impact on enhancing consumer purchasing behavior. Video bloggers are regarded as influential when they possess the expertise to enhance consumer confidence in a specific product [35]. If consumers view a source of information as well informed, they will trust that the source is better able to provide useful information for evaluating service performance and quality. Consumers increasingly rely on expert reviews as a trusted source of information when making purchase decisions, which are more likely to influence their behavior [14]. Hence, the following hypothesis is proposed:

Hypothesis H1b. 

Video bloggers’ expertise is positively associated with consumers’ purchase intentions.

2.3. Mediating Effects of Perceived Source Credibility

The trustworthiness, truthfulness, and lack of bias perceived by consumers in a source of information refer to source credibility. The credibility of a blogger is measured by the consumer’s belief that the blogger “knows everything” about the content of the video he or she is making and perceives the blogger’s advice as positive, honest, and trustworthy [19]. Relevant research on eWOM suggests that its effectiveness in influencing consumer attitudes toward products and consumers’ intentions to make a purchase are impacted by the reliability and trustworthiness of the endorser [40,41]. Arguably, recipients of information will take heed of the communicated messages if they believe that the source is credible. Hence, the sources’ credibility could affect the perceptions of the usefulness of the communicated content [42]. Bloggers’ credibility positively influences consumer attitudes toward advertisements and brands, indicating that consumers are more prone to trust and believe information from marketing sources that are perceived as credible [13]. Consumers are less skeptical about the credibility of video content when they view the source of information as reliable. Information on the Web is often heterogeneous. Therefore, consumers typically assess the reliability of the publisher before accepting or dismissing the provided information. Consumers perceive the information provided as useful when it is sourced from highly credible publishers, and they are more inclined to accept the recommendations provided [13]. Therefore, we propose the following hypotheses:

Hypothesis H2a. 

Perceived source credibility mediates the relationship between consumers’ perceptions of video bloggers’ homophily and consumers’ purchase intentions.

Hypothesis H2b. 

Perceived source credibility mediates the effect of video bloggers’ expertise on consumers’ purchase intentions.

Hypothesis H2c. 

The mediating effect of perceived source credibility between video bloggers’ expertise and consumers’ purchase intentions is greater than the mediating effect between video bloggers’ homophily and consumers’ purchase intentions.

2.4. Mediating Effects of Emotional Attachment

Attachment theory suggests that when consumers have prior knowledge of a product endorser, it is probable that they will form an emotional connection with both the endorser and the product, which has an impact on their purchasing intent [26]. Because of the presence of emotions, most decision-making processes take place unconsciously and, in most cases, within a relatively short time period [20]. Individuals’ attachment experiences influence their thoughts, interactions with others, and their actions [43]. Feelings and emotional states play a significant role in shaping consumer behavior, including motivation, judgment, and purchase decisions. Emotions serve as a driving factor that encourages consumers to form a connection with a product or brand [2]. Social media facilitates a sense of greater proximity and psychological closeness to video bloggers, and the closeness that consumers may feel toward video bloggers also leads to emotional connections or strong feelings toward them [44]. The persuasive power of video bloggers over consumers can be attributed to the emotional attachment they foster [45]. Many studies have shown that consumers often form emotional bonds with celebrities and social media influencers, and that consumers connect with endorsers through emotional attachment to generate positive attitudes that directly affect purchase intentions [2,18,46]. Additionally, some scholars have concluded that recognizing the importance of positive emotions toward celebrities and further exploring the dynamics of fan-celebrity relationships in order to facilitate successful celebrity endorsements [47,48,49]. Therefore, this study proposes the following hypotheses:

Hypothesis H3a. 

Emotional attachment mediates the relationship between consumers’ perceptions of video bloggers’ homophily and consumers’ purchase intentions.

Hypothesis H3b. 

Emotional attachment mediates the relationship between video bloggers’ expertise and consumers’ purchase intentions.

Hypothesis H3c. 

The mediating effect of affective attachment between video bloggers’ homophily and consumers’ purchase intention is greater than the mediating effect between video bloggers’ expertise and consumers’ purchase intention.

2.5. Mediating Effects of Advertising Intention

Advertising intention refers to the clarity of the content of an advertisement in terms of its willingness to induce purchasing behavior [50]. Generally speaking, the content of the advertising message may take the form of a “hard sell” or “soft sell” style [50]. Some video bloggers market their products as regular video content updates to their accounts, which makes it difficult to differentiate them from regular videos (unclear branding and unclear advertising intent). Marketing with excessive advertising intent can create skepticism and negative feelings toward videos among consumers [51]. In other words, the ways in which video bloggers market advertisements vary, with content designed to be more effective when it has implicit advertising intent than when it has explicit advertising intent. The absence of strong advertising intent and maintenance of impartiality are key factors in improving content credibility (ad credibility) [52]. Consumers who watch video content make certain assessments of the content, including credibility, accuracy, and whether there is an ulterior motive for the video content [53]. Theoretically, video blog viewers require that the source of the video blog be credible and that the intent of the behavior be positive and unbiased. The value of the message conveyed by a video blog can be enhanced by its positive characteristics; on the contrary, the lack of credibility of the source. Consumers carry a degree of resistance if they perceive the video content as an advertisement as suspicious [6]. This also suggests that it is difficult for consumers to generate relevant behavioral intentions toward an advertised product when video content advertising intentions are clearly expressed. In contrast, if the video content or a certain type of advertising placement does not explicitly express the advertising intent, the advertising intent is implicit. Therefore, we can see that there are only explicit and implicit advertising intentions. If video bloggers do not show persuasive intentions to consumers, they are implicit and can avoid arousing consumers’ suspicion and resistance. In a past study, the effect of advertising intent on the effectiveness of blog advertising (the act of promoting a product or service through persuasive messages embedded in blog posts or website banners is what we mean by the term) was examined to examine consumer buying attitudes and behaviors [4]. In addition, one scholar found that implicit brand advertisements inserted in games produce more favorable brand attitudes than obvious brand advertisements for low-involvement products [54]. From the above, obvious advertising intent in blog content tends to reduce the perceived credibility of the source, leading to lower consumer purchase behavior intentions than when implicit advertising intent is present [4]. When influencers provide explicit advertising intention as opposed to subtle advertising intention, consumers are likely to infer that influencers have a less sincere intent to recommend products. As a consequence, consumers are less likely to purchase the advertised brand items. Influencer marketing is based on the relationship and sense of credibility that exists between influencers and followers [55]. Thus, this research study suggests the subsequent hypotheses:

Hypothesis H4a. 

Advertising intention weakens the effect of perceived source credibility on consumers’ purchase intentions.

Hypothesis H4b. 

Advertising intention weakens the effect of emotional attachment on consumers’ purchase intentions.

The proposed research conceptual model is shown in Figure 1.

3. Research Design

3.1. Sample and Data Collection

Before the formal preparation of the questionnaire, 10 video blogging channel users were randomly selected for the interview. Of the 10 interviewees, 7 were women, and 3 were men. Of those aged between 20 and 25, 6 have bachelor’s degrees, and 4 have master’s degrees. All 10 interviewees are deep users of video blogs and have purchased advertising products from video bloggers. Based on the SOR model, the paper summarizes the key variables of the video platform advertising products mentioned by the interviewees, as shown in

Table 1. Statistics of variables mentioned by respondents.

Interviewer Number	The Appearance of the Vlogger	The Vlogger’s Resemblance to Myself	Whether Vlogger Can Be Trusted	The Level of Professionalism of the Vloggers	The Marketing Intent of the Implant
1	√	√	√	√	√
2		√	√	√	√
3		√	√	√	√
4			√	√	√
5		√	√	√	√
6	√	√	√		√
7			√
8		√	√	√	√
9		√	√	√	√
10			√	√	√
Interviewer Number	Whether the Vlogger Has Enough Interaction with Me	How Much I Love Vlogger	The Personality of the Vlogger Is Good or Bad	Whether Advertising Products Have a Great Role in Me
1	√	√	√
2		√		√
3		√
4	√	√		√
5		√		√
6		√		√
7	√	√		√
8		√
9
10			√

Note: “√” indicates that respondents think that they will consider when buying vlogger advertising products.

, and determines the final research variables combined with the high-frequency variables previously studied. Firstly, we created a pre-survey questionnaire. Based on the respondents’ understanding of the measurement items and suggestions for improvement, we revised the language expression. In subsequent formal surveys, respondents rarely did not understand, dislike, or avoid the measurement items. This study was conducted from early May to mid-May 2023 through the questionnaire design from the “Questionnaire Star website”; the survey was conducted through the WeChat, Little Red Book (www.xiaohongshu.com, 10 October 2023) and MicroBlog websites to collect data related to the topic from Chinese video blogging users. This study focused on consumers who had purchased products advertised by video bloggers, and the data collected were only for this dissertation. The survey was conducted anonymously, and the confidentiality of the respondents’ personal information was maintained. The survey collected 621 video blogging users who completed the questionnaire; there were 56 questionnaires with an option selected for all scales, 14 repeated responses, and 33 too-long responses. Five hundred eighteen valid questionnaires were retained for the next step in the analysis.

Table 2. Demographic profiles of the respondents.

Demographics		Frequency	Percentage
Gender	1. Male	141	27.20
	2. Female	377	72.80
Age	1. Below 18	12	2.30
	2. 18–25	436	84.20
	3. 26–30	58	11.20
	4. 31–40	4	0.80
	5. 41–50	4	0.80
	6. 51 and over	4	0.80
Education	1. High school and below	24	4.60
	2. Technical secondary degree	21	4.10
	3. Undergraduate degree	325	62.70
	4. Master’s degree or above	148	28.60
Household income	1. Below 5000	93.00	18
(CNY)	2. 5000–10,000	186.00	35.9
	3. 10,000–20,000	158.00	30.5
	4. Over 20,000	81.00	15.6

Gender: 1 = male, 2 = female. Age: 1 = Below 18 years’ old, 2 = 18–25 years’ old; 3 = 26–30 years’ old; 4 = 31–40 years’ old; 5 = 41–50 years’ old; 6 = over 50 years’ old. Education: 1 = High school and below; 2 = Technical secondary degree; 3 = Undergraduate degree; 4 = Master’s degree or above. Household income (CNY): 1 = Below 5000, 2 = 5000–10,000; 3 = 10,000–20,000; 4 = Over 20,000—UCLI, upper level of the 95% confidence interval.

provides an outline of the statistical summary of the fundamental demographic information of individuals who participated in the questionnaire.

More than 70% of the respondents identified as female (72.80%). Meanwhile, almost 95% of the participants fell within the age range of 18 to 30. Of the respondents, 62.70% held a bachelor’s degree. Of the respondents, 66.4% earned between CNY5000 and CNY20,000 per month. According to the 2022 CNNIC report, most Chinese short video users are aged 18–44 years, and the number of survey respondents in this study who are aged 18–50 years accounts for more than 50%, accounting for a large proportion of the survey respondents. Based on the descriptive analysis of a previous study, it was found that 73% of the survey respondents have a bachelor’s degree or higher [2]. In addition, in previous relevant studies, it can be seen from the descriptive statistics of respondents that most of them are young people with a high education level, and the majority of them are female, which is consistent with our sample [1,3,11]; therefore, the data in this paper are representative.

3.2. Measures

All scale item groupings in the questionnaire were obtained from the original academic study questionnaire scales. A five-point Likert scale was used for all measures (1 = low degree, 5 = high degree).

Homophily was measured using three question items from the scale of Riadh Ladhari et al. [2]. An example of one of them is “My favorite vlogger thinks like me”. The level of expertise was assessed by employing four items derived from the scale developed by Riadh Ladhari et al. scale, to each of which we added: “for advertising products”. One example is: “For advertising products, my favorite vlogger is an expert”. The extent of emotional attachment was assessed using three items adapted from the study conducted by Riadh Ladhari et al., each of which took the form of “My XX level for/with favorite video bloggers”. An example of one of these items is “My association level with favorite video bloggers”.

Perceived source credibility was measured using four question items from a scale by Chung, Set al. [11]. An example of this measure is, “My favorite video blogger is sincere”.

Purchase intention was measured using three items from the scale developed by Sally Rao Hill et al. [1]. An example of one of these items is: “It is very likely that I will buy the advertised product in this vlog”.

Advertising intention was measured using three items adapted from the scale by Chan et al. [50]. An example of one of these items is “The fact that my favorite video blogger inserted an advertisement into the video showing the product”.

Control variables. To conduct the tests for mediation and moderation, we controlled for participants’ gender and age. Previous academic research suggests that gender and age may have different starting points for being persuaded to buy, thus producing differences in the willingness to buy a product [21,56].

4. Data Analysis and Results

4.1. Measurement Reliability and Validity

The AMOS 24.0 and SPSSAU web tools were used to analyze the study data. The CFA results showed that the measurement model had a good overall fit, as indicated by the acceptable fit indices, including the normed ${{\displaystyle \chi }}^2$ (466.806) (normed chi-square, designed to reduce the impact of sample size on chi-square values. The smaller the chi-square degree of freedom ratio, the higher the model fit, preferably less than 5, but not less than 1, indicating that the researcher has overcorrected the model) = 3.01, GFI (goodness-of-fit index, ranges from 0 to 1, and the larger the number of samples, the larger the GFI value. The closer the GFI value is to 1, the higher the model fit. GFI > 0.9 indicates that the model has a good model fit) = 0.915, CFI (comparative fit index, whose value is between 0 and 1, the closer to 0 indicates the worse the fit and the closer to 1 indicates the better the fit. It is generally believed that the model fits well if CFI is ≥0.9) = 0.955 and TLI (Tuck−Lewis index, ranges from 0 to 1. The closer to 0 indicates the worse the fit and the closer to 1 indicates the better the fit) = 0.944, Both CFI and TLI rely on chi-squares and degrees of freedom for model estimates, which measure the gap between the model’s assumptions and the actual data, and RMSEA (Root mean square error of approximation, which is a missing degree index, and the larger the value, the lower the fitting degree between the hypothesis model and the data) = 0.062, suggesting that the model performed well.

Table 3. Confirmatory factor analysis.

Construct	Items	Standard Loading	Cronbach’s Alpha Value	Composite Reliability	AVE
Purchase intention (PI)	PI1	0.79	0.88	0.88	0.64
	PI2	0.75
	PI3	0.79
	PI4	0.87
Advertising intent (AI)	AI1	0.75	0.75	0.75	0.60
	AI2	0.80
Perceived source credibility (PSC)	PSC1	0.83	0.92	0.92	0.75
	PSC2	0.87
	PSC3	0.88
	PSC4	0.88
Emotional attachment (EA)	EA1	0.81	0.83	0.83	0.62
	EA2	0.78
	EA3	0.77
Homophily (H)	H1	0.71	0.80	0.81	0.59
	H2	0.85
	H3	0.74
Expertise (E)	E1	0.76	0.87	0.87	0.63
	E2	0.86
	E3	0.78
	E4	0.78

Note: AVE means average variance extracted.

presents the results of the reliability analysis, discriminant validity, and convergent validity tests conducted on the constructs; Cronbach’s alpha values ranged between 0.75 and 0.92, both above the required threshold of 0.70 [57]. The composite reliability values were above the required threshold (0.70) [57]. These findings indicate that the measurements exhibited high internal consistency and reliability. All standard loadings were above the recommended threshold of 0.70 [58], ranging from 0.71 to 0.88. The AVE values ranged from 0.59 to 0.75, indicating that the measurements exceeded the required threshold of 0.5 [59]. This implies that the convergent validity of the aforementioned measurements is satisfactory. The results of the Confirmatory Factor Analysis (CFA) can be found in Table 3.

The square roots of the AVE values are listed in

Table 4. Correlations.

	H	E	EA	PSC	AI	PI
H	0.77
E	0.64	0.80
EA	0.63	0.58	0.79
PSC	0.57	0.61	0.66	0.86
AI	0.55	0.62	0.62	0.66	0.78
PI	0.50	0.54	0.56	0.57	0.54	0.80

Note: H, homophily; E, Expertise; EA, Emotional attachment; PSC, Perceived source credibility; AI, Advertising intent; PI, Purchase intention.

. To ensure discriminant validity, the correlation between each pair of constructs was compared with the square root of the Average Variance Extracted. The analysis revealed strong discriminant validity among the constructs, as the square roots of the AVE were larger than the correlations between the constructs [60]. Thus, the measure demonstrated strong convergent validity. Table 4 displays the correlation coefficients for all study variables, excluding demographic factors. Therefore, this study exclusively uses one source of data; it was crucial to examine potential issues related to Common Method Variance (CMV) in this study. We employed the approach recommended by Harman [61] for all measures loaded with an additional factor. The results indicated that the influence of Common Method Variance (CMV) was minimal, as the variance accounted for by a single factor was below the required threshold of 50%.

4.2. Hypothesis Testing

4.2.1. Structural Model

AMOS is a commonly used Structural Equation Modeling (SEM) software for building and analyzing complex statistical models. The core principle of Amos is to evaluate the relationship and influence of variables through path analysis, factor analysis, and regression analysis. We analyzed the Structural Equation Regression relationships using the AMOS 24.0 software package. To minimize the likelihood of multicollinearity (the model estimation is distorted or difficult to estimate accurately due to the existence of precise correlation or high correlation among explanatory variables), all research variables were centered on the mean when analyzing data. The overall goodness-of-fit indexes for the structural model are ${{\displaystyle \chi }}^2$ (450.592) = 3.52, GFI = 0.910, CFI = 0.948, TLI = 0.937, and RMSEA= 0.07, suggesting that the model performed well. Hypothesis 1 proposed that video bloggers’ homophily (1a) and expertise (1b) were positively related to consumer purchase intentions. As depicted in the first model in

Table 5. Path analysis results of the direct and indirect effects (n = 518).

Dependent Variables	Model 1		Model 2
	Purchase Intention		Emotional Attachment		Perceived Source Credibility		Purchase Intention
Measures	B (SE)	p	B (SE)	p	B (SE)	p	B (SE)	p
a constant (math.)	1.168 **	0	0.561 *	0.018	0.918 **	0	0.831 **	0.001
Gender	−0.201 **	0.002	0.053	0.373	−0.095	0.106	−0.192 **	0.001
Age	0.067	0.17	0.016	0.726	0.042	0.351	0.054	0.238
Place of Residence	0.023	0.729	0.027	0.663	0.001	0.988	0.016	0.79
Educational level	0.011	0.776	−0.009	0.81	0.05	0.175	0.002	0.952
Monthly household income	0.002	0.954	0.038	0.188	0.046	0.105	−0.017	0.55
Homophily	0.283 **	0	0.452 **	0	0.296 **	0	0.112 *	0.025
Expertise	0.382 **	0	0.309 **	0	0.426 **	0	0.214 **	0
Perceived source credibility							0.225 **	0
Emotional attachment							0.232 **	0
R2	0.351		0.45		0.436		0.433
Mediating effects					Indirect effect		LLCI	UCLI
homophily → Perceived source credibility → Purchase intention					0.067 (0.001)		0.027	0.108
Homophily → Emotional attachment → Purchase intention					0.105 (0.000)		0.053	0.155
Expertise → Perceived source credibility → Purchase intention					0.096 (0.000)		0.041	0.146
Expertise → Emotional attachment → Purchase intention					0.071 (0.001)		0.033	0.115

* p < 0.05, ** p < 0.01.

, the results show that video bloggers’ homophily is positively related to consumers’ purchase intention (β = 0.283, p < 0.001) as is video bloggers’ expertise (β = 0.382, p < 0.001). Therefore, hypotheses 1a and 1b received empirical support. The contributions of the direct effects model to the variance in participants’ purchase intentions were 39.58% and 56.02%, respectively.

4.2.2. Mediating Effect

We used the SPSSAU web tool to verify the mediation effect [62]. We tested the mediating role of perceived source credibility and emotional attachment on homophily, expertise, and consumers’ purchase intentions using Model 2. As shown in Table 5, the findings demonstrate that perceived source credibility plays a positive mediating role between homophily and consumers’ purchase intention (indirect effect = 0.067, 95% CI = [0.027, 0.108]), supporting Hypothesis 2a. Perceived source credibility positively mediated between expertise and consumer purchase intention (indirect effect = 0.105, 95% CI = [0.053, 0.155]), thus supporting Hypothesis 2b. In addition, the mediating role of perceived source credibility between expertise and consumers’ purchase intention was more significant than the mediating role between video bloggers’ homophily and consumers’ purchase intention, supporting Hypothesis 2c.

Meanwhile, the findings of model 2 demonstrated that emotional attachment acted as a mediator in the positive association between homophily and consumers’ purchase intention (indirect effect = 0.105, 95% CI = [0.053, 0.155]), supporting Hypothesis 3a. In addition, we discovered that emotional attachment played a mediating role in the positive correlation between expertise and consumers’ purchase intention (indirect effect = 0.071, 95% CI = [ 0.033, 0.115]). Therefore, Hypothesis 3b is supported.

Further comparison of the mediating role of emotional attachment in the relationship between homophily, expertise, and consumers’ purchase intention revealed that the mediating effect of affective attachment between video bloggers’ homophily and consumers’ purchase intention was greater than the mediating effect between video bloggers’ expertise and consumers’ purchase intention. The parallel mediation model explains the variance in consumers’ purchase intentions. Consequently, it can be concluded that Hypothesis 3c is upheld. Figure 2 shows the standardized coefficients for the overall model.

4.2.3. Moderating Effect

The moderating effect was also analyzed using the SPSSAU web tool. Hypothesis 4a suggests that the explicit intent of an advertisement weakens the effect of perceived source credibility on consumers’ purchase intentions. The results indicated that advertisement intention had a significant negative effect on consumer purchase intention (β = −0.281, t = −5.876, p < 0.01) and that the interaction term between perceived source credibility and advertisement intention did not yield a significant impact on consumers’ purchase intention (β = −0.034, t = −0.901, p = 0.368 > 0.05). These results suggest that the obvious intention of an advertisement has a direct effect on consumers’ purchase intention, whereas the relationship between perceived source credibility and consumer purchase intention remains unaffected by any moderating factor. Therefore, Hypothesis 4a was not confirmed.

Hypothesis 4b suggests that the obvious intention of the advertisement weakens the effect of emotional attachment on consumers’ purchase intention and has a negative moderating effect. Consumers’ purchase intention was found to be significantly impacted in a negative manner by advertising intention (β = −0.290, t = −6.398, p < 0.01); the presence of a significant negative effect on consumers’ purchase intention was identified when considering the interaction between emotional attachment and advertising intention (β = −0.072, t = −2.032, p < 0.05). These results indicate that advertising intention has a direct effect on consumers’ purchase intentions and negatively moderates the relationship between emotional attachment and consumer purchase intentions. Figure 3 illustrates the moderating effect, showing that the strength of the relationship between emotional attachment and consumers’ purchase intentions varies when the level of explanation of advertising intentions is one standard deviation above or below the mean. Thus, Hypothesis 4b is confirmed.

5. Research Discussion

5.1. Conclusions

We have basically achieved the expected research objectives. Through data analysis, this study draws the following conclusions for vlogging platform users living in China: homophily and expertise of video bloggers are positively correlated with consumers’ purchase intention. The positive mediating effect of perceived source credibility between expertise and consumers’ purchase intention is more significant than that between vlogger homophily and consumers’ purchase intention. The positive mediating effect between homophily of video bloggers and consumers’ purchase intention is greater than that between expertise of video bloggers and consumers’ purchase intention. All mediating effects were significant. Advertising intention directly and negatively affects consumers’ purchase intention; however, the relationship between perceived source credibility and consumers’ purchase intention is not regulated by advertising intention. At the same time, advertising intention negatively regulates the intensity of the relationship between emotional attachment and consumers’ purchase intention.

5.2. Discussion

First, the homophily of video bloggers was directly linked to consumers’ intentions to make a purchase. This validation is consistent with previous academic research. For example, in a study of male and female impulse shopping, Dittmar et al. found that consumers tend to emulate online celebrities of comparable value to themselves and buy the goods they use and recommend [63]. Homophily can positively influence consumers’ purchase intentions by strengthening their perceptions of emotional value and emotional trust [64].

Second, video bloggers’ expertise significantly affects consumers’ purchase intentions for advertised products. This study’s findings align with previous research, underscoring the substantial link between expertise and consumers’ purchase intentions [36]. In the context of e-commerce, consumers’ trust and purchase intentions were positively influenced by the expertise of the speakers. In this study, video bloggers were found to influence consumers’ purchase decisions by infecting them with expertise and mobilizing their emotional attachment and perceived trustworthiness.

Third, this study found that the mediating effect of perceived source credibility between video bloggers’ expertise and consumers’ purchase intentions was more pronounced than the mediating effect between video bloggers’ homophily and consumers’ purchase intentions. This is consistent with previous research [37]. Overall, when consumers perceive that the video blogger they watch is more specialized, it enhances their perceived trust in the video blogger, and the more they trust the product they are promoting, directly promoting consumers’ willingness to purchase the product [13].

Fourth, when emotional attachment was used as a mediating variable, the homophily of the video blogger, compared to the video blogger’s expertise, caused consumers to feel stronger empathy for the video blogger, generating shared emotions and stronger connections, which increased the persuasive effect of the product purchase. These results are in agreement with those of existing papers [22]. In simpler terms, when consumers perceive a video blogger as similar to themselves, they develop a strong and enduring bond, friendship, trust, recognition, and attachment to the video blogger, and the video blogger’s use of emotional persuasion techniques to promote the advertised product may be more effective [2,5].

Finally, obvious advertising intention weakens the effect of emotional attachment on consumers’ purchase intentions. In concrete terms, under a certain level of emotional attachment, the higher the perceived explicit intent of an advertisement, the lower the consumers’ purchase intention. This finding has been verified in previous studies [6]. For example, the obvious disclosure of a video containing sponsored content significantly reduces a video’s intended effect on consumers [51]. Rozendaal et al. [51] found that sponsored messages may negatively affect persuasion; if consumers detect any biased commercial motive, it often leads to a high level of skepticism and a more discerning approach toward the message, thus decreasing consumers’ willingness to buy, even if there is some emotional connection or attachment to the video blogger. However, contrary to the research hypothesis, the obvious intention of the advertisement did not diminish the effect of perceived source credibility on consumers’ purchase intentions. This is consistent with previous research showing that the more professional the information source is perceived to be, the more trustworthy and behavioral consumers are about the advertised product [38]. When the communicator has a high degree of credibility, the impact of advertising intent will be reduced [4].This is probably because the explicit intention of the advertisement does not cause consumers’ aversion or suspicion toward video content if they have a certain level of perceived credibility toward the video blogger. Even if the video’s advertising tendency is very obvious, it does not weaken consumers’ purchase intention if they trust the video blogger enough to believe in the quality of the advertised product of the trusted video blogger.

This study makes the following research contributions. First, we examine the extent to which consumers’ intention to purchase advertised products is affected in today’s data-rich environment, thus adding to the existing knowledge on the impact of video blogging channels for social media on consumers’ buying behavior. China has a huge population base, and the number of individuals accessing the Internet has reached 1.079 billion people; while research on Internet users’ purchase intention of advertising products in terms of video blogging channels for social media is relatively blank and not clear and systematic enough, this study fills the gap by analyzing Chinese Internet users as the object and Chinese social media as the background. Second, this study proposes the variable of advertising intention for the first time in the field of video blog marketing research, explores the key factors influencing the purchase intention of video blog advertisements, and reveals the potential moderating mechanism of the influence of apparent or inconspicuous advertising intention on purchase intention, which helps to enrich the research and literature in related fields. Third, mediated by perceived source credibility and emotional attachment, this study provides insights into the communication channels that impact consumers’ intention to purchase and sheds light on the mechanism behind the “celebrity” effect. Finally, by examining how the transparency of advertising intentions affects consumer psychology, this study emphasizes the use of appropriate advertising methods in videos and the utilization of suitable language rhetoric and marketing methods in advertisements. This can help brands and video bloggers to fully understand consumer psychology, captivate potential clientele, steer consumers toward boosting their inclination to make a purchase, and increase the purchase rate of business or brand advertising and marketing.

5.3. Theoretical Implications

First, this study drew on several consumer psychology-related theories and related studies, extracted key variables that facilitate quantitative analysis to construct an analytical model, and conducted a more comprehensive study on the factors influencing product purchase intention in video blog advertising. This study provides a reference point for future studies on video blogging channels for social media consumption.

Second, this study used perceived source credibility and emotional attachment as dual mediators to create a parallel mediation between homophily, expertise, and consumers’ purchase intention. To a certain extent, the research field of consumer purchase intention is the study of consumer psychology and emotions. Based on the SOR model, this study introduces two consumer emotion variables to study the influence of consumers’ rational judgment on the characteristics of video bloggers and the formation of their emotional cognition, forming a complete research mechanism that is different from previous related studies. Therefore, this study combines the analysis of consumer rationality with the formation of consumer emotional cognition to provide new ideas and research support for subsequent related studies. In a way, it contributes to a better understanding of consumer psychology in social media video blogging channels from various perspectives.

Finally, synthesizing previous academic research, it was found that few scholars have studied the marketing content itself for research because the way that video bloggers conduct marketing is mainly based on the display of video advertisements. The content of the advertisements itself has a more important research position. Therefore, this study introduced the variable of advertising intention, adding it to the framework of previous research and correlating it with other key variables that have been used as research objects as key moderators in the study of video blog advertising purchase intent. Boundary conditions were explored, in which the strength of the relationship between perceived source credibility, emotional attachment, and purchase intentions varied according to advertising intentions. This extends the research scope of video blog advertising-related studies and provides new research ideas for other scholars.

5.4. Practical Implications

This study helps video bloggers understand that they should moderately control the marketing intention of videos, select marketing sponsorship, pay attention to the smooth pronunciation of advertisements and video content, innovate advertising content, make consumers more accepting and non-resistant, enhance the attractiveness and persuasion of advertisements, and evaluate the impact of various marketing methods on consumer psychology. For video bloggers whose perceived source credibility is not high enough, the brand should focus on the delivery method of the advertisement, minimize the intention of the advertisement, design clever and sincere displays, and minimize the suspicion and disgust of consumers. At the same time, brands should try to cooperate with video bloggers with high perceived source credibility, provide them with free services and products, and improve their willingness to cooperate because they can flexibly deploy advertising, and the degree of advertising intent will not particularly affect their marketing effect. In addition, the use of appropriate language and rhetoric in advertising can significantly attract potential customers and influence their willingness to buy.

Second, although nowadays, the information on the Internet is mixed and influences people to make the right decision for themselves, consumers have improved their screening criteria and independent judgment of information sources, so the basic professional knowledge of video bloggers in the marketing process and the homophily of the account shown with the audience are very important. For brands and companies to choose to partner with video bloggers who have a certain level of influence and solid product-related expertise, marketers should strive to encourage experienced vloggers to recommend products and services to others. At the same time, in the early stages of cooperation, video bloggers should actively conduct professional training on product knowledge, which can improve brand awareness and sales. In addition, brands should understand the values that influencers convey and those that potential customers value and choose video bloggers that match the brand image, as bloggers will be more influential on followers who think like them [5]. However, at the same time, it should be noted that the market situation is changing in different ways; some products have the use of universality, and some products are just needed and do not require a high degree of professionalism or too much emotional attachment of consumers to the video blogger, according to the product category to analyze the specifics.

Finally, the research findings provide reliable and useful information for consumers. This can make their consumption more rational, reduce impulse buying, identify the marketing tactics of video bloggers when watching video blogs, see the essence of the phenomenon, and accurately obtain information about the advertised products after analyzing the marketing methods and marketing language of the video bloggers to reduce the risk of purchasing and buying products to their liking. At the same time, the brand should pay attention not to rely too much on the results of this study; too much attention to the cooperation of video bloggers, product quality, service quality, and product positioning is the basic guarantee of sales; otherwise, it is easy to reverse the effect.

5.5. Limitations

This study has some limitations that need to be further discussed and expanded. First, there is a great variety of variables influencing consumer purchasing decisions based on different perspectives and theories in the identified studies. Because the range of theories and variables involved in the relevant research is too large, we can only interpret the research results from the perspective of our model and framework combined with past research and experience. Secondly, for the same reason, the hypothesis verified in the study may be that other potential variables play a role in the relationship between variables, which leads to the final result, rather than a direct relationship between variables, which needs to be discussed in a more systematic and detailed way with advanced theories. Third, because advertising intent has rarely been discussed in previous academic research related to video bloggers, the scale used to collect data on advertising intent was an adapted scale borrowed from other fields. Future research could develop a solid scale or other means of data collection to test this more rigorously. Fourth, our data collection was limited to consumers in China, and our sample profile was mostly female. Therefore, the questionnaire could be expanded to a global scale in future studies to make the study more objective and profound. Finally, the advertising intentions discussed in this study may also be influenced by other factors, such as the linguistic rhetoric of the video blogger (e.g., two-sided messages) and the form of the advertisement (e.g., product evaluations, ad insertions), which we did not discuss here and can be expanded and explored in depth in subsequent studies based on this point.