Social Network Emotional Marketing Influence Model of Consumers’ Purchase Behavior

Abstract

With the deepening application of Internet technology, social network emotional marketing has become a new way of sustainability marketing. However, most of the existing emotional marketing research belongs to the field of qualitative research, and there is a lack of data analysis and empirical research between social network emotional marketing and consumers’ purchase behavior. In this paper, firstly the influencing factors of consumers’ purchase behavior are extracted from a massive social network emotional marketing data set, and the Delphi method is adopted to interview experts to revise and improve the influencing factors. Then, a model simulating the influence of social network emotional marketing on consumers’ purchasing behavior is constructed. The proposed model explores the mechanism of the influence of social network emotional marketing on consumers’ purchase behavior through trust, attachment and other psychological factors from the perspective of emotion. Finally, a questionnaire is used to obtain survey data, and statistical methods are used to analyze the relevant data, so as to verify the correctness of the proposed model and related research hypothesis.

1. Introduction

Emotional marketing comes from the emotional needs of consumers; it can induce an emotional resonance in consumers, and integrate emotions into marketing. In the era of emotional consumption, consumers not only care about the quantity, quality and price of products, but also need emotional satisfaction and psychological identification when shopping. Emotional marketing is not only very important for establishing the business relationship between enterprises and consumers, but also crucial for the sustainable development of enterprises and their brands. Therefore, the emotional marketing environment is very important for enterprises to establish a good enterprise image and achieve sustainable development.

With the continuous development and application of the Internet, social networks have become an important platform for consumers to obtain information on products or services [1]. Compared with traditional e-commerce platforms that only help brands and sellers open the beginning of consumers’ purchase process, social networks have constructed a complete closed loop of consumers’ purchase decision-making process on their own platforms [2,3]. Therefore, many enterprises have attracted consumers’ attention by initiating topics and holding activities on social networks, so as to strengthen consumers’ loyalty and improve the quantity and quality of “honest” consumers. Due to the openness and interactivity of social networks, great changes have taken place in the marketing mode and brand communication mode in social networks. Social network brand communication is a kind of integration and utilization of the deep, long-term and close emotional connection between consumers and brands. It is a recessive means for enterprises to promote their brands and establish powerful brand images. Unlike traditional interest-driven marketing methods, emotional marketing focuses on the interaction between the core connotation of the brand and the stories, experiences and memories of consumers’ daily life, so as to stimulate consumers’ desire to buy. From the perspective of psychology, the operation of emotional marketing can bring a special kind of consumer trust based on consumer satisfaction [4]. Therefore, emotional marketing is also considered as a separate dialogue between enterprises and consumers. In this way, the reputation and credibility of enterprises is greatly improved in the minds of consumers. At the same time, cultivating loyal brand consumers is also very important for the sustainable development of enterprises.

Because emotional content belongs to a relatively private expression, and it is also a common way that groups can communicate in a specific cultural environment, emotional content is easier to present and transmit in social networks. In addition, due to the strong driving force of emotion and the psychological state associated with emotion, the consumers’ willingness to participate in the interactive process of communication is more likely to be excited. Once the emotional connection between consumers and brands is established, whether concerned with positive or negative information, consumers will quickly express their emotional state through social networks. These emotions become the evaluation content of the brand, and it is reprocessed in social networks. Therefore, social network emotional marketing is a typical sustainable marketing mode for enterprise development [5].

Making use of emotional marketing can not only enrich product connotation and accelerate product popularity, but also strengthen an enterprise’s sustainable development capabilities. Whereas traditional marketing aims at increasing business revenues, sustainable marketing aims at reaching sustainability and business goals. Some reviews and pieces of research show that businesses operating online have to implement changes from traditional marketing to sustainable marketing to move forward in the competitive space. However, at present, there is little research on the influence of social network emotional marketing on consumers’ purchase behavior. Some existing studies have only analyzed the relationship between emotional marketing and consumers’ purchase behavior, without detecting direct or indirect influencing factors from a deeper level, especially from the psychological level.

The sustainability of products or brands requires the persistent attention of consumers, and the purchasing desire and behavior of consumers are affected by many factors, among which emotional factors of consumers often play a crucial role. Therefore, if emotional factors are introduced into social network marketing, it will greatly increase the persistent attention of social network users given to brands or products.

Therefore, this paper takes social network emotional marketing as the research object and analyzes the relationship between social network emotional marketing, psychological trust, psychological attachment and consumers’ purchase behavior, in order to enrich the social network marketing model in practice.

The contributions of our research are as follows:

(1)

The influence model of social network emotional marketing’s effect on consumers’ purchasing behavior is constructed.

(2)

The mechanism of the influence of social network emotional marketing on consumers’ purchase behavior is found.

(3)

We found that trust and attachment all play an intermediary role between social network emotional marketing and consumers’ purchase intention.

2. Related Works

Social network marketing refers to a kind of marketing behavior in which enterprises, individuals or other groups use social network platforms to spread and share information, collect feedback, and interact, so as to influence their consumers and conveniently conduct their market research, product publicity and promotion, consumer relationship maintenance, and other behaviors [6]. With the expansion and popularization of social network applications, social network marketing has become a new mode of enterprise marketing in the era of social media. This Internet-based marketing model is quite different from the traditional marketing model. Yang et al. [7] found that the social network marketing mode has changed the unidirectional transmission of traditional marketing, and the precision marketing function of social networks has greatly exceeded the traditional marketing mode. This is not only conducive to increasing the interaction between both sides in marketing, but can also reduce the marketing costs for enterprises. Zhang et al. [8] pointed out that the relationship influencers have with other social media users is the foundation for the success of influencer marketing. Ajorlou et al. [9] studied word-of-mouth communication and the mode of word-of-mouth marketing on social networks, analyzed more than 1.5 million posts containing brand content, emotions and opinions, and found that social networks are an effective tool for consumers to communicate through word-of-mouth. Yan et al. [10] discussed the operating mechanism and influencing factors of social interaction, and concluded that interaction based on social networks can produce effective cognition, but too much interaction can lead to cognitive confusion. William et al. [11] found through a survey of more than 1000 social network users that the longer they used social networks, the more frequently they used them, and the stronger their emotional connection with each other became. Anastasiei et al. [12] investigated the effects of emotional marketing on the persuasiveness of social media messages related to products and brands. Sobhanifard et al. [13] explored a model of viral marketing emotional methods to increase greener purchase intentions in social networks. They discovered that viral-marketing-based joy and surprise have a significant positive effect on the social value of green product consumption. Wang et al. [14] explored the impact of social network marketing influencing factors (i.e., event marketing, product placement advertising and customer service platforms) on consumers’ purchase behavior, and found that these social network marketing methods need to affect consumers’ purchase behavior through intermediary variables. Camoiras et al. [15] believed that the opportunities brought by social network marketing are mainly that social networks have a large user group and have a certain consumption capacity, and social network users have high loyalty and stickiness. Bakar et al. [16] extracted promotion, brand, opinion leaders, interaction, information and interest as the influencing factors of social network marketing, and studied cognition, emotion and trust as intermediary variables. The analysis showed that brand has a direct impact on consumer behavior, and cognition, emotion and trust indirectly affect sharing intention and purchase intention. Sandip et al. [17] summarized and analyzed the strategies that enterprises need to consider when carrying out social network marketing. Through the research and analysis of the influencing factors of social network marketing, they found that brand and promotion have a positive impact on search intention, but there is a negative correlation between information interest and search intention. Sun et al. [18] proposed a new marketing model for social networks, which relies on the accumulation of big data and affects the target consumers through opinion leaders.

At present, there is less existing research on social network emotional marketing. Elsäßer et al. [19] conducted in-depth research and analysis on the emotional factors in the marketing process, found the influence degree of emotional factors in consumer purchase, and constructed a model of the interaction between them. Nisreen et al. [20] found that consumers in a luxury store setting expressed stronger emotional attachment and brand loyalty than consumers in a non-luxury store environment. Amyx et al. [21] used data analysis to study a large number of social networks, and found that 95% of them mentioned more than one brand, whereas about 75% of them made an emotional evaluation of some brands, which indicates that consumers can make emotional evaluations of brands through social networks, and enterprises can also affect consumers’ evaluation content through social network marketing, and affect people’s emotional tendency to a certain extent. Halkiopoulos et al. [22] used machine learning and data mining methods for evaluating the effect of emotional influence factors on social network consumers. They found that the influencing factors of social network marketing and intermediary variables have a positive effect on the brand purchase intention of consumers, and understood that how users behave when they connect to social networking sites creates opportunities for better interface design and richer studies of social interactions. Li et al. [23] applied the user’s emotion and group features to study the influence of multi-dimensional characteristics on information propagation in social networks. Lee et al. [24] analyzed the importance of emotional marketing in the context of the experience economy, and proposed a marketing mix strategy from the perspective of emotional products, emotional prices, emotional communication and emotional services. Zhang et al. [25] developed a theoretical framework and carried out a meta-analysis of empirical studies to investigate the relationships within the business model innovation. Shareef et al. [26,27] conceptualized advertising value and consumer attitudes towards advertisements. Based on the analysis of existing research, we can see that the existing social network emotional marketing research mainly focuses on the characteristics and mechanism of social network emotional marketing, most of which belong to the field of qualitative research. There is a lack of objective and quantitative research on the influence of social network emotional marketing on consumers’ purchase behavior. Some existing studies have revealed that there was a close relationship between consumers’ purchase intention and trust/attachment, but the specific influencing factors and the role of these factors have not been studied.

3. Materials and Methods

3.1. Data Collection and Analysis

According to the main research objective of this paper, the data acquisition object of this paper will be the registered users in the social network and the relevant data on the users’ purchase behavior for a specific brand. Specifically, it is necessary to collect relevant data such as personal information, social network relationships and the subjective attitudes and behavior intentions of registered users in social networks. This paper chooses RED, a well-known social e-commerce network in China, as the research object. RED has developed into a lifestyle social platform and consumption decision-making portal with more than 200 million users; the number of daily active users is more than 24 million. It has become the most representative marketing social network in China.

According to the characteristics of emotional marketing, experiential products can best reflect the emotional relationship between consumers and brands. At the same time, most of the users in RED are young women consumers. Eye cream has become a kind of product that is widely discussed and sold in RED. Therefore, this paper selects eye cream as the research object. In order to make the brand selection more reasonable, six brands are selected in the high-end, middle and low-end brand categories in the social network. High-end brands are GUERLAIN, Estee Lauder and Sisley; middle and low-end brands are OLAY, MARUMI and Loreal. This paper uses crawler software (Charles 4.5) to collect the purchase data and comment data on relevant brands on RED; a total of 372,067 complete and valid items of data were obtained. The collected data mainly include the number of comment likes, the number of comment collections, the content of comments, the release time of comments, the content of comment feedback, the number of comment feedback, the individual information of commentators, the social network information of commentators and the response content of commentators. During data analysis, we first used natural language processing technology and a classification algorithm to analyze and classify the acquired data through sentiment analysis. For pre-processed data, through constructing a binary logistic regression model, the influencing factors of social network emotional marketing were extracted as follows: emotional brand, emotional price, emotional communication and emotional service. Emotional brand refers to products or brands with emotional packaging, concepts and functions mentioned in social network marketing, which are marketing products that pay more attention to the emotional needs of consumers and meeting the psychological needs of consumers. Emotional price is a reasonable product price that matches the quality, brand and reputation of emotional products and is reasonably satisfactory to consumers. Emotional communication refers to the process of differentially transferring product features or service concepts to different consumers in the process of social network marketing, so that consumers can understand and identify with products or services, while paying attention to benign two-way interaction with consumers, fully listening to the opinions of consumers, so as to stimulate them to make an emotional evaluation. Emotional service refers to the process in which marketers use emotional factors to sell products in the process of social network marketing.

3.2. Impact Model Construction

A consumer’s purchase intention refers to the degree of the consumer’s reasonable preference for a certain product or brand, which reflects the purchase tendency of individual behavior. The purchase intention directly determines consumers’ purchase behavior. Therefore, in the model, consumers’ purchase intention is studied as a single variable. According to the theory of consumer purchase intention and trust [28], the purchase intention is based on trust, and trust guides action. Attachment refers to the emotional and specific attachment relationship between an individual and a specific object. It is a lasting and firm connection between the individual and the object to which he is attached, both psychologically and emotionally. A stronger attachment to a specific object can make individuals produce more emotional and psychological responses to the object, and allocate their own emotional, cognitive and behavioral resources to the object [29]. In social networks, attachment refers to the emotional connection between users of social networks and specific objects. This kind of attachment is defined as the emotional attachment of users to other users and the attachment of users to social networking sites.

The Delphi method is also called the expert survey method [30]. It uses correspondence to transmit the problems that need to be solved to each expert separately for consultation. Then, the suggestions of all experts are collected and summarized, and the integrated opinions are sorted out. The Delphi method begins with the definition of the problem and the choice of a panel of experts. It can absorb the opinions of experts in different fields, make full use of the experience and knowledge of experts, and ensure the reliability of the final conclusion. Because the analysis of the influence of social network emotional marketing on consumers’ purchasing behavior involves many fields such as e-commerce, marketing, psychology, computer science, etc., it is very suitable for this research. Although it provides the benefits of anonymity and the possibility for reevaluation and reflection, the Delphi method does not result in the same sort of interactions as a live discussion [31].

In this paper, 11 experts were selected to form an expert survey team, a questionnaire was designed, and a questionnaire survey on each expert member of the expert team was conducted. Three turns of consultation were required. Several rounds of questionnaires were sent out to the group of experts, and the anonymous responses were aggregated and shared with the group after each round. Finally, the statistical results of the questionnaire survey reached a consensus and then a conclusion. Therefore, this paper concludes that trust and attachment play a mediating role between social network emotional marketing and consumers’ purchase intention, and studies them as variables that affect consumers’ purchase behavior.

The impact model of social network emotional marketing on consumers’ purchase behavior in this paper is shown in Figure 1.

Figure 1. The framework of impact model.

Therefore, in this paper social network emotional marketing and its influencing factors are taken as the antecedent variables of consumers’ purchase intention, and the following research hypotheses are proposed:

H1: 

Social network emotional marketing has a positive impact on consumers’ purchase intention.

H2: 

Four influence factors of social network emotional marketing all have a positive impact on consumers’ purchase intention.

In addition, social network emotional marketing and its influencing factors are taken as the antecedent variables of consumer trust and attachment, and the following research hypotheses are proposed:

H3: 

Social network emotional marketing has a positive impact on consumer trust.

H4: 

Four influence factors of social network emotional marketing all have a positive impact on consumer trust.

H5: 

Social network emotional marketing has a positive impact on consumer attachment.

H6: 

Four influence factors of social network emotional marketing all have a positive impact on consumer attachment.

Only when consumers trust a specific product will they be willing to further understand and communicate, and will be likely to have positive behavioral intentions. Therefore, trust can be said to control and guide actions. Trust is the premise and basis for generating purchase intentions. Therefore, in this paper consumer trust is taken as the antecedent variable of purchase intention and the following research hypothesis is proposed:

H7: 

Trust has a positive impact on consumers’ purchase intention.

The attachment of consumers to a specific product can increase their understanding of the product, reduce their concerns about various possible risks and uncertainties, and thus promote their purchase intention in social networks. Therefore, in this paper consumer attachment is taken as the antecedent variable of purchase intention and the following research hypothesis is proposed:

H8: 

Attachment has a positive impact on consumers’ purchase intention.

When consumers trust or attach themselves to a specific product or brand, they have a positive purchasing intention towards the product, and this is likely to further change into actual purchasing behavior. On the contrary, if consumers distrust or detach from a specific product, they will have negative purchasing intentions towards the product, and then will be less likely to buy the product, and will even feel dissatisfied with the enterprise that owns the product. Therefore, the following research hypotheses are proposed:

H9: 

Trust plays an intermediary role between social network emotional marketing and consumers’ purchase intention.

H10: 

Attachment plays an intermediary role between social network emotional marketing and consumers’ purchase intention.

4. Results

This study first designed a questionnaire based on relevant literature [32,33,34,35], small-scale in-depth interviews and relevant mature scales. The questionnaire was mainly conducted among the active users of RED. The questionnaire is composed of three parts: (1) the use of RED, the main purpose of which is to detect effective research samples, (2) demographic questionnaire, which mainly measures the demographic characteristics of the participants such as gender, age and education background, (3) the questionnaire of independent variables, intermediary variables and dependent variables, which is the core of the questionnaire design. The specific contents of this part are shown in Table 1.

Table 1. The core contents of the questionnaire.

Variables		Question
Social network emotional marketing	Emotional products	Q1: I like brand names and products in social network marketing advertisements
		Q2: I think the product quality in social network marketing ads is guaranteed
		Q3: The products in social network marketing advertisements are novel
	Emotional prices	Q4: The product price in the social network meets my expected pricing
		Q5: I pay attention to the promotion, forwarding lottery and other activities held in the social network
		Q6: I will participate in promotion activities such as forwarding lottery in social networks
	Emotional communication	Q7: I can understand the service concept or product features of the enterprise from social network interaction
		Q8: I am willing to participate in sharing activities in social networks
		Q9: I can learn more about the enterprise or product from communication with social network users
	Emotional services	Q10: I think the terms of social networking product introductions are accurate and concise, and the expressions are vivid and easy to understand
		Q11: I think product introduction on the social network is timely and appropriate
		Q12: I think the social network interface is simple, which makes users feel comfortable
Intermediary role	Trust	Q13: I trust the social network platform
		Q14: I think the information conveyed or activities held by social networks are authentic
		Q15: I think the brand or enterprise promoted by the social network has credibility
	Attachment	Q16: I have attachment to brands or products I care about in social networks
		Q17: I have attachment to the social networks that promote products
		Q18: I have attachment to other users of the social network
Consumer purchase intention	Purchase intention	Q19: I am very likely to buy products recommended in social networks
		Q20: When I need a product, I will first consider buying from products recommended by social networks
		Q21: I am very likely to buy products in social network marketing advertisements

In Table 1, we divide the variables into three types, namely, independent variable, intermediate variable and dependent variable, and each variable corresponds to a series of questions. These questions use the Likert scale: there are five options for each question, namely “strongly agree”, “agree”, “neutral”, “disagree” and “strongly disagree”, which, respectively, correspond to 5, 4, 3, 2 and 1. The total score of the respondent is the sum of his/her scores for each question, which represents the degree of agreement with the statement.

4.1. Reliability and Validity of Questionnaire

To ensure the reliability of the questionnaire, this paper adopted the pre-commissioning method to ensure the consistency and effectiveness of the questionnaire. In this stage, 100 questionnaires were distributed in total and 100 questionnaires were returned, including 98 valid questionnaires. In the samples, there were 63 females and 35 males, and the age structure was mainly between 18 and 43 years old.

The questionnaire data was imported into the SPSS 12.0 software and descriptive statistics on the variables were made. When analyzing the reliability of the sample, the measure statistic used was Cronbach’s Alpha coefficient, namely α coefficient. The value range of α coefficient was 0–1. The higher the coefficient value, the higher the reliability. The reliability analysis of the data collected from the pre-commissioning questionnaire is shown in Table 2.

Table 2. The reliability analysis of the pre-commissioning questionnaire.

Variables		Question	α Coefficient
Social network emotional marketing	Emotional products	Q1	0.757
		Q2
		Q3
	Emotional prices	Q4	0.768
		Q5
		Q6
	Emotional communication	Q7	0.809
		Q8
		Q9
	Emotional services	Q10	0.795
		Q11
		Q12
Intermediary variable	Trust	Q13	0.887
		Q14
		Q15
	Attachment	Q16	0.892
		Q17
		Q18
Dependent variable	Purchase intention	Q19	0.779
		Q20
		Q21

Generally speaking, if the value of α coefficient is greater than 0.9, it means that the reliability is very good, and if it is greater than 0.7, it means that the reliability is acceptable [36]. As shown in Table 2, the overall α coefficient value of the independent variable social network emotional marketing reached 0.782. The α coefficient values of two intermediary variables reached 0.887 and 0.892, respectively, and the α coefficient value of the dependent variable consumers’ purchase intention was 0.779. They all have good reliability, and thus the questionnaire is highly reliable and stable.

Validity analysis was used to investigate whether the measurement results of the questionnaire are true and effective, and whether the test results can really reflect the characteristics intended to be measured. The KMO test and Bartlett’s spherical test are the most commonly used methods for validity analysis. KMO mainly detects the correlation between each variable. Its value range is 0–1; a KMO value greater than 0.5 is the minimum standard for factor analysis, and less than 0.5 indicates poor validity [37]. Bartlett’s spherical test is considered from the whole correlation coefficient matrix. If the sig. value is less than 0.5, it indicates that the index has reached the significance level and is suitable for factor analysis [38]. The validity analysis of the data collected from the pre-commissioning questionnaire is shown in Table 3.

Table 3. The validity analysis of the pre-commissioning questionnaire.

Variables		KMO Test	Bartlett’s Spherical Test
			Chi-Square	Free Degree	Significance Level
Social network emotional marketing	Emotional products	0.652	35.872	3	0.000
	Emotional prices	0.533	37.375	3	0.000
	Emotional communication	0.616	50.858	3	0.000
	Emotional services	0.672	43.893	3	0.000
Intermediary variable	Trust	0.826	118.368	3	0.000
	Attachment	0.768	77.426	3	0.000
Dependent variable	Purchase intention	0.635	38.682	3	0.000

As shown in Table 3, the KMO value of the independent variable, intermediary variable and dependent variable also reached more than 0.5, which indicates that each variable is suitable for factor analysis. In addition, the sig. value of Bartlett’s spherical test for all variables is 0.000, with a high level of significance, indicating that the variables are not independent of each other, but there is a certain correlation, and thus it is more suitable for factor analysis.

4.2. Statistical Analysis of Questionnaire

The survey started in November 2021 and ended in June 2022. A total of 5200 questionnaires were distributed and 4996 were returned. Finally, 4762 valid questionnaires were obtained, with an effective recovery rate of 91.57%. The sample characteristics are as follows: females account for 89.5%, the age structure is mainly between 23 and 34 years old, and the education background is undergraduate or above accounting for 82.75%.

In order to further verify the theoretical model and research hypothesis of this paper, we tested the correlation between various variables, and conducted regression analysis between the influencing factors, intermediary variables and consumers’ purchase intention concerned with social network emotional marketing, to verify the causal relationship between them, and to verify and analyze the intermediary role of trust and attachment.

4.2.1. Correlation Analysis

In this paper, Pearson correlation analysis was used to analyze the correlation between social network emotional marketing, intermediary variables (trust and attachment) and consumers’ purchase intention. The correlation test was carried out according to the following standard: if the correlation coefficient between two variables was less than 0.85 and the significance level was below 0.5, then it meant that there was a certain degree of discrimination validity between the variables [39].

The Pearson correlation coefficient was used to measure the correlation between two independent variables. Firstly, the correlation and significant level between social network emotional marketing and its four influencing factors and consumers’ purchase intention were analyzed. The analysis results are shown in Table 4.

Table 4. Correlation analysis between social network emotional marketing and consumers’ purchase intention.

Independent Variable	Index	Consumers’ Purchase Intention
Emotional products	Pearson correlation	0.655 **
	Significance (bilateral)	0.000
Emotional prices	Pearson correlation	0.626 **
	Significance (bilateral)	0.000
Emotional communication	Pearson correlation	0.646 **
	Significance (bilateral)	0.000
Emotional services	Pearson correlation	0.530 **
	Significance (bilateral)	0.000
Social network emotional marketing	Pearson correlation	0.738 **
	Significance (bilateral)	0.000

** represents p < 0.01.

According to the analysis results in Table 4 and validation criteria, at the significance level of 0.01, the correlation coefficient of each variable is less than 0.85, which shows that social network emotional marketing has a significant positive correlation with consumers’ purchase intention, and its four factors also have a significant positive correlation with consumers’ purchase intention. Therefore, the research hypotheses H1 and H2 proposed in this paper can be preliminarily verified.

Then, the correlation and significant level between social network emotional marketing and its four influencing factors and two intermediary variables were analyzed. The analysis results are shown in Table 5.

Table 5. Correlation analysis between social network emotional marketing and intermediary variable.

Intermediary Variable	Independent Variable	Index	Consumers’ Purchase Intention
Trust	Emotional products	Pearson correlation	0.655 **
		Significance (bilateral)	0.000
	Emotional prices	Pearson correlation	0.626 **
		Significance (bilateral)	0.000
	Emotional communication	Pearson correlation	0.646 **
		Significance (bilateral)	0.000
	Emotional services	Pearson correlation	0.530 **
		Significance (bilateral)	0.000
	Social network emotional marketing	Pearson correlation	0.738 **
		Significance (bilateral)	0.000
Attachment	Emotional products	Pearson correlation	0.662 **
		Significance (bilateral)	0.000
	Emotional prices	Pearson correlation	0.682 **
		Significance (bilateral)	0.000
	Emotional communication	Pearson correlation	0.685 **
		Significance (bilateral)	0.000
	Emotional services	Pearson correlation	0.558 **
		Significance (bilateral)	0.000
	Social network emotional marketing	Pearson correlation	0.775 **
		Significance (bilateral)	0.000

** represents p < 0.01.

According to the analysis results in Table 5, at the significance level of 0.01, the correlation coefficient of each variable is less than 0.85, which shows that social network emotional marketing has a significant positive correlation with trust and attachment. Therefore, the research hypotheses from H3 to H6 proposed in this paper can be preliminarily verified.

Table 6 shows the correlation and significance level analysis between trust, attachment and consumers’ purchase intention.

Table 6. Correlation analysis between consumers’ purchase intention and intermediary variables.

Independent Variable	Index	Consumers’ Purchase Intention
Trust	Pearson correlation	0.602 **
	Significance (bilateral)	0.000
Attachment	Pearson correlation	0.616 **
	Significance (bilateral)	0.000

** represents p < 0.01.

According to the analysis results in Table 6, there is a significant positive correlation between trust and consumers’ purchase intention, and thus the research hypothesis H7 proposed in this paper can be preliminarily verified. In the same way, there is a significant positive correlation between attachment and consumers’ purchase intention, and thus the research hypothesis H8 proposed in this paper can be preliminarily verified.

4.2.2. Regression Analysis

Correlation analysis has verified that there is a certain correlation between the variables in the model, but it cannot explain the causal relationship between the variables, and thus it is necessary to study the in-depth causal relationship through regression analysis.

Firstly, social network emotional marketing was taken as the independent variable and consumer purchase intention was taken as the dependent variable, and a Linear Regression model (LR) and Polynomial Regression (PR) model were used to test these, respectively. Linear regression uses the least squares function, which is called the linear regression equation, to model the relationship between one or more independent variables and dependent variables [40]. Polynomial regression illustrates a general strategy for extending linear regression so as to fit curved lines to response data. Regardless of the relationship between dependent variables and other independent variables, polynomial regression can always be used in analysis. In the regression analysis, gender, age, degree and browsing time were taken as control variables and social network emotional marketing as the independent variable. The regression results of the effect of social network emotional marketing on consumer purchase intention, trust and attachment are shown in Table 7.

Table 7. Regression results of the effect of social network emotional marketing on consumer purchase intention, trust and attachment.

Variables	Consumer Purchase Intention		Trust		Attachment
Control Variable	LR	PR	LR	PR	LR	PR
Gender	−0.009	−0.018	0.043	0.038	0.041	0.039
Age	−0.029	−0.046	0.029	0.017	0.026	0.018
Degree	0.051	−0.038	0.096	0.015	0.075	0.013
Browsing time	0.292 **	0.131 **	0.265 **	0.117 *	0.237 **	0.112 *
Independent variable
Social network emotional marketing		0.751 **		0.702 **		0.689 **
R2	0.086	0.532	0.082	0.527	0.078	0.483
Adj-R2	0.061	0.543	0.062	0.502	0.059	0.506
F	3.389 *	35.068 **	3.316 *	28.903 **	2.985 *	29.836 **

** represents p < 0.01, * represents p < 0.05.

R2 and Adj-R2 indicate the fitting degree of the regression experiments. It can be seen from the regression results in Table 7 that the F value of each model is significant at a certain level, which means that the regression effect of each model is significant. The time that consumers spend browsing social networks every day has a significant predictive effect on consumers’ purchase intention. After adding control variables, social network emotional marketing has a significant positive impact on consumers’ purchase intention. Therefore, the hypotheses H1 and H2 proposed in this paper are verified. This means that social and emotional network marketing has a significant impact on consumers’ purchase intention, and emotional marketing has a positive effect on improving product sales. The time that consumers spend browsing social networks every day has a significant predictive effect on trust and attachment. After adding control variables, social network emotional marketing has a significant positive impact on trust and attachment. Therefore, the hypotheses H3 to H6 proposed in this paper are verified. This shows that social network emotional marketing can increase consumers’ trust and attachment to products and brands.

Trust was taken as the antecedent variable and consumer purchase intention as the result variable, and regression analysis was conducted. The specific regression results are shown in Table 8.

Table 8. The regression results of the effect of trust on consumers’ purchase intention.

Variables	Consumers’ Purchase Intention
Control Variable	LR	PR
Gender	−0.012	−0.039
Age	−0.031	−0.057
Degree	0.052	−0.008
Browsing time	0.293 **	0.139 **
Independent variable
Trust		0.598 **
R2	0.087	0.361
Adj-R2	0.067	0.359
F	3.358 **	17.666 **

** represents p < 0.01.

Attachment was taken as the antecedent variable and consumer purchase intention as the result variable, and regression analysis was conducted. The specific regression results are shown in Table 9.

Table 9. The regression results of the effect of attachment on consumers’ purchase intention.

Variables	Consumers’ Purchase Intention
Control Variable	LR	PR
Gender	−0.008	−0.033
Age	−0.029	−0.052
Degree	0.049	−0.006
Browsing time	0.268 **	0.141 **
Independent variable
Attachment		0.576 **
R2	0.081	0.367
Adj-R2	0.057	0.343
F	4.156 **	16.931 **

** represents p < 0.01.

It can be seen from the results in Table 8 and Table 9 that the F value of each model is significant at a certain level, which means that the regression effect of each model is relatively significant. After adding the control variable, trust and attachment have a significant positive impact on consumers’ purchase intention, and thus the hypotheses H7 and H8 proposed in this paper have also been verified. This means that when consumers have trust in and attachment to the product, the purchase intention is significantly enhanced and has sustainability.

Intermediary variables can be further divided into complete intermediary and partial intermediary. Partial intermediary means that after the introduction of the causal relationship between variables, the explanatory power of independent variables on dependent variables will decrease, that is, the significance will decrease. Complete intermediary means that after the introduction of the causal relationship between variables, independent variables will not have explanatory power on dependent variables, that is, it is not significant. This paper examines the intermediary role of trust and attachment in social network emotional marketing and consumer purchase intention.

Firstly, social network emotional marketing was taken as the independent variable, and trust as the dependent variable, and control variables were added to conduct regression analysis. Then, social network emotional marketing, trust and control variables were taken as independent variables, and consumer purchase intention was taken as the dependent variable to conduct regression analysis. Regression results are shown in Table 10.

Table 10. Intermediary role of trust in social network emotional marketing and consumer purchase intention.

Variables	Consumer Purchase Intention		Trust
Control Variable	LR	PR	LR	PR
Gender	−0.019	−0.028	0.043	0.038
Age	−0.049	−0.053	0.029	0.017
Degree	−0.041	−0.043	0.096	0.015
Browsing time	0.132 **	0.117	0.265 **	0.117 *
Independent variable
Social network emotional marketing	0.743 **	0.651 **		0.702 **
Trust		0.165 *
R2	0.576	0.572	0.082	0.527
Adj-R2	0.551	0.559	0.062	0.502
F	34.277 **	31.286 **	3.316 *	28.903 **

** represents p < 0.01, * represents p < 0.05.

It can be seen from the regression results in Table 10 that the F value of each model is significant at a certain level, which indicates that the overall regression effect of each model is relatively significant. The regression results of the effect of social network emotional marketing on trust show that social network emotional marketing has a positive predictive effect on trust at a significance level of 0.01. The regression results of social network emotional marketing, trust and consumer purchase intention show that social network emotional marketing has a positive predictive effect on consumer purchase intention at a significance level of 0.01. After controlling the variable of trust, the regression coefficient of social network emotional marketing to consumers’ purchase intention is 0.651, which is significant at the level of 0.01. According to the criteria for intermediary effect, trust plays a partial intermediary role between the two. Therefore, the hypothesis H9 proposed in this paper is verified. This means that when consumers show trust in a particular product, they will have a positive behavioral attitude towards that product, which is likely to be further transformed into actual purchase behavior. On the contrary, if consumers show distrust in a particular product, they will have a negative behavioral attitude towards that product, and the possibility of consumers buying that product will be reduced.

In the same way, regression results for the intermediary role of attachment in social network emotional marketing and consumer purchase intention are shown in Table 11.

Table 11. Intermediary role of attachment in social network emotional marketing and consumer purchase intention.

Variables	Consumer Purchase Intention		Attachment
Control Variable	LR	PR	LR	PR
Gender	−0.017	−0.026	0.041	0.039
Age	−0.051	−0.052	0.026	0.018
Degree	−0.039	−0.041	0.075	0.013
Browsing time	0.127 **	0.119	0.237 **	0.112 *
Independent variable
Social network emotional marketing	0.727 **	0.238 **		0.689 **
Attachment		0.172 *
R2	0.583	0.569	0.078	0.483
Adj-R2	0.543	0.560	0.059	0.506
F	33.262 **	30.187 **	2.985 *	29.836 **

** represents p < 0.01, * represents p < 0.05.

It can be seen from the regression results in Table 10 that the regression results for the effect of social network emotional marketing on attachment show that social network emotional marketing has a positive predictive effect on attachment at a significance level of 0.01. The regression results of social network emotional marketing, attachment and consumer purchase intention show that social network emotional marketing has a positive predictive effect on consumer purchase intention at a significance level of 0.01. After controlling the variable of attachment, the regression coefficient of social network emotional marketing to consumers’ purchase intention is 0.238, which is not significant at the level of 0.01. According to the criteria for intermediary effect, attachment plays a complete intermediary role between the two. Therefore, the hypothesis H10 proposed in this paper is verified. This shows that social network emotional marketing not only has a positive impact on attachment, but also has a positive impact on consumers’ purchase intention. In other words, enterprises can not only enhance consumers’ attachment to the brand, but also improve consumers’ willingness to buy through reasonable pricing, promotion strategies and timely communication when conducting social emotional marketing.

Based on the above, through the reliability and validity analysis, correlation analysis, regression analysis, and other empirical analysis results, all the hypotheses proposed in this paper are valid.

5. Discussions

In order to clarify the influence mechanism of social network emotional marketing on consumers’ purchase behavior, this paper proposes the following three research issues from the perspective of information systems, social psychology and marketing: How does consumers’ social network trust affect their intention to purchase in social networks? How does consumers’ social network attachment affect their intention to purchase in social networks? What are the factors that affect consumers’ purchases in social networks?

According to the hypothesis test of social network emotional marketing on trust, the following conclusions are drawn: social network emotional marketing has a significant positive correlation with consumers’ trust, and the four influencing factors of social network emotional marketing will have a positive impact on consumers’ trust, which indicates that the better social network emotional marketing is carried out, the stronger consumers’ trust in related products or enterprises will be. It is consistent with the conclusions of previous studies [41,42]. In addition, the research results also show that consumer attachment, as expected, has a positive impact on social network users’ purchase intention. It shows that consumers’ attachment to brands or products on social networks leads to more purchase-related behaviors. Guèvremont [43] thought that brand attachment could help consumers to improve their eating habits. Dwivedi et al. [44] examined the effect of consumers’ emotional attachment on social media consumer-based brand equity from a “brand” perspective. Although these existing studies also recognized the role of consumers’ brand attachment in emotional marketing, they have not conducted in-depth analysis of attachment and consumer social network purchase behavior.

According to the hypothesis test of social network emotional marketing on consumers’ purchase intention, the following conclusions are drawn: there is a significant positive correlation between social network emotional marketing and consumers’ purchase intention, and the four influencing factors of social network emotional marketing will have a positive impact on consumers’ purchase intention, indicating that the better the social network emotional marketing is conducted, the more conducive it is to improving and promoting consumers’ purchase intention. Yuan et al. [45] proposed that word-of-mouth in social networks would affect the purchase intentions of consumers. Alalwan [46] identified and tested the main factors related to social media advertising that could predict customer’ purchase intention. However, these studies have not clearly pointed out the specific influencing factors of social networks’ emotional impact on consumers’ purchase intention. Therefore, the research results have practical guiding significance for enterprises carrying out social network emotional marketing. Enterprises should perform well in social network emotional marketing using the four aspects of emotional products, emotional prices, emotional communication and emotional services. On the premise of ensuring product quality, enterprises should use social network word-of-mouth publicity and emotional experience to enhance consumers’ goodwill and trust, and improve consumers’ willingness to buy.

Trust and attachment both play an intermediary role between social network emotional marketing and consumers’ purchase intention, but the intermediary effects are different. Trust plays a partial intermediary role between social network emotional marketing and consumers’ purchase intention, whereas attachment plays a complete intermediary role. This is because social network emotional marketing can increase consumers’ recognition of the corporate image, and also win consumers’ favor and trust in products and information. Consumers’ trust is the premise and basis for their willingness to buy, and thus trust plays an intermediary role between social network emotional marketing and consumers’ willingness to buy. However, for emotional products in social network emotional marketing, the products in emotional marketing can only act on consumers’ purchase intention through consumer attachment. This may be because consumers can only obtain basic cognition through the introduction of marketing advertisements and the evaluation of other consumers before they actually buy, and they cannot feel the effect of the product personally. Therefore, attachment is particularly important between social network emotional products and consumers’ purchase intentions. Some existing studies [47,48,49] investigated the role of attachment or trust in consumers’ evaluation of new products, and pointed out that social networks have emerged as powerful platforms in addressing the issue of consumers’ trust and brand loyalty. However, they did not elaborate the intermediary role of trust and attachment between social network emotional marketing and consumers’ purchase intention, and there is no further distinction made between the intermediary role of trust and attachment.

It can be seen from the above discussion that the research results of this paper reveal the influence mechanism of consumers’ purchase intention in social network emotional marketing. The proposed model in this paper studies the relationships among social network emotional marketing, trust/attachment and consumer purchase intention, and discusses the intermediary effect of trust/attachment. In the marketing process of enterprises, winning the trust of consumers is the key to continuing marketing activities, and establishing consumers’ attachment to the brand is the key to bringing long-term value to enterprises. This is of great significance to the sustainable development of enterprises. Social network emotional marketing both moves and retains people with emotion. The relationship with consumers is sustainable. At the same time, it can also strengthen the stickiness and loyalty of social network users, and it provides the possibility for enterprises to develop high-quality consumers.

6. Conclusions

In this paper, social network emotional marketing is taken as the research object, and quantitative research methods and the Delphi method are used to explore the influence of social network emotional marketing on consumers’ purchase behavior. Four influencing factors of social network emotional marketing are extracted through data analysis, and an impact model of the effect of social network emotional marketing on consumers’ purchase intentions is constructed. Data is collected by means of a questionnaire survey, and empirical work is conducted by using data statistics and analysis methods to explore and analyze the mechanism of social network emotional marketing and influencing factors. The Delphi method is used to further confirm that trust and attachment influence consumers’ purchase intention in social network emotional marketing. The empirical analysis results prove the correctness of the hypotheses and models proposed in this paper, and provide an important reference for further enriching research in social network marketing and related fields. This study is useful for marketing managers to establish sustainable marketing strategies on social network platforms to encourage more consumers to buy their products. At the same time, it also provides enlightenment for social network marketing enterprises to improve their own influence. Due to the limitations of the amount of data obtained from the questionnaire and the source platform in the study, it cannot guarantee the comprehensiveness of the influencing factors of social network emotional marketing on consumers’ purchasing behavior. In the future, we hope to further improve the model through the collection of more comprehensive social network emotional marketing data from various social network platforms. In addition to trust and attachment, there are many factors that can affect consumers’ purchase behavior. Future research will further explore whether there are other intermediary variables between social network emotional marketing and consumers’ purchase behavior, so as to gain a more comprehensive understanding of social network emotional marketing.