Investigating the Drivers to Covert Social Media Members to Customers

information and guiding purchases, making it easier for consumers to obtain goods and gradually getting used to this consumption and purchase mode. This study takes Facebook and Instagram social fan groups as examples to explore the impact of social characteristics and social participation on guided purchase behavior. This study classifies and statistically analyzes fans in Facebook and Instagram communities. A total of 212 valid samples were collected for statistical analysis. Use ANOVA and Scheffe's to verify. The results show that there are differences in social characteristics (uniqueness, similarity, informativeness) and social participation and behaviors (social participation, Parasocial Interaction (PSI)) among members of the guidance/purchase subgroup. The Influence of social characteristics and social participation on guided purchase behavior.


Motivation
The rise of social networks has made them the most popular network media rapidly thanks to the fast-evolving virtual world.Sharing texts or posting pictures over social networks has become a daily routine to connect with others.Social network fan club is established by opinion leaders and user fans, where the opinion leaders shale their trial experience, thoughts, photos and information with fans through the fan club and slowly bring fans close to them, while fans are rewarded with more response and interactions due to the participation and in turn feel accepted and recognized.Gangadharbatla (2008) mentioned that people's attitudes and behaviors toward social networks may come from the need for belongingness.Hajli (2014) pointed out that the consumption patterns of today's consumers are gradually changing, as social business is one of the key corporate driving forces.Social commerce works by working with commercial activities via social relationship for powerful marketing.Relationship with fan clubs is developed through social activities, where opinion leaders share their experience, answer all fans' questions and communicate with them.This has evolved into an important consumption pattern today.

Purpose of study
Most of previous social commerce studies were focused on investigating the correlation between social commerce and consumers' purchase willingness as well as fans' participation and consumption behaviors.Seeing that most of the social marketing technique are based on guided purchase, the purposes of the study are aimed to study the causes of guided purchase/purchase behaviors found in social network fan clubs and their effects, as follows: 1. Develop dimensional measurement questions based on literature collection and review; 2. Perform validity analysis, reliability analysis and one-way ANOVA on the results from official questionnaire survey, and verify the existence of difference between guided purchase behavior groups in the study; 3. This study presents a value as reference for practice and for future studies.

Literature review
The literature review is focused on the three dimensions selected for the study, namely social characteristics, consisting of "uniqueness," "similarity" and "informativeness;" social participation behaviors, including "social participation" and "parasocial interaction (PSI);" and guided purchase behavior, including "guided behavior and purchase behavior."

Uniqueness
Online businesses advertise and publish posts with mostly similar contents in social commerce.
According to Derbaix and Vanhamme (2003), contents that are unique and original are easier to catch consumers' attention and make them interested, which in turn results in greater responses.Because of this, consumers become more willing to comment on a published post or share the post as they grow increasingly interested (Peters et al, 2009).In their study, Moldovan et al (2011) pointed out that the more creative and unique a product is, the more words of mouth are generated for the product.Obviously, greater uniqueness seems to catch more attention of consumers.Gökerik et al (2018) suggested that social platform marketing through social media is full of creativity and uniqueness and, therefore, appealing to more consumers and better catch their attention.Clearly, unique contents are something that businesses can exploit for better marketing.

Similarity
People have been satisfying their social needs, from making friends, joining clubs to today's red-hot social websites.The greater similarity a social platform has, the more people participate and the more they are willing to interact, share information, express themselves and receive recognition from others.A distinct feature of social network is that people tend to make friends with those similar to themselves, and are willing to share their experience and information.Gangadharbatla (2008) mentioned that people's attitudes and behaviors toward a social website may come from their need for similarity belongingness.The need for similarity belongingness is a basic human instinct to form and maintain interpersonal relationships.It is learned from this, therefore, that people who sing up for a social website have a certain need for similarity belongingness.Liu et al (2018) pointed out that the similarity belongingness has a certain influence on the user' habits; the more frequent use of social network, the greater the similarity belongingness, and vice versa.Lou and Yuan (2019) argued that the greater the value of information in a post, social trust, attractiveness and follower similarity are, the greater the positive influence the followers have on the trust in social network, which in turn has influence on the recognition of brand social network and purchase willingness.Delone and McLean (2003) pointed out in their study that informativeness is the subjective opinion that a user has on an information system.The contents of information have four dimensions: usefulness, correctness, timeliness and completeness.They are evaluation indicators, where the usefulness means that a social community provides its members with useful information; correctness indicates that the information provided by social community is correct; timeliness suggests that a social community provides its members with updated information; and completeness represents that a social community provides its members with complete information.If a social community provides its members with the information contents above, and the information is valuable and meaningful, it will make users more willing to participate in the social community.Karakaya and Barnes (2010) found a correlation between reliability of information source and the participation of social community members; the more complete the information is, the more willing social community members are to participate.In their study, Wen et al. (2016) pointed out that the more similar social community members find that the conditions described in a post is to them, and the more informative the post is, the better resonance the members feel, and the greater trust there is in the social community.Noguti and Waller (2020) showed that users have an active relationship in searching for information in social media, suggesting that members look for information that is useful for them in social communities.

Social participation
Social participation is of the utmost importance for social marketing.However, not all participation behaviors lead to the desired results.Social participation means mutual benefits between businesses and consumers in general.Rishika et al (2013) proposed three influences of social participation: firstly, it encourages greater interactions between consumers and businesses, thus creating a stronger bond; secondly, it is easier for consumers to learn others' opinions and experience of a business through social participation; and thirdly, social participation makes it easier to access the products of a business and other information.Therefore, social participation helps business establish strong bonding with consumers, accelerate content spreading, improve purchase willingness, increase the quantity of purchases, and increase the frequency of users' participation, all of which contribute to businesses.However, it is not as easy as creating a Facebook fan page and the result presents itself.It requires careful management and design opportunities to develop a good relationship with customers.A business should have a specific target consumer group and investigate deep into the purpose of their behaviors and, thus, increase social participation and its frequency effectively.The point is to allow the social participation behavior and its results to generate values, instead of blindly chasing how many likes you get for a post.John et al (2017) expressed a similar point of view, as they analyzed the likes that social media users gave businesses on their Facebook fan pages.This may mean something for a business in terms of development relationships with customers, improving participation, encouraging understanding and catching attention.However, there is a lot more to do for social community if a business wishes to improve consumers' attitudes toward its brand or improve their purchases.It is necessary to focus on the recognition of brand image, develop the sense of belonging and establish a friendly relationship in order to create a power that vitalizes social participation.Shiau et al (2017) pointed out that a factor affecting a company's reputation is words of mouth through social participation, and it is suggested to pay more attention to social network users, improve the interactive relationship with users and establish a good relationship, which is one of the purposes that a modern business needs to work hard to maintain social participation.Hartmann (2008) believed that parasocial interaction (PSI) is an asymmetric interaction.The interactions induced and the level of interactions depend not on the readers but the original posters.For the use between communities, it refers to the process where the readers generate their feelings about their dialogues with the posters through texts.

Parasocial interaction
The consumption patterns are gradually changing these days, as consumers' purchase patterns have evolved.Products can be bought easily in a social community or over an online platform.Xiang et al (2016) discovered that the more professional knowledge and affinity a poster has, the greater level of PSI the consumers have, and the greater the willingness to buy.It is an important part of marketing strategy to create good PSI with consumers.
The rise of virtual social community allows consumers for easy access to massive information.It has become a very important business topic to attract existing consumers and uncover potential ones.Sokolova and Kefi (2020) found that the attitude affinity and PSI are positively correlated, which is a concept of extensive connection between the appearance attractiveness and PSI.This means that the attitude affinity and attractiveness of an opinion leader have a certain degree of influence on potential consumer groups.When it comes to building a good relationship with consumers, the greater the level of PSI in consumer participation, the more helpful for improving consumers' purchase willingness.

Guiding link and purchase behavior
What is guided purchase?Guided purchase is the successful translation of network traffic into transactions.Guided purchase comes in two stages.In Stage 1, consumers are led to a website, platform or store for purchase, whereas Stage 2 is to encourage consumers to place an order successfully.The guiding link leads to purchase behavior.A successful guided purchase marketing is the successful completion of purchase.
The key to the guided purchase is to attach the link to a post that allows consumers to see the product very quickly.The faster the product is seen by consumers, the better.It is imperative not to give anyone any chance to read the post without making a purchase!The key to social community guided purchase is keep publishing contents that interest consumers and keep probing their willingness to shop and purchase.With the understanding of their willingness, the next is to encourage their purchase desires step by step, making them interested and then willing to actually pay for the purchase.Yahia et al (2018) indicated that the ease of use, convenience, consumer habits and operation motivation of social platform have significant influence on social business, where the convenience and ease of use have the greatest influence on social business.A purchase model that is more convenient and an operation behavior that is easier are the marketing technique that social business platforms need to pay increasing attention to.
What is the concept of social community guided purchase?The term "social community" is a group formed by people sharing the same interests and a platform where its members communicate and share information with others, while "guided purchase" is a marketing technique that leads consumers to purchase.By combining both, it is the "marketing technique to bring orders to social community."Unlike conventional social e-business, the key to social community guided purchase is not about "transaction pattern," but the "marketing technique" in it.How to publish a post that interests potential customers or how to attract new fans with the help of advertising becomes a topic that all social community managers and e-business platforms have to study on, innovate and learn hard, which also indicates the importance of "social community guided purchase."

Conceptual framework
For the purpose of the study, fan group members were grouped by their guided purchase / purchase behaviors, the guided purchase groups compared in terms of social characteristics (uniqueness, similarity and informativeness) and social participation behaviors (social participation and PSI), and an analysis performed to determine any difference among these groups.The following conceptual framework was built based on social characteristics, social participation behavior and guided purchase/purchase behavior for the purpose of the study: At the rear of the store, consumers' purchase records were accessed to identify whether they came to the store through the links of posts at those social communities, before any purchase behavior took place.Also, the consumers' "social characteristics" and "PSI" were analyzed.

Hypotheses
Listed in Table 1

Definition of variables
The survey questionnaire developed based on literature review and variables proposed for the study consisted of two parts, the social characteristics and social participation behavior, where the former included uniqueness (with 4 questions), similarity (with 3 questions) and informativeness (with 3 questions), while the social participation behavior covered social participation (with 3 questions) and PSI (with 3 questions); in total 16 questions.The questions proposed were modified and adjusted according to the study purpose and needs.The measure scale was developed using the Likert's 6-point scale was adopted to prevent central tendency, ranging from 1 for "very disagreed" to 6 for "very agreed."

Hypotheses and variable definition
A post was published to the Facebook fan clubs on Facebook, as well as Facebook and Instagram for the purpose of the study.The post told fans that an online questionnaire survey was performed through the link provided in the post.At the end of answering the questionnaire, The Shopee IDs and respondents' accounts were accessed and compared with consumers' purchase records based on whether they received or used a discount code and made purchase.The respondents were compared against the Facebook or Instagram fan club list and grouped into A, B and C by their guided purchase behavior.
The survey required the respondents to provide their Shopee IDs and Facebook or Instagram account names.The account names collected were compared against the information of Facebook or Instagram fan club and Shopee buyers to determine that the data collected were correct, improve the effectiveness and accuracy of questionnaire, and ensured that the respondents were Facebook or Instagram fan club members.The respondents were grouped into Group A, B and C by their guided purchase behavior, as shown in the following table and figure:

Data analysis method
Collected using the specified data collection method, the data was subject to several analyses.The first analysis performed was the descriptive statistical analysis, followed by validity and reliability analyses and discriminant validity test, and finally the one-way ANOVA.

Descriptive statistical analysis
Gender, age and occupation were distributed based on number of times and percentage, as to investigate the attributes of sample data.The data in the collected questionnaires was identified quickly and in detail for a clear picture of the survey profile.

Reliability and validity analyses
The means and standard deviation were determined for the questions after the questionnaire was collected.The confidence coefficient of Cronbach's α was introduced as the measurement indicator to shed some light on the consistency and stability of survey results.For the purpose of reliability test, Cronbach's α > 0.7 indicates that the reliability reaches the minimum threshold; α > 0.8 suggests good reliability and α > 0.9 means extremely high reliability (Cooper et al 2006).The greater the confidence coefficient, the more reliable the survey questions.

Discriminant validity
The discriminant validity analysis is performed to check the truthfulness and accuracy of the measurement dimension studied.For this study, the measurement was made based on the correlation coefficient of discriminant validity.The composite reliability (CR) was examined for compliance with construct validity in order to ensure that the survey concept was correct.

One-way ANOVA
One-way ANOVA (Heiberger and Neuwirth, 2009) was adopted as the statistical method of the study to examine the existence of difference between guided purchase behavior groups for social characteristics and social participation behavior as well as to verify whether the hypotheses held true.

Empirical findings and analysis
Social community posts served as notification based on previous study backgrounds and purposes.Members of Shopee store were selected as the subjects of study, who were mostly members of fan clubs.The link to questionnaire was provided, as samples were collected by distributing online questionnaire.For the study background, the personal parts business was the focus of study, and the fan club of Xiao Yi Parts was chosen as the target of study.For the target social community, the Facebook fan club was selected.To ensure that the questionnaires collected were correct and valid, a post link was published over the Facebook fan page and notification over Facebook and Instagram.The respondents were asked to provide their Facebook or Instagram and Shopee accounts for a discount coupon.The accounts of social media and Shopee provided by the respondents were checked against the list of social community member accounts and buyer accounts collected at the rear of the Shopee store, thus eliminating those who were not members of social media fan club and grouping them by their actual purchase behaviors.The respondents were grouped into A, B and C by the guided purchase behavior.The process is shown in Figure 3 below.5.

Descriptive statistical analysis
The means and standard deviation were determined for the questions developed for the studied dimensions, including social characteristics (uniqueness, similarity and informativeness) and social participation behavior (social participation and PSI), to present how these questions were answered.For most questions, the means was greater than 4 and standard deviation between 0.7 and 1.2.

Reliability and validity analyses
For the reliability of survey questions, Cronbach's α, the most commonly used indicator in the Likert's 6-point scale was used to determine the reliability of these questions.The analysis result showed that α was greater than 0.7 for both dimensions, suggesting acceptable reliability for the questionnaires.The values of Cronbach's α were collected in Table 7.For discriminant validity, the values along the diagonal are the square root of AVE, while the others are the correlation coefficient values between dimensions.The discriminant validity exists if the square root values along the diagonal are all greater than the correlation values in the columns and rows for all dimensions (Fornell and Larcker, 1981).As mentioned above, the square root values of uniqueness, similarity, informativeness, social participation and PSI along the diagonal were all greater than the correlation coefficient values between dimensions below, signifying difference between dimensions and good discriminant validity.By summarizing the above, the dimensions exhibited good discriminant validity and, therefore, the discriminant validity is effective for this study, as shown in Table 8.Note: the figures in bold along the diagonal are square roots of AVE.
The one-way ANOVA applies mostly to the difference in means between multiple groups.For a group to achieve significance, it should be tested using the Scheffe's posterior comparisons to determine the difference between groups.The difference test was performed on guided purchase for the understanding of variables of social participation behavior.The variables of guided behavior groups were grouped into A, B and C, as shown in Table 9 below.One-way ANOVA was conducted for means testing.The one-way ANOVA was carried out to determine any significant difference in the influence social community members and guided purchase, before Scheffe's posterior comparisons were performed for testing.The results are presented as follows.

Influence of uniqueness in social characteristics on guided purchase of social community members
The test result in Table 10 indicates that the significance, P, was smaller than 0.05 for uniqueness for all social community members, suggesting significant difference (F=34.224,p =.000).This means that the uniqueness has a certain degree of influence on the guided purchase in social community members.It is learned from Table 11 that Group A > Group B > Group C, and Group B > Group C. Obviously, difference existed between guided behavior groups.Therefore, the hypothesis that the uniqueness in social characteristics creates difference between guided purchase behavior groups stands.

Influence of similarity in social characteristics on guided purchase of social community members
The test result in Table 12 indicates that the significance, P, was smaller than 0.05 for similarity for all social community members, suggesting significant difference (F=26.319,p =.000).This means that the similarity has a certain degree of influence on the guided purchase in social community members.It is learned from Table 13 that Group A > Group B > Group C, and Group B > Group C. It is clear that difference existed between guided behavior groups.Therefore, the hypothesis that the similarity in social characteristics creates difference between guided purchase behavior groups stands.

Influence of informativeness in social characteristics on guided purchase of social community members
The test result in Table 14 indicates that the significance, P, was smaller than 0.05 for similarity for all social community members, suggesting significant difference (F=22.536,p =.000).This means that the informativeness has a certain degree of influence on the guided purchase in social community members.It is learned from Table 15 that Group A > Group B > Group C, and Group B > Group C. It is clear that difference existed between guided behavior groups.Therefore, the hypothesis that the informativeness in social characteristics creates difference between guided purchase behavior groups stands.

Influence of social participation in social characteristics on guided purchase of social community members
The test result in Table 16indicates that significant difference exists in the influence of social participation on social community members and guided purchase.The significance, P, was smaller than 0.05 for similarity for all social community members, suggesting significant (F=31.041,p =.000).This means that the social participation has a certain degree of influence on the guided purchase in social community members.It is learned from Table 17 that Group A > Group B > Group C and significance exists between groups.It is obvious that difference existed between guided behavior groups.Therefore, the hypothesis that the social participation leads to difference between guided purchase behavior groups stands.

Influence of PSI on guided purchase of social community members
The test result in Table 18 indicates that the significance, P, was smaller than 0.05 for similarity for all social community members, suggesting significant difference (F=40.261,p =.000).This means that PSI has a certain degree of influence on the guided purchase in social community members.It is learned from Table 19 that Group A > Group B > Group C, and Group B > Group C. This suggests that difference existed between guided behavior groups.Therefore, the hypothesis that PSI leads to difference between guided purchase behavior groups stands.

Study findings and implication
The convenience introduced by today's online shopping has changed the consumption patterns of consumers, as they started gradually to enjoy the convenience that comes with online shopping.In addition to that, how to shop online cheaply, fast and reliably is now one of the considerations in consumers.The online marketing in recent years is slowly evolving and the guided purchase is seen through social participation.For example, guided purchase by live stream, fan club manager and post publishing are some of the common online marketing techniques.It is believed in this study that it would be greatly helpful for customer management to exploit the characteristics of brand community, make consumers recognize social community and become fans of the community that are actively participating in community activities.Yahia et al. (2018) pointed out that the ease of use, convenience, consumer habits and operation motivation of social platforms have significant influence on social marketing.With the convenience and ease of use of social community, guided purchase can be created by posing product information.Convenient purchase patterns and easy operations are the means to encourage consumers' motivation to buy and change their use habits.These are marketing technique valued increasingly by social business platforms and a topic focused on frequently.

Theoretical contributions
By comparing the theories and perspectives of social guided purchase and previous social marketing, the following theoretical contributions are achieved.Firstly, guided purchase by posting in social community is an ongoing marketing technique that has a certain degree of influence on social marketing.One of the key points for social community managers to contemplate on is how to publish posts to catch consumers' attention and create resonance.
Secondly, each of the uniqueness, similarity and informativeness in social characteristics has difference between the members of guided purchase groups.Thirdly, it is proven in this study that difference exists in social participation behavior in guided purchase groups, which means that social participation or PSI has significant importance to social communities.Therefore, apart from paying attention to establish a good interactive relationship with community members and working hard to manage social community relationship, it is more important to covert a consumer who purchases under the influence of guided purchase by posting product information into one of the social community members.This is an important topic for social community leaders and mangers to pay attention to and work hard on.

Practical contributions
This study was conducted to find out consumers' consumption behaviors and willingness in addition to data collection.For this, a simple subsequent interview was performed with consumers in Group B and C, where Group B consumers were only directed to the online store but did not make any purchase.The following were found in the interview with Group B consumers: (1) It is suggested to have posts in forms of unboxing presentation or physical photos of product, as physical items and functional introductions are more appealing to consumers.(2) The actual sales amount is included in the post to give consumers a clear picture and allow them to compare for price and, thus, increase their purchase willingness.
On the other hand, Group C consumers went directly to the online store without participating any social community behavior.The comments of Group C consumers are summarized as follows: (1) The contents in the online store or product introduction can be accompanied by information of relevant social communities to improve consumers' perceptions.
(2) There can be more categories or types of products in social community posts or introductions, not only high-priced products but also much more affordable ones.
The following practical suggestions are proposed for social community managers, social community participants and interested third parties after analysis on data collected from questionnaire survey and consumers' comments in interviews: 1.For social community managers (1) Posts in social community in the form of unboxing presentation or physical product photos are closer to consumers' needs.
(2) It is suggested to provide product information in detail, such as product size and specifications, for sales by social community posts.(3) It should be considered to show the sales amount in posts, allowing consumers for a clear picture.2. For social community participants (1) Participation in social communities, such as following a fan club, helps obtain more product information.
(2) It is suggested to leave comments under social community posts and interact with not only the manager but also other members for experience of using a certain product, for example.

For interested third parties
(1) Social marketing may be performed in a more direct and interactive way, such as through live streaming.
(2) Successful sales rate may be improved by replying the question or need of consumers or customers at the first available time.
Repurchase rate can be improved by providing thorough after-sale service and developing consumption habits in consumers.

Study limits and future suggestions
There is room for improvement for the study due to specific product types social community types and traits, as described below: (1) The samples for this study were collected from a specific social community type; that is, the conclusions of this study apply only to the subjects of the same properties and may cover other types of social communities.(2) A specific type of products was selected to investigate the topic of the study, and an extensive study on other product types is yet to be conducted.Therefore, the conclusions may not apply on other products.

Figure 1 .
Figure 1.Conceptual framework A fan club on Facebook called Xiao Yi Parts (小隱部品) was selected as the subject of study.Posts on Facebook and Instagram were actually links to its Shopee page -Xiao Yi Parts Store.At the rear of the store, consumers' purchase records were accessed to identify whether they came to the store through the links of posts at those social communities, before any purchase behavior took place.Also, the consumers' "social characteristics" and "PSI" were analyzed.
I often interact with fans \ moderator of the fan club.PSI I always follow the posts in the fan club.attention to posts by fans \ moderator.I feel fan club moderator \ fans are my friends.

Table 1 .
below, hypotheses were proposed based on social characteristics and social participation behavior: Hypotheses proposed for the study

Table 2 .
Operational definition of social community members for social characteristics

Table 3 .
Operational definition of social community members for social participation

Table 4 .
Guided purchase behavior groups of respondents groups describe A Those who are directed to Xiao Yi Parts Store on Shopee through Facebook or Instagram fan page and make purchase.B Those who are directed to Xiao Yi Parts Store on Shopee through Facebook or Instagram fan page but do not make purchase.C Those who make purchase directly at Xiao Yi Parts Store on Shopee.

Table 5 .
Sample structure

Table 6 .
Sample structure

Table 7 .
Questionnaire reliability analysis

Table 8 .
Discriminant coefficient vs. correlation coefficient for combinations of dimensions Dimension

Table 9 .
Guided purchase behavior groups of respondents groups describe A Those who are directed to Xiao Yi Parts Store on Shopee through Facebook or Instagram fan page and make purchase.B Those who are directed to Xiao Yi Parts Store on Shopee through Facebook or Instagram fan page but do not make purchase.C Those who make purchase directly at Xiao Yi Parts Store on Shopee.

Table 10 .
Uniqueness vs. guided purchase Note: The asterisk "*" denotes the means difference of P < α＝0.05 is significant.

Table 12 .
Similarity vs. guided purchase Note:The asterisk "*" denotes the means difference of P < α＝0.05 is significant.

Table 16 .
Social participation vs. guided purchase

Table 18 .
PSI vs. guided purchase Note: The asterisk "*" denotes the means difference of P < α＝0.05 is significant.