Will you purchase what I recommend? The role of interaction orientation, parasocial relation, and product involvement

In recent years, sales in online platforms have received huge attention from generation Z in Indonesia, along with live streaming business model which has undergone rapid development in recent years. Although the way live streaming with consumers has been known as the key to influence consumer behavior, only a few research has studied the style of streaming communication in affecting purchase intentions. Based on the social influence theory, this study attempts to examine the influence of interaction orientation, parasocial relation, and digital influencer credibility on purchase intention. In addition, the present study also investigates the role of product involvement advertised by streamers as moderator. The quantitative approach with the survey is used through samples of 200 respondents selected by purposive sampling. The data is collected online through disseminating questionnaires using Google Form, which is then analyzed using SEM-PLS technique. The findings indicate that interaction orientation positively influences purchase intention and parasocial relation, while parasocial relation positively influences digital influencer credibility, and digital influencer credibility positively influences purchase intention. In addition, product involvement has a moderating role that can strengthen the influence of interaction orientation and purchase intention. This study provides implications for how an interaction can be delivered online and how the involvement of a credible influencer can affect purchase intention.


Introduction
The increasingly massive growth of technology and information has changed the way business operate in many aspects.In an era of globalization and digital transformation, consumer behavior has changed and become increasingly complex, making it a major concern for practitioners and academics.Dash et al. (2021) and Wang et al. (2022a) stated that the factors that influence consumers purchase intention on a product or service has gained attention in many literature, especially on the context of online business (Dash et al., 2021;Wang et al., 2022a).By definition, purchase intention refers to a person's desire to buy a particular product or service.In the theory of planned behaviour (Ajzen, 1991), purchase intention can be affected or formed by attitudes, subjective norms, and intentions that are either directly or indirectly involved in the activity to choose before making a purchase decision.In addition, selling practice on a digital platform has resulted in an interaction in discussing a product or service which can also drive consumer purchase intention (Salmiah et al., 2024).
Currently, Indonesia has become one of the countries with the largest online shopping activities.This is signified by the increasing trend of online shopping using e-commerce application which can be accessed easily through smartphones.According to Guo et al. (2021) and Rahim et al. (2022), the development of online sales has given rise to a new sales model, such as live streaming, thus many vendors are using it as a tool to raise attention and attract interest.This is also mentioned by Sun et al. (2019) and Todd and Melancon (2018) that live streaming, which refers to business practices where vendors sell products while streaming is going on, is trending.Wang et al. (2022b) and Zhu et al. (2022) asserted that sales through apps and online live streaming can be done when the seller has a good interaction orientation.Similarly, Li et al. (2022) suggested that the feature of live streaming can be an interactive communication between streamer and consumer that can attract purchase intention.Consumers who use social media can often be influenced by online figures who provide information related to products or services offered either through Instagram, YouTube, TikTok, and etc (Balaban et al., 2022;Yuan & Lou, 2020).The popularity of influencers on social media has been widely utilized by companies as a strategy in attracting consumer interest (Appel et al., 2020).This business model is built on collaboration to promote products (Melnychuk, 2023).Besides, another important part of influencer marketing is they can be a sales force hired to interact with their followers to offer products or services through content on social media (Arsenyan & Mirowska, 2021;Pradhan et al., 2023).
Today, social media platforms have a considerable influence in providing opportunities to initiate two-way communication between brands, firms, or influencers and their followers.As a result, this raises the question of how a communication style can affect consumer behavior, especially regarding purchase intention.As stated by Herzallah et al. (2022), purchase intention reflects the extent to which consumers are willing or inclined to buy a particular product or service.Therefore, understanding it can be an important key for business players in designing an effective and relevant marketing strategy.In addition, understanding purchase intention can provide benefits for companies in formulating strategic planning as well as developing relationships with customers.Through digital platforms, consumer interaction can run efficiently and easily accessible anywhere and anytime.
Although previous research has found that the way sellers communicate with consumers is key to influencing purchasing behaviour, only a few research has studied how the style of streaming communication can affect purchase intentions.In addition, existing studies often explain the role of influencer credibility in raising consumer reactions.As stated by Liao et al. (2023), when the seller has a good interaction, this will produce a good bond.Align with this statement, the outstanding feature of live streaming for product or service sales is the presence of interactive communication between the streamer and the prospective consumer.Sokolova and Kefi (2020) asserted that influencer effectiveness includes attraction and attitude, trust and expertise, the ability to build parasocial relationship (PSR) with followers (Ferchaud et al., 2018), and influencer credibility (Pöyry et al., 2022).Thus, this study fills the gaps in existing research (Balaban et al., 2022;Rokonuzzaman et al., 2020;Salmiah et al., 2024), by exploring the factors that influence consumer purchasing intentions on online application platforms.Specifically, this study is carried out to examine the influence of interaction orientation, parasocial relation, and digital influencer credibility on purchase intention using the framework of social influence theory.Additionally, the present study also considers the moderating role of product involvement as the moderator in the relationship between interaction orientation and purchase intention.

Social Influence Theory
As stated by Book and Tanford (2020) and Venkatesh et al. (2011), social influence theory is a comprehensive framework that seeks to elucidate how individuals' thoughts, feelings, and behaviors are shaped by the influence of others within their social environment.This theory explores various facets of social influence, from the number and strength of influence sources, to the mechanisms through which individuals conform, comply, or resist persuasive messages (Kwon et al., 2021).The existing study have asserted that there are three primary types of social influence processes, namely normative influence, informational influence, and minority influence (Kwon et al., 2021;Zhou, 2022).In the context of this study, this theory can be used to understand how the presence, behavior, and communication style between digital influencers and its followers can impact consumer decision-making processes.Viewers, or followers, often conform to digital influencers' preferences, representing the normative influence.They also often seek guidance for product choices and contribute to the market dynamics influenced by the digital influencers, which represents informational influence.With this in mind, this study can help to understand how social forces in the digital space can influence consumers' online purchase intention.

Interaction Orientation, Parasocial Relation, and Purchase Intention
The presence of social media and the development of information technology have changed the way business operate (Palmié et al., 2022).Nowadays, sales through live streaming on apps characterized by interaction between streamer and audience have grown and become a necessity (Chen & Xiong, 2019;Ma et al., 2022).The communication style copied by the streamer is an interaction orientation.It has been mentioned that a seller has three orientation approaches that can be taken when establishing a relationship with a customer, namely profit orientation, market orientation, and social orientation (Efrat & Øyna, 2021;Song et al., 2019).According to Aparicio et al. (2021), in the context of e-commerce, interaction orientation refers to the tendency of one's approach to engaging with others in an environment.Song et al. (2019) also stated that interaction orientation includes attitudes, behaviors, and preferences that shape the way a person interacts in the form of active communication between the streamer and the viewer as well as providing answers to consumer questions.
Likewise, Lacap et al. (2023) suggested that the direct dialogue between the seller and the audience on social media can be conducive when the interaction is well constructed as it will also enhance intimacy.Dialogue and direct interaction with the audience on social media are very conducive to enhancing the relationship between the streamer and the audience.Interactionoriented sales forces can show their attitude that they are warm, friendly and accessible through verbal and non-verbal behavior that enhances physical and psychological intimacy (Ambroise et al., 2020).According to Reinikainen et al. (2020) and Xu et al. (2023), parasocial interaction occurs when there is an interlinked interaction.This interaction occurs when individuals or consumers feels connected to a media persona or a character.The orientation of interaction becomes relevant, thus creating a consciousness of mutual attention.Like a streamer adopting a conversation style or a physical signal to start two-way communication.Thus, the first hypothesis is submitted as follows: H1: Interaction orientation has a positive influence on parasocial relation.
Interactions between media users and influencers are portrayed as mediators, actors, which can be used as marketing tools.This interaction is a mutual relationship characterized by a sense of mutual understanding, attention, and adaptation (Ambroise et al., 2020).People in online interactions often express feelings of intimacy (Xu et al., 2023).In turn, individual attachment, involvement in relationships, and interaction lead to relationships of a bond, thus increasing the intention of purchase.The interaction that occurs on social media refers to how the interaction is built between the customer and the element in the marketing environment, which then influences the purchase intention.Bharadwaj and Shipley (2020) noted that there are several factors that can influence the interaction orientation, namely by involving customers in the product or brand and other information.Itani et al. (2020) also found that interaction with consumers is part of the marketing mix that marketers can use to boost purchase intention.H2: Interaction orientation has a positive effect on purchase intention.

Parasocial Relation and Digital Influencer Credibility
Parasocial interaction refers to consumer communication when interacting with others as if they were present and engaged in mutual relationships (Xu et al., 2023).Parasocial interactions can be seen in attitudes of mutual understanding, attention, and adaptation.People engaged in parasocial relation will interact with each other which in turn leads to individual attachment, relationship engagement, and loyalty.Previous research has shown that current parasocial interactions have entered the online environment (Penttinen et al., 2022).Social media features have shown that everyone can interact with others without limit so that audiences interacting on social media often form a strong emotional bond with social media influencers (Munnukka et al., 2019).Thus, his/her influence can boost the credibility of influencers in social media.H3: Parasocial relation has a positive influence on digital influencer credibility.

Digital Influencer Credibility and Purchase Intention
Despite the relationship built by the user with the influencer, uncertainty can still occur on consumer behavior in making purchases.According to Crisafulli et al. (2022), influencer credibility must be coupled with good comments given by others.Today, companies have recognized the role of influencers as the most effective strategic tool in the online business environment to build close relationships with customers.According to Torres et al. (2019), the role of the influencer itself has a positive impact on building the brand of the company and marketing.The presence of social media today has facilitated various aspects of marketing to attract consumer interest in making purchases.Thus, the role of digital influencers has a positive relationship in explaining purchase intentions.H4: Digital influencer credibility has a positive influence on purchase intention.

Product Involvement as Moderating Variable
According to Mittal (1995), involvement is a condition that is considered important in the formation of interests caused by stimuli and circumstances.Kautish et al. (2022) and Rokonuzzaman et al. (2020) stated that product involvement is a consumer perception of the importance of product categories based on customer values, needs, and interests.Rokonuzzaman et al. (2020) added that understanding product involvement is crucial in increasing consumer purchasing intentions.Generally speaking, Mittal (1995) divided product involvement into situational and permanent involvement.Situational product involvement reflects that consumers have involvement that can depend on situational factors or loyalty.In the context of online shops, advertised products play a moderate role in strengthening consumer purchasing intentions.Thus, the existence of influencers can have a positive impact in increasing consumer purchasing intentions.Richins and Bloch (1986) stated that product involvement contains consensus on an interest.If a product feels unimportant to buy, then it does not involve involvement and consumers feel that it is irrelevant, generating an attitude of indifference.Product involvement is an internal aspect that can be affected by a particular situation.A product can attract the attention of the consumer when the seller builds a good interaction through various platforms thus creating awareness, interest, and encouraging action for purchase.Balaban et al. (2022) assumed that product involvement plays a moderating role in strengthening the orientation of influencer interaction on purchase intention.Thus, the hypothesis proposed is as follows: H5: Product involvement moderates the relationship between interaction orientation and purchase intention.

Research Methods
This study uses a quantitative approach with survey as a data collection tool to see how and under what conditions interaction orientation affects purchase intention and parasocial interaction on digital influencer credibility.This study also seeks how digital influencer credibility affects purchase intentions as well as considers the role of product involvement as moderating variable.The quantitative approach is chosen as it allows researchers to know the magnitude of the causal relationship of each variable studied (Cooper & Schindler, 2014).According to Cresswell and Cresswell (2017), survey research can be used to describe and measure the relationship between two or more variables.
To see the impact, the study is conducted on users of e-commerce applications in the city of Bandung, which was taken using non-probability method, as it allows researcher to collect data from determined samples (Cooper & Schindler, 2014), through purposive sampling.The criteria for the respondents are: (1) the respondents have influencers they follow for at least 1 month and interact with them in social media; and (2) the respondents are e-commerce application users who have watch live streaming shopping shows from the influencers for more than once.Based on these criteria, the sample size obtained is as many as 200 respondents.The data is collected around three months, from October to December 2023.Then, the data collection was done using a questionnaire distributed using Google Form platform.The data obtained is then analyzed using structural equation modeling with partial least square (SEM-PLS) using Smart-PLS software.This method consists of two steps to examine the conceptual model developed in the study, by estimating the measurement model for validity and reliability in the first step, and then estimating the structural model for hypothesis test in the second step (Ayuni, 2019).
In the questionnaires used for data collection, respondents were asked to give a rating on their perception of the five variables present in the study using a 5-point Likert scale.The measurements of each variable in this study were adopted and modified from the previous study.
1. Interaction orientation is measured with six items adopted from the study of Liao et al. (2023).2. Parasocial relation is measured with seven items adapted from Yuan et al. (2016).3. Digital influencer credibility is measured with four items adopted from Sesar et al. (2022) and Weismueller et al. (2020).4. Product involvement is measured by five items adopted from Ferreira and Coelho (2015).5. Purchase intention is measured by three items adopted from Ma et al. (2022).

Results and Discussion
The sample profile revealed that most of the sample consisted of female respondents of 53%, and belongs to the younger generation (generation Z) with the age ranging from 18-25 years old (43%).Based on education, most of the respondents have diploma degree (23%).Based on income, the majority of respondents have income of IDR 3,000,000 -4,000,000 (56%).Furthermore, most of the respondents watch live streaming shopping for 1-2 times a week (35%), and they last made online purchase one month ago (34%).The detail is presented in Table 1.

Validity and Reliability
Before evaluating the structural model, each measurement model is examined for the reliability (Cronbach's Alpha (CA), Composite Reliability (CR), and Average Variance Extract (AVE) (Urbach & Ahlemann, 2010) and validity, both convergent and discriminant (Fornell & Larcker, 1981).The cut off value for Cronbach's Alpha and CR were 0.7 whereas the cut-off value for AVE was 0.5 (Fornell & Larcker, 1981).The reliability test result can be seen in Table 2. Minimum result was obtained on Cronbach's Alpha of 0.720, Composite Reliability of 0.824 and AVE of 0.540.The resulting values meet the recommended requirements, so it can be concluded that the measurement model is reliable for testing.Discriminant validity is tested by comparing the square root value of AVE to the correlation value between constructs (Fornell and Larcker, 1981).Meanwhile, convergent validity is tested using factor loading values with a cut off value of 0.7.Another alternative to test the validity of the measurement model is to test the Heterotrait-Monotrait correlation ratio (HTMT) with the requirement that it must have a value below 0.85 (Henseler et al., 2014).The convergent validity values can be seen in Table 2, in the factor loading column which indicates that all values exceed 0.7 with a minimum value of 0.709.The discriminant validity and HTMT test values can be seen in Table 3.The result indicates that all values have meet the recommended requirements, thus it can be stated that the measurement model is valid for testing.The influencer was easy to talk with.

0.813
The influencer was friendly.0.774 The influencer likes to help audiences.

0.853
The influencer was a cooperative person.

0.879
The influencer tried to establish a personal relationship.(Yuan et al., 2016) The influencer personalizes the product information.0.828 0.890 0.915 0.606 0.778 The influencer makes people feel more closely related to the product information.

0.851
The influencer provides product information that suits my personal style.

0.717
The influencer provides product information that interests me.

0.822
The influencer provides product information that is reflected in my purchase decision.The correlation matrix shows the relationship between five variables, namely Interaction Orientation (IO), Parasocial Relation (PR), Digital Influencer Credibility (DIC), Product Involvement (PI), and Purchase Intention (PUI).Each cell in the matrix presents the value of the Pearson's correlation coefficient between two variables.All correlation coefficients contained in Table 4 have statistically significant values (Sig.(2-tailed) < 0.01), indicating that the relationship between these variables provides insight into the direction and strength of the relationship between the variables measured in the research.The higher the correlation coefficient value, the stronger the relationship between these variables.Furthermore, from the result of descriptive statistical test, the average value of each variable indicate that the respondents tend to choose to agree with the statements contained in the study, with variations in the real scores not very variable.It can be seen from the standard value of a small deviation.Hypothesis model testing uses the bootstrapping method with 500 sub-samples.The path coefficients, t-values, and p-values of all hypotheses can be seen in Table 5 and Figure 2-3.Based on the result listed in Table 5, it can be concluded that Interaction Orientation (β = 0.636; t = 12.534; p = 0.000) has a positive and significant influence on Parasocial Relation, so H1 can be accepted.Apart from that, Interaction Orientation (β = 0.546; t = 7.950; p = 0.000) also has a positive and significant effect on Purchase Intention, so H2 is accepted.Parasocial Relation (β = 0.720; t = 21.535;p = 0.000) has a positive and significant effect on Digital Influencer Credibility, so H3 is accepted.Digital Influencer Credibility (β = 0.209; t = 2.969; p = 0.003) has a positive and significant effect on Purchase Intention, so H4 is accepted.Furthermore, Product Involvement (β = 0.159; t = 4.208; p = 0.000) as a moderating mechanism has a significant influence on the relationship between Interaction Orientation and Purchase Intention.The result of the moderation interaction are depicted in Table 5 and Figure 4, which shows that Product Involvement strengthen relationship between Interaction Orientation and Purchase Intention, so that H5 can be accepted.In evaluating model quality, Hair et al. (2012) suggest that researchers utilize PLS-SEM by using a measure of the relevance of model predictions.Sarstedt et al. (2019) recommend a process for testing the predictive relevance (Q 2 ) of structural models.According to test conducted by Hair et al. (2013), when the cross-validated Q 2 (redundancy measure) value is greater than 0 (zero), it indicates that the predictive relevance of the model has been confirmed.Based on this standard, the value of the cross-validation redundancy measure (Q 2 ) for the Digital Influencer Credibility is 0.2663, Parasocial Relation is 0.239 while for Purchase Intention is 0.345, indicating that the model has predictive relevance (see Table 6).

Discussion
This study aims to explore the factors that influence purchase intention by looking at the role of interaction orientation, parasocial relation, digital influencer credibility, and product involvement as moderation.As stated by Itani et al. (2020), the relationship between the seller and the buyer embedded in social media has become a strategic means in promoting a product or service.The findings suggest that the orientation of the interaction carried out by the seller can produce a proximity, emotional, and attractive attraction to the streamer or audience.This is in line with Aparicio et al. (2021) who stated that the sales force showing warm, friendly, and approachable attitude can enhance physical and psychological intimacy, thus resulting in mutual attention and awareness.The findings are consistent with the study from Liao et al. (2023), that communication with interaction orientation from streamers have a positive effect in building an intimacy (H1 accepted).Furthermore, this study also found that the interaction orientation influenced the purchase intention.The results of the study are consistent with Bharadwaj and Shipley (2020).The interaction that occurs on social media refers to how the interaction is built between the customer and the element in the marketing environment, thus influencing the purchase intention.According to Liao et al. (2023), direct interaction-oriented communication can increase consumer purchase intentions.Sellers who have a good interaction orientation can build relationships with consumers, thus generating positive responses.According to Yuan and Lou (2020), a positive impact of communication is a sustainable relationship.The findings show that when a relationship is well established, the new product being promoted can attract consumers' purchase intentions.In turn, individual engagement and involvement in relationships will increase, resulting in consumer purchase intentions (H2 is accepted).Furthermore, the results of this study also show that parasocial relation influence digital influencer credibility.The findings suggest that people involved in parasocial relation will interact with each other, and this will not happen when the credibility of influencers is not well built.Features of social media have shown that any person who interacts can form a powerful emotional impulse, and this bond can build the trust and credibility of influencers.Thus, parasocial relation and influencer credibility are two concepts that has an important linkage.Besides, Reinikainen et al. (2020) stated that when someone has good experience with an influencer, this can increase credibility.Munnukka et al. (2019) mentioned that parasocial relations with an are expected to build credibility (H3 is accepted).
Regardless of the relationship that exists between interaction-oriented communication and parasocial relation, the role of influencer credibility can affect consumer purchase intentions.The findings suggest that today's use of social media has become an inseparable consumption for generation Z (Lou, 2022).Their use of communication on social media (e.g.post comments, live messages on Instagram, write reviews, etc.) can increase interest in making purchases when streamers have good credibility.As an influencer, credibility is a relevant asset within an online environment.The findings are in line with the research conducted by Balaban et al. (2022) which states that influencer credibility is important factor that can increase purchase intention.Thus, the fourth hypothesis is accepted (H4 is accepted).
Finally, this study also found that product involvement becomes a moderator that can strengthen the relationship between interaction orientation and purchase intention.The findings indicate that consumers can have an involvement to the advertised products thus strengthening their purchase intention.The presence of streamers that have a good interaction orientation can have a positive impact in increasing purchase intention, and this influence can be strengthened when consumers have a strong product involvement.As stated by Balaban et al. (2022), product involvement is internal aspect of individuals that can be affected by certain situations.Gong and Jiang (2023) stated that a product can attract the attention of consumers if the seller builds good interactions through various platforms to create awareness, interest, and encourage purchasing action.Streamers who publish content will be able to provide good results in influencing purchase intentions if the streamer has a good interaction orientation.High and low levels of involvement can influence consumer perceptions in increasing their purchasing intentions, thereby resulting in behavior.Thus, the results of the present study and previous research have a correlation (H5 is accepted).

Implication and Conclusion
In alignment with the social influence theory, this study delves into the evolving landscape of online sales, particularly within the live streaming business model, capturing the attention of generation Z in Indonesia.The findings underscore the pivotal roles of interaction orientation, parasocial relationship, digital influencer credibility, and product involvement in shaping purchase intentions.Furthermore, the study explores the moderating influence of product involvement on the relationship between interaction orientation and purchase intention.This study contributes to the social influence literature by empirically validating the interconnected relationships between digital influencers and consumers in the e-commerce domain.Our results affirm the social influence theory's premise that the content delivered by credible influencers holds substantial sway over consumer purchase intentions.The identified factors offer a nuanced understanding of the mechanisms influencing consumer behavior in the context of live streaming.
This study offers valuable insights for practitioners in the field.Businesses can leverage these findings to strengthen collaborations with influencers, recognizing the influential role they play in cultivating robust consumer relationships.However, the importance of aligning advertising channels with consumer interests, considering potential product involvement, becomes evident for effective marketing strategies.However, despite the contributions made, this study acknowledges certain limitations that should be addressed in future research endeavors.The focus on a specific geographical area, namely users of e-commerce applications in Bandung, necessitates further exploration of cultural and industry nuances for broader generalizability.The cross-sectional nature of our data presents an opportunity for future longitudinal studies to deepen the understanding of evolving consumer behaviors over time.Additionally, the study prompts further investigation into the adaptability of content to consumer interests, adding a layer of complexity to the evolving field of influencer marketing.
In conclusion, this study not only advances theoretical understanding but also brings practical value to businesses navigating the intricacies of online consumer influence.The identified factors and their interplay contribute a nuanced perspective to the ongoing discourse on ecommerce, digital influencers, and consumer behavior, laying the groundwork for future explorations in this dynamic and evolving field.

Figure
Figure 1.Conceptual framework

Figure
Figure 2. Outer Model

Figure 4 .
Figure 4. Product Involvement Moderation Effect Predictive Relevance Q 2

Table 2 .
Result of Validity and Reliability Test

Table 5 .
Result of Hypothesis Test

Table 6 .
Cross Validation Redundancy Result