# Sequential mediation of parasocial relationships for purchase intention: PLS-SEM and machine learning approach

**Authors:** Nora Sharkasi, Saeid Rezakhah  
**arXiv:** [2307.00005](https://arxiv.org/abs/2307.00005)  
**Date:** 2023-06-08  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2307.00005v1/

## Key points

- Social media influencers (SMIs) have the power to influence consumer behavior and drive purchase intentions
- A study investigated the impact of source influence (SI) on consumers' intention to purchase (I2P)
- Parasocial relationships (PSR) with factors like benign envy (BE) and perceived brand influencer fit (BIF) mediate this relationship
- The study used two independent samples from Europe and Southeast Asia, with a total of 411 and 355 participants respectively
- Partial Least Squares Structural Equation Modeling (PLS-SEM) analysis and classification-based machine learning (ML) techniques were employed for data analysis
- Trustworthiness and expertise are significant in forming source influence
- Benign envy has a more significant role in predicting consumer behavior than perceived brand influencer fit
- Building trustworthiness and expertise are key attributes for successful influencer campaigns
- Benign envy plays a significant role in driving consumer purchase intentions
- The research expands knowledge on social media influencers' effectiveness in advertising campaigns.

## Summary

In the world of advertising, companies have recognized the power and effectiveness of social media influencers (SMIs). These influencers have the ability to sway consumer behavior and drive purchase intentions. To address this gap in understanding the factors that contribute to their success, a study was conducted to investigate the impact of source influence (SI) on consumers' intention to purchase (I2P). The researchers examined how parasocial relationships (PSR) with two key factors - benign envy (BE) and perceived brand influencer fit (BIF) - sequentially mediate this relationship. The study utilized two independent samples from Europe and Southeast Asia, with a total of 411 and 355 participants respectively. To analyze the data, a combination of Partial Least Squares Structural Equation Modeling (PLS-SEM) analysis and classification-based machine learning (ML) techniques were employed. Comparative analyses were also conducted within and between studies. The findings revealed that trustworthiness and expertise played significant roles in forming source influence. Moreover, when comparing the impact of BE and BIF on followers' intention to purchase, it was found that BE had a more vital role in affecting/predicting consumer behavior than BIF. These results contribute valuable insights into the credibility of social media influencers and provide a better understanding for marketing practitioners and researchers. The study highlights the importance of building trustworthiness and expertise as key attributes for successful influencer campaigns. Additionally, it emphasizes the significance of audience-related features such as benign envy in driving consumer purchase intentions. Overall, this research expands our knowledge on social media influencers' effectiveness in advertising campaigns and offers comparative analysis paradigms for future studies in this field.