“Will You Buy It If They Recommend It?” Exploring the Antecedents of Intention to Purchase Podcaster-Endorsed Items

Abstract

The diverse content of and ease of listening to podcasts have made podcasts popular, particularly during the COVID-19 pandemic. Advertisers have begun to recognize their marketing potential and are now hiring podcasters to recommend their products. This study sought to determine the factors influencing podcast commitment, parasocial interaction (PSI), and the intention to purchase podcaster-endorsed items. It was conducted in Taiwan with a sample size of 578 participants and an online questionnaire. Structural equation modeling and mediation analysis were applied to the collected data from the perspective of uses and gratifications theory. We found that podcast commitment is positively related to edutainment, storytelling transportation, and social engagement. Social engagement is positively related to PSI, while storytelling transportation has a negative relationship with PSI. Additionally, there is a strong positive correlation between podcast commitment and PSI and both factors positively influence the intention to purchase podcaster-endorsed items. PSI also significantly moderates the positive relationship between podcast commitment and the intent to buy podcaster-endorsed items.

1. Introduction

On 18 February 2022, nearly 10,000 listeners gathered on the busy streets of Taipei to attend an offline event hosted by Bailingguo News, Taiwan’s most popular podcast program. Although the most popular medium on online social platforms is videos, the lengths of which are becoming shorter and shorter, the popularity of podcasts continues to rise. Indeed, podcasts are now widely recognized as having successfully transitioned from a niche market, with COVID-19 further contributing to the industry’s growth [1]. The market performance and growing audience base of podcasts have attracted the attention of producers, investors, and advertisers, such that brand advertising budgets for podcasts have grown exponentially in recent years [2]. According to the Internet Advertising Bureau, by the end of 2023, podcasts in the United States were expected to generate more than USD 2 billion in advertising revenue, representing a growth rate of 237% from 2020 [3]. According to their report, 41% of the world’s internet users listened to podcasts in 2020. These listeners are typically young, well-educated, and financially secure and thus are more likely to pay for trustworthy content [4]. Due to the variety of podcast genres and ease of listening, the range of listeners reached by the medium continues to expand. The audio nature makes podcast listeners more attentive than users of other media platforms, and the trust listeners have in podcasters makes them more likely to accept recommendations [2].

Podcasts are an essential source of current affairs information as well as entertainment and are popular worldwide [4]. Podcasts have the qualities of not only traditional media but also new media channels and have become the new favorite of advertisers [5]. The “listening economy” is on the rise in Taiwan. The outbreak of the COVID-19 pandemic prompted people to look for new ways to entertain themselves. Taiwanese celebrities and influencers have also begun to use podcasts to communicate with their fans [6]. However, despite the boom in the podcast market, academic research on podcasts and their audiences in Asian countries has yet to catch up with the research on Western markets and the empirical data collected from Western audiences (e.g., [7]). This gap means that the industry lacks a theoretical perspective on market strategies. Thus, the current study focuses on Taiwanese podcast listeners, elucidating why the Taiwanese audience exhibits a strong commitment to podcasts and which factors promote listeners’ intention to purchase podcaster-endorsed items.

What is the source of podcasts’ far-reaching appeal to audiences? Podcasts are usually produced by the podcasters themselves and distributed through diverse social media platforms. Podcasters can freely and authentically express their opinions and personalities through podcasts, which has shaped the unique styles of the different programs [8]. Podcasts meet the needs of specific audiences through diverse content. When listeners find that the program or podcasters are appealing and meet their expectations, they are more likely to become emotionally invested in the program and identify with the podcasts or podcasters, which promotes long-lasting interest in the program and a deeper commitment [8,9].

The current study is based on the assumption that the emotional experience of listening explains audience commitment and consumption decision making. It is likely that this emotional experience is driven by the characteristics of the podcasts as well as the audience’s need for media value. We propose an audience-centered research framework based on uses and gratifications (U&G) theory and parasocial theory. For listeners motivated by emotional needs, parasocial interaction (PSI) is likely to impact listeners’ intention to purchase podcaster-endorsed items. PSI refers to the interpersonal interaction and emotional connection between media characters and their audiences during media use. The development of digital technology has expanded the scope and targets of PSI [10,11]. A study on microblogging found that PSI positively affects endorsement effectiveness and brand credibility [12]. A study on Instagram found that PSI reinforces the perceived credibility of influencers and consumers’ willingness to purchase products that influencers recommend [13]. The emotional bond that is nurtured between podcasters and their listeners presents opportunities to increase the effectiveness of podcasters’ recommendations [1].

Both the media and the characteristics of the audience influence PSI, and scholars have argued that the audience’s traits, such as motive to use the media, are more likely to dominate PSI than the media content itself [14,15]. Thus, listeners’ motives for listening to podcasts were included as a significant independent variable. Research based on U&G theory has found that the value offered by a form of media triggers an intimate relationship between the audience and the media, indirectly affecting the audience’s attitudes toward the brand and consumption decisions [16,17]. This means that understanding the audience’s motive to engage in PSI can help to elucidate the mechanisms underlying this phenomenon [15]. The current study also considers the role of podcast commitment, which reflects listeners’ attitudes towards establishing a continuous and stable relationship with podcast programs [18]. This variable is an essential predictor of PSI and a vital prerequisite for shaping consumers’ purchase intentions [19,20].

The current study was based on three important research objectives. In Taiwan, podcasts are a new media form that is under rapid development; however, it has yet to garner attention from local researchers. Therefore, this study focused on podcast listeners in Taiwan for user profiling and usage behavior characterization of the Taiwanese market. Although podcast advertising is favored by many brands, the mechanisms underlying this popularity have yet to be elucidated. Under the framework of U&G theory and parasocial theory, this study puts forward a novel audience-centered perspective on the audience’s motives and their effect on intentions to purchase podcaster-endorsed items. The value that listeners perceive in listening to podcasts contributes to their emotional experience, which then transforms into behavioral intentions. This conceptual framework extends the scope of existing research by linking listeners’ commitment to purchase intentions. Our findings confirm the significance of podcasts as a medium for marketing campaigns, increasing the knowledge for both podcast production teams and consumers of audience psychographics of this new media form. To achieve the study objectives, we conducted an empirical study with 578 samples in Taiwan. We collected data through an online questionnaire and analyzed the data using structural equation modeling (SEM) and mediation analysis.

2. Literature Review and Hypothesis Development

2.1. Theoretical Basis of the Study

This study utilizes uses and gratifications (U&G) theory and parasocial theory to explore the psychological factors influencing podcast listeners’ commitment and purchase intentions. The U&G theory is a research framework focused on explaining individual psychological and social needs when using media [21]. U&G theory posits that people make deliberate and purposeful choices when selecting media. Viewing media through the perspective of U&G theory helps explore how individuals use and consume media, aiding in understanding the functions and services that media provide to audiences (Kim, 2020 [22]). Over the past half-century, U&G theory has provided crucial insights into numerous media studies, spanning from traditional media to the recent surge in mobile media, contributing significantly to academic research. For instance, Kim [22], guided by the U&G theory, examined consumers’ motives for watching YouTube unboxing videos and how the motives influence purchase intentions. Madan and Kapoor [17] applied U&G theory to study the motives for using Instagram, finding that user needs significantly influence their willingness to engage with brands and indirectly affect consumer decisions.

Parasocial theory also plays a crucial role in examining users’ emotional interactions with new media and their consumption decision-making process. PSI refers to a type of interpersonal interaction that occurs during media exposure [10]. Dibble et al. [23] argued that PSI occurs when “media performers acknowledge the audience’s presence in their performance, adapt the conversational style of informal face-to-face gatherings, and bodily and verbally address their users.” Users can engage in PSI with any media figure, such as politicians, movie actors, pop stars, news anchors, athletes, and even avatars [24]. While the user does not know these figures personally, the user may build trust in and become dependent on them in various ways [25]. Through the mediation of computers, the emotional relationships between media audiences and media characters have become a widespread phenomenon, attracting significant attention from scholars. For instance, Farivar et al. [26] emphasized that parasocial relationships in the field of influencer marketing have emerged as a key factor influencing purchase intentions, demonstrating a significant impact.

Both theories have a rich history and have consistently contributed significant research efforts in the evolving landscape of media and human behavior, proving to be robust frameworks applicable to the new media environment (e.g., [22,26]). This study acknowledges the potential role of podcasters as influencers in shaping consumer intention within the podcast consumption environment. Based on empirical research findings, it is clear that the audience’s psychological needs and their emotional connections with podcasters exert a notable influence in the podcast context. Therefore, integrating the U&G theory with parasocial theory establishes a strong foundation for advancing this study.

2.2. Podcasts and Parasocial Interaction

Technology is constantly changing the way that information is communicated, as well as changing the content, production methods, business models, and industrial structure of the media. As an audio-based presentation, podcasts have allowed users to focus more on the process of information dissemination [2]. Scholars have found that the use of podcasts is straightforward and is thus applied by listeners in a wide variety of scenarios, such as at home, during transportation, at work, and while exercising [27]. Podcasts were initially defined as audio programs that are available on the web as downloadable audio programs [28]. Nowadays, podcasts also use streaming media, such as YouTube, as distribution platforms. The concept of video podcasting has been embraced, and the definition of podcasting is changing as the form of the media evolves [7].

With this growth, the parasocial interaction (PSI) between the podcaster and the listener has become a concept of interest. Schlütz and Hedder [29] endeavored to demonstrate that podcasts facilitate the establishment of an intimate relationship between listeners and hosts. For example, podcasters often utilize an informal narrative style that creates for listeners an intimate audio environment that keeps them following the program as the stories unfold [30]. Podcasts can provide listeners with customizable experiences through diverse program formats that allow listeners to choose preferred topics, narratives, and host personalities; furthermore, the technological features of podcasts breakdown time and space constraints, allowing for more autonomy in listeners’ media consumption [29,31]. The rise of social media has also created the conditions for deepening PSI between listeners and podcasters, and podcasters can continue to interact with listeners and deepen their understanding of each other through social media [32]. As listeners continue to tune in, podcasters can build impressions, allow listeners to get to know them, and even see them as friends [33]. When a sense of intimacy and identification develops between the listener and the blog host, PSI can further develop into a parasocial relationship (PSR) [34].

2.3. Podcast Commitment and PSI

In social psychology, commitment refers to an individual’s long-term attitude toward a relationship, including psychological attachment and a willingness to continue the relationship [35]. Podcast commitment in this study is defined as viewers’ long-term attitudinal dispositions toward podcast programs, the desire to develop a stable relationship with a podcast program, the willingness to maintain this relationship, and confidence in the stability of the relationship [36]. Commitment is crucial in establishing behavioral motives in ongoing relationships [35]. Podcast commitment is conceptually similar to the continued use of media [37], and it is as such likely to reflect listeners’ intention to continue listening to podcasts. Listeners’ commitment to podcast programs and ongoing listening behavior are essential foundations for building PSI with podcasters. Several studies have suggested that increased contact between followers and online celebrities is likely to increase their willingness to continue following celebrities, increase perceived celebrity authenticity and intimacy, and trigger more robust PSI [38]. Based on the above discussion, the following hypothesis is proposed:

H1. 

PSI is positively correlated with podcast commitment.

2.4. Motives to Listen, Podcast Commitment, and PSI

Motives to listen to podcasts may have a significant effect on podcast commitment and PSI with podcasters. The U&G theory suggests that media users satisfy their needs through the platforms and messages that they choose and that they choose media that will meet their expectations [21]. Users have different patterns of behavior when it comes to media use, which is due to the varied purposes and psychological or social needs that they aim to fulfill [39]. In other words, the motive to use media directly affects attitudes toward media [7,40]. For example, Lin and Huang [8] found that users’ stickiness to podcasts was more substantial when they perceived that podcasts had features that met their needs for entertainment, storytelling transportation, and multitasking. Research on motive and PSI has revealed that people are more likely to use media and generate PSI with their favorite media characters when they have needs for information, entertainment, and relationship building [15,41]. The U&G perspective is not only applicable to traditional media but has also been widely applied in studies on new media. For example, Perks et al. [27] proposed the following four motives for listening to podcasts based on the U&G theory: edutainment, storytelling transportation, social engagement, and multitasking. Building on this research, the current study further explores how usage motives affect listeners’ commitment to their favorite podcasts and PSI with podcasters.

2.4.1. Edutainment

Edutainment refers to the combination of educational and entertaining features that allows listeners to acquire new information in an enjoyable way [27]. In the last century, it has been suggested that entertainment, information, and education are important components of broadcasting services [42]. While some view the growth of knowledge as distinct from entertainment, which is a pleasurable experience, many forms of media seek to simultaneously provide education and entertainment [43]. The combination of entertainment and education is widely valued in the podcast context, with market reports from Taiwan suggesting that entertainment, interests, and professional knowledge are all highly valued by podcast listeners [6]. Thus, edutainment is an important listening motive. Research suggests that podcast programs that satisfy listeners’ entertainment needs can increase consumers’ stickiness to podcasts [8]. There is also evidence that integrating entertainment techniques into media content as a tool for teaching and learning can help to increase users’ attention and engagement, as well as induce positive emotional reactions and emotional experiences in listeners [43]. This study thus proposes the following hypotheses:

H2 and H3.

Podcast commitment (H2) and PSI (H3) are positively correlated with edutainment.

2.4.2. Storytelling Transportation

Storytelling transportation occurs when listeners are immersed in the narration of the podcast to the point that they can escape from their current reality [27]. When listeners are immersed in the narrative content presented by podcasts, their attention to reality is diverted, resulting in a sense of transcendence (Van Laer et al., 2014 [44]). The medium of sound maximizes listeners’ immersion [27]. Scholars note that “stories, and their ability to transport their audience, constitute a central part of human life and the consumption experience” [44]. This means that if podcasts provide content that meets listeners’ storytelling transportation needs, then there is an opportunity to increase listener stickiness and further increase the audience’s intention to continue listening to podcasts [8]. Empathy is an important mechanism of storytelling transportation; it enables listeners to feel the experiences of the characters in the story and understand the world from their perspective [44]. This study thus proposes the following hypotheses:

H4 and H5.

Podcast commitment (H4) and PSI (H5) are positively correlated with storytelling transportation.

2.4.3. Social Engagement

Listeners motivated by social engagement seek to acquire information and stories that they can share with others, fulfilling their need to build social relationships [27]. Today’s social media platforms provide users with the opportunity to build and maintain a wide range of relationships and social interactions [45]. Scholars have highlighted companionship as a key motive for media use, which can be satisfied through the development of PSI, especially for people who lack real-world social connections [45]. Companion and social support are also important drivers of audience engagement and social connections positively moderate the relationship between audience engagement and listener stickiness to podcast programs [46]. This study thus proposes the following hypotheses:

H6 and H7.

Podcast commitment (H6) and PSI (H7) are positively correlated with social engagement.

2.4.4. Multitasking

Listeners practice multitasking when they engage in at least one other activity at the same time as they listen to podcasts to enhance their work or life productivity [27]. Research suggests that media multitasking is influenced by users’ cognitive, affective, or usage habits [47]. Thus, usage profiles change as media formats change. For example, research on traditional television media found that many viewers watched television programs while doing housework [48]. Many listeners listen to podcasts while doing routine or boring tasks, thus increasing their productivity [31]. This study thus proposes the following hypotheses:

H8 and H9.

Podcast commitment (H8) and PSI (H9) are positively correlated with multitasking.

2.5. Podcaster Endorsement and Purchase Intentions

2.5.1. Podcast Commitment and Intention to Purchase Podcaster-Endorsed Items

While new media and technology have improved the efficiency of marketing activities, the establishment of sustained relationships between consumers and brands remains important. Commitment plays a significant positive role in the establishment of consumer intentions. Research has shown that commitment to a brand can improve customer loyalty and brand perception and even influence purchase decisions in settings such as online shopping, fitness apps, and social media [49,50,51]. Thus, the following hypothesis is proposed:

H10. 

Podcast commitment is positively correlated with intention to purchase podcaster-endorsed items.

2.5.2. PSI and Intention to Purchase Podcaster-Endorsed Items

Social media has significantly expanded the role of influencers in advertising. The response of consumers towards advertising messages can greatly affect the outcome of marketing campaigns. Research has found that PSI plays a crucial role in consumers’ attitudes towards advertising messages. When consumers have an ongoing PSR with celebrities, their attitudes and behaviors towards brands or goods endorsed by these celebrities are likely positive [12,13]. For example, it has been found that when listeners are unfamiliar with the podcaster (i.e., when there is no PSI), consumers usually perceive the goods recommended by the podcaster as lacking credibility and authenticity [52], whereas high levels of PSI have a positive impact on endorsement effectiveness and brand credibility and further promote consumers’ purchase intentions [12,53]. Therefore, to enhance the effectiveness of marketing communication, advertisers are willing to spend a significant amount of their budget on hiring media celebrities to recommend their products [52]. The synergy between PSI and podcaster endorsement has made a significant contribution to marketing campaigns due to the focused nature of the podcast audience and the down-to-earth nature of podcasters, which are more likely to garner a level of trust [2]. Considering the possible positive relationship between podcast commitment and PSI and the significant role of PSI in enhancing consumption intentions, we suggest that PSI mediates the relationship between podcast commitment and intention to purchase podcaster-endorsed items. Therefore, the following hypothesis is proposed:

H11. 

PSI is positively correlated with intention to purchase podcaster-endorsed items.

H12. 

PSI mediates the relationship between podcast commitment and intention to purchase podcaster-endorsed items.

3. Methods

3.1. Participants and Procedure

To achieve our research objectives, we designed an audience-centered conceptual framework to examine the relationships among the motives for listening to podcasts (i.e., edutainment, storytelling transportation, social engagement, and multitasking), podcast commitment, PSI with podcasters, and intention to purchase podcaster-endorsed items. This conceptual framework is shown in Figure 1.

Our study population was drawn from Taiwanese podcast listeners. We used an online questionnaire to obtain sample data. Before formal data collection, we conducted semi-structured in-depth interviews with 20 experienced podcast listeners in December 2021 and invited them to complete the questionnaire and make suggestions for revisions. Participants in the interviews were required to be at least 20 years old, have listened to podcasts every week for at least two years, and have participated in offline activities organized by podcast programs. During the interviews, we discussed the following topics with the participants: (1) their podcast listening habits; (2) their motives for listening to podcasts; (3) the influence of the content of podcasts and podcasters on their perceptions of the programs; (4) the behaviors that they use in addition to listening to express their support for podcasts or podcasters; and (5) suggestions for questionnaire design. To thank the respondents for their participation in the study, we provided each participant with a merchandise coupon worth TWD 200. The most important goal of this phase was to understand the characteristics of podcast listeners and to adjust the questionnaire design to ensure that the questionnaire was contextualized and easy to complete.

Formal data collection was conducted from 5 January to 28 February 2022. Random sampling was used to ensure that as many respondents as possible were reached. A variety of means were used to facilitate the spread of the online questionnaire. First, we created the questionnaire through Google Forms and then reposted the link to the questionnaire to podcast fan communities on Facebook, Clubhouse, and Telegram. In February 2022, our research assistants distributed cards with the QR code linked to the questionnaire at Bailingguo News’ offline event in Taipei, and many influencers shared our cards in their Instagram Stories, which further expanded the impact of the questionnaire. In addition, to incentivize listeners to participate in the study, we prepared prizes such as SONY WH-1000XM4 wireless noise-canceling headphones and 10 merchandise vouchers valued at TWD 500, which were given to randomly selected participants. A total of 578 valid samples were obtained.

Each participant was informed of the process and purpose of the study and their permission was obtained through consent forms. The study design, questionnaire, and informed consent were reviewed by the Research Ethics Committee of National Taiwan University. The questionnaire asked the respondents a series of questions about themselves and their participation in podcast programs under the following five categories: (1) their demographic characteristics such as gender, age, and education level; (2) their habits of podcast use, such as frequency of listening, length of time of listening, platforms used, types of programs in which they are interested, and the number of programs to which they have subscribed; (3) their motives for listening to podcasts and podcast commitment; (4) the degree of PSI between the respondents and podcasters; and (5) intentions to purchase podcaster-endorsed items.

Of the sample of 578 cases, 44.3% were male subjects. The largest age group was 20–29 years old (53.6%), followed by 30–39 years old (24.9%). The education level of the respondents was generally high, with 85.8% of respondents university educated. The respondents indicated that they listened to podcasts most frequently when they were commuting (57.8%), working (26.0%), or exercising (24.0%). The types of podcasts that the respondents listened to most frequently were general chatting (88.6%), knowledge (48.1%), and news (30.8%). The analysis showed that 46.4% of the respondents have been listening to podcasts for one to two years, 19.4% have been listening to podcasts for more than two years, and 46.7% of the respondents are loyal listeners of their favorite podcast, regularly tuning in to the latest releases.

Table 1. Demographic profile (n = 578).

Variables		Distribution	%
Gender	Male	256	44.3
	Female	322	55.7
Age	<20 years old	78	13.5
	20–29 years old	310	53.6
	30–39 years old	144	24.9
	40–49 years old	43	7.4
	50–59 years old	3	0.5
Education level	High school or below	82	14.2
	College or university	390	67.5
	Master’s degree or above	106	18.3
Occupation	Student	227	39.3
	Employee	279	48.3
	Freelance	26	4.5
	Others	46	8.0
When do you listen to podcasts?	While commuting	334	57.8
	While working	150	26.0
	While exercising	139	24.0
	While doing housework	102	17.6
	At any time	48	8.3
	During other activities not mentioned here	31	5.4
What types of podcasts do you usually listen to?	Chatting	512	88.6
	Knowledge	278	48.1
	News	178	30.8
	Storytelling	167	28.9
	Others	96	16.6
How many years have you been listening to podcasts?	<1 year	198	34.3
	1–2 years	268	46.4
	2–3 years	96	16.6
	>3 years	16	2.8
How often do you listen to podcasts?	Occasionally	114	19.7
	I listen to topics I like, but I don’t have a regular habit.	194	33.6
	I regularly tune in to keep up to date with the latest releases.	270	46.7

summarizes the demographic characteristics of the sample.

3.2. Measures and Analysis

Relevant scales were selected from the literature to measure each study variable. To ensure consistency, all items were measured using a seven-point Likert scale, from (1) “strongly disagree” to (7) “strongly agree”. All items were first discussed and translated by the research team to produce a preliminary Chinese version of the questionnaire. Next, we used the back translation method, inviting translators who were unfamiliar with the original scale to re-translate the Chinese questionnaire into English. We applied this method in several rounds to ensure that the Chinese and English versions of the questionnaire were semantically consistent. As mentioned above, we also invited 20 experienced podcast listeners to complete the questionnaire, and based on their feedback, we further revised the scale to ensure that the measurement was appropriate.

We adapted the Podcast U&G Scale developed by Perks and Turner [27] to measure the four listening motives with three items each. It is important to note that in the original scale, edutainment comprises seven items and storytelling transportation comprises five items. When we conducted the pre-test, we found that respondents perceived the individual items of edutainment and storytelling transportation to be repetitive and they were unable to recognize the difference between these items. In addition, some of the items for edutainment only described the meaning of entertainment; thus, based on the respondents’ comments, we removed the measurement items that were not consistent with the study objectives. We used the ten-item version of the PSI Scale [14,54] to measure the PSI between participants and podcasters. This scale was initially applied to measure the degree of PSI between TV viewers and TV news anchors; therefore, we adapted the wording of the items to fit the context of this study. Podcast commitment was measured using the six items proposed by Lin et al. [55], with adaption to fit the context of the study. We selected and adapted three items from the scale proposed by Tran et al. [56] to measure the intention to purchase podcaster-endorsed items.

This study used SPSS and AMOS for data analysis. First, we analyzed the frequency distribution of the demographic profile and descriptive statistics of the research variables using SPSS. We then applied the two-step structural equation modeling (SEM) approach proposed by Anderson and Gerbing [57] for hypothesis validation. Using AMOS, we conducted confirmatory factor analysis (CFA) to ensure the validity and reliability of each construct. Then, to investigate the correlations among the study variables and confirm the research hypotheses, SEM was applied. Finally, we followed the bootstrapping method proposed by Zhao et al. [58] to analyze for mediation effects.

4. Results

4.1. Testing of Measurement Model

To evaluate the reliability and validity of the measurement constructs, CFA was conducted using AMOS and reliability analysis was performed using SPSS.

Table 2. Summary of CFA results.

Constructs and Items	Standardized Factor Loading	Mean	SD
Edutainment (α = 0.835)
CE1	0.835	5.59	1.21
CE2	0.858	5.83	1.15
Storytelling transportation (α = 0.859)
ST1	0.784	5.90	1.17
ST2	0.843	6.22	0.97
ST3	0.852	6.15	1.04
Social engagement (α = 0.869)
SE1	0.888	4.71	1.83
SE2	0.924	4.59	1.85
SE3	0.700	4.14	1.99
Multitasking (α = 0.805)
MT1	0.638	4.91	1.74
MT2	0.825	4.99	1.72
MT3	0.824	4.95	1.72
Parasocial interaction (α = 0.921)
PSI1	0.684	4.15	1.80
PSI2	0.634	4.46	1.85
PSI3	0.763	5.00	1.61
PSI4	0.762	4.27	1.81
PSI5	0.742	3.78	1.79
PSI6	0.731	5.12	1.50
PSI7	0.799	4.40	1.81
PSI8	0.713	3.65	1.72
PSI9	0.745	4.16	1.69
PSI10	0.719	4.89	1.62
Podcast commitment (α = 0.926)
PC1	0.837	5.69	1.24
PC2	0.834	5.74	1.23
PC3	0.791	5.20	1.54
PC4	0.810	5.79	1.31
PC5	0.850	5.57	1.29
PC6	0.810	5.87	1.17
Intention to purchase podcaster-endorsed items (α = 0.731)
IPPI1	0.725	5.27	1.29
IPPI2	0.694	4.66	1.44
IPPI3	0.886	4.28	1.73

χ²=1144.471, d.f. = 379, χ²/d.f. = 3.020, p < 0.001, RMSEA = 0.059, GFI = 0.873, AGFI = 0.844, NFI = 0.906, RFI = 0.892, IFI = 0.935, TLI = 0.925, CFI = 0.935.

and

Table 3. Correlation matrix for variables.

	EDU	ST	SE	MT	PSI	PC	IPPI
Edutainment	0.847
Storytelling transportation	0.771 ***	0.827
Social engagement	0.501 ***	0.485 ***	0.843
Multitasking	0.309 ***	0.314 ***	0.306 ***	0.767
Parasocial interaction	0.518 ***	0.521 ***	0.624 ***	0.377 ***	0.731
Podcast commitment	0.704 ***	0.798 ***	0.534 ***	0.332 ***	0.718 ***	0.822
Intention to purchase podcaster-endorsed items	0.545 ***	0.519 ***	0.535 ***	0.323 ***	0.727 ***	0.634 ***	0.773
CR	0.835	0.866	0.879	0.809	0.919	0.926	0.815
AVE	0.717	0.684	0.711	0.589	0.534	0.676	0.597

The numbers on the diagonal are the square roots of the AVE values. CE: edutainment; ST: storytelling transportation; SE: social engagement; MT: multitasking; PSI: parasocial interaction; PC: podcast commitment; IPPI: intention to purchase podcaster-endorsed items. *** p < 0.001.

summarize the results. As recommended by Chin et al. [59] and Hair et al. [60], the standardized factor loadings for all items measured in CFA should exceed 0.6. Thus, we deleted one item for edutainment due to a factor loading of less than 0.60; factor loadings for the remaining items passed the threshold. The results of the analyses showed that the adjusted measurement model had a better-fit index: χ² = 1144.471, d.f. = 379, χ²/d.f. = 3.020, p < 0.001, RMSEA = 0.059, GFI = 0.873, AGFI = 0.844, NFI = 0.906, RFI = 0.892, IFI = 0.935, TLI = 0.925, and CFI = 0.935. As shown in Table 3, the construct reliability (CR) of the study variables is greater than the desired value of 0.70. In addition, the average variance extraction (AVE) for all variables is greater than the desired value of 0.50, indicating that the measurement model has good convergent validity [61]. Further, the square root of the AVE for all variables is greater than the correlation coefficients between the variable and other variables, which indicates that the measurement model has good discriminant validity. Based on the above information, it can be inferred that the measurement model has sufficient validity and reliability.

4.2. Hypothesis Testing

SEM has the advantages of considering multiple variables simultaneously, allowing measurement errors in independent and dependent variables as well as integrating factor analysis and path analysis. It is also more reasonable than traditional regression analysis. The SEM results indicated that the measurement model fits well: χ² = 1117.870, d.f. = 383, p < 0.001, RMSEA = 0.058, GFI = 0.8. 1117.870, d.f. = 383, χ²/d.f. = 2.919, p < 0.001, RMSEA = 0.058, GFI = 0.878, AGFI = 0.851, NFI = 0.908, RFI = 0.896, IFI = 0.938, TLI = 0.929, and CFI = 0.938. Figure 2 summarizes these results.

The findings showed that there was a significant positive correlation between podcast commitment and PSI (β = 0.611, p < 0.001); this supports H1. There was a significant positive correlation between edutainment and podcast commitment (β = 0.168, p = 0.006); however, there was no significant correlation between edutainment and PSI; thus, the findings support H2 but not H3. There was a significant positive correlation between storytelling transportation and podcast commitment (β = 0.579, p < 0.001) and a significant negative correlation between storytelling transportation and PSI (β = −0.208, p = 0.005), which supports H4 but not H5. There was a significant positive correlation between social engagement and podcast commitment (β = 0.123, p < 0.001) as well as between social engagement and PSI (β = 0.406, p < 0.001), which supports H6 and H7. There was no significant correlation between multitasking and podcast commitment and multitasking showed a weak positive correlation with PSI (β = 0.094, p = 0.010), thus supporting H9 but not H8. In addition, the study data showed that there was a significant positive correlation between podcast commitment and intention to purchase podcaster-endorsed items (β = 0.154, p = 0.002) and between PSI (β = 0.731, p < 0.001) and intention to purchase podcaster-endorsed items, supporting H10 and H11.

In order to examine H12, we referred to the procedure proposed by Zhao et al. [58] and followed the bootstrapping method proposed by Preacher and Hayes [62] to examine mediation effects. With a sample size of 5,000 and a 95% confidence interval, we found that PSI significantly mediates the effect of podcast commitment on the intention to purchase podcaster-endorsed items. The analysis showed that the total effect (β = 0.601, p = 0.001) and direct effect (β = 0.154, p = 0.010) of podcast commitment on the intention to purchase podcaster-endorsed items were significant. The indirect effect (β = 0.447, p = 0.001) through PSI was also significant and relatively strong.

5. Discussion and Conclusions

5.1. Discussion

Podcasts are an audio medium that has received increasing attention from marketers due to their market performance and growing user base [2]. With strong influences and loyal audiences, podcasters are recognized as media influencers. The ongoing relationship between podcasters and their audiences increases the effectiveness of advertisements; thus, marketing budgets increasingly include the hiring of podcasters to recommend products and engage consumers [52,53]. The current study proposes an audience-centered framework that integrates U&G theory with motives to listen as well as PSI and podcast commitment. This framework examines how the perceived value of media use affects consumers’ affective experience and thereby their intentions to purchase podcaster-endorsed items.

We found a strong positive relationship between podcast commitment and PSI. This result suggests that listeners’ commitment to a podcast program has the potential to significantly improve their relationship with the program’s podcaster. The use of media is a prerequisite for PSI; thus, through ongoing podcast use, the exposure and connection between listeners and podcasters increases and, eventually, listeners develop emotions such as trust in, dependence on, and support for the podcaster, at which point PSI occurs [23,25].

This study also found correlations between audience motives for podcast listening and podcast commitment and PSI. Specifically, podcast commitment was positively correlated with edutainment, storytelling transportation, and social engagement; however, there was no clear relationship between podcast commitment and multitasking. We also found that PSI between podcasters and listeners was negatively correlated with storytelling transportation and positively correlated with social engagement and multitasking; however, there was no significant correlation between PSI and edutainment. It is important to note that the correlation coefficient between multitasking and PSI was low, with a negligible effect. The literature suggests that the motives for media use can be categorized as ritualistic or instrumental. Ritualistic listeners believe that the form of the media is more important than the content in the media, and they may be less active in listening activities and less involved in the program [63]. Listeners motivated by instrumental motives are more selective and engaged with podcasts, and they value the media content more, which triggers more purposeful use behaviors [64]. Multitasking can thus be classified as ritually motivated; listeners motivated by multitasking view podcasts as a way to enhance their efficiency and will typically engage in other activities or tasks while listening [8,27]. Listeners who listen to podcasts so that they can engage in multitasking are more likely to prioritize the more active task (e.g., work or exercise) [31]. Similar findings were found by Lin and Huang [8], whose study noted that the ability to manage other tasks while listening to podcasts was not an important determinant of listeners’ stickiness to the podcasts and had no significant effect on the intention to continue listening. Thus, the effect of multitasking on podcast commitment is insignificant and its relationship with PSI is weak.

Edutainment, storytelling transportation, and social engagement are instrumental motives that positively predict podcast listening based on cognitive demand [65]. Listeners who are motivated by these factors have clear preferences with regard to podcast programs. For example, listeners motivated by edutainment actively choose to listen to particular podcasts to enhance their knowledge as well as have an enjoyable experience [27]. Within this context, listeners make active choices; however, PSI with the podcaster is not the main driver. This explains our findings of a significant positive relationship between edutainment and podcast commitment but not between edutainment and PSI. Listeners motivated by storytelling transportation are keen on narratives that create a high-quality immersive experience, which means that these listeners value the content of the media much more than the podcasters themselves. Joining the virtual worlds created by podcasts, listeners develop a sense of immersion, and this further motivates listeners’ commitment to and intention to continue using the podcast program [8,44]. This explains the significant positive correlation between storytelling transportation and podcast commitment and its negative correlation with PSI. Listeners motivated by social engagement have a stronger intention to build interpersonal relationships [15]; they seek satisfaction both through interaction with podcasters and fulfilling interpersonal relationships by actively participating in the podcast program to build connections with other listeners. Therefore, there are positive relationships among social engagement, PSI, and podcast commitment [15,41].

Finally, we found that podcast commitment and PSI had a significant positive effect on the intention to purchase podcaster-endorsed items and on the relationship between podcast commitment and purchase intentions. PSI mediates the relationship between podcaster commitment and intention to purchase podcaster-endorsed items. Cases of utilizing media personalities to extend marketing objectives and enhance marketing effectiveness are common in industry and academia. Podcast listeners are generally well educated and economically well-off. They want to consume trustworthy content [4], and podcasters are perceived to be more approachable and are thus more likely to gain consumer trust [2]. In the presence of a high level of PSI between the listener and the podcaster, consumers’ willingness to purchase goods is enhanced by credibility and authenticity [12,13,52]. Podcast commitment not only enhances listeners’ trust in podcasts but also makes listeners more willing to believe in the information present in the podcasts, further enhancing the willingness of listeners to purchase goods recommended by podcasters [49,50,51]. We found that the effect of commitment on purchase intention is often not direct. This is supported by previous studies; for example, Wang et al. [66] found that commitment does not have an immediate impact on purchase intention but can change consumers’ attitudes toward brands. This means that through the intermediary of PSI, the effect of commitment on purchase intention is likely to be amplified.

5.2. Implications

Given the growing size of the podcast market and the associated increase in marketing activities and budgets, it is important to gain a deeper understanding of the psychological needs of podcast listeners. This study was conducted from a user-centered perspective. Motives to listen were combined with PSI and podcast commitment to explain listeners’ willingness to purchase products recommended by podcasters. Because most related studies are based on the Western market, this study is a rare example of the needs and habits of the Asian market. We focused specifically on Taiwan, which has experienced rapid development in its podcast industry. Theoretically, we put the user at the center of our framework, combining U&G and parasocial theory to propose a model of the perceived value of media use and a listener’s affective experiences and their purchase intentions. This model not only reveals the influence of motives on the direct use of media but also considers their impact on third-party products. Thus, this model appropriately describes the current market trends, such as increasing cross-media usage, composite advertising methods, and the growing influence of digital celebrities and influencers in consumer decision making.

In addition to theoretical expansion, this study provides an essential reference and insights for marketers. First, we elucidate the relationships among podcast listening motives, edutainment, storytelling transportation, social engagement, multitasking, podcast commitment, and PSI. This helps to pinpoint the differences in the needs of different user groups and to discover how these differences affect their podcast engagement and interpersonal relationships with podcasters. Specifically, listeners motivated by edutainment, storytelling transportation, and social engagement were more motivated to choose programs whose content satisfies their psychological needs and demonstrated stronger dependence and commitment to their chosen programs. However, only viewers motivated by social engagement paid attention to the interaction with podcasters in addition to the program. This implies that podcast production teams should seek to understand the needs of their audience. For example, for knowledge, news, and storytelling programs, content planning is more important, while for chat programs, it is the charisma of the podcasters that satisfies the audience’s needs for social interaction and emotional connection. In addition, we confirm that podcast commitment and PSI are essential factors influencing listeners’ consumption decisions in the Taiwanese podcast market. This implies that marketers should select podcast programs with higher audience stickiness, engagement, and emotional commitment for product placement.

5.3. Limitations and Suggestions for Future Study

Despite its valuable contributions, this study remains subject to certain limitations. First, this is a cross-sectional study with a sample size of 578 cases in the Taiwanese context, which means it cannot reflect the dynamic process of changes in the market and users’ psychology. Thus, including time as a factor would help to expand the scope of this study. In addition, all the data in this study were obtained through self-reporting, which makes it difficult to exclude research bias. The marketing activities generated by podcasts are rich and diverse and this study only focuses on podcaster endorsement. This is notable as podcast advertising is set to become more complex. Thus, exploring the synergies between podcasts and other social media would be a worthwhile avenue for future research. We hope that this study inspires further research into audio-media advertising from an audience-centered perspective. Furthermore, we examined podcasts in the Taiwanese market as a whole, which makes our findings less precise to specific programs and audiences. Future research could categorize podcasts and examine advertising contexts in more detail, as well as consider the effects of advert format, time of appearance, content, and program type.