Sage Journals: Discover world-class research

Abstract

Subscription to multiple subscription video-on-demand (SVOD) services, known as multi-homing, is increasingly prevalent. This research explores the relationship between multi-homing and traditional media consumption and identifies the key factors that predict multi-homing tendencies among SVOD users. We analyzed data from 2019 to 2021 across the U.S., Canada, China, Japan, Germany, and France to investigate the impact of SVOD usage on legacy media viewing time, and then explored the predictors of multi-homing behaviors. Our findings indicate that multi-homing users are more inclined to consume content on legacy media platforms, such as terrestrial or cable television, compared with single-homing users. Critical differentiators between multi-homing and single-homing users include demographic profiles, lifestyle preferences, content expectations, and usage habits, particularly in terms of account sharing and primary devices for content consumption. These results are crucial for both SVOD and traditional media providers, as they offer strategic guidance for companies to effectively operate in various international markets. Such a comprehensive analysis of multi-homing behavior also offers a new perspective on the dynamics of global media consumption.

Plain language summary

This study delves into the growing trend of “multi-homing” in media consumption, where individuals subscribe to multiple online streaming platforms alongside their traditional TV viewing. We sought to uncover who these multi-homers are and how their behavior impacts their engagement with conventional television. Over 3 years, we surveyed participants across six countries: the USA, Canada, China, Japan, Germany, and France. Our analysis reveals that multi-homing is not just about having more choices; it’s about a shift in how people engage with media. Those who subscribe to several streaming services also tend to watch more traditional TV, suggesting that rather than replacing old with new, digital and traditional media consumption can coexist and complement each other. Demographically, younger individuals, those with higher education levels, and people with greater household incomes are more inclined to be multi-homers. This group values diversity in their content and expects a high-quality viewing experience. Lifestyle choices, content expectations, and viewing habits also play significant roles in their propensity to subscribe to multiple services. Our findings offer valuable insights for streaming service providers and traditional TV broadcasters, highlighting the importance of understanding viewer preferences and consumption patterns in the evolving media landscape. For academia, this research contributes to the broader discourse on digital media’s impact on traditional media consumption, offering a nuanced understanding of viewer behavior in the age of streaming services.

Keywords

subscription video-on-demand (SVOD)multi-homing multiple-subscription pay-television media industry

Introduction

With the surge in streaming service usage, “multi-homing”—that is, subscription to multiple subscription video-on-demand (SVOD) services simultaneously—is on the rise (Bakos & Halaburda, 2020; Belleflamme & Peitz, 2019; Wu & Chiu, 2023). People in the U.S. subscribe to an average of 4.5 SVODs per month (Gruenwedel, 2022), while in Germany, the average is 3 SVODs per month (Prange, 2021). According to The Guardian, many SVOD subscribers believe that canceling cable television and using only some SVOD services is more economical (Aratani, 2022). However, the monthly cost of subscribing to multiple SVOD services such as Netflix, Disney Plus, Hulu, and HBO Max can be as high as $50, which is comparable to a basic cable plan’s price. On average, U.S. consumers spend $552 annually on streaming services (Orentas & Allen, 2024). Despite the high cost of subscribing to multiple SVOD services, a significant number of global users are subscribed to at least two SVOD services (Gruenwedel, 2022; Orentas & Allen, 2024; Prange, 2021). However, subscriber retention requires SVOD providers to offer reasonable prices. In North America, for instance, Netflix recently increased its prices to increase profits, which resulted in its first decline in subscribers in a decade (Aratani, 2022; Sperling, 2022). Netflix’s attempt to stop account sharing has also been received negatively by users (Orentas & Allen, 2024). Netflix’s case shows that SVOD providers must attract new subscribers while retaining their existing base by grasping the traits of multi-homing users.

In today’s digital media landscape, consumer behavior generally falls into one of two distinct patterns: “single-homing,” that is, preferring a single service, and “multi-homing,” that is, subscribing to several services to fulfill diverse needs. Multi-homing users differ from single-homing users in several ways. People who socialize with others of different genders and ages are more likely to be multi-homers who also use multiple social network services (Kwon et al., 2017). Gender, age, education, and monthly frequency of platform use are predictors of multi-homing (Singh et al., 2022). While numerous studies have explored multi-homing and single-homing behaviors in the context of social media and online shopping (Gu et al., 2016; B. C. Kim et al., 2017; Kwon et al., 2017; Singh et al., 2022; Wan et al., 2020), few target users of streaming services such as Netflix. Although Bravo and Farro-Mejía (2023) utilized semi-structured interviews to explore perceptions of multi-homers, empirical research on media usage among SVOD multi-homers is still limited. We thus explore the prevalent patterns of media use among SVOD subscribers and identify the predictors of multi-homing to better understand such users’ behaviors.

People who subscribe to multiple SVODs are “heavy users” who appreciate the diverse content offered by each SVOD provider and are willing to pay the cost of accessing such variety (Kimura, 2016). Multi-homing users differ in their service utilization compared with single-homers. Goode (2020) investigated the factors influencing users’ likelihood of switching services to identify distinctions between multi-homers and single-homers. SVOD platforms provide users with the flexibility to watch content at their convenience, offering a versatile viewing experience (Li, 2020; Mulla, 2022). The convenience and variety of choices in SVOD services can complement users’ consumption of legacy media, resulting in a balanced viewing experience encompassing both digital and traditional media. Many countries have shown simultaneous growth in both SVOD services and traditional pay TV, which confirms their complementary relationship (S. Lee et al., 2021).

Nonetheless, research exploring how users’ multi-homing behavior regarding online streaming services influences their consumption of legacy media is scarce. This gap in knowledge makes it challenging to determine whether the use of multiple streaming services leads users to substitute traditional media or use both services. The limited knowledge of factors influencing multi-homing makes it difficult for SVOD services to develop business strategies for subscribers. We thus examined the relationship between SVOD multi-homing and legacy media viewing times to understand users’ characteristics and media consumption patterns. The results of this study will provide operators with insights into multi-homing subscriber acquisition and retention strategies.

While research exists on the relationship between demographics and multi-homing (Amendola et al., 2015; Byun & Baek, 2022; Singh et al., 2022), there is no clear conclusion about how users’ lifestyles, SVOD expectations, and usage habits affect multi-homing. Accordingly, we conducted surveys between 2019 and 2021 targeting users in the U.S., the U.K., China, Japan, Germany, and France—regions with high penetration of pay TV (both traditional and virtual; Stoll, 2023). As a higher proportion of households in these regions watch TV than in other regions, we expect our sample to provide us valuable insights into the legacy media usage habits of multi-homing users. These multiyear and cross-national data analyses are further crucial for identifying those most likely to subscribe to multiple SVODs. Based on the number of SVODs used, we categorized SVOD users as single-homing, heavy multi-homing, or light multi-homing users. We also investigated (1) how multi-homing affects terrestrial or cable network viewing and (2) how user characteristics, such as demographics, lifestyle, quality expectations, and usage patterns, influence multi-homing. To investigate the differentiated impact of multi-homing among SVOD users on terrestrial and paid broadcast viewing time across various countries, we applied the seemingly unrelated regression (SUR) model, and then conducted a logistic regression analysis to identify predictors of multi-homing behavior among users.

In summary, we conducted a detailed examination of multi-homing behaviors among SVOD users by utilizing an extensive dataset that spans multiple regions to deliver a comprehensive perspective on global media consumption trends. The insights thereof will unravel the interaction between traditional and digital media and offer SVOD providers critical knowledge to strategically navigate the competitive market.

Literature Review

SVOD Multi-Homing

Multi-homing refers to the practice of utilizing multiple technologies to fulfill one’s media consumption needs (Bravo & Farro-Mejía, 2023). In SVOD multi-homing, users subscribe to multiple SVODs simultaneously (Bakos & Halaburda, 2020; Belleflamme & Peitz, 2019; Wu & Chiu, 2023). The emergence of diversified subscription-based video streaming services has made SVOD multi-homing increasingly common. On average, individuals in the U.S. maintain subscriptions to 4.5 streaming services per month (Gruenwedel, 2022), while in Germany, people subscribe to at least 3 (Prange, 2021). Multi-homing is marked by a desire to access a variety of entertainment and information; such multi-homers actively engage across multiple platforms (Hyrynsalmi et al., 2016) and are willing to pay for a wider range of content provided by different SVOD platforms (Kimura, 2016). Unlike single-homing subscribers, who remain loyal to one service, multi-homers have flexible viewing habits and an interest in diverse media content. This change in media consumption is noteworthy (Bravo & Farro-Mejía, 2023), and it emphasizes the need for service providers to gain a comprehensive understanding of multi-homing users.

As noted earlier, digital media consumers are either single-homers or multi-homers, where the latter tend to socialize with diverse age groups and genders, engage with various social networking services (Kwon et al., 2017), and often use multiple credit cards in the case of older multi-homers with higher incomes (Amendola et al., 2015). Gender, age, education, and monthly frequency of food platform use are predictors of multi-homing on food platforms (Singh et al., 2022). Prior studies comparing multi- and single-homers mostly focus on social media usage (Gu et al., 2016; Kwon et al., 2017) and commerce (B. C. Kim et al., 2017; Singh et al., 2022; Wan et al., 2020), with few studies on SVOD services. Although Bravo and Farro-Mejía (2023) used semi-structured interviews to examine perceptions of multi-homing, the media usage patterns of SVOD multi-homers are less known. We fill this gap in knowledge by understanding how SVOD users consume mainstream media and what drives their multi-homing behavior.

Media Substitution Theory

Media substitution theory suggests that various media types compete for our time and attention; one implication is that new media replace older media (McCombs, 1972; Yamatsu & Lee, 2023). In the framework of media substitution theory, individuals allocate their time to consume media within certain constraints. That is, when people engage with a specific media, they naturally reduce their usage time of other media. This behavior makes it crucial to know how people manage their media time under the condition of multiple choices. McCombs (1972) suggests that new media’s entry into the market leads audiences to redistribute their financial and time resources between both old and new platforms. As mass media fulfill similar roles, new platforms with comparable features might displace older ones. Media substitution theory posits that new media advancements lead to the decline of traditional media, while complementary theory argues that both can coexist and potentially enhance one another, especially when offering different functions and satisfying various needs (S. J. Kim et al., 2020).

Numerous studies in this area investigate whether new media serves to replace or enhance traditional media. Studies based on the media replacement theory argue that people have limited resources of time and money (McCombs, 1972) and that using one type of media reduces the resources necessary for using other existing media (McCombs, 1972). These studies primarily compare usage times between new and traditional media (H. Kim et al., 2013; P. S. Lee & Leung, 2008; S. Y. Lee et al., 2016). The literature on media consumption yields diverse findings: a substitution pattern between traditional and new media usage (Budzinski et al., 2021; S. Y. Lee & Lee, 2013; Nimrod, 2019; Yamatsu & Lee, 2023); a complementary relationship between traditional and new media (S. J. Kim et al., 2020; Udoakpan & Tengeh, 2020), or neither (Hall et al., 2019).

Subscribing to two or more SVOD services is becoming increasingly common all over the world (Gruenwedel, 2022; Prange, 2021). Owing to time constraints, multi-homing users might watch fewer terrestrial or cable channels and spend more time watching SVOD than single-homing users. Scholars have also examined the relationship between genre and home television viewing time of a single SVOD service (Han & Lee, 2012) and the effect of paid broadcast viewing (e.g., cable TV, Internet Protocol Television [IPTV], satellite TV) and paid VOD expenditure on SVOD usage (S. Lee, 2021). Their studies have mainly investigated the substitutive or complementary relationship between cable television and SVOD.

Despite the growing number of multi-homing users globally, very few studies have examined the relationship between SVOD multi-homing and legacy media. Jung and Melguizo (2023) found that the period from 2015 to 2020 witnessed a significant rise in the adoption of over-the-top streaming services in Latin America, aligning with a notable trend toward cord-cutting—terminating traditional cable or satellite TV subscriptions—and cord-shaving—diminishing usage without full termination. Despite the surge in multiple SVOD subscriptions, there exists no definitive evidence to assert that multi-homing is replacing traditional media consumption.

We thus investigated how multi-homing influences variations in legacy media usage by focusing on SVOD multi-homing users, which has not been previously studied. Discussing how SVODs have displaced or supplemented legacy media across countries is challenging, as most research has been limited to a single nation. Hence, we examined data collected from six countries between 2019 and 2021 to address this research gap. Examining the relationship between SVOD multi-homing and legacy media consumption across countries will reveal global media substitution or supplementation trends.

Research Question 1. Does multi-homing cause an increase or decrease in the time spent watching terrestrial and cable television? Do these relationships differ across countries?

Predictors of SVOD Multi-Homing

Lifestyle and Demographics

A lifestyle is the pattern of a person’s life, shaped by the employment of personal resources such as time and money in specific circumstances (Kaynak & Kara, 2001). Lifestyle is a crucial concept for market segmentation and identifying target customers (Kahle & Kennedy, 1988; Novak & MacEvoy, 1990; Plummer, 1974). People’s lifestyles, which reflect their values, opinions, and interests, influence their media consumption (Chan & Leung, 2005; Kwak et al., 2021).

Lifestyles may be categorized using the list of values (LoV) model (Kahle & Kennedy, 1988). This model distinguishes lifestyles based on how people prioritize internal values such as excitement, achievement, and self-esteem; external values such as respect, belonging, and safety; or those centered on interpersonal relationships and warmth. In this context, “yuppies” are characterized by a focus on success and social status, and are more inclined to adopt new media compared with “strugglers,” who face financial constraints and prioritize basic needs over technological adoption (Wei, 2006). Those who seek fun and enjoyment in life and value warm relationships with others tend to spend more on SVOD services than those who do not (Kwak et al., 2021). People with traditional, family-oriented lifestyles tend to use multiple media forms more than those in other lifestyle groups (Shim et al., 2008), indicating that SVOD multi-homing may also be influenced by one’s lifestyle.

Plummer (1974) suggests that demographics and lifestyles should be considered together to comprehend people’s purchasing behavior. Demographics are a fundamental factor in describing an individual’s media consumption patterns (Dupagne, 1999; Kwak et al., 2021; Palomba, 2020). SVOD multi-homing is likely influenced by demographic factors, such as age and education, which are associated with an individual’s technology acceptance and willingness to pay (Goyanes, 2014; Kwak et al., 2021; Singh et al., 2022). Additionally, income and educational levels might significantly impact multi-homing, as users with multiple SVOD subscriptions typically incur higher costs than those who single-home.

While it is reasonable to assume that multi-homing behaviors may differ based on users’ lifestyle choices and demographic factors, we have limited extant findings on these potential variations. We thus investigated whether the lifestyles and demographic characteristics of SVOD users in each country predict their tendencies towards multi-homing. Comparing these relationships between SVOD subscribers in key regions—such as North America (U.S., Canada), East Asia (China, Japan), and Western Europe (Germany, France)—could have substantial implications for market segmentation.

Research Question 2. Do user demographics and lifestyles affect SVOD multi-homing? Do these relationships differ across countries?

Expectations for SVOD

According to expectation-disconfirmation theory, satisfaction fluctuates based on the discrepancy between an individual’s prior expectations and the actual performance of a product or service (Oliver, 1980). Positive confirmation arises when the product or service exceeds expectations, leading to user satisfaction. Conversely, negative disconfirmation, stemming from unmet expectations, diminishes user satisfaction (McCollough et al., 2000). This theory has been applied across diverse use contexts, including on websites (Cheung & Lee, 2011), social networks (Ha et al., 2021; Kourouthanassis et al., 2015), e-learning platforms (Dai et al., 2020; Dangaiso et al., 2022), and mobile applications (Wang & Zhou, 2023; Wen et al., 2023). Indeed, we do find a consistent link between heightened expectations of system and service quality and increased user satisfaction (Chen & Chang, 2018; Dangaiso et al., 2022; Park, 2020). Despite extensive studies on expectations and satisfaction, we know little of how expectations impact multi-homing behavior. Users who pay the monthly membership cost for SVOD are likely to anticipate a high level of quality. Especially, multi-homers may expect far higher quality, because they pay higher subscription prices than single-homers. Thus, we sought to answer whether user expectations regarding content quality, system reliability, service excellence, and cost-effectiveness can predict multi-homing behavior.

Research Question 3. Do users’ SVOD expectations (e.g., content, systems, services, and costs) affect SVOD multi-homing? Do these relationships differ across countries?

SVOD Usage Habits

Users struggling with the cost of monthly SVOD subscriptions often resort to sharing their accounts with others, including strangers, family, or friends. Subscription sharing is particularly prevalent among multi-homers, whose media consumption patterns are costlier to sustain (Orentas & Allen, 2024). This behavior, driven by a desire for diverse content, suggests that account sharing may be an important predictor of multi-homing.

The type of device used can also influence media consumption patterns. The rise of the Internet has led to a decrease in the use of legacy media (S. Y. Lee et al., 2016; Woo et al., 2014), with an uptick in online video consumption correspondingly reducing the time spent on traditional video media (Jung & Melguizo, 2023; S. Y. Lee et al., 2016). The more content users watch on smartphones, the less they watch it on television (Jang & Park, 2016). Smartphone usage may further affect time spent on personal computers (PCs) and televisions for media consumption (Hwang & Lee, 2011). The convenience of mobile devices, which allows users to access content anytime and anywhere, appears to particularly appeal to multi-homers. Consequently, the choice of a primary viewing device—especially the preference for smartphones over traditional devices—may serve as an important predictor of multi-homing behavior.

Users who predominantly consume content at home are likely to engage with media for prolonged durations, creating a favorable environment for subscribing to multiple SVOD services to satisfy their varied content needs (Starosta et al., 2020). In contrast, individuals who primarily watch content outside the home might exhibit different consumption patterns and motivations. Given the distractions and time constraints associated with external environments, these users may prefer shorter viewing sessions, diminishing the necessity for a broad variety of content. Consequently, they might perceive a single SVOD platform as sufficient for their entertainment needs, as it provides adequate variety within the limited time they allocate for content consumption outside the home. This distinction suggests that the primary viewing location where users engage with content could be a significant predictor in determining their likelihood to adopt multi-homing behavior.

While there is potential for users’ media consumption patterns to distinguish between single-homing and multi-homing behaviors, empirical research in this area remains scarce. We fill this gap by examining whether factors such as account sharing, the primary device for content consumption, and the location of content consumption serve as predictors of multi-homing behavior among SVOD users (Figure 1).

Research Question 4. Do user usage habits (e.g., account sharing, primary viewing device, and viewing location) affect SVOD multi-homing? Do these relationships differ across countries?

Figure 1.

Research model.

Methodology

Sample and Data Collection

To gather user samples, we collaborated with the market research firm Macromill Embrain to conduct quota sampling, which was based on country, gender, and age criteria. Responses were obtained from 1,400 terrestrial or cable TV users, with an even distribution of 200 individuals per country, covering a wide age range from the 20s to the 70s. We collected responses from 7,200 participants over 3 years. Considering the year-over-year growth of the SVOD market in North America (U.S., Canada), East Asia (China, Japan), and Western Europe (Germany, France) (Statista, 2023), we targeted six countries with a substantial SVOD user base. The final analysis, aimed at understanding multi-homing users, incorporated 3,032 individuals who subscribed to more than one SVOD service and also consumed terrestrial or pay TV. Table 1 reports the distribution of these respondents across the six countries.

Table 1.

Country-Wise SVOD Use.

Single/multi-homing	Total	U.S.	U.K.	China	Japan	Germany	France
Single-homing	1,035	216	177	196	131	134	181
Two SVOD subscriptions	641	83	129	145	87	114	83
More than three SVOD subscriptions	1,356	221	300	235	123	222	255
Total	3,032	520	606	576	341	470	519

Measure

To assess the impact of users’ multi-homing behavior on legacy media consumption, we investigated the number of SVOD services participants subscribed to and their viewing habits for both terrestrial and pay TV. Participants reported the number of SVOD services they are currently subscribed to, with the maximum possible response capped at three. To classify SVOD users, we divided participants into two main categories based on their subscription patterns: single-homing and multi-homing. Single-homers are defined as those subscribed to only one SVOD service, while multi-homers are characterized by subscriptions to two or more SVOD services. We further segmented multi-homers into two subcategories to capture varying levels of service usage: light multi-homers (subscribed to exactly two SVOD services) and heavy multi-homers (subscribed to three SVOD services). For analytical purposes, single-homing was coded as 1, light multi-homing as 2, and heavy multi-homing as 3.

Legacy media usage refers to the average time spent per day watching videos on terrestrial or pay TV. This variable was measured using a 7-point Likert scale (1 = less than 30 min to 7 = more than 5 hr).

To find the predictors of multi-homing, we investigated the demographics, lifestyle, expectations for SVOD, and usage habits of SVOD users. Demographic data were collected from participants, which included details on their gender, age, educational attainment, occupational status, household income, and the size of their household. For analysis, respondents’ employment status was classified into two categories: employed and unemployed.

Lifestyles in this study were categorized into six distinct dimensions using LoV and employed as predictors. The concept of “lifestyles” was understood as a cognitive manifestation of users’ underlying fundamental needs and objectives (Kwak et al., 2021). These persistent beliefs shape specific choices, attitudes, and behaviors in users (Homer & Kahle, 1988; Rokeach, 1969; Schwartz, 1994). Measurements based on the LoV scale can be better explained when used with demographic variables (Novak & MacEvoy, 1990). Items that focus on LoV were adapted from Kwak et al. (2021).

Social relationships refer to the desire to interact with others, while family-oriented is the desire to support or spend time with family members. Self-confidence refers to one’s trust in oneself. Self-management refers to having one’s responsibilities, whereas self-fulfillment refers to gaining satisfaction from indulging in creative activities. Culture and consumption-focused refers to the tendency to be passionate about sports and popular culture. Individuality refers to the inclination to value one’s personality. Finally, achievement refers to the disposition to appreciate one’s success and accomplishments. All dimensions were measured using a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree), which showed adequate reliability (Nunnally, 1978).

Expectations for SVOD are the extent to which users expect SVOD to provide a certain level of content, services, systems, and price. Content expectations relate to the content provided by an SVOD, whereas system expectations are related to an SVODS’s operation and usage procedures. Finally, service expectations refer to SVOD service quality expectations, whereas cost expectations refer to SVOD price expectations. All items were adapted from Al-Fraihat et al. (2020) and Shin and Park (2021). All dimensions were measured using a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree), which showed adequate reliability (Nunnally, 1978).

SVOD usage habits were categorized based on account sharing, primary viewing devices, and viewing location. Account sharing was determined by whether an SVOD account is used by multiple individuals, with non-sharers coded as 0 and sharers as 1. The primary viewing device was identified as the device predominantly used to access SVOD content, categorizing users of stationary devices (e.g., desktop PCs, TVs) as 0 and those using portable devices (e.g., smartphones, laptops, tablets) as 1. Primary viewing location was defined by the preferred location for viewing SVOD content, coding users who predominantly watch outside their home as 0 and those who watch primarily at home as 1. Table 2 shows the items used in the study, Cronbach’s alpha for the measures, and the means and standard deviations of the variables.

Table 2.

Measurement of Items.

Construct	Items	Cronbach’s α	Mean (SD)
SVOD multi-homing	Number of SVOD subscriptions (single-homing: 1, light multi-homing: 2, and heavy multi-homing: 3)	—	2.03 (0.89)
Legacy media usage	Daily average time spent watching terrestrial or cable television (less than 30 min: 1 ∼ More than 5 hr: 7)	—	4.12 (1.65)
Gender	Respondent’s gender (male: 1, female: 2)	—	—
Age	Respondents’ ages (20–29 years old: 1 ∼ over 50 years old: 4)	—	2.32 (1.08)
Education	Respondents’ educational status (unschooled: 1 ∼ Above graduate school: 4)	—	2.81 (0.75)
Occupation	Respondents’ employment status (employed: 0, unemployed: 1)	—	—
Household income	Respondents’ monthly household incomes ($1,000 or less: 1 ∼ $10,000 or more: 11)	—	4.73 (2.73)
Household size	Respondents’ current number of family members (live alone: 1 ∼ more than five: 6)	—	3.17 (1.40)
Social relationships	I get along well with people I am sociable I am actively involved in various gatherings	.667	3.42 (0.86)
Family-oriented	I (will) support my aged parents/parents-in-law (in the future) I (will) sacrifice everything for my children (in the future) I enjoy going out with my family	.668	3.66 (0.82)
Self-confidence	I think I am the most precious thing in the world I swear a bright future lies before me I respect my personality	.680	3.47 (0.83)
Self-management	I accomplish all I aim to do I intend to watch/know the news every day I intend to complete assigned tasks	.657	3.75 (0.73)
Self-fulfillment	I like to write or make something by hand I am satisfied with my sincere report at school or work I like changes in my daily life	.667	3.49 (0.79)
Culture and consumption-focused	I intend to buy a brand-new product before others I am passionate about sports and pop music I enthusiastically support the local baseball/soccer team	.703	3.21 (1.00)
Individuality	I live my way regardless of other’s judgments I do not care much about others’ assessments of me I want to live a life different from others	.660	3.55 (0.77)
Achievement	I spare no effort to get ahead in the world I can do what I really want to do if I rise in the world through my effort I work hard to enjoy my life	.686	3.49 (0.80)
Content expectations	SVOD should provide varied content SVOD should provide regularly updated and latest content SVOD should have high quality content	.796	3.95 (0.71)
System expectations	SVOD should be easy to use SVOD should be stable SVOD should be convenient to use	.777	3.99 (0.77)
Service expectations	SVOD should be available anywhere and at any time SVOD should recommend content that suits my preferences well SVOD should quickly respond to my opinions and complaints	.696	3.77 (0.74)
Price expectations	SVOD should be reasonably priced SVOD should not be too financially burdened SVOD should show little advertising	.693	3.88 (0.80)
Account sharing	Sharing subscription (not share: 0, share: 1)	—	—
Primary viewing device	Devices used primarily for SVOD watching (immobile: 0, mobile: 1)	—	—
Primary viewing location	Location for SVOD watching (outside: 0, home: 1)	—	—

Analytical Strategies

The objective of this study is twofold: (1) to investigate the impact of multi-homing on legacy media viewing time and (2) to explore the predictors of users’ multi-homing behaviors. To this end, SVOD users were categorized into three segments: single-homing, light multi-homing, and heavy multi-homing. Multiple regression analysis was conducted to examine the effects of these categorical segments on legacy media viewing time. To uncover how these impacts vary across different countries, we employed SUR. This method is particularly beneficial for our analysis, as it allows for the estimation of multiple, interrelated regression equations, thus enhancing the efficiency and accuracy of our findings by accounting for the potential correlation of error terms across the different country-specific models (Budzinski et al., 2021). To investigate the predictors of users’ multi-homing behaviors, we employed a multinomial logistic regression model, as it facilitates the prediction of the probabilities of the different categories of multi-homing behavior based on the independent variables (H. Kim et al., 2021). Light multi-homing and heavy multi-homing were set as dependent variables, while demographics, lifestyles, expectations for SVOD, and usage habits were set as independent variables. A multinomial logistic analysis was conducted with single-homing as the reference category.

Results

Impact of Multi-Homing on Legacy Media Viewing Time

Multiple linear regression was used to examine the effect of SVOD users’ multi-homing on legacy media usage (Research Question 1). The analysis, as detailed in Table 3, indicated that demographic factors—age (β = .017, p < .001), occupation (β = .219, p < .01), and household income (β = .023, p < .05)—positively influenced legacy media usage. Specifically, older individuals, those not currently employed, and individuals with higher household incomes were found to spend more time viewing legacy media. Multi-homing, as a predictor, also positively affected legacy media usage. Compared with users who engage in single-homing, those categorized as light (β = .213, p < .01) and heavy (β = .542, p < .001) multi-homers demonstrated a significantly higher propensity to watch terrestrial or cable television.

Table 3.

Regression Analysis Results on Legacy Media Usage.

			DV: Legacy media usage (N = 3,032)
Category	Variables	Items	B	SE	β	t	VIF
Constant			3.358	0.198		16.976***
Control	Nation ref. U.S.	U.K.	0.272	0.098	.068	2.784**	1.958
		China	−0.682	0.098	−.167	−6.981***	1.892
		Japan	−0.249	0.109	−.049	−2.283*	1.528
		Germany	−0.198	0.103	−.045	−1.917	1.801
		France	−0.077	0.101	−.018	−0.762	1.880
	Year ref. 2019	2020	0.111	0.082	.030	1.342	1.595
	Year ref. 2019	2021	−0.072	0.071	−.022	−1.007	1.621
	Demographics	Gender (female = 1)	−0.007	0.057	−.002	−0.115	1.029
		Age	0.017	0.002	.133	7.245***	1.113
		Education	−0.023	0.039	−.011	−0.588	1.114
		Occupation (unemployed = 1)	0.219	0.077	.053	2.846**	1.151
		Household income	0.023	0.012	.040	1.971*	1.363
		Household size	−0.002	0.021	−.002	−0.118	1.107
Predictor	Multi-homing ref. single-homing	Light multi-homing	0.213	0.078	.054	2.733**	1.304
Predictor	Multi-homing ref. single-homing	Heavy multi-homing	0.542	0.066	.168	8.267***	1.366
R² = .086, △R² = .082; F(15, 3016) = 19.018***

p < .05. **p < .01. ***p < .001.

The SUR model was employed to examine the effects of SVOD multi-homing on legacy media viewing times across various countries. This model considers the correlation between the error terms of the dependent variable, allowing for more efficient estimation than ordinary least squares (OLS) (Koo & Choo, 2021). The SUR model’s findings, which align with the OLS results (see Table 4). Light (β = .469, p < .05) and heavy multi-homers (β = .868, p < .001) in the U.S. and heavy multi-homers in the U.K. (β = .527, p < .001), China (β = .466, p < .001), Japan (β = .492, p < .05), and France (β = .532, p < .01) were more inclined to watch terrestrial or cable television than single-homers were. These results suggest a complementary relationship between SVOD and terrestrial or cable television for all countries, except Germany.

Table 4.

Country-Wise SUR Analysis Results on Legacy Media Usage.

	U.S.		U.K.		China		Japan		Germany		France
	(N = 520)		(N = 606)		(N = 576)		(N = 341)		(N = 470)		(N = 519)
Variables	B	t	B	t	B	t	B	t	B	t	B	T
Constant	3.421 (0.458)	7.464***	4.093 (0.402)	10.192***	2.735 (0.486)	5.623***	3.909 (0.638)	6.130***	2.496 (0.408)	6.118***	3.671 (0.458)	8.007***
Gender (female = 1)	−0.218 (0.148)	−1.479	0.006 (0.125)	0.052	0.202 (0.107)	1.898	0.322 (0.186)	1.729	−0.207 (0.145)	−1.422	−0.177 (0.148)	−1.198
Age	0.015 (0.006)	2.501*	0.020 (0.005)	4.066***	0.005 (0.005)	0.925	0.009 (0.007)	1.280	0.029 (0.006)	4.826	0.010 (0.006)	1.600
Education	0.007 (0.109)	0.068	−0.251 (0.085)	−2.964**	−0.067 (0.118)	−0.566	−0.197 (0.159)	−1.240	0.117 (0.079)	1.484	−0.001 (0.091)	−0.013
Occupation (unemployed = 1)	0.517 (0.205)	2.524*	0.172 (0.162)	1.059	0.044 (0.216)	0.204	0.354 (0.242)	1.461	0.086 (0.181)	0.475	−0.081 (0.182)	−0.445
Household income	0.048 (0.026)	1.856	−0.009 (0.027)	−0.352	0.102 (0.022)	4.606***	0.021 (0.036)	0.570	−0.056 (0.034)	−1.649	0.025 (0.039)	0.648
Household size	−0.139 (0.050)	−2.771**	0.023 (0.043)	0.532	0.058 (0.054)	1.075	0.006 (0.067)	0.083	0.142 (0.050)	2.823**	−0.077 (0.057)	−1.357
2020	0.297 (0.242)	1.227	0.341 (0.184)	1.852	−0.122 (0.146)	−0.838	−0.035 (0.264)	−0.133	0.003 (0.197)	0.017	0.196 (0.219)	0.895
2021	−0.091 (0.206)	−0.442	0.044 (0.158)	0.281	−0.266 (0.131)	−2.020*	−0.215 (0.222)	−492.968	−0.093 (0.178)	−0.520	0.281 (0.188)	1.492
Light multi-homing	0.469 (0.215)	2.181*	0.080 (0.173)	0.460	0.198 (0.140)	1.412	−0.064 (0.235)	−0.272	0.208 (0.195)	1.069	0.259 (0.219)	1.183
Heavy multi-homing	0.868 (0.165)	5.256***	0.527 (0.144)	3.670***	0.466 (0.127)	3.667***	0.492 (0.222)	2.217*	0.276 (0.172)	1.605	0.532 (0.165)	3.226**

Note. DV: legacy media usage.

p < .05. **p < .01. ***p < .001.

Predictors of Multi-Homing: Demographic and Lifestyle

To explore the predictors of users’ multi-homing behaviors, we employed a multinomial logistic regression model. Multi-homing segment variables were set as the dependent variable, while demographic and lifestyle variables were used as predictors. Light multi-homing differed from single-homing by gender (B = −0.306, p < .01), age (B = −0.017, p < .001), and education (B = 0.148, p < .05) (see Table 5). Men, young people, and highly educated people were more likely to be light multi-homers than single-homers.

Table 5.

Multinomial Logistic Regression Analysis Results: Demographics and Lifestyles.

	Total (N = 3,032)
	Light multi-homing (N = 641)				Heavy multi-homing (N = 1,356)
Variables	B	Wald	Sig.	OR	B	Wald	Sig.	OR
Country
U.K.	0.704***	14.214	.000	2.022	0.919***	35.801	.000	2.508
China	0.583**	10.097	.001	1.791	0.244	2.555	.110	1.277
Japan	0.586**	8.608	.003	1.797	0.258	2.197	.138	1.294
Germany	0.830***	18.399	.000	2.293	0.934***	32.743	.000	2.545
France	0.200	0.993	.319	1.222	0.846***	28.235	.000	2.331
Year
2020	−0.098	0.433	.511	0.906	0.273*	4.463	.035	1.313
2021	−0.071	0.318	.573	0.931	0.153	1.878	.171	1.165
Demographics
Gender (female = 1)	−0.306**	8.303	.004	0.737	−0.117	1.690	.194	0.890
Age	−0.017***	14.698	.000	0.983	−0.033***	73.648	.000	0.968
Education	0.148*	4.029	.045	1.160	0.220***	12.575	.000	1.246
Occupation (unemployed = 1)	0.147	1.105	.293	1.159	0.165	1.844	.174	1.179
Household income	−0.004	0.034	.854	0.996	0.069***	13.406	.000	1.071
Household size	0.024	0.373	.541	1.025	0.038	1.289	.256	1.039
Lifestyles
Social relationships	0.012	0.016	.899	1.012	0.065	0.653	.419	1.067
Family-oriented	0.072	0.730	.393	1.075	0.212**	8.213	.004	1.236
Self-confidence	0.109	1.163	.281	1.115	−0.017	0.037	.847	0.983
Self-management	0.115	1.285	.257	1.122	0.065	0.543	.461	1.067
Self-fulfillment	0.053	0.259	.611	1.054	0.147	2.643	.104	1.158
Culture and consumption-focused	−0.013	0.040	.841	0.987	0.085	2.340	.126	1.089
Individuality	0.016	0.035	.852	1.016	0.178*	5.475	.019	1.195
Achievement	−0.173	2.868	.090	0.841	−0.056	0.404	.525	0.945
Nagelkerke R ²				.125
χ²(df)				353.612(42)***

The reference category was “single-homing (N = 1,035).”

p < .05. **p < .01. ***p < .001.

Heavy multi-homing differed from single-homing by age (B = −0.033, p < .001), education (B = 0.220, p < .001), and household income (B = 0.069, p < .001). Young and highly educated individuals with a high household income were more likely to be heavy multi-homers than single-homers. People with high family-oriented (B = 0.212, p < .01) or individuality (B = 0.178, p < .05) lifestyle scores were more likely to be heavy multi-homers than single-homers.

In conclusion, light multi-homers were more likely to be male, young, and highly educated than single-homers were; and heavy multi-homers were more likely to be young, highly educated, with high household incomes, family-oriented, and individuality lifestyle scores. The test results based on all predictors explained 12.5% of the variance in users’ multi-homing (χ ²(42) = 353.612, p < .001).

Appendix A presents the country-wise results of the multi-logistic analysis. Light multi-homing differed from single-homing by gender only in France, that is, women were more inclined to be single-homing than light multi-homing users (B = −0.707, p < .05). Light multi-homing differed from single-homing by the user’s age in all six countries. Young users in the U.S. (B = −0.044, p < .001), the U.K. (B = −0.027, p < .01), China (B = −0.047, p < .001), Japan (B = −0.034, p < .01), Germany (B = −0.037, p < .01), and France (B = −0.030, p < .01) tended to be heavy multi-homers. Highly educated users were more likely to be light (B = 0.494, p < .05) or heavy multi-homers (B = 0.685, p < .001) than single-homers only in the U.S. In the U.K., people with high household incomes were more likely to be single-homers than light multi-homers (B = −0.119, p < .05). In contrast, people with high household incomes were likely to be heavy multi-homers in China (B = 0.113, p < .05), Japan (B = 0.179, p < .01), and Germany (B = 0.174, p < .01).

The impact of lifestyle varies by country. People with a family-oriented lifestyle were more likely to be heavy multi-homers than single-homers in Western European countries such as the U.K. (B = 0.348, p < .05) and Germany (B = 0.365, p < .05). The Chinese with a self-fulfillment lifestyle (valuing accomplishing something themselves) were more likely to be heavy multi-homers than single-homers (B = 0.495, p < .05). Asians with a culture and consumption-focused lifestyle were more likely to be heavy multi-homers than single-homers (China: B = 0.582, p < .01; Japan: B = 0.502, p < .05). The Chinese with a lifestyle valuing individuality were less likely to be light (B = −0.589, p < .01) or heavy multi-homers (B = −0.414, p < .05) than single-homers. However, the Japanese with a lifestyle prioritizing individuality were more likely to be heavy multi-homers (B = .742, p < .01) than single-homers. Lastly, the French pursuing an achievement lifestyle were more likely to be single-homers than heavy multi-homers (B = −0.443, p < .05).

Predictors of Multi-Homing: Expectations for SVOD and Usage Habits

Research Questions 3 and 4 investigated whether the degree of multi-homing among users differed depending on users’ SVOD expectations and usage habits. We conducted a multinomial logistic analysis with expectations for SVOD and users’ usage habits as predictors, and multi-homing as the dependent variable. Single-homing users were the reference group. Table 6 presents the analysis results.

Table 6.

Multinomial Logistic Regression Analysis Results: Expectations for SVOD and SVOD Usage Habits.

	Total (N = 3,032)
	Light multi-homing (N = 641)				Heavy multi-homing (N = 1,356)
Variables	B	Wald	Sig.	OR	B	Wald	Sig.	OR
Country
U.K.	0.671	14.647	.000	1.956	0.630	20.253	.000	1.878
China	0.605	12.012	.001	1.831	0.223	2.408	.121	1.250
Japan	0.585	9.028	.003	1.795	0.150	0.810	.368	1.162
Germany	0.890	22.652	.000	2.435	0.814	27.844	.000	2.257
France	0.244	1.663	.197	1.277	0.583	16.275	.000	1.792
Year
2020	−0.147	0.986	.321	0.863	0.172	1.844	.174	1.187
2021	−0.105	0.710	.400	0.900	0.103	0.897	.344	1.109
Expectations for SVOD
Content	0.166	1.981	.159	1.181	0.317**	10.013	.002	1.373
System	−0.046	0.181	.670	0.955	−0.037	0.162	.687	0.964
Service	0.091	0.777	.378	1.095	0.264**	9.240	.002	1.302
Price	−0.071	0.524	.469	0.931	0.004	0.002	.961	1.004
SVOD usage habits
Account sharing	0.189	3.146	.076	1.207	0.694***	59.406	.000	2.002
Primary viewing device (share = 1)	0.223*	4.321	.038	1.250	0.265**	8.613	.003	1.303
Primary viewing location (home = 1)	0.158	0.904	.342	1.171	−0.155	1.375	.241	0.857
Nagelkerke R²					.125
χ²(df)					353.612(42)***

The reference category was “single-homing (N = 1,035).”

p < .05. **p < .01. ***p < .001.

Heavy multi-homing differed from single-homing in terms of users’ content (B = 0.317, p < .01) and service expectations (B = 0.264, p < .01). People with high expectations for content and services were more likely to be heavy multi-homers than single-homers. People sharing an account with others were more likely to be heavy multi-homers than single-homers (B = 0.694, p < .001). People using mobile devices (e.g., smartphones, laptops, tablet PCs) to watch videos tended to be light (B = 0.223, p < .05) or heavy multi-homers (B = 0.265, p < .01) compared with single-homers. The test results based on all predictors explained 12.5% of the variance in users’ multi-homing (χ²(28) = 230.039, p < .001).

Appendix B details the outcomes of the country-specific multi-logistic analysis, demonstrating how users’ expectations for SVOD services influence their light and heavy multi-homing behaviors. In Japan, individuals with heightened expectations for SVOD content were more likely to engage in both light (B = 0.987, p < .01) and heavy multi-homing (B = 1.096, p < .01) relative to single-homing. In Germany, users with more significant expectations of SVOD services exhibited a stronger tendency toward light multi-homing (B = 0.616, p < .01) than toward single-homing. Conversely, in France, users with high expectations for the SVOD system were less inclined to participate in either light (B = −0.942, p < .01) or heavy multi-homing (B = −0.486, p < .05) than single-homing.

Heavy multi-homing differed from single-homing in terms of users’ SVOD account sharing in all countries except France. People sharing accounts with others were more likely to be heavy multi-homers than single-homers (U.S.: B = 0.732, p < .01; U.K.: B = 0.701, p < .01; China: B = 0.943, p < .001; Japan: B = 0.862, p < .01; Germany: B = 0.635, p < .05). In China and France, individuals who predominantly used mobile devices for viewing SVOD content demonstrated a higher propensity for both light (China: B = 0.501, p < .05; France: B = 0.624, p < .05) and heavy multi-homing (China: B = 0.524, p < .05) than did single-homers. In the U.S., individuals who predominantly viewed SVOD content at home were less likely to engage in heavy multi-homing than were single-homing users (B = −0.598, p < .05). Conversely, in the U.K., those primarily watching SVOD at home showed a greater likelihood of being light multi-homers than single-homers (B = 1.235, p < .01).

Table 7 summarizes the main results for each research question. First, multi-homers watched more terrestrial or cable television than single-homers, indicating that SVOD services and legacy media were complementary (Research Question 1). Second, young and highly educated people with large household incomes were more likely to be multi-homers than single-homers. Those with family-oriented or individuality-focused lifestyles were more likely to be multi-homers than single-homers (Research Question 2). Third, people with considerable expectations regarding SVOD content or services were more likely to be multi-homers than single-homers (Research Question 3). Finally, people who shared their SVOD accounts or often watched SVOD on their mobile devices (e.g., mobiles or tablet PCs) were more likely to be multi-homers than single-homers (Research Question 4).

Table 7.

Comparison of Multi-Homing Users Based on the Predictors.

RQ	Construct	Light multi-homing	Heavy multi-homing	Single-homing
RQ1	Legacy media use	Terrestrial or cable TVviewing (+)	Terrestrial or cable TVviewing (+)	Reference group
RQ2	Demographics	Gender (−)Age (−)Education (+)	Age (−)Education (+)Household income (+)
RQ2	Lifestyles	—	Family-oriented (+)Individuality (+)
RQ3	Expectations for SVOD	—	Content (+)Service (+)
RQ4	SVOD usage habits	Primary viewing device (+)	Account sharing (+)Primary viewing device (+)

Table 8 summarizes the country-wise results. Multi-homers in the U.S., U.K., China, Japan, and France watched more terrestrial or cable television than single-homers.

Table 8.

Country-Wise Comparison of Multi-Homing Subscribers.

Construct	Light multi-homing	Heavy multi-homing	Single-homing
U.S.	Terrestrial or cable TV viewing (+)Education (+)	Terrestrial or cable TV viewing (+)Age (−)Education (+)	Reference group
U.K.	Household income (−)	Terrestrial or cable TV viewing (+)Age (−)Family-oriented (+)
China	Culture consumption-focused (+)Individuality (−)	Terrestrial or cable TV viewing (+)Age (−)Household income (+)Self-fulfillment (+)Culture consumption-focused (+)Individuality (−)
Japan	Age (−)	Terrestrial or cable TV viewing (+)Age (−)Household Income (+)Culture consumption-focused (+)Individuality (+)
Germany	Age (−)	Age (−)Household income (+)Family-oriented (+)
France	Gender (−)Age (−)	Terrestrial or cable TV viewing (+)Age (−)Achievement (−)

Young and highly educated Americans were more likely to be multi-homers than single-homers. Low-income British were more likely to be light multi-homers, and young and family-oriented British were more likely to be heavy multi-homers than single-homers.

Young Chinese with high household income and self-fulfillment or culture and consumption-focused lifestyles were more likely to be heavy multi-homers than single-homers. Individuality-oriented Chinese people were more likely to be single-homers. Young Japanese were more likely to be light multi-homers than single-homers, and those young, well-earning with culture and consumption-focused or individuality-oriented lifestyles were more likely to be heavy multi-homers than single-homers.

Young Germans were more likely to be light multi-homers than single-homers, and those who were young and well-earning with family-oriented lifestyles were more likely to be heavy multi-homers than single-homers. Males and young French were more likely to be light multi-homers than single-homers, and older people with achievement-focused lifestyles were more likely to be single-homers than heavy multi-homers.

Discussion

Let us now discuss our findings. First, multi-homers showed higher terrestrial or cable television viewing times than single-homers (Research Question 1). Similar findings were found for all countries, with heavy multi-homers being more inclined to watch terrestrial or cable television than single-homers. The result that multi-homers viewed legacy media more than single-homers indicates that legacy media and SVOD are complementary (S. J. Kim et al., 2020; S. Lee et al., 2021; Udoakpan & Tengeh, 2020).

According to this study, multi-homers can be regarded as cord couplers who use both pay TV and SVOD services concurrently depending on their needs and preferences (Jung & Melguizo, 2023), as well as heavy users who enjoy a variety of content via legacy media and SVOD services. Recently, however, many SVOD providers have introduced password-sharing restriction policies to enhance their profitability. Providers such as Netflix, Disney, and Hulu have eliminated password sharing, resulting in increased revenue (Pequeño, 2024). However, the increased costs of anti-sharing policies have led to an uptick in the cancellation of subscriptions (Orentas & Allen, 2024). Although Netflix recorded its highest revenue after implementing its anti-sharing policy, the longevity of this success remains uncertain. Similarly, the outlook for other SVOD providers such as Disney, Hulu, and MAX is not very optimistic (Saul, 2024). SVOD operators need to adopt diversified pricing strategies to lock multi-homers into their services. Currently, terrestrial and cable broadcasters provide their programs to SVOD via partnerships, a strategy anticipated to attract multi-homers successfully. SVOD services must retain existing subscribers by bundling current terrestrial/cable plans with a subscription SVOD service plan or by combining multiple SVODs to offer services at a slightly discounted rate to reduce multi-homing consumers’ financial burdens.

Second, SVOD multi-homing differed from single-homing in terms of their lifestyles and demographics (Research Question 2). People who were young, highly educated, and high-income earners with family-oriented or individuality-focused lifestyles were more likely to be multi-homers than single-homers. In Western countries (U.K. and Germany), those with a family-oriented lifestyle were more likely to be multi-homers than single-homers. This may be because Disney Plus, which offers family-friendly programming, was first launched in the West. Therefore, operators should consider offering family-friendly content that families can access internationally. Users in Asian countries (China and Japan) with lifestyles valuing sports and pop culture are more likely to be multi-homers than single-homers, satisfying their need to consume such content via multi-homing. Asian operators can retain multi-homers on their platforms by rapidly acquiring and distributing sports and pop culture content based on users’ preferences. U-next, a prominent Japanese over-the-top platform, exemplifies how understanding and catering to local cultural preferences can lead to success in the SVOD market (Frater, 2024). Boasting an extensive library of Korean content, including K-pop artist performances, dramas, and variety shows, U-next has tapped into the growing popularity of Hallyu (the Korean Wave) in Japan. Besides Korean content, the platform offers a wide array of Japanese animations, comics, novels, and more, showcasing its commitment to content diversification.

The pursuit of individuality was also significant in Asia, but in a different way. In China, people with an individualized lifestyle are less likely to be multi- than single-homers, while the opposite is true in Japan. These results can be interpreted considering the sociocultural features of these countries. People who value their individuality tend to accept the new culture well while hesitating to accept the dominant culture (Michael, 2015). Individuality-seeking individuals appreciate experimental culture and embrace diversity to express their genuine selves. (Michael, 2015). Therefore, they will likely prefer watching new content from other countries to consuming the existing mainstream culture. However, abroad video streaming platforms have never been launched in China until now (Kharpal, 2019), making it difficult for Chinese viewers to access Netflix and Disney Plus. In recent years, the lack of free content has made Chinese users more openly express concerns about the excessive fee structures of SVOD providers (Zhanhang & Zuer, 2023). As domestic platforms in China lack content diversity compared with global platforms, Chinese users who pursue individuality are more likely to choose single-homing over subscribing to multiple domestic platforms. In Japan, in contrast, foreign operators such as Netflix, Amazon Plus, Hulu, and Disney Plus established themselves relatively quickly. Consequently, those valuing individuality might be more likely to be multi-homers than single-homers to satisfy their tastes.

Third, multiple-and single-homing differed according to user expectations for SVODs (Research Question 3). Users with considerable expectations for SVODs’ content or services are more likely to be multi-homers than single-homers. Hence, the overall SVODs’ content and service quality must be improved to attract multi-homers. SVODs must provide a wide range of content, improve the algorithms of the recommended systems, and respond quickly to negative user feedback.

Fourth, multi- and single-homing differed according to the users’ usage habits (Research Question 4). The results showed that people sharing SVOD accounts with others and primarily watching videos using mobile devices (e.g., mobile phones and laptops) were more likely to be heavy multi-homers than single-homers in all countries, except France. Netflix’s strategy of restricting account sharing to improve revenue might negatively impact users all over the world, as evidenced by the rise in user complaints (Kelly, 2022; Orentas & Allen, 2024). Instead of prohibiting users from sharing their accounts unilaterally, operators should consider creating an advertising plan or diversifying their rate plan services for multi-homers in partnership with SVOD.

Finally, users who primarily watched SVOD on their mobile devices were more likely to be multi-homers. To retain multi-homers’ subscriptions, a strategy that differentiates SVOD plans for mobile device users must be implemented.

Implications

Theoretical Implications

Our findings have several interesting implications. First, we empirically examined the relationship between multi-homing and terrestrial or cable television viewing from the perspective of media replacement and complementarity. Media replacement theory argues that, if new media even partially compete with current media, the risk of competitive displacement heightens (Dimmick et al., 2004). However, research also shows that SVOD use could be complementary to traditional media (S. J. Kim et al., 2020; Udoakpan & Tengeh, 2020). Our findings add a significant dimension to this discussion by confirming that multi-homing behavior is also complementary to traditional media.

Second, our study is noteworthy because we investigated the predictors of multi-homers. Although some studies have examined the relationship between lifestyle and SVOD use (Kwak et al., 2021), few studies have evaluated whether lifestyle can predict multi-homers. We found that multi-homing and single-homing differed in terms of demographics, lifestyle, expectations for SVOD, and SVOD viewing patterns.

Practical Implications

First, our results suggest that multi-homers watch more terrestrial or cable televisions than single-homers. This implies that multi-homers see and use the two media as complementary, and not as one replacing the other. If operators combine terrestrial or cable broadcasting plans with their services and make both media affordable, SVOD operators could successfully retain multi-homers.

Second, this study can help SVOD operators develop strategies to retain and recruit multi-homers. Those who valued family and individuality were more likely to be multi-homers than single-homers. Multi-homers are likely to use several SVODs to spend their leisure time with their families and watch unique content. For multi-homers, SVOD operators should provide family-friendly content. Operators must obtain content with various ratings, production countries, genres, and formats for users who value their individuality.

Lastly, operators must make efforts to meet users’ expectations of content or services, and SVOD providers should develop strategies for providing high-quality content and services. Operators must provide up-to-date and varied content, and creating a recommendation service that accurately reflects user preferences is also necessary. Providing services that respond quickly to user complaints can also be beneficial. Differentiating multi-homers’ rate plans based on the primary viewing device can also help prevent subscription cancellation. Multi-homers are more likely to share accounts than single-homers; therefore, the regulation of charging an additional fee to account-sharing users must be carefully implemented.

Limitations

This study has some limitations. First, it was impossible to quantify the total number of subscriptions because users were only questioned about the first three SVOD services they were using during the study. As U.S. users have been shown to subscribe to an average of 4.5 video streaming services (Gruenwedel, 2022), most multi-homers are most likely to use more than two SVODs. Further research is required to determine the number of SVOD subscriptions indicating “light” or “heavy” use based on averages.

Second, our study analyzed data collected from 2019 to 2021 and did not survey the same individual at each time point. Analyzing panel data in future studies would allow us to grasp the characteristics of multi-homers more precisely.

At the time of the study, the COVID-19 pandemic increased time spent at home, which correlated with a surge in SVOD subscriptions. Our findings could be specific to the pandemic period, so caution should be exercised when interpreting and generalizing these results. We expect a more detailed investigation into the volatility of SVOD multi-homing subscriber numbers through time-series analysis that encompasses the periods before, during, and after the pandemic.

Third, we utilized quota sampling based on gender and age, which might have limited the representativeness of our findings. To mitigate this issue, we recommend employing a more systematic approach, such as stratified or cluster sampling techniques, to obtain less biased samples.

Lastly, our study did not consider cultural factors such as religion and language in the analysis of users’ demographic characteristics. With the global expansion of SVOD services, it is becoming increasingly important to collect and analyze more detailed demographic information, including users’ cultural backgrounds. This could play a crucial role in understanding the usage patterns of SVOD services. Including these cultural variables would offer opportunities for a more thorough investigation of user characteristics.

Footnotes

Appendix

Appendix B.

Country-Wise Multinomial Logistic Analysis: Relationship Between Expectations for SVOD, SVOD Usage Habits, and Multi-Homing.

	U.S.		U.K.		China		Japan		Germany		France
	Light	Heavy	Light	Heavy	Light	Heavy	Light	Heavy	Light	Heavy	Light	Heavy
	(N = 83)	(N = 221)	(N = 129)	(N = 300)	(N = 145)	(N = 235)	(N = 83)	(N = 221)	(N = 129)	(N = 300)	(N = 145)	(N = 235)
Year
2020	−.119 (.414) [0.888]	−.072 (.351) [0.931]	.194 (.356) [1.214]	.418 (.294) [1.520]	−.885** (.318) [0.413]	.029 (.281) [1.030]	−.407 (.454) [0.665]	.103 (.391) [1.108]	−.051 (.347) [0.950]	−.013 (.307) [0.987]	.448 (.407) [1.566]	.313 (.300) [1.368]
2021	−.882* (.357) [0.414]	−.501 (.292) [0.606]	.120 (.295) [1.128]	.136 (.245) [1.146]	−.527* (.263) [0.590]	−.121 (.256) [0.886]	.305 (.344) [1.356]	.134 (.334) [1.144]	.204 (.319) [1.227]	.460 (.280) [1.584]	.371 (.358) [1.449]	.583* (.257) [1.792]
Expectations for SVOD
Content	.095 (.283) [1.110]	.134 (.215) [1.143]	.047 (.323) [1.048]	−.022 (.264) [0.978]	−.117 (.252) [0.889]	.251 (.236) [1.285]	.987** (.350) [2.682]	1.096** (.340) [2.993]	−.579 (.324) [0.561]	.194 (.282) [1.214]	.590 (.329) [1.804]	.466 (.240) [1.593]
System	.479 (.261) [1.615]	.211 (.197) [1.235]	−.200 (.252) [0.818]	.001 (.214) [1.001]	.111 (.246) [1.118]	.235 (.230) [1.265]	−.394 (.351) [0.674]	−.747 (.333) [0.474]	.219 (.289) [1.245]	.168 (.249) [1.183]	−.942** (.302) [0.390]	−.486* (.231) [0.615]
Service	.048 (.235) [1.050]	.281 (.186) [1.324]	.052 (.243) [1.053]	.318 (.203) [1.374]	.200 (.259) [1.222]	.224 (.238) [1.251]	−.372 (.336) [0.689]	.123 (.321) [1.131]	.616** (.240) [1.851]	.312 (.203) [1.367]	−.166 (.286) [0.847]	.368 (.214) [1.445]
Price	−.248 (.232) [0.780]	.090 (.183) [1.094]	−.158 (.233) [0.853]	.019 (.194) [1.019]	−.053 (.206) [0.949]	.142 (.192) [1.152]	−.140 (.303) [0.869]	.158 (.289) [1.171]	.060 (.296) [1.062]	−.157 (.254) [0.854]	.497 (.283) [1.644]	−.072 (.209) [0.931]
SVOD usage habits
Account sharing	.215 (.288) [1.240]	.732** (.235) [2.079]	.247 (.243) [1.280]	.701** (.205) [2.015]	.277 (.228) [1.320]	.943*** (.207) [2.567]	.211 (.311) [1.235]	.862** (.284) [2.368]	−.203 (.300) [0.816]	.635* (.253) [1.888]	.457 (.280) [1.579]	.379 (.205) [1.461]
Primary viewing device (share = 1)	.162 (.279) [1.176]	.206 (.211) [1.229]	−.188 (.255) [0.829]	.074 (.207) [1.076]	.501* (.250) [1.650]	.524* (.226) [1.688]	.533 (.287) [1.704]	.482 (.268) [1.619]	.202 (.287) [1.224]	.197 (.250) [1.217]	.624* (.287) [1.867]	.348 (.214) [1.416]
Primary viewing location (home = 1)	.064 (.363) [1.067]	−.598* (.248) [0.550]	1.235** (.442) [3.440]	.172 (.286) [1.188]	−.567 (.321) [0.567]	−.528 (.300) [0.590]	.207 (.702) [1.230]	1.131 (.869) [3.098]	.088 (.448) [1.092]	.064 (.369) [1.066]	.288 (.531) [1.333]	.205 (.402) [1.228]
R² Nagelkerke	.127		.074		.129		.156		.091		.087
χ²(df)	60.851(18)***		40.578(18)**		69.582(18)***		50.834(18)***		38.973(18)**		40.878(18)**

The reference category is “Single-homing.”

p < .05. **p < .01. ***p < .001.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the MSIT (Ministry of Science and ICT), Korea, under the ITRC (Information Technology Research Center) support program (IITP-2024-2020-0-01749) supervised by the IITP (Institute of Information & Communications Technology Planning & Evaluation).

ORCID iD

Sang Woo Lee

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

References

Al-Fraihat

Joy

Sinclair

(2020). Evaluating e-learning systems success: An empirical study. Computers in Human Behavior, 102, 67–86. https://doi.org/10.1016/j.chb.2019.08.004

Amendola

Pellecchia

Sensini

(2015). The usage of credit cards: An empirical analysis on Italian households panel data. European Journal of Business and Management, 7(14), 132–139. https://scholar.google.co.kr/citations?view_op=view_citation&hl=ko&user=Xeg749YAAAAJ&cstart=20&pagesize=80&citation_for_view=Xeg749YAAAAJ:zYLM7Y9cAGgC

Aratani

(2022, March 8). “Cable cutters” wonder whether cheaper alternative is over as Netflix raises price. The Guardian. https://www.theguardian.com/media/2022/mar/08/cable-cutters-netflix-raises-price-streaming-cheaper-alternative

Bakos

Halaburda

(2020). Platform competition with multi-homing on both sides: Subsidize or not? Management Science, 66(12), 5599–5607. https://doi.org/10.1287/mnsc.2020.3636

Belleflamme

Peitz

(2019). Platform competition: Who benefits from multi-homing? International Journal of Industrial Organization, 64, 1–26. https://doi.org/10.1016/j.ijindorg.2018.03.014

Bravo

E. R.

Farro-Mejía

(2023). Exploring multi-homing behavior: User perceptions of subscribing to multiple streaming platforms. Issues in Information Systems, 24(3), 232–244. https://doi.org/10.48009/3_iis_2023_120

Budzinski

Gaenssle

Lindstädt-Dreusicke

(2021). The battle of YouTube, TV and Netflix: An empirical analysis of competition in audiovisual media markets. SN Business & Economics, 1(9), 116. https://doi.org/10.1007/s43546-021-00122-0

Byun

Baek

(2022). An exploratory study on the current state of multiple subscriptions to OTT services and the characteristics of users. Journal of Broadcasting and Telecommunications Research, 117, 43–69. https://doi.org/10.22876/KJBTR.2022.117.002

Chan

J. K. C.

Leung

(2005). Lifestyles, reliance on traditional news media and online news adoption. New Media & Society, 7(3), 357–382. https://doi.org/10.1177/1461444805052281

10.

Chen

C. C.

Chang

Y. C.

(2018). What drives purchase intention on Airbnb? Perspectives of consumer reviews, information quality, and media richness. Telematics and Informatics, 35(5), 1512–1523. https://doi.org/10.1016/j.tele.2018.03.019

11.

Cheung

C. M.

Lee

M. K.

(2011). Antecedents and consequences of user satisfaction with an e-learning portal. International Journal of Digital Society, 2(1), 373–380. https://doi.org/10.20533/ijds.2040.2570.2011.0045

12.

Dai

H. M.

Teo

Rappa

N. A.

Huang

(2020). Explaining Chinese university students’ continuance learning intention in the MOOC setting: A modified expectation confirmation model perspective. Computers & Education, 150, 103850. https://doi.org/10.1016/j.compedu.2020.103850

13.

Dangaiso

Makudza

Hogo

(2022). Modelling perceived e-learning service quality, student satisfaction and loyalty. A higher education perspective. Cogent Education, 9(1), 2145805. https://doi.org/10.1080/2331186X.2022.2145805

14.

Dimmick

Chen

(2004). Competition between the internet and traditional news media: The gratification-opportunities niche dimension. The Journal of Media Economics, 17(1), 19–33. https://doi.org/10.1207/s15327736me1701_2

15.

Dupagne

(1999). Exploring the characteristics of potential high-definition television adopters. Journal of Media Economics, 12(1), 35–50. https://doi.org/10.1207/s15327736me1201_3

16.

Frater

(2024, February 5). SVOD Subscriptions in Japan exceed 52 million, says report. Variety. https://variety.com/2024/tv/news/japan-svod-subscriptions-1235897898/

17.

Goode

(2020). Understanding single homing and multihoming user switching propensity in cloud file hosting service relationships. e-Service Journal, 11(2), 34–73. https://doi.org/10.2979/eservicej.11.2.02

18.

Goyanes

(2014). An empirical study of factors that influence the willingness to pay for online news. Journalism Practice, 8(6), 742–757. https://doi.org/10.1080/17512786.2014.882056

19.

Gruenwedel

(2022, May 23). Use of multiple SVOD services in the US is rebounding but remains below mid-pandemic highs. MediaPlayNews. https://www.mediaplaynews.com/npd-use-of-multiple-svod-services-in-the-u-s-rebounds-remains-below-mid-pandemic-high/

20.

L. B.

Wang

(2016). Multi-homing on SNSs: The role of optimum stimulation level and perceived complementarity in need gratification. Information & Management, 53(6), 752–766.

21.

Q. A.

Chen

J. V.

Nguyen

T. H. T.

(2021). Continuance use of enterprise social network sites as knowledge sharing platform: Perspectives of tasks-technology fit and expectation disconfirmation theory. International Journal of Knowledge Management Studies, 12(4), 429–451. https://doi.org/10.1504/IJKMS.2021.118349

22.

Hall

J. A.

Kearney

M. W.

Xing

(2019). Two tests of social displacement through social media use. Information, Communication & Society, 22(10), 1396–1413. https://doi.org/10.1080/1369118X.2018.1430162

23.

Han

Lee

S. W.

(2012). Relationship between N-screen service and home TV: Focusing on the Korean representative N-screen service, TVing. Journal of the Korea Contents Association, 12(5), 144–153. https://doi.org/10.5392/JKCA.2012.12.05.144

24.

Homer

P. M.

Kahle

L. R.

(1988). A structural equation test of the value-attitude-behavior hierarchy. Journal of Personality and Social Psychology, 54(4), 638–646. https://doi.org/10.1037/0022-3514.54.4.638

25.

Hwang

J. S.

Lee

J. H.

(2011). Analysis of media use of smartphone adopters. Journal of Cybercommunication Academic Society, 28(4), 303–351.

26.

Hyrynsalmi

Suominen

Mäntymäki

(2016). The influence of developer multi-homing on competition between software ecosystems. Journal of Systems and Software, 111, 119–127. https://doi.org/10.1016/j.jss.2015.08.053

27.

Jang

Park

(2016). Do new media substitute for old media?: A panel analysis of daily media use. Journal of Media Economics, 29(2), 73–91. https://doi.org/10.1080/08997764.2016.1170021

28.

Jung

Melguizo

. (2023). Is your Netflix a substitute for your Telefunken? Evidence on the dynamics of traditional pay TV and OTT in Latin America. Telecommunications Policy, 47(1), 102397. https://doi.org/10.1016/j.telpol.2022.102397

29.

Kahle

L. R.

Kennedy

(1988). Using the list of values (LOV) to understand consumers. Journal of Services Marketing, 2(4), 49–56. https://doi.org/10.1108/eb024742

30.

Kaynak

Kara

(2001). An examination of the relationship among consumer lifestyles, ethnocentrism, knowledge structures, attitudes and behavioural tendencies: A comparative study in two CIS states. International Journal of Advertising, 20(4), 455–482. https://doi.org/10.1080/02650487.2001.11104906

31.

Kelly

(2022, June 22). Streamers worry the end is coming for lax password-sharing rules. The Washington Post. https://www.washingtonpost.com/technology/2022/06/22/streaming-password-sharing/

32.

Kharpal

(2019, May 10). Netflix has a China strategy—But it doesn’t involve launching there soon. CNBC. https://www.cnbc.com/2019/05/10/netflix-has-a-china-strategy-it-doesnt-involve-launching-there-soon.html

33.

Kim

B. C.

Lee

Park

(2017). Two-sided platform competition with multihoming agents: An empirical study on the daily deals market. Information Economics and Policy, 41, 36–53. https://doi.org/10.1016/j.infoecopol.2017.06.006

34.

Kim

Chan-Olmsted

S. M.

Hwang

K. H.

Chang

B. H.

(2021). Examining the use, perception, and motivation of cord-cutting: A consumer segment approach. Journalism & Mass Communication Quarterly, 98(1), 126–147.

35.

Kim

Jeong

S. H.

(2013). Effect of smartphone use on traditional media uses: Application of media substitution hypothesis. Journal of Media Economy and Culture, 11(1), 88–119.

36.

Kim

S. J.

Viswanathan

Lee

H. M.

(2020). Platform war vs. platform synergy? A longitudinal analysis of media substitution between personal computers and mobile devices. Journal of Broadcasting & Electronic Media, 64(2), 65–88. https://doi.org/10.1080/08838151.2020.1718986

37.

Kimura

(2016). A simulation of consumer-side multi-homing of original and derivative digital games: Evidence from Japan. Universal Journal of Management, 4(5), 234–245. https://doi.org/10.13189/ujm.2016.040503

38.

Koo

Choo

(2021). Analysis of factors affecting travel time considering travel purpose. Journal of Korean Society of Transportation, 39(5), 697–709.

39.

Kourouthanassis

Lekakos

Gerakis

(2015). Should I stay or should I go? The moderating effect of self-image congruity and trust on social networking continued use. Behaviour & Information Technology, 34(2), 190–203. https://doi.org/10.1080/0144929X.2014.948489

40.

Kwak

K. T.

C. J.

Lee

S. W.

(2021). Who uses paid over-the-top services and why? Cross-national comparisons of consumer demographics and values. Telecommunications Policy, 45(7), 102168. https://doi.org/10.1016/j.telpol.2021.102168

41.

Kwon

H. E.

Kim

(2017). Platform structures, homing preferences, and homophilous propensities in online social networks. Journal of Management Information Systems, 34(3), 768–802.

42.

Lee

P. S.

Leung

(2008). Assessing the displacement effects of the internet. Telematics and Informatics, 25(3), 145–155. https://doi.org/10.1016/j.tele.2006.08.002

43.

Lee

(2021). An exploratory analysis of contributors to smart TV usage. The Journal of the Korea Contents Association, 21(6), 524–532. https://doi.org/10.5392/JKCA.2021.21.06.524

44.

Lee

Joo

Nam

(2021). Examining factors influencing early paid over-the-top video streaming market growth: A cross-country empirical study. Sustainability, 13(10), 5702. https://doi.org/10.3390/su13105702

45.

Lee

S. Y.

Lee

S. W.

(2013). Time displacement effect of online video service on pay TV. International Telecommunications Policy Review, 20(4), 1–21.

46.

Lee

S. Y.

Lee

S. W.

Kim

(2016). Time displacement effect of online video services on other media in South Korea. Telematics and Informatics, 33(2), 247–255. https://doi.org/10.1016/j.tele.2015.08.002

47.

S. S.

(2020). Lifestyles, technology clustering, and the adoption of over-the-top television and internet protocol television in Taiwan. International Journal of Communication, 14, 1–19. https://ijoc.org/index.php/ijoc/article/view/12259

48.

McCollough

M. A.

Berry

L. L.

Yadav

M. S.

(2000). An empirical investigation of customer satisfaction after service failure and recovery. Journal of Service Research, 3(2), 121–137. https://doi.org/10.1177/109467050032002

49.

McCombs

M. E.

(1972). Mass media in the marketplace. Journalism Monographs, 24, 1–106.

50.

Michael

(2015). It’s really not hip to be a hipster: Negotiating trends and authenticity in the cultural field. Journal of Consumer Culture, 15(2), 163–182. https://doi.org/10.1177/14695405134932

51.

Mulla

(2022). Assessing the factors influencing the adoption of over-the-top streaming platforms: A literature review from 2007 to 2021. Telematics and Informatics, 69, 101797. https://doi.org/10.1016/j.tele.2022.101797

52.

Nimrod

(2019). Selective motion: Media displacement among older internet users. Information, Communication & Society, 22(9), 1269–1280. https://doi.org/10.1080/1369118X.2017.1414865

53.

Novak

T. P.

MacEvoy

(1990). On comparing alternative segmentation schemes: The list of values (LOV) and values and life styles (VALS). Journal of Consumer Research, 17(1), 105–109. https://doi.org/10.1086/208541

54.

Nunnally

J. C.

(1978). An overview of psychological measurement. In Wolman

B. B.

(Ed.), Clinical diagnosis of mental disorders: A handbook (pp. 97–146). Springer.

55.

Oliver

R. L.

(1980). A cognitive model of the antecedents and consequences of satisfaction decisions. Journal of marketing research, 17(4), 460–469. https://doi.org/10.1177/002224378001700405

56.

Orentas

Allen

(2024, February 15). Streaming trends for 2024: 44% report streaming costs increasing over the last year. Forbes HOME. https://www.forbes.com/home-improvement/internet/streaming-survey/

57.

Palomba

(2020). Consumer personality and lifestyles at the box office and beyond: How demographics, lifestyles and personalities predict movie consumption. Journal of Retailing and Consumer Services, 55, 102083. https://doi.org/10.1016/j.jretconser.2020.102083

58.

Park

(2020). User acceptance of smart wearable devices: An expectation-confirmation model approach. Telematics and Informatics, 47, 101318. https://doi.org/10.1016/j.tele.2019.101318

59.

Pequeño

(2024, March 5). Max users will start losing password sharing late this year—The latest streaming platform crackdown. Forbes. https://www.forbes.com/sites/antoniopequenoiv/2024/03/05/max-users-will-start-losing-password-sharing-late-this-year-the-latest-steaming-platform-crackdown/?sh=74c5c31c20e8

60.

Plummer

J. T.

(1974). The concept and application of life style segmentation: The combination of two useful concepts provides a unique and important view of the market. Journal of Marketing, 38(1), 33–37. https://doi.org/10.1177/002224297403800106

61.

Prange

(2021, December 16). Whip survey: European viewers average three SVOD subscriptions. Media Play News. https://www.mediaplaynews.com/whip-survey-european-viewers-average-three-SVOD-subscriptions/

62.

Rokeach

(1969). Part I. Value systems in religion. Review of Religious Research, 11(1), 3–23.

63.

Saul

(2024, January 24). Netflix stock soars 14%—But some analysts have doubts. Forbes. https://www.forbes.com/sites/dereksaul/2024/01/24/netflix-stock-soars-14-but-some-analysts-have-doubts/?sh=76fd799a7f01

64.

Schwartz

S. H.

(1994). Are there universal aspects in the structure and contents of human values? Journal of Social Issues, 50(4), 19–45.

65.

Shim

M. S.

Kim

E. M.

Rhee

J. W.

(2008). Lifestyle, multiple media use, and cross-media genre use. Korean Journal of Journalism & Communication Studies, 52(6), 189–216.

66.

Shin

Park

(2021). Factors affecting users’ satisfaction and dissatisfaction of OTT services in South Korea. Telecommunications Policy, 45(9), 102203. https://doi.org/10.1016/j.telpol.2021.102203

67.

Singh

Kumar

Kapoor

(2022). Consumer multihoming predisposition on food platforms: Does gender matter? Journal of Retailing and Consumer Services, 67, 103029. https://doi.org/10.1016/j.jretconser.2022.103029

68.

Sperling

(2022, April 19). Netflix loses subscribers for first time in 10 years. The New York Times. https://www.nytimes.com/2022/04/19/business/netflix-earnings-q1.html

69.

Starosta

Izydorczyk

Dobrowolska

(2020). Personality traits and motivation as factors associated with symptoms of problematic binge-watching. Sustainability, 12(14), 5810. https://doi.org/10.3390/su12145810

70.

Statista. (2023, May 8). Video streaming (SVOD). https://www.statista.com/outlook/dmo/digital-media/video-on-demand/video-streaming-SVOD/worldwide?currency=usd

71.

Stoll

(2023, December 12). Penetration rate of pay TV worldwide from 2020 to 2026, by region. Statista. https://www-statista-com-ssl.access.yonsei.ac.kr/statistics/1228729/penetration-rate-pay-tv-worldwide/#statisticContainer

72.

Udoakpan

Tengeh

R. K.

(2020). The impact of over-the-top television services on pay-television subscription services in South Africa. Journal of Open Innovation: Technology, Market, and Complexity, 6(4), 139. https://doi.org/10.3390/joitmc6040139

73.

Wan

Wang

Liu

B. S. C.

(2020). Impact of O2O platform multihoming and vertical integration on performance of local service firms—A quantile regression approach. Internet Research, 30(5), 1583–1610.

74.

Wang

Zhou

(2023). Impacts of user expectation and disconfirmation on satisfaction and behavior intention: The moderating effect of expectation levels. International Journal of Human–Computer Interaction, 39(15), 3127–3140. https://doi.org/10.1080/10447318.2022.2095479

75.

Wei

(2006). Lifestyles and new media: Adoption and use of wireless communication technologies in China. New Media & Society, 8(6), 991–1008. https://doi.org/10.1177/1461444806069879

76.

Wen

Wang

Fang

Huang

(2023). An integrated model of continued M-commerce applications usage. Journal of Computer Information Systems, 63(3), 632–647. https://doi.org/10.1080/08874417.2022.2091682

77.

Woo

Choi

J. Y.

Shin

Lee

(2014). The effect of new media on consumer media usage: An empirical study in South Korea. Technological Forecasting and Social Change, 89, 3–11. https://doi.org/10.1016/j.techfore.2014.09.001

78.

C. H.

Chiu

Y. Y.

(2023). Pricing and content development for online media platforms regarding consumer homing choices. European Journal of Operational Research, 305(1), 312–328. https://doi.org/10.1016/j.ejor.2022.05.030

79.

Yamatsu

Lee

S. G.

(2023). Multiple relationships between streaming and linear TV: Examining media substitution theory using big data. Journal of Broadcasting & Electronic Media, 67(4), 442–465. https://doi.org/10.1080/08838151.2023.2250039

80.

Zhanhang

Zuer

(2023, August 23). As viewers fret, China vows to streamline fees for OTT content. Sixth Tone. https://www.sixthtone.com/news/1013585

SVOD Multi-Homing Users in Six Countries: Analyzing Legacy Media Usage,Demographics,Lifestyle,Expectations,and Usage Habits

Abstract

Plain language summary

Keywords

Introduction

Literature Review

SVOD Multi-Homing

Media Substitution Theory

Predictors of SVOD Multi-Homing

Lifestyle and Demographics

Expectations for SVOD

SVOD Usage Habits

Methodology

Sample and Data Collection

Measure

Analytical Strategies

Results

Impact of Multi-Homing on Legacy Media Viewing Time

Predictors of Multi-Homing: Demographic and Lifestyle

Predictors of Multi-Homing: Expectations for SVOD and Usage Habits

Discussion

Implications

Theoretical Implications

Practical Implications

Limitations

Footnotes

Appendix

Declaration of Conflicting Interests

Funding

ORCID iD

Data Availability Statement

References