Keywords

1 Introduction

Consumers are moving toward being permanently connected to the Internet, with a growing number of embedded systems transforming our physical, analog world into one of hyperconnectivity. According to Cisco’s estimation, there will be 50 billion Internet connections by 2020 [8]. This increased connectivity is experienced by consumers through many Internet of Things (IoT) services, which extend the benefits of regular Internet use to every aspect of consumers’ everyday lives.

The many smart devices connected to the Internet not only interact with people but are also interconnected with one another [16]. Based on this extended connectivity, ICT companies have developed and launched various IoT services that equip consumers with an automated, elaborately customized living environment. These services are expected to support consumers’ daily activities by increasing convenience and efficiency. They are mainly delivered through wearable devices, smart homes and buildings, smart vehicles, and healthcare devices or services [14]. The remarkable growth of the IoT-related market is expected to continue with increasing corporate investment in IoT-related hardware, software, and services [17].

Although IoT technology has advanced rapidly, supported by a huge investment, its applications in consumer products and services remain primitive [13]. Such limitations may be related to consumers’ diffident acceptance, as well as structural aspects of the technology itself. As IoT services are mainly based on ICT, the infrastructure of high-speed wired and wireless Internet is critical, and consumers need to be equipped with embedded devices. Furthermore, due to the hyperconnective nature of IoT services, it is inevitable that personal information is collected, transferred, and analyzed, which may provoke consumer concerns regarding information security. For user-oriented IoT applications, therefore, it is important to identify not only the benefits of the technology but also such major challenges for consumer acceptance.

Since IoT technologies are in the initial stages of development for practical application, their usefulness has been mainly discussed from technological perspectives. Previous research on user-oriented IoT services has mainly addressed the technological features and related concepts of IoT and business models [5, 19, 25, 27, 30, 31]. Only a few empirical studies have examined the determinants of home IoT service adoption from the user’s perspective [4, 20, 24]. In this regard, the direction of designing applicable IoT products or services might be determined by the technology’s feasible attributes. However, such a provider-driven perspective may lack insights into the opportunities for and challenges to services’ adoption. Consumers’ evaluation of the benefits and costs of their experiences in using technological services does not entirely depend on the usefulness of the technology itself. Therefore, it is important to understand consumers’ perceptions of IoT services and how these shapes their attitudes toward the services.

In this sense, this study investigated the perceived benefits and costs of user-oriented home IoT services from consumers’ perspectives, as well as how consumers feel about such IoT services. This study focused on South Korea since the service quality of the country’s Internet is considered to be among the best in the world [18]. Accordingly, we attempted to provide insights to practitioners planning and designing user-oriented home IoT services by understanding consumer perceptions of the benefits and costs of these services.

According to IDC’s [17] “Worldwide Semiannual Internet of Things Spending Guide,” the compound annual growth rate of corporate investment in home IoT technologies is expected at 19.8% which is the highest among IoT-related industries. In particular, home IoT services – also called “smart home services” – have been advancing rapidly and related services have been extensively developed [7]. Most home IoT services provide functions to monitor or manage one’s home environment, including home appliances, energy systems, and security systems [36]. In this study, we classify home IoT services into three types: home security, control, and saving services.

Home security includes services for 24-h monitoring via a control center, message alerts, and monitoring inside and outside the home via smartphones and tablet PCs. Home control includes home-automation systems, such as those controlling lights and thermostats and providing real-time video streaming and video storage services. Energy saving systems measure and report the temperature and electricity usage.

For the purposes of this study, we adopted the theoretical framework of the Value-based Adoption Model (VAM) [21, 22] and the Elaboration Likelihood Model (ELM) [28]. According to the VAM, which stems from the Technology Acceptance Model, consumers decide whether to adopt a particular technology by balancing the perceived sacrifice, including technicality and anticipated fees, against the perceived benefit, including usefulness and enjoyment. The ELM provides a framework to understand how persuasive messages that are about to change a receiver’s attitude affect their acceptance of personal information technology [15]. Based on these frameworks, perceived sacrifice can be distinguished as non-monetary and monetary costs [26]. Usefulness is extrinsic and is characterized by perceived benefits, while enjoyment is intrinsic and characterized by emotional benefits [34].

From a consumer perspective, this study examines how consumers perceive the potential benefits and costs of home IoT technology-based services, and whether they are inclined to accept as desirable home IoT services. Finally, we seek to enable practitioners planning and designing home IoT services to deeply understand consumer perceptions and, accordingly, evolve toward providing user-centered services.

2 Method

2.1 Participants

Applying quota sampling by gender and age, 300 Korean consumers in their 30s, 40s, or 50s were recruited as participants. They all registered for a panel through a professional market research organization. The sample characteristics are presented in Table 1. An online survey was conducted on January 10 and 11, 2018. An online questionnaire was sent to randomly selected participants who accepted the email invitation to participate. To ensure that only non-users’ acceptance of home IoT services were investigated, a screening question preceded the questionnaire, asking potential participants to report their experience of IoT services. Only those with no experience of using IoT services were surveyed.

Table 1. Description of the respondents
Full size table

2.2 Measurements

The measures were pilot tested in January 2018 by collecting 100 responses from an online survey institution’s panel. After performing a reliability check, the final items were generated. Consumers’ perception of the costs and benefits and their attitude toward three types of home IoT services were measured by a self-report questionnaire using a 5-point Likert scale (1 = not at all to 5 = perfectly). Before answering questions on the three different types of home IoT service, participants were introduced to each of them.

Based on variables from the VAM, we identified the costs and benefits of home IoT services as perceived by consumers and aimed to understand how these affect consumers’ attitude. Twelve items for measuring consumers’ perceptions of the costs of using home IoT services were adapted from previous studies by Kim et al. [21], Angst and Agarwal [2], Davis [10], DeLone and McLean [11], Voss et al. [35], and Wang and Wang [37]. Based on these previous studies, we hypothetically structured consumers’ costs perceptions with three constructs: monetary, non-monetary, and privacy costs. Each construct comprised four items, examples of which are as follows: monetary costs: “The fee for this home IoT service will be expensive”; “The price for this home IoT service will be affordable for me”; non-monetary costs: “Using this home IoT service seems inconvenient”; “For me, this home IoT service seems difficult to use effectively”; privacy risks: “If I use this home IoT service, my personal information is likely to be leaked and used for unintended purposes”; “If I use this home IoT service, it will be difficult to prevent my personal information from leaking.”

To measure consumers’ perceptions of the benefits of using home IoT services, 12 items were generated by drawing on Kim et al. [21], Agarwal and Karahanna [1], Bhattacherjee and Sanford [3], Cheung and Vogel [6], Davis [10], and Reychav and Wu [29]. These 12 items were divided into three hypothetical constructs: efficiency, effectiveness, and emotional benefits (enjoyment). Efficiency was measured by five items (e.g., “If I use this home IoT service, I can (manage the safety of my home/control my home environment/save energy) more efficiently”; “This home IoT service will be useful to (manage the safety of my home/control my home environment/save energy)”). Effectiveness comprised three items (e.g., “This home IoT service seems to have excellent functions”; “This home IoT service will perform well”). Finally, enjoyment comprised four items (e.g., “This home IoT service will give me pleasure”; “I will enjoy using this home IoT service”).

Attitude was defined as beliefs about the consequences of individual behavior, and measured by six items (e.g., “Using this home IoT service is desirable”; Using this home IoT service is valuable”). These six items were adapted from Fishbein and Ajzen [12]. To validate the scales and test their reliability, exploratory factor analyses (EFAs) and Cronbach’s α tests were performed. Results showed acceptable scale reliability (see Tables 2, 3 and 4).

Table 2. Exploratory factor analysis: cost
Full size table
Table 3. Exploratory factor analysis: benefit
Full size table
Table 4. Exploratory factor analysis: attitude
Full size table

2.3 Analysis

For the purposes of this study, the SPSS 22.0 program was used. Descriptive analysis was employed to evaluate the sample’s demographic characteristics and calculate the mean scores of consumers’ perceptions and attitude. To examine consumers’ perceptions of three types of home IoT services, repeated ANOVA was performed. Regression analyses were conducted to examine how consumers’ perceptions affect their attitude toward the three types of home IoT services.

3 Results

3.1 Consumer Perceptions of Costs and Benefits: Exploratory Factor Analysis

To investigate how consumers perceive the costs and benefits of three different home IoT services, EFAs using a Varimax rotation were adopted, as shown in Tables 2, 3 and 4. Prior to conducting an EFA on each scale, the Kaiser-Meyer-Olkin (KMO) and Bartlett’s test statistics were checked.

For consumers’ perceptions of the costs of using home IoT services, different dimensions were extracted according to the service type. For security and saving services, four factors were extracted, namely, privacy risk, non-monetary cost, costliness, and unaffordability, which cumulatively explain 71% (security) and 76% (saving) of the data variation for cost perception. The hypothetical construct of monetary cost was divided into two different dimensions: costliness and unaffordability. For control services, two items measuring costliness were excluded due to the problems of cross-loadings and low factor loadings (under 0.5). Finally, three of the four factors extracted for security and saving services were also extracted for control services, the exception being costliness; these three factors cumulatively explain 73% of the data variation. For consumers’ benefit perception, three factors were extracted in accordance with the hypothesized dimensions, indicating efficiency, effectiveness, and emotional benefits. All factor loadings are above .60, and no items were excluded. The cumulative explained variances of security, control, and saving services were 68%, 66%, and 70%, respectively. Consumers’ attitude was a single dimension, explaining 62%, 62%, and 66% of variance for security, control, and saving services, respectively.

3.2 Consumer Perceptions of Costs and Benefits by Home IoT Service Type

As shown in Table 5, repeated ANOVAs were conducted to analyze the differences in consumers’ perceptions of the costs and benefits of the three different types of home IoT services. For the cost perceptions, the mean scores of privacy risk, costliness, and unaffordability differed according to the type of IoT services. The results indicate that consumers perceived a higher privacy risk for security services, at a 0.1% significance level. Consumers’ perceptions of costliness were compared between security and saving services, because this factor’s items were excluded from the EFA of control services. The mean score of costliness was higher for security than for saving services (p < .001). Moreover, the mean score of unaffordability was also higher for security services than for the other two types, at a 0.1% significance level.

Table 5. Repeated measures ANOVAs
Full size table

For consumers’ benefit perceptions, the mean differences of perceived usefulness and effectiveness were statistically significant. Consumers’ perceptions of perceived usefulness were lower for security services than for the other two types (p < .01). The mean score of effectiveness was lower for security services than for saving services, at a 5% significance level.

3.3 Consumer Attitude Toward Home IoT Services

Consumers’ attitude toward home IoT services was analyzed using regression models, as shown in Table 6. Gender, age, education, and household income were employed as sociodemographic variables. Factor scores of perceptions of costs and benefits were also used. Three regression models were statistically significant overall, and the variance inflation factor (VIF) for each independent variable was less than 2, indicating that multicollinearity was not present.

Table 6. Consumers’ attitude toward home IoT services
Full size table

Security Services.

Gender and age had statistically significant effects on consumers’ attitude toward security services, indicating that male consumers have a more positive attitude than female consumers and that older consumers have a more positive attitude than younger consumers. Among perceptions of costs and benefits, privacy risk and unaffordability had negative effects on consumer attitude toward security services, whereas efficiency, enjoyment, and effectiveness had positive effects. Among the predicting variables, the three benefit perceptions showed higher standardized coefficients than any of the other variables, the highest being for enjoyment. The predicting variables in this model explain 57.6% of the variance in consumers’ attitude toward security services (adjusted R2 = .576).

Control Services.

Among the sociodemographic variables, only age had a statistically significant effect on attitude. The four cost perception variables were non-significant, while three benefit perception variables were statistically significant. Among the benefit perceptions, enjoyment showed the highest standardized coefficient (β = .525, p < .001). The adjusted R2 of this model was .560.

Saving Services.

Age and education had statistically significant effects on consumers’ attitude toward saving services. As for security and control services, age showed a positive effect on attitude. Regarding education level, consumers who completed graduate school education indicated a more positive attitude toward saving services than consumers who progressed no further than high school (or below). Except for privacy risk, consumers’ cost perceptions were all statistically significant in relation to consumers’ attitude toward saving services. The three variables of benefit perceptions were all statistically significant, indicating higher standardized coefficients than any of the other variables, with enjoyment showing the highest standardized coefficient. This model explains 67.1% of the variance in consumers’ attitude toward saving services (R2 = .671).

4 Discussion and Conclusions

Consumers’ lives are becoming increasingly hyperconnected, allowing them to enjoy convenient, efficient services by adopting IoT services, which place them within a mutually interconnected environment. Because IoT services, based on new technology, change many aspects of consumer lives, there are inherent costs for consumers that embrace such services, which may cause some to be reluctant to do so.

This study investigated the benefits and costs of home IoT services from consumers’ viewpoint, aiming to advance understanding beyond providers’ perspectives on the technological requirements to increase service penetration. We analyzed consumers’ perceptions and attitudes regarding home IoT services in South Korea. Specifically, this study investigated how consumers perceive the benefits and costs of home IoT services and the effects of consumers’ perceptions on their attitude toward three types of home IoT services. In particular, this study attempted to provide insights into the design and development of home IoT services for consumers. The main results are outlined as follows.

First, for consumers’ perceptions of the benefits of three types of IoT services, three common factors were extracted: efficiency, effectiveness, and emotional benefits. For the security and saving services, four factors were extracted for consumers’ perceptions of costs: privacy risk, non-monetary cost, costliness, and unaffordability. Three of these four factors (the exception being “costliness”) were also extracted for control services. This result shows that consumers have different perception structures on costs according to the service type. Of the three types of services, control services are usually in easy reach [23].

Second, consumers were found to perceive higher privacy risk and monetary cost for security services than for the other two service types. Conversely, consumers’ recognition of benefits indicated that efficiency is perceived to be lower for security services than for the other two categories. Such perceptions show that, for security type home IoT services, consumers are more concerned about costs and less inclined to recognize the benefits. According to previous research, consumers could perceive security services as a cost [4]. Consumers’ perceptions of costs were found to be nonsignificant for control services (compared to the finding of significance for the other two service types). This suggests that consumers may not be aware of the costs of control services. Therefore, service providers need to understand consumers’ different perceptions of each type of home IoT service.

Third, the regression models indicated the effects of consumers’ perceptions of the costs and benefits on their attitude toward the home IoT services. Age was found to affect attitude toward home IoT services, with older consumers found to have a more positive attitude toward accepting home IoT services. For control services, the benefit perceptions of efficiency, enjoyment, and effectiveness had significant effects on attitude. For saving services, the cost perceptions of non-monetary and monetary factors had negative effects on consumers’ attitude, while the benefit perceptions of efficiency, effectiveness, and enjoyment had positive effects thereon. For security services, the perceptions of privacy risk and affordability had a statistically significant effect on attitude, as did the three subsets of benefit perceptions. These results show that consumers’ perceptions of benefits have significant effects on their attitude toward three types of IoT services. For all three models, the standardized effect of enjoyment had the greatest effect on attitude. This suggests that consumers may be more receptive to home IoT services when they have strong beliefs on the benefits, especially those of an emotional nature.

This study identified the structures of consumers’ perceptions of the costs and benefits of home IoT services. In particular, we analyzed how such perceptions affect consumers’ receptive attitudes toward three different types of home IoT services. The study’s findings provide meaningful insights into the consumer factors that affect acceptance of each service. For the three considered types of home IoT services, the benefit perceptions had greater effects on attitude than the cost perceptions. Although these services are based on highly advanced technologies, consumers are most affected by emotional benefits, which have the greatest effect on attitudes toward all three types of home IoT services. This finding may reflect the context in which these services are experienced. As home IoT services are pervasively experienced in consumers’ daily lives, consumers’ service experiences might be critical to forming a positive attitude. In previous research, perceived enjoyment has been shown to be a strong predictor for consumers’ attitudes [9, 38, 39] and purchase intentions [32, 33].

These results provide insights to develop more consumer-oriented services, which will allow potential demands in the market to come to the fore. Although technology has driven such services to date, by developing new functions and improving performance, it is now necessary to attune the services more closely to consumers’ expectations and to address their concerns. Each technology can only be meaningful through its adoption by consumers to whose lives it contributes. In this sense, based on this study’s results, home IoT service design must be directed in consonance with consumers’ perceptions and expectations.