**3. Methodology**

### **3.1 Interviews as survey follow-up**

In previous work [3], we confirmed one of the social informatics' key lessons, that is, the effects of technology are always unequal, in other words, "that some social groups will benefit more than others from the uses of digital technologies" [12]. We also found that the privacy and security-preserving capability of devices are the most significant challenges to smart home adoption. Hence, we further investigated how organizations in the sector perceive and respond to these challenges. This study involving Human Subject Research received full approval by the University of Warwick ethics committee on May 29, 2019 (ref. no BSREC 51/18-19). The methodology consisted of a series of semi-structured interviews: seven experts from six organizations involved in the design, development, or review of smart home devices and services (**Table 1**) 2 . We adopted a semi-structured interview format as this is more likely to ensure that valid and reliable data can be obtained from interviewees [36]. Also, semi-structured interviews provide respondents with enough flexibility to build and expand on the initial guiding topic, which, in turn, allows the researcher to analyze the dataset with different degrees of depth.

In order to achieve a balance of views, a broadly equal proportion of business and nonbusiness organizations were included in the sample with four experts from NGOs and three experts from businesses. Respondents from interviews [2, 3, 7] are business respondents, whereas those from interviews [4, 5, 8] are NGO respondents. Interview [5] includes two NGO experts from the same organization (see **Table 1**).

<sup>2</sup> Ethical approval for the study was secured from the Biomedical and Scientific Research Ethics Committee (BSREC) at the University of Warwick on 29 May 2019.


*How Is the Internet of Things Industry Responding to the Cybersecurity Challenges of the Smart… DOI: http://dx.doi.org/10.5772/intechopen.106012*

#### **Table 1.**

*Details of sample composition consisting of business (product provided and respondent role) and NGO (target represented and respondent role) organizations.*

We sought to include policy-side views on the security threats in smart home adoption. The majority of our respondents were from large organizations but we sought to include at least one small business among them.

The sample of interviewees arrived through suggestions made by Petras project colleagues. Interviews were conducted face-to-face and by video conference call and lasted between 30 min and 1 h.

To ensure rigorous data collection, we followed the guidelines set by Braun and Clarke [37] concerning planning thematic analysis. Hence, in devising the interview questions, we first clarified the scientific method upon which the analysis would rest and opted for a broadly deductive approach, constrained by the survey findings and adoption challenges of the smart home identified in Cannizzaro et al. [3] and listed below:

i.Overall, fairly low levels of trust.


Our questionnaire was developed based on these challenges that allowed us to formulate a question guide. Each question topic was formed of guiding questions as well as some follow-up questions [38] (see Appendix A). The interviews were transcribed and a thematic analysis via coding was conducted on the transcripts. The thematic analysis was based on Braun and Clarke's [37] principle of *realism*. These questions were rotated according to the background of the respondents, particularly whether they were businesses or NGO organizations. Topics forming the questions guide included general background questions to allow participants to respond to questions about their roles within the organization and break the ice; followed by the topics pertaining to the most significant factors affecting IoT adoption, as previously explored quantitatively in [3, 39], such as IoT and smart home *awareness*, *risks and benefits* of IoT for both organizations and consumers, *trust, digital* divide in IoT adoption, *future and change* in the sector, including *responses to IoT challenges* (see Appendix A). The interviews were transcribed and a thematic analysis via coding (based on the topics above) was conducted on the transcripts. In order to reach the saturation point, we then examined the transcripts further using Social Construction of Technology (SCOT), a mid-ground theoretical framework, which is outlined below.

## **3.2 Theoretical framework: SCOT's interpretive flexibility**

Within innovation studies, approaches to understanding meanings range from technological determinist (e.g., [40]) to social constructivist (e.g., [41]). Occupying a conceptual middle ground is the SCOT framework [42]. In SCOT, a key concept is "interpretive flexibility" [43], which recognizes that the "meaning" of an innovation may be initially contested by different stakeholders or social groups before "closure"—and hence its use-value—is reached [44]. According to Orlikowski [43], interpretive flexibility is an attribute of the relationship between people and technology, a function of the material artifact, the characteristics of the human agents, and the institutional context in which technology is being introduced [45]. The social groups involved in interpreting the meanings of the technology include producers, engineers, designers, marketers, and investors; those who have a direct relationship with technology and develop an artifact—advocates—policymakers, lobbyists, and academics; those who are indirectly related with technology and work on policymaking, lobbying, and research; and also, users and bystanders [46]. Elle et al. [47] contend that, in most cases, interpretive flexibility diminishes when the social groups reach an agreement on an interpretation.

Initially, SCOT perspectives originated in studies of organizational innovation processes. Unsurprisingly, Rowland [48] argues that SCOT emphasizes the role of large business corporations, whereas Burns et al. [49] see innovation within a context of receptivity and institutionalization. However, some argue that in the current context where digital innovation is a largely available consumer commodity, SCOT needs to be translated to the consumer digital technology marketplace, and hence it requires a new framework variant, Social Construction of Digital Technologies (SCODT). The SCODT framework posits that dimensions of innovation ought to be considered in light of digital advances [50, 51]. This implies that the social groups involved interact in different ways from those involved with technological innovation—traditional employees-employers' hierarchies typical of the workplace are replaced by consumerseller relationships, where power relationships occur in an always connected, and competitive, digital context. Wellman et al. [52] argue that digital technology users

*How Is the Internet of Things Industry Responding to the Cybersecurity Challenges of the Smart… DOI: http://dx.doi.org/10.5772/intechopen.106012*

are connected in a specific way, that is, by means of networked individualism: fragmented, opportunistic, fast connecting individuals, and organizations forming temporary relevant social groups. Furthermore, SCODT posits that interaction switches from interpersonal to interpersonal, person-technology, technology-technology, and technology-physical environment interactions [50], where it is also artificial agents (*sensu* [53]) in addition to human agents that take decisions within such relationships.
