Exploring perceptions of smart, modular living in the UK: a think aloud study

ABSTRACT

While there is growing interest in the design and deployment of smart and modular homes in the UK, there remain questions about the public’s readiness and willingness to live in them. Understanding what conditions prospective residents might place upon the decisions to live in such homes stands to improve their design, helping them to meet with the expectations, and requirements of their residents. Through direct interaction with a prototype of a smart and modular home within a university context, the current study investigated how people negotiate the prospect of smart and modular living, and the conditions they would place on doing so. The study explores the short observational experiences of 20 staff and students within a UK university context, using think aloud interviews. Findings indicate that whilst participants were able to identify the benefits of smart and modular homes, there were nuanced responses when they negotiated the challenges of living. Further, a framework of considerations and recommendations are presented which could support practitioners and policy makers in making more informed, citizen-led decisions on ways to adapt and improve these home solutions.

KEYWORDS:

• Home solutions
• think aloud method
• conditionality
• acceptance
• public involvement

1. Introduction

Smart and modular homes have risen-up the UK national political agenda in recent years, and has become a priority area of in UK housing policy (e.g. Planning for the future 2020; Powering Up Britain for Energy Security and Net Zero 2023). Such homes are deemed to present ‘solutions’ for the evolving residential sector, in that they can improve people’s quality of life and provide the flexibility to meet the changing needs of residents. They can also be seen to address wider societal and sustainability challenges in this sector (Ahmad et al. 2022; Zimmermann et al. 2023). Consistent with this vision, there has been recent public investment in demonstrators of homes that are both smart and modular by design (e.g. the REFIT Smart Home project, MyGlobalHome demonstrator project; Department for Business, Energy and Industrial Strategy 2022).

Despite growing interest, the public uptake of these smart and modular home solutions in the UK remains low, compared to their global uptake which are more variable amongst potential consumers (Ferreira, Oliveira, and Neves 2023; Jagannath 2023; Windl, Schmidt, and Feger 2023). Key drivers and barriers shaping the uptake of these home solutions include, the ethics of data sharing, affordability, and emotional wellbeing, investing in future-proof living spaces, as well as surveillance and privacy concerns associated with sensor types used (Chalhoub et al. 2021; D'Alessandro et al. 2020; Koolen 2020; Windl, Schmidt, and Feger 2023). Notable complexities involved in understanding peoples’ acceptance are largely influenced by several factors and characteristics, i.e. social, contextual, and psychological (Gerli et al. 2022). However, existing studies can often be technologist-driven rather than fully reflecting the needs and concerns of potential users (Kwon 2022). Having an in-depth understanding of what conditions people choose when deciding whether to use these home solutions through qualitative elicitation methods (e.g. the Think Aloud method) has been recommended in order to improve their design, so that they are more responsive to people's own requirements (Jagannath 2023; Jaspers and Pearson 2022; Katunský et al. 2020; Quach et al. 2022).

The current paper reports the results of a ‘think aloud’ interview study designed to understand how people negotiate the prospect of living in a smart and modular home. To our knowledge no study has used this approach in the context of understanding the acceptability of home-concepts that are designed to be simultaneously smart and modular (D'Alessandro et al. 2020; Shin et al. 2022). The study context is a demonstrator modular smart home constructed on a university campus in the southeast of England, known as the Innovation Centre. The study participants were staff and students at the university. Exploring digital transformation opportunities within university campus contexts is an emerging area of research. Digitally enabled ‘smart’ campuses have been regarded as miniature replicas of smart cities (Dong et al. 2020; Polin et al. 2023; Valks et al. 2021), yielding opportunities for the systematic study of the factors (e.g. technological orientations, shared living context) that can shape people’s willingness of people to interact with smart buildings, including smart homes (Baudier, Ammi, and Deboeuf-Rouchon 2020). It is also noteworthy that this study’s findings were later used to help shape a later pre-occupancy study where participants stayed longer in the Innovation Centre.

A framework of considerations and recommendations are then presented for research, designers, and industry developers in related fields. These highlight the importance for understanding the conditions people make for both smart and modular home designs to better informed by their potential users own perspectives.

This paper is divided into four sections. An overview of the literature relevant to this study is presented in Section 2. Next, Section 3 outlines the study design, data collection and analysis are described. After presenting our results in Section 4, discussion and conclusions are presented in Sections 5 and 6, the final sections of the paper.

2. Literature review

2.1. Public acceptance of smart and modular living environments

2.1.1. Smart homes

Smart homes take many forms, but they can generally be defined as being a domestic residence where there is a convenient set-up of internet-connect devices and appliances (e.g. thermostat, entertainment systems, security and lighting systems; Sovacool et al. 2022). This ‘internet of things’ within the home enables the various devices and appliances to ‘communicate’ and share data with one another, in some instances automating actions between them (Shafi and Mallinson 2023; Zimmermann et al. 2023). It also gives the householder(s) the ability to remotely interact with and control them from a mobile or other networked device (Latikka et al. 2021).

The uptake and use of smart home technologies can offer an array of benefits for users. For instance, encouraging energy-efficient behaviours or increasing the independence of older adults or those with disabilities (Shafi and Mallinson 2023; Zimmermann et al. 2023). Proponents of these technologies – which include national governments, smart home technology developers and third-party service providers – also attest to the wider socio-economic value that can be obtained from the extensive and rich data collected by them (e.g. data monetisation via data sharing; Quach et al. 2022; Sarker 2022). Despite this, adoption of smart home technologies (e.g. Amazon Echo, Google Nest) among consumers globally has been variable (e.g. Ling et al. 2021; Windl, Schmidt, and Feger 2023). This has prompted research into the drivers and barriers to uptake, which highlights a diverse array of contributing factors. These include concerns about the difficulty of using the technologies (Jamwal et al. 2022), the ethics of data sharing and consent (Koolen 2020), concerns about surveillance and invasion of privacy (Chalhoub et al. 2021; Crabtree and Mortier 2015; Windl, Schmidt, and Feger 2023), as well as the upfront cost of purchasing and installing the technologies (Nascimento et al. 2022; Shafi and Mallinson 2023).

2.1.2. Modular homes

The concept of the modular home is older than that of the smart home and is broadly described as a home where the living spaces are flexible, changing to the needs and preferences of residents across their lifecycles (Jagannath 2023; Katunský et al. 2020). The ‘modularity’ of the home encompasses both the adaptability of the living space and the multifunctional and/or space optimising nature of the furniture (D'Alessandro et al. 2020). For example, modular homes can incorporate room dividers that can be physically or automatically adjusted to create different spatial arrangements (Jagannath 2023; Katunský et al. 2020).

Akin to the proclaimed benefits of smart home technologies, there are also mooted benefits to living within modular environments. Modular homes are flexible and can be adapted to suit the changing needs or preferences of residents (e.g. adapting spaces for living, working, learning and entertainment) (Appolloni and D’Alessandro 2021; Doling and Arundel 2022; Harrouk 2021). Modern modular homes also tend to incorporate new materials (e.g. sustainably sourced), design elements (e.g. space-optimising furniture) as well as manufacturing approaches (e.g. off-site manufacturing) that proffer further economic, social, and ecological benefits compared to more standard building techniques (Appolloni and D’Alessandro 2021; GOV.UK 2019; Zhao and Riffat 2019).

As with smart home technologies, the commercial success of modular homes relies upon their uptake and use among consumers. Uptake of modular homes in the UK is, however, currently very low (Gerli et al. 2022; Pirzada et al. 2022). While less developed than the literature on smart home technologies, research into perceptions of modular living among developers and publics identify the barriers as including the need to invest in future-proof living spaces (e.g. sustainable, durable, and upgradable furniture) and managing control issues in shared accommodation as being key considerations (D'Alessandro et al. 2020; Jagannath 2023). Like research into smart home technologies, studies in this space have predominantly used experimental, attitudinal, and descriptive methodological approaches (e.g. lab in situ observations; D'Alessandro et al. 2020; Till and Schneider 2016). For example, Jagannath (2023) used online mixed-method surveys to investigate potential associations, if any, between people’s wellbeing and the flexibility of the home in the UK. Further, Ye, Li, and Yang (2021) researched children’s exploratory behaviours and satisfaction when engaging with modular furniture using scene theory, field surveys and observations. Nonetheless, in contrast with smart home technologies, less is fully known about people’s perceptions and reservations of modular homes as well as the underlying factors which help to influence and shape these perceptions (Doling and Arundel 2022).

2.1.3. Smart and modular homes

The prospect of combining both ‘smartness’ and ‘modularity’ into homes could offer further practical advantages to enhance people’s quality of living (Husein 2021; Khangura and Haney 2022) – particularly in the face of socio-demographic change (e.g. an aging but more computer literate population), new digital ways of learning and working from home environments (e.g. remote working and distance learning), and environmental change (e.g. growing needs to mitigate and/or adapt to climate change). The ability to control or automate the physical size or arrangement of spaces (e.g. moving walls; Jagannath 2023) within a smart, connected environment, presents new opportunities for people to personalise their living environment. It also offers opportunities to optimise the living conditions (e.g. thermal comfort, air quality) to promote resource-efficiency and enhance wellbeing (Husein 2021). The combination of smart and modular elements could, however, come with additional concerns and considerations for residents (e.g. providing those external to homes with deeper insight and data on how people utilise and adapt their living spaces). Currently, little is known about perceptions of smart and modular home environments, including what physical and digital technologies (and any optional, requisite, or implied data-sharing requirements) would be considered ‘acceptable’ and by whom (Appolloni and D’Alessandro 2021; Husein 2021; Khangura and Haney 2022). Research in this space is thus warranted, with the findings standing to inform the future design and promotion of smart, modular living concepts. Equally, there is limited research exploring modular home solutions in parallel with these technological innovations, nor is the shared student living environment explored within a smart and modular university context amongst those regarded as the ‘digital native’ population (Baudier, Ammi, and Deboeuf-Rouchon 2020; Dong et al. 2020; Polin et al. 2023; Valks et al. 2021). Our contribution also thus lies in understanding how people perceive and view the possibility of living in a smart and modular home as acceptable in a university campus context.

2.2. Related work and methodological approaches

There is complexity in developing a firm understanding of peoples’ acceptance of new technologies, given that perspectives of such innovation can be influenced by many social, contextual, and psychological characteristics (Gerli et al. 2022). Without a firm regard for methodology, there is a risk that studies will assess ‘pseudo-opinions’ (Andersen et al. 2023) or could be overly ‘technologist-driven’ rather than fully reflecting the genuine needs and concerns of potential users (Kwon 2022). As such, having an in-depth understanding of people’s barriers for accepting and conditions for opting to use these home solutions (whether individually or in parallel) is deemed important if we are to improve their future designs, (Jagannath 2023; Jaspers and Pearson 2022; Katunský et al. 2020; Quach et al. 2022).

Recent years have seen a growth in sociotechnical studies examining people’s preferences for housing attributes relating to smart home technologies. These studies have used attitudinal and descriptive methodological approaches (e.g. questionnaire surveys), drawing on a variety of theoretical models to help understand people’s overall acceptance (e.g. technology acceptance model; Attiéa and Meyer-Waarden 2022; Gerli et al. 2022; Shah et al. 2020). For instance, Ferreira, Oliveira, and Neves (2023)’s study surveyed 800 Hong Kong residents to explore people’s general attitudes towards and barriers for market uptake of smart home technologies, as well as their potential future developments. Additionally, Seo and Yang (2023) surveyed end-users’ satisfactions of using smart home technologies in real home environments whilst they engaged in a living laboratory experimental study.

Across each of these research fields, few studies have used theoretical models, and qualitative elicitation methods to explore the determinants of and sense making processes associated with the acceptance of smart and modular homes ‘solutions’ (e.g. D'Alessandro et al. 2020; Shin et al. 2022). Qualitative elicitation methods (e.g. the ‘think aloud’ method) are widely used across the social sciences in order to provide in-depth insight into people’s perspectives on topics and issues, including new technologies (Hope et al. 2018). Such qualitative elicitation method, on account of their more discursive nature, are also useful in exploring ‘conditional acceptance’ and the caveats that people might place upon their headline support for technological innovation as well as the persistence of any concerns held (Heyerdahl et al. 2022; Sarker 2022).

Conditional acceptance can be described as a person’s willingness to agree or accept something when conditions are either met, or changes are made (Heyerdahl et al. 2022). It can be understood through examining a person’s process of drawing conclusions from inferences made about a given topic or item, termed conditional statements (Casini, Meyer, and Varzinczak 2021). For instance, ‘if a occurs, then b will happen as a consequence’ (Evans, Thompson, and Over 2015). Research has widely established that when people reason on everyday conditions, they spontaneously bring relevant contextual knowledge and factors into account (Casini, Meyer, and Varzinczak 2021). Understanding everyday conditions could provide valuable, real-time insight into how people negotiate and deliberate topics and issues (Jagannath 2023; Jaspers and Pearson 2022; Katunský et al. 2020; Quach et al. 2022). Such approach can – particularly with regards to smart and modular home solutions – be used to design products and services that are more reflective of consumers preferences to people’s own requirements. However, it remains understudied area of research (Heyerdahl et al. 2022).

2.3. Aims of the current study

Despite the growing interest in the design, demonstration and retail of homes that are both smart and modular, there are few studies that have sought to systematically examine the opinions of prospective residents (Ferreira, Oliveira, and Neves 2023; Jagannath 2023; Windl, Schmidt, and Feger 2023). While some insight into the drivers of consumers’ willingness to purchase or rent these homes can be gleaned from to the literature on smart living (e.g. the ethics of data sharing, relative affordability, surveillance and privacy concerns) (Chalhoub et al. 2021; D'Alessandro et al. 2020; Koolen 2020; Windl, Schmidt, and Feger 2023), there is a need for more targeted research into homes that unite the dual concepts of smartness and modularity.

The current study sought to address this gap, utilising a ‘think aloud’ interview method. Think aloud is a widely used qualitative elicitation method, which has been applied in a range of social science disciplines (e.g. psychology, sociology) to gain an understanding of the conscious processes that occur whilst completing a cognitive task (Alhadreti 2021; Thompson et al. 2018). While the think aloud method is used to feedback peoples’ experiences during or after completing a task (Shin et al. 2022), the method also has the potential to gather useful insights about the processes and content of people’s everyday conditions on a given topic or item (Heyerdahl et al. 2022). For this reason, it was deemed suitable for gathering insights about how people negotiate the pros and cons of living in a smart and modular home, as well as the conditions that might promote or inhibit this acceptability of the concept.

The think aloud interviews were conducted within a pilot demonstrator of an innovative smart and modular home, located on a university campus in the southeast of England. While it is the case that smart home technologies and modular home solutions are being developed, tested, and implemented in a ‘city’ context, the chosen context had many strengths. First, a key anticipated market for the end-product was the student accommodation sector, thus building the demonstrator and running the study on a university campus made sense. Second, and relatedly, the demonstrator home fits with the emergence of the smart campus concept (Baudier, Ammi, and Deboeuf-Rouchon 2020; Dong et al. 2020). A smart campus is where digital transformations can help in facilitating space management and resource allocation practices on campus, as well as creating new teaching and learning experiences (Valks et al. 2021). A smart campus in many ways can be likened to a miniature replica of a smart city, providing a built environment within which new ideas and technologies can be developed, deployed and (safely) tested among ‘digital native’ populations of staff and students (Polin et al. 2023). Third, the pragmatic accessibility of the demonstrator building also made it appealing as a study context. The contribution of this study thus lies in developing an understanding of how people perceive the prospect of living in a smart and modular home, particularly within a university context. This study’s findings are therefore considered novel in that they could stimulate design transformations that are derived bottom-up, including participants of specific educational levels, that are likely ‘adopters’, and may be engage in in-depth discussions and interrogations of the technologies offered.

3. Methodology

The contribution of this think aloud study is to develop an understanding of how people perceive the prospect of living in a smart and modular home, particularly within a university context. Accordingly, the study investigated the following inter-related research questions:

1. How acceptable do people consider the prospect of living in a smart and modular home within a ‘smart campus’ context to be?

2. What caveats or conditions would people place upon their willingness to live within a smart and modular home, within a ‘smart campus’ context?

3. How can qualitative elicitation methods, such as the Think Aloud method, be used to engage people in discussions about what they view as ‘acceptable’ when considering a smart and modular home?c

3.1. Sampling population, recruitment, and ethical consideration

Twenty participants (n = 20) over 18 years from the University of Surrey were recruited, using a non-probabilistic quota sampling and snowballing approaches (Zack, Kennedy, and Long 2019). Participants were university staff, undergraduate students, and postdoctoral researchers. Participants were recruited from nine academic departments (e.g. medical sciences and literature studies) and other basic diversity categories: gender, ethnicity, and educational attainment (Xu et al. 2020). Figure 1 summarises the demographic make-up of participants.

Figure 1. Demographic make-up of participants. Participants voluntarily consented to being involved in the study and received a £25 shopping voucher as an incentive for contributing. The project received ethical approval from the Research Integrity and Governance Office at the University of Surrey (Ref: FASS 20-21 118EGA).

Our sampling strategy was to recruit participants representing a range of attitudes. This was achieved using participant responses gathered from a pre-questionnaire which asked for participants’ demographic information and use of smart home technologies (Parasuraman and Colby 2015; Shin et al. 2022). It also included the standardised measure for measuring a person’s willingness to use technology, the shortened Technology Readiness Index, used to categorise participants as techno-sceptic or techno-enthusiast based on four measurement dimensions: innovativeness, optimism, discomfort, and insecurity (see Mels 2018). By techno-sceptic we refer to a person who is sceptical about the benefits of modern technology as opposed to a techno-enthusiast who presents a strong liking for modern technology (Yang et al. 2022).

3.2. Structure of the think aloud interviews

Twenty 1-h interviews were held during the COVID-19 pandemic between October and December 2021, using a concurrent think aloud approach. Participants were asked to verbalise their thoughts and reflections of the household items, controlling mobile applications (MyGlobalHome’s Ultra app), as well as the layout found within the Innovation Centre space, thus enhancing the contextual veracity of study outcomes (Alhadreti 2021).

A general scope of existing empirical research was used to identify key thematic topics, sub-topics, and key questions of relevance to the study’s aim (Knott et al. 2022). A tabular representation of the key thematic topics, sub-themes and questions identified from this scoping exercise can be found in supplementary materials, Table A. In short, topic areas included: (a) the usability of smart and modular homes, (b) concept of a home, (c) sense of control within a home, as well as (d) issues of data sharing, data management and related ethical issues.

The think aloud interviews proceeded in the manner outlined in Table B within the supplementary materials. The inclusion of tasks as part of the interview process and/or the use of items as prompts is a common feature of the think aloud procedure. These techniques were used during this study in order to elicit participants’ tacit knowledge and opinions, while also serving to guide and progress the interview discussions (Alhadreti 2021). A full description of the interview topic guide, interview questions and think aloud tasks/items used during the interview can be found in supplementary materials, Table B.

The interview process consisted of three main parts: (1) a tour of the Innovation Centre; (2) an ice-breaker session; and (3) an interview discussion (Thompson et al. 2018).

The first two authors gave each participant a 15-min tour of the Innovation Centre, providing an overview of the smart technologies and modular home living environments at Innovation Centre. This ensured participants were made aware of and could experience the smart and modular home living environment available to assist with later interview discussions (O’Neill and Roberts 2019). An icebreaker session then followed, which asked participants to practice a ‘think aloud’ exercise to introduce the approach, exploring for a few minutes their previous experiences with smart home technologies, as well as their reflections of the MyGlobalHome Innovation Centre’s smart and modular home solutions present (O’Neill and Roberts 2019). Participants were then asked to engage in the main think aloud interview discussion. This included walking around and exploring the Innovation Centre presenting a smart, modular home prototype as they engaged in interview discussions, tasks, and interacted with the household items and features (Thompson et al. 2018).

3.3. Data analysis

Because 20 participants took part, quantitative data analysis of participants’ responses to the Technology Readiness Index 2.0 (TRI) was restricted to descriptive statistics, including simple counts and percentages. To assess participant’s technological readiness further, participants were classified into one of five behavioural profile types, based on different combinations of innovativeness, optimism, discomfort, and insecurity (Mels 2018). Each profile type has a distinct personality and background: explorers, pioneers, sceptical, hesitant and avoiders (see Table 1). This classification approach was achieved using the segmentation model proposed by Parasuraman and Colby (2015), grouping participants by average TRI 2.0 scores.

Table 1. TRI 2.0 five behavioural profile types and associated characteristics.

Behavioural Profile Type	Characteristics
Sceptics	Have a detached view of technology, though have less extreme positive and negative beliefs.
Explorers	Have a high degree of motivation, but with a low degree of resistance towards technology.
Avoiders	Have a high degree of resistance and low degree of motivation towards technology.
Pioneers	Hold both a strong positive and negative views about technology.
Hesitators	Have a low degree of innovativeness towards technology.

Qualitative analysis was conducted using NVivo (Version 12). Think aloud interview discussions were transcribed and analysed using thematic methods as outlined by Braun and Clark (2006), coded in NVivo Version 12. This approach was used to identify the relevant themes, categories and linkages emerging from the transcripts of the interviews (Williams and Moser 2019).

Coding was conducted by the first and second authors, each with an in-depth knowledge of smart home technology, and related data governance issues. Intercoder reliability was measured using a simple proportion agreement method (see Campbell et al. 2013). This reliability approach increased consistency in coding, enhanced intercoder reliability and provided additional training and support to all coders (Campbell et al. 2013).

3.4. Study site

This study was carried out within the MyGlobalHome’s Innovation Centre located on a university campus in the southeast of England, UK. MyGlobalHome are a smart home design and installation company, and the Innovation Centre comprises a testbed for their modular smart home concept and associated technologies. Figure 2 provides an illustration of the internal space within the Innovation Centre.

Figure 2. MyGlobalHome Innovation Centre at the University of Surrey, Guildford, UK.

The MyGlobalHome concept is that each home incorporates a multitude of smart sensor and control technologies, as well as modular furniture and dynamic wall arrangements (e.g. moving walls made from ethically resourced wood and can be manually manipulated or controlled via a Smart phone app). It enables inhabitants to control all aspects of the home environment (e.g. smart phone enabled and manual control systems for home heating, lighting, and entertainment), while at the same time gathering information on how the home is performing in terms of energy and resource consumption, maintaining air quality, etc. These details are fed back to the end-user via a smart phone app called ‘Ultra’. The Ultra app can also be used to operate smart technology within the building, as well as other home features (e.g. controlling the light switches and a smart home hub). The Ultra app was used as a prompt within the study.

4. Results

4.1. Technology Readiness Index 2.0

Descriptive measurable summaries for each of the Technology Readiness Index 2.0 statements (TRI 2.0; Mels 2018) are presented in Table C in the supplementary materials. Scores between the individual statements within a construct did not differ much for Optimism, Discomfort, and Insecurity, with noticeable differences for innovativeness. Optimism, Discomfort, and Insecurity mean scores had scores above 3, whilst Innovativeness mean scores were 3 or below. Average TRI 2.0 score was 3.23 on a scale that runs from 1 till 5. The highest score was 4.03 and the lowest score was 2.1. Of the four TRI measurement dimensions, Optimism had the overall highest score, and Discomfort had the lowest overall score, with only slight variations between statement scores within them. For instance, participants largely agreed with statement OPT2 and somewhat disagreed with statement DIS1, with an overall neutral response to all other statements (i.e. ‘Neither Agree nor Disagree’).

A summary of the average scores (SD) for the TRI constructs across the behavioural profile types are presented in Table 2. Most study participants were classified as either hesitators (n = 8; 40%), or explorers (n = 6; 30%), with fewer classified as either sceptics (n = 3; 15%), pioneers (n = 2; 10%), or avoiders (n = 1; 5%). The split of staff and students across the different profiles are outlined in Figure 3. Further, those participants classified as sceptics were predominantly university staff, whilst the one participant classified as an avoider was a student. It is noteworthy that those participants classified as hesitators came from a variety of disciplinary backgrounds (See Figure 4). Similarly, sceptics and explorers tended to come from the natural and social sciences disciplinary backgrounds, with the one avoider coming for an arts and humanities background.

Figure 3. Number of participants per behavioural profile type specified per subgroup.

Figure 4. Number of participants per behavioural profile type specified per disciplinary subject.

Table 2. A summary of the average scores per behavioural profile type and overall scores.

	Sceptics	Explorers	Avoiders	Pioneers	Hesitators
N	3	6	1	2	8
%	15	30	5	10	40
Average Optimism Scores	2.9	3.8	1.5	4.25	4.3
Average Innovative Scores	3.3	3.8	1	4.25	2.5
Average Discomfort Scores	2.2	2.2	3	2.75	2.68
Average Insecurity Scores	3.5	2.75	4.75	4.5	3.56
Average overall TRI Scores	2.87	3.14	2.56	3.94	3.26
Standard Deviation for overall TRI Scores	1.14	0.80	1.69	0.80	0.83

4.2. Summarised think aloud discussions

Three primary themes were identified through the qualitative analysis of Think Aloud interview transcripts. These themes emerged in relation to the commentary provided as people toured the Innovation Centre, and during the ice-breaker session and interview discussion. The three primary themes related to: (1) Public Good, Values and Trade-offs; (2) Data Ownership, Trust, and Governance; and (3) Social Living Arrangements. Central to participants’ discussions on these topics were negotiations of acceptance and the formulation of conditions for opting to live in a smart and modular home. These negotiations are reflected in the participant quotes presented below.

4.2.1. Public good, values and trade-offs

When reflecting on the values of smart and modular living environments, prompted by the interview tasks they engaged in, participants (n = 18) distinguished between the ‘public good’ these types of environment would generate for the wider society and their own personal and monetary values (see Table 3 for theme summary). It is noteworthy that this ‘persona’ approach was not prompted by the interview design, but by the participants themselves.

Table 3. Example summary of the theme and sub-themes relating to Public Good, Values and Trade-offs.

Theme	Sub-themes	Description	Participants
Trade-offs between personal and public benefits	Improve people’s quality of living (including health and lifestyle).	Focused on the benefits of smart home technologies and how they can assist lifestyle changes, particularly those vulnerable, in their everyday lives.	P1, P2, P4, P6, P7, P8, P9, P11, P13, P14, P15, P17.
	Supportive and adaptive living.	The benefits of modularity furniture and how homes can be adapted to people’s needs at different stages of their lives.	P1, P2, P3, P5, P6, P7, P8, P9, P10, P11, P13, P13, P14, P15, P16, P17, P19, P20.
	Wider community and society benefits (including social care requirements).	The benefits of monitoring the health of elderly relatives or vulnerable adults for diagnostic purposes.	P1, P2, P11.
	Modularity as a solution for families.	Home environments can be adapted to support families as they grow.	P12, P16, P17.
	Between affording and best way to spend money.	Weighing up the subjective and personal value of buying smart home technologies versus their long-term benefits.	P2, P3, P8, P20.
	Affordability of and smart home living (monetary value).	Affordability of smart home technologies can create a barrier for adopting them.	P3, P4, P6, P7, P8, P9, P11, P12, P13, P14, P15, P16, P17, P18, P19, P20.
	Ability to control smart home.	Focused on the degree to which someone can fully or partially control a smart home.	P1, P3, P4, P5, P6, P7, P8, P9, P10, P11, P12, P13, P14, P15, P16, P17, P18, P19, P20.
	Love to try new things/Factors influencing my willingness to live in a smart home.	The ability to trial smart home technology before opting to buy it.	P2, P7, P8, P12, P20.

When considering wider society, most participants (n = 18, all behavioural profile types except ‘avoider’) saw benefits to smart home technologies for helping those with assistive living requirements and healthcare issues (e.g. system reminders, Powers of Attorney scenarios, and caring for others). Participants (n = 3; classified as explorers and hesitators) also welcomed the ability to ‘manage and adapt home living spaces’ to support people’s changing needs according to lifecycles (e.g. single living or raising a family). These included its ability to support independent living for vulnerable people, aiding home working opportunities, and providing solutions for small and communal shared spaces (e.g. wall dividers). P17, for example, said:

You actually do want to change the way you use space quite often, almost at an hourly basis. But then also anything at a longer term, as the children grow and their needs and their wants and the space, the way they use space kind of changes. I really like that.

Conversely, when considering the personal benefits and challenges of living in a smart and modular environment, participants framed their conditional statements according to their personal values. For example, when discussing the possibility for opting to live in a smart and modular living environment, most participants (n = 19, all behavioural profile types) said that they would consider this option. However, they advocated for the need to develop a manual override system to enable the ability to exercise control over the smart technologies, particularly amongst those users who claimed less experience with smart systems or in the event of an emergency (e.g. system failures and security breaches).

When considering the ‘affordability’ of living in a smart and modular home environment, participants reflected on this in two ways: the value for money (exchange of money) and the subjective worth (based on personal beliefs). Relating to this, most participants (n = 16, all behavioural profile types) felt that existing costs of smart home technologies needed to be lowered if they were to be made more accessible to prospective end-users (e.g. subsidising student rented accommodation or a tiered costing system) and to increase their general appeal. Others (n = 4, classified as explorers and hesitators) also discussed weighing up the trade-offs between initial costs of living in a smart and modular home environment (which were perceived to be quite high) and their long-term beneficial value (e.g. of health and environmental monitoring). P02, for example, said:

Yes. Or it is the same when you need to choose if you want to live in a smart home or not, you need to decide the trade-off between the value they give you, that kind of technology, and of course money that you are willing to pay. So, there are different types of smart fridge. There are other ones that just alert you that you need to put some water in so you can make ice (laughter). There are levels.

Nonetheless, whilst some acknowledged (n = 5, classified as explorers, pioneers, and hesitators) the benefits of smart and modular living environments, some argued that the only way for them to truly consider it was to try living in such an environment.

4.2.2. Data ownership, trust, and governance

Participants acknowledged (n = 19, all behavioural profile types) the role of good data governance when discussing their data sharing expectations in a smart and modular home environment (see Table 4 below). Data sharing expectations related to which data would be shared, with whom and for what purpose. Expectations were conveyed through a series of conditional statements regarding data ownership, data sharing, consent management, and best practices in data management. Comments were principally made in response to a conceptual ‘data sharing’ table produced by MyGlobalHome, which was used as a prompt in the interview.

Table 4. Example summary of the theme and sub-themes relating to Data Ownership, Trust, and Governance.

Theme	Sub-themes	Description	Participants
Data sharing and governance management.	Gaining user’s consent.	The process of gaining user’s informed consent.	P2, P3, P4, P5, P6, P7, P8, P9, P10, P11, P12, P13, P14, P15, P16, P17, P18, P19, P20.
	Increasing knowledge about data sharing and governance.	Enabling people to obtain data sharing information in an accessible format.	P1, P2, P3, P4, P5, P6, P8, P9, P10, P11, P12, P13, P14, P15, P16, P17, P20.
	Consent upon clarity and explanation.	Providing people with all information relating to the nature of data being shared and with whom as an open dialogue with smart home technology developers.	P12, P15.
	Being able to decide/Data people are comfortable sharing (including health data).	The type of data participants was willing to share that had been collected by smart and modular living environments.	P1, P3, P4, P5, P6, P7, P8, P9, P10, P11, P12, P13, P14, P15, P16, P17, P19, P20.
	Owners versus renters.	Basing conditions about whether to use smart technologies based upon whether someone is renting or buying their property.	P3, P8, P20.
	Credibility and trust in the data sharing system.	The degree to which the existing system is both credible in its ability to protect users and manage their data ethically.	P2, P4, P6, P8, P9, P11, P12, P13, P14, P15, P16, P17, P18, P19, P20.
	Personal communications (Building Trust).	Smart home technology developers adopting a personalised approach when asking users to consent to sharing data collected by smart home technologies.	P11, P12.

Discussions indicated mixed views amongst participants over the types of data being collected and shared by technologies within the smart and modular home environment with third-party stakeholders. The exact views conveyed depended largely on contextual factors. For instance, three participants identified data sharing to be more acceptable if they were renting the accommodation (e.g. with data on building performance being shared with landlords) than if they were a property owner (e.g. with data on building performance being shared with home insurers). The notion of sharing personal health data was met with strong disagreement by most participants (n = 18, all behavioural profile types), particularly where the data was to be used for anything beyond providing healthcare support (e.g. patient diagnostic purposes).

Trustworthiness and data management related issues were intertwined and discussed by most participants (n = 15, all behavioural profile types). The main focus of the discussion was on the factors affecting their trust in the developers of smart and modular homes. The potential for misuses of data and/or vulnerability to data-security breaches (e.g. hacking attempts) negatively affected these participants’s trust in, and perceived credibility of, developers. This led participants to discuss the need for greater online protection, with some (n = 2, classified as explorers) advocating for a more personalised approach from developers regarding the management of data.

Whilst there was general agreement that anyone choosing to live in smart and modular home environments should own the any data collected by the home, a strong emphasis was also place upon the public having a role in the consent processes associated with data management (n = 18, all behavioural profile types). P19, for example, said:

I think being able to go away, get as much research as you can, understand what you’re getting yourself into and then coming back and having a discussion, and coming to an agreement, I think that would be a really ideal way to get consent.

A key reason for this was to reduce the risk caused by data sharing (e.g. financial risk) posed by the actions of those operating within the smart home industry. Approaches for managing this suggestion included adding a built-in data-management application to user’s mobile devices, or using personalised written agreements with potential users. Participants expressed differing views on what the design principles should be for consent management (e.g. one-off versus regularly reviewing consent approvals).

Some participants (n = 2, classified as explorers) expressed a need for ‘clarity’ in all information provided by smart home industry developers to potential users, as a condition for seeking informed consent when collecting their data. Specifically, the ability to improve the design and comprehension of information used in Terms of Agreement (e.g. length of documents), as well as ensuring its accessibility for a diverse range of users. P017, for example, said:

When you go on the website and it asks you about cookies and you’re supposed to read like six pages of text, generally most people just accept because they really can’t be arsed to read all the text.

There were also variations in how participants wanted to acquire data sharing and governance related information. Some participants (n = 17, all behavioural profile types) reported the need for users to be ‘self-learners’, educating themselves on the benefits and challenges of data sharing (e.g. what data is shared, why and with whom). Others (n = 2, classified as explorers) wanted to be able to have an open dialogue with developers within the smart home industry. Both approaches were felt to not only increase the sense of trust between the public and the developers, but also enabled people to make better informed decisions when considering their adoption and use of smart technologies within a household context.

4.2.3. Social living arrangements

Most participants (n = 12, all behavioural profile types except avoiders) acknowledged that shared accommodation would present unique functioning and relational challenges to those living in shared housing conditions and to the developers of smart and modular homes (see Table 5 below). Specifically, participants discussed their reduced sense of control in shared home environments, and the ethical considerations needed for guests in situations where the living environment was being monitored (e.g. rights to privacy and consent management processes needed).

Table 5. Example summary of the theme and sub-themes relating to Social Living Arrangements.

Theme	Sub-themes	Description	Participants
Need to control and manage smart and modular living environments.	Managing shared modular living spaces.	Focused on how participants would envisage managing share modular living spaces, based on their reflections of the Innovation Centre.	P5, P6, P7, P8, P11, P12, P13, P15, P16, P17, P19, P20.
	Manual and ‘app’ control.	The ability to control a home via an ‘app’ as well as through a manual override system.	P11, P12, P13, P15, P16, P18, P19, P20.
	Tensons with multiple residents.	The conflicting relationship dynamics between those sharing smart and modular living environments. This included the potential for an app to negotiate the living environment between residents (e.g. use of smart devices).	P11, P12, P13, P14, P15.
	Relationship dynamics of shared living spaces (Families and student halls of residence).	The potential dynamics of students or families living within a smart and modular living environment.	P1, P2, P3, P4, P5, P6, P7, P8, P9, P10, P11, P12, P13, P15, P16, P17, P19, P20.
	Factors influencing my willingness to live in a smart home.	The ability to trial smart home technology before opting to buy it.	P2, P7, P8, P12, P20.
	Need to be more ethical (Considerations for guests and children).	Ethical concerns regarding the consent measure sin place for guests and children in a smart and modular living environment.	P11, P12, P13, P15, P16.

First, some participants (n = 5, classified as explorers, and sceptics) expressed concern over the possibility that smart home technologies could foster conflict amongst people living in shared living accommodation. For example, disagreements and power imbalances between people living in shared accommodation could lead arise where smart-technology profiles and preferences are set up and controlled by one user to the exclusion of others. Participants (n = 18, all behavioural profile types), however, expressed humour when reflecting on previous family and/or student conflicts in similar shared living environments, noting the need to create private soundproofed zones. This prior experience seemed to instill future living preferences among some of the participants (n = 5, classified as explorers, pioneers, and hesitators), with most indicating that they would choose to either live alone or with known individuals if they had the option to live in a smart and modular home. Suggested ways of managing these functional and relational challenges were also put forth by some participants. The option of ‘an App’ was discussed, which would enable people to negotiate environmental conditions and organise equity of access to shared facilities and resources. P11, for example, said:

Well clearly there are always issues with a shared space, and an app is no different to the idea of having a two-way light-switch with a switch at either end of the room. If somebody wants the lights on and somebody wants the lights off, then there is a scope for conflict.

Other participants (n = 5, classified as explorers, pioneers, and sceptics) mentioned ethical complexities which are challenging to control. These included privacy and equity of access (e.g. guardianship for children and vulnerable adults), admitting visitors, providing access for social gatherings, and granting select access to share facilities (e.g. heating and laundry facilities). These participants provided solutions through the following practices: the development of a co-living App was again mentioned, as were parental control mechanisms, creating override systems, and consent management systems for guests.

Finally, whilst interview discussions did not explicitly explore the implications of the COVID-19 pandemic, some participants (n = 4) made direct references to the pandemic when weighing up the benefits and challenges associated with working from home in shared smart and modular living environments. P12, for example, discussed the modularity concept:

I mean I’m also trying to think maybe how much of this was done pre-COVID and conceptualise. So, that idea of having work and home, there might be a stronger drive for that.

5. Discussion

Our study echoes recent calls for more research, using theoretical models and in-depth qualitative elicitation methods (e.g. Think Aloud approach; Shin et al. 2022), to help advance the future design of smart and modular homes, such that they better address the requirements of prospective end-users. Accordingly, by engaging participants in a realistic scenario (i.e. providing access to a smart modular home) and via the use of a Think Aloud interview approach, we were able to gain insights into the sense-making, deliberative processes and conditions that people might place upon their acceptance of smart and modular living in a university ‘smart campus’ context (Jagannath 2023; Shin et al. 2022). Therefore, the study findings offer a unique opportunity to contribute to the ongoing practical designs of smart and modular living environments by focusing on the agency of potential residents and users to co-create these designs, including in other similar higher education institutions (Baudier, Ammi, and Deboeuf-Rouchon 2020; Polin et al. 2023). The knowledge generated by this study could also be of use to practitioners, developers, and policymakers to make more informed, citizen-led design decisions on how to adapt and improve smart and modular homes, including in other Higher Education settings (Doling and Arundel 2022; Jagannath 2023; Jaspers and Pearson 2022; Quach et al. 2022). For example, the findings provide greater clarity over data ownership preferences in shared living environments. A suggested framework of considerations and recommendations emerging from this study are presented in Table D in the supplementary materials.

Consistent with other smart home research (e.g. Shafi and Mallinson 2023; Sovacool et al. 2022; Zimmermann et al. 2023), participants also acknowledged general benefits that would be associated with people’s willingness to share data at a societal level (e.g. assistive living for vulnerable adults). It is noteworthy, however, that this persona approach was not prompted by the interview design when considering the wider societal benefits, but by the participants themselves and may reflect the elicit nature of the interview discussions (e.g. participants asked to verbalise their thoughts and reflections that come to mind; Boyle et al. 2023). Participants’ own willingness to live in smart and modular living environments was negotiated in relation to their personal values (Gerli et al. 2022; Park et al. 2022; Shin et al. 2022). For example, our findings indicate that affordability could be viewed from two perspectives: subjective value and monetary value (Park et al. 2022). For some, the long-term quality of life benefits associated with smart and modular living were seen to outweigh (or justify) the initial financial outlay. For others, despite identifying societal and personal benefits, the monetary costs were seen as a prohibitive financial barrier. This latter view has been expressed in previous research (see Sovacool et al. 2022). Such differing assessments of affordability could be reflective of our participants’ relative socioeconomic status and educational background (see Baudier, Ammi, and Deboeuf-Rouchon 2020; Polin et al. 2023), or how they perceive society based on life experiences more broadly (Park et al. 2022). Yet despite calls by participants to promote schemes to help incentivise access to smart living environments (e.g. subsidising rent, tiered pricing model, and free-trial practices; Sovacool and Furszyfer Del Rio 2020), there was also an acknowledgement that financial accessibility would not guarantee people’s willingness to live in such environments. For this reason, not only does the objective affordability of smart and modular living environments need to be considered, but so do the roles that consumer choice, personal values and the beliefs that people hold play (Baudier, Ammi, and Deboeuf-Rouchon 2020; Park et al. 2022; Polin et al. 2023).

Hosting the think aloud interviews at the Innovation Centre enabled participants to more easily verbalise their opinions on privacy and data-management considerations relating to smart and module living. Thus, the Think Aloud method could be an approach that the designers of smart and/or modular living environments could use to support citizen engagement activities to inform their future designs. Some of the topics raised included the need for improvements to the clarity of information about data sharing with third-party stakeholders (e.g. in relation to sharing of health data). This extends prior research (Runkle et al. 2019), where situational factors were key to some conditional statements participants made in practice (e.g. type of data collected). For instance, sharing building performance data with landlords but not home insurers. Study findings also resonated with research exploring the ethical challenges raised about guest privacy (e.g. the need for consent management processes to cover visitors to monitored environments), Powers of Attorney, and equity of access for shared facilities (e.g. heating facilities) (D'Alessandro et al. 2020). The ethical issues around data-sharing, particularly in shared-living environments emphasise the need to shift the balance of power over use of data to those who have shared it, i.e. enabling people to control access to data (Bolton et al. 2018; Jaspers and Pearson 2022).

Interview discussions also enabled participants to deliberate the importance of public involvement in data governance processes, and trusted mechanisms to facilitate these processes (e.g. an ‘App’ to audit data shared; Ling et al. 2021). These discussion are consistent with recent data governance frameworks created to ensure responsible data sharing practices (e.g. participatory data stewardship and the Gemini Principles; Ada Lovelace Institute 2021; Chowdhury and Oredo 2022; Phillips-Wren, Daly, and Burstein 2022; Shafi and Mallinson 2023). Building on from existing sociotechnical research (Crabtree and Mortier 2015; Sarker 2022), interviews revealed knowledge of how and in what ways participants wanted to have further involvement in data governance practices not only helps researchers and home solution industry developers to create such frameworks in the future (e.g. framework for participatory data stewardship). These findings also give further weight to our understandings on how having representative and inclusive frameworks can be important to the public as are future collaborations between the public (as both citizens and consumers) and industry; particularly if these practices are to reflect public wants and needs (Pirzada et al. 2022).

Additionally, the interviews enabled the relationships between prospective co-habitants to be further examined, and under what conditions potential users would perceive smart and modular living to be acceptable in a smart campus context (Debrah, Chan, and Darko 2022; Sovacool et al. 2022). For instance, whilst some participants welcomed the benefits of having modular features to sub-divide living spaces (e.g. moving walls, and system reminders), others acknowledged the unique relational challenges within shared accommodation (e.g. conflict between people relating to household noise). This led them to point to the need for a manual override for the automated systems and a co-living management system to manage these issues (Jagannath 2023; Ye, Li, and Yang 2021). Whilst these findings resonate with research exploring issues of autonomy and domestic life in shared accommodation (D'Alessandro et al. 2020; Ehrenberg and Keinonen 2021; Hargreaves, Wilson, and Hauxwell-Baldwin 2015; Jaspers and Pearson 2022), the need for a manual override or co-living management system had only been recommended for smart home technologies. These features had not previously modular home solutions as our study indicates (Jagannath 2023; Katunský et al. 2020; Quach et al. 2022). This should draw the attention of those in the design of smart and modular living environments to some of the emerging design challenges and research questions associated with the interaction of smart and modular design elements in the context of communal living (Alhadreti 2021; Knott et al. 2022; Thompson et al. 2018).

Finally, the TRI 2.0 revealed participants’ general attitudes towards technology prior to the study, with the majority classified as either ‘explorers’ or ‘hesitators’, meaning that most people had a high sense of optimism about technology and experienced lower degrees of discomfort and insecurity about technology (Mels 2018). This could have affected the responses within the study; shaping the conditional arguments made about the technology and the nuances in the observations made about the design elements of the Innovation Centre (see Mels 2018; Park et al. 2022). Similarly, the fact that our participants were university staff and students, who tend to be from higher-income groups in the UK, might have affected the comments made within our study (e.g. in relation to affordability), particularly as higher-income groups tend to be earlier adopters of smart home technologies (Baudier, Ammi, and Deboeuf-Rouchon 2020; Polin et al. 2023). That said, income was not directly measured in this study and so this latter conclusion remains somewhat speculative. While our study arguably succeeded in recruiting the kinds of people who would likely be the ones to consider living in a smart and modular home environment (particularly in the campus context studied), there is a need for further research to explore people’s conditional arguments made by those beyond this study’s sampling population (see Mels 2018; Park et al. 2022).

5.1. Study strengths and limitations

There are several key strengths to this study. For example, the Think Aloud approach conducted within the Innovation Centre enabled participants to reflect deeply on the challenges and benefits of smart, modular living while in situ. Also, the study design enabled participants to engage in a ‘real time’ and contextualised negotiation of the kinds of conditions they would place upon the prospect of living in a smart and modular home, as well as what they would view as acceptable or not regarding things like data-sharing.

There are also some noteworthy limitations. First, it should be noted that the key findings are tied to a UK university context where study participants tended to be of a higher socioeconomic background and educational attainment. The study also occurred during the COVID-19 pandemic. Care should therefore be taken when transferring the findings to other populations both within the UK and more widely as well as making comparisons with empirical research occurring pre-and post-COVID-19 pandemic. For instance, as most participants studied or worked remotely from home or student halls of residence and might account for these deeper insights into co-living experiences (Self 2021). Replicating the study and/or scaling up and widening the pool of participants recruited would be useful in examining the robustness of our findings. Furthermore, our study engaged participants for a short duration in the Innovation Centre. Study conclusions therefore did not go beyond participant’s short-term observational perceptions and initial reflections. Future research exploring long-term experiences with other and fully functioning units is therefore recommended.

6. Conclusion

Understanding how people negotiate the possibility of living in a smart and modular home, as well as what they deem to be ‘acceptable’ in such contexts (e.g. in terms of data-sharing in such environments), is important but understudied research area. This study used a Think Aloud approach – conducted within a smart, modular home situated on a university campus – in order to shed light on this research question. Knowledge generated by this study could assist practitioners, developers, and policymakers working in this field to make more informed, citizen-led design decisions on ways to adapt and improve future smart, modular home concepts. Our use of a situated Think Aloud method, enabled participants to ‘perspective take’ in a realistic context and scenario, adding credibility to our findings. The nuanced accounts produced by our participants recognised the public benefits of automated smart and modular environments, whilst acknowledging some of the unique functioning and relational challenges, particularly within shared living. Our study provides new insights into the bottom-up understanding of smart and modular living spaces, considered in situ, among a group of participants who would be likely adopters of the technology. These insights could be transferable to other contexts and populations, but there is the necessity for further research in this regard.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Innovate UK.