Understanding Trust in the Smart City Through Focus Groups in Hong Kong

Abstract

This paper examines the role of focus groups in mixed-methods research, focusing on trust dynamics within the smart city. It takes the Asian metropolis of Hong Kong as its empirical case. It successfully validates focus groups as a research method across three dimensions. Firstly, focus groups facilitate the comparison of in-depth discussions across various demographic groups; secondly, focus groups enable collective deliberation to uncover hidden relationships; and, thirdly, when correctly employed, focus groups can bridge qualitative (interviews) and quantitative (surveys) research to enhance triangulation. In this study, the findings of focus groups substantiate the data trust paradox identified in previous surveys and interviews: high public support for technology in a low-trust environment, with explicit privacy concerns related to data management. The focus group results reinforce distinct insights in relation to birth origins and political identities, reconfirming differential trust among demographic groups. This paper demonstrates tailored focus groups as an effective approach to unraveling the trust dynamics in smart city strategies.

1. Introduction

This paper examines the role of focus groups in investigating a complex research field centered on the concepts of trust and the smart city, taking Hong Kong, the Asian metropolis, as the empirical case. Hong Kong has consistently appeared close to the top of various global rankings covering the digital domain, especially since 2006, when the Hong Kong Special Administrative Region (HKSAR) government introduced the One-Stop Government Portal (GovHK), which has received significant professional acclaim internationally [1]. In 2020, the Hong Kong Smart City Blueprint 2.0 was released by the HKSAR government [2], which outlines strategies across six smart city dimensions, including the economy, the environment, the government, living, mobility and people, seeking to transform Hong Kong into a top-tier smart city internationally. Hong Kong, a special administrative region of China with almost 7,500,000 inhabitants, is understood as an exemplar of a smart city in a regional and international context based on cooperation and strategic rivalry [3,4,5,6]. This Asian metropolis is currently experiencing a transformative shift.

Although historically defined by the arrangement of One Country, Two Systems, Hong Kong is increasingly aligned with Mainland China’s political structure and economic strategy (through the Greater Bay Area initiative) [7]. Studying Hong Kong’s urban planning strategies uncovers the complex policy implementation in a city with an escalating sense of fear and mistrust between the people and the government and where certain citizens are reluctant to be further integrated with Mainland China, highlighting the numerous disparities in defining ‘smart’ between the government and citizens.

Irrespective of place, the ‘smart city’ raises important issues of trust [8,9,10]. These dilemmas are captured across three principal dimensions. The first is that of trustworthiness in actors, providers or outcomes. Trustworthiness is widely recognized in the literature as a suitable descriptor for the dimensions of honesty, integrity and benevolence [11,12]. Cole and Tran [8] consider trustworthiness in terms of providers, regardless of their public, private or hybrid status, conceptualizing it as the form of a relationship that extends from interpersonal ties to more conceptual objects such as technology, utility providers and institutions.

The second dimension revolves around trust and technology. Smart cities are sometimes described as being both unavoidable and beneficial, with technological development as an unstoppable force bringing us richer, safer and more sustainable urban futures. Yet the implementation of technology extends beyond engineering, also linking to strategic, political and value-driven choices. Wong [13] differentiates between trust-enhancing and trust-enabling technology, with the former staying within the public policy framework and including attributes such as government transparency and freedom of information, while the latter depends on citizens providing high-quality data within a trustworthy environment. As the smart city is principally a data-driven city, the success of the smart city relies on personal data collected from mobile devices and environmental sensors installed by the government and, subsequently, trust in data usage [14,15,16].

The governance challenge [17] lies in developing legitimate frameworks for perceptions, power relations and decision-making in smart cities. Smart city governance, however, suggests that the smart city is an interplay between the government, market and citizens, shaped by the politics of data management, including collection, storage, visualization and usage. Despite the widespread rhetoric that citizen participation is essential to smart city development, critics argue that the dominant corporate model tends to exclude citizens, reducing them to only passive roles as consumers of smart technology or data producers for corporations and governments [18,19,20]. Scholars claim that smart devices, data collection and the digitalization of the government fail to translate into empowerment strategies and meaningful citizen participation in urban governance [21,22]. As Gooch et al. [18] state, dominant smart city visions leave little room for citizens to take control of their own communities or to collaborate in shaping policy, governance and solutions to particular challenges.

The third dimension concerns whether the dimension of trust is a proxy for deeper beliefs. Why might the smart city be distrusted? Could this be explained by distinct characteristic-based trust profiles? Is it linked to data trust issues stemming from the application of smart technologies and artificial intelligence, or is data trust quite simply an epiphenomenon, with trust or mistrust in smart cities reflecting broader social and political dimensions? This effort of interpretation is important, as it defines the nature of trust or mistrust—whether it is in technology per se, in providers or in technology as a process.

The originality of this article is its linkage of trust, technology and perceptions of urban service delivery in a Chinese or Asian context (specifically, Hong Kong and the Greater Bay Area). This study is a single case study using Hong Kong as the empirical case, but it is embedded within the dynamics of the Greater Bay Area and China more generally. The single case study provides a detailed examination of the complexities shaped by the unique history and influences of the area [23]. This advantage enables an in-depth analysis of how the sociopolitical context of Hong Kong affects citizens’ trust in smart city development.

2. Methods and Materials

The smart city is a notably slippery concept, especially when linked with trust in data. Mobilizing distinct ways of knowing can assist in methodological triangulation and enhanced understanding [24]. Because of Hong Kong’s consistent endeavors to evolve, remain and further develop as a smart city, its regular appearance at the top of the rankings of the best smart cities worldwide [10] and its self-proclamation as a smart city, Hong Kong is an appropriate setting for this study. Aiming at better managing its resources for both citizens and visitors, Hong Kong endorses the utilization of smart technologies and instruments. Having operationalized various smart city projects, like iAM Smart, eHealth, LeaveHomeSafe and FPS, and having widely advertised the Hong Kong Smart City Blueprint 2.0, one would expect not only that officials would consider their city as ‘smart’ but that this would also be the case for other key smart city stakeholders, like the Hong Kong residents, businessmen, civil society actors and academics. What do Hong Kongers from diverse segments of the society and varied backgrounds know about Hong Kong as a smart city? Do they understand the contents of the Smart City Blueprint? To what extent do they support (or oppose) the development of Hong Kong into a smart city? Are they proud of Hong Kong’s performance as a smart city? We asked these questions in a territory-wide survey (N = 808). We also discussed these issues in purposive interviews with key stakeholders from the business sector, academia and civil society (N = 25). We debated these topics with four special convened focus groups (N = 42). The perspectives of Hong Kong residents about Hong Kong as a smart city were primarily collected from a purposive sample interviewed between July 2020 and December 2021. The views of the residents were further validated in the territory-wide survey and focus groups. Focus groups, as a participatory method, have been reintroduced as a research method recently, aiming to understand the meanings, beliefs and cultures of social issues that influence the feelings, attitudes, perceptions and behaviors of individuals [25,26,27,28].

In our research, we assessed and compared the knowledge of residents about Hong Kong as a smart city, their familiarity with the Smart City Blueprint and their support of (or opposition to) the smart city development of Hong Kong. We gathered primary data from three different sources (survey, interviews, focus groups), and we analyzed the collected data quantitatively and qualitatively (see Figure 1). Statistical relationships have produced interesting findings about explanatory variables determining support for the smart city [8] or identifying the foundations of pride in Hong Kong’s smart city [29]. More interpretative accounts have retraced the smart city as an urban narrative over decades in Hong Kong [9]. This article continues to explore how to understand and interpret the complex relationship between trust and the smart city by introducing an interactive and deliberative dimension.

2.1. Step 1: Survey

Utilizing Hong Kong as a case study, we conducted a randomized telephone survey with 13 questions (see Table S1 for examples) and a sample size of 808 respondents. The survey was conducted by the Hong Kong Public Opinion Research Institute (HKPORI) [30] in Hong Kong between 24 March and 16 April 2021. The target population of the survey was that of Cantonese-speaking Hong Kong citizens of age 18 or above.

The survey produced four main findings. First, although Hong Kong citizens perceived that they had very little understanding of the government’s smart city plan, they supported the general idea of the smart city. Second, Hong Kong citizens—the younger generations in particular—were generally unsatisfied with the smart city services in Hong Kong. Privacy was one factor associated with the level of support. Third, citizens exhibited different levels of trust among different government bodies in providing smart city services. Fourth, Hong Kong citizens had a generally higher level of trust towards private companies and non-governmental organizations (NGOs) in providing smart city services. The results suggested that citizens had genuine trust in smart technology but mistrust towards potential smart city providers, especially the government. We interpret this in terms of the data trust paradox, indicating that there is high support for technology in a low-trust environment. These factors emphasize the fundamental challenge of a smart city—that it struggles to fulfill its intended vision. Our survey reveals the challenges facing government policy.

2.2. Step 2: Concurrent Interviews

In addition to the territory-wide survey (n. 808), the team carried out 25 individual interviews (see Table S2) between July 2020 and December 2021 to collect diverse data from three sectors—the private sector, public sector and civil society—allowing deep and penetrating insights. The results revealed the presence of mixed views towards the smart city concept in Hong Kong. Overall, two salient findings stand out: the first is a lack of enthusiasm about the technical narrative of the smart city, and the second is the diverse opinions on the effectiveness of ‘smart’ environmental policies, with the link between smart city initiatives and sustainability being a topic of debate among the interviewees [9].

In support of the first theme, we observed the prevalence of critical accounts. For interviewee HK17, ‘The narrative is overwhelmingly tech-driven, but we rarely consider what technology means in context of smart city development’. For interviewee HK20, the smart city concept seemed fixated on the ‘technical and technological dimensions, like the gadgets, or installing the latest technologies’. Representative of the sample was interviewee HK15: ‘These different actions seems lacking cohesion to me. All the Internet of Things (IoT), big data, artificial intelligence (AI), fifth-generation (5G), are on the table without clarifications how they connect’. Notably, one interviewee even criticized the focus on STEM, contrasting the technologically emphasized smart city with art technology: ‘Art technology could power Hong Kong like a STEAM engine—Science, Technology, Engineering, Arts, Mathematics—beyond just STEM. We need more than only STEM-oriented talent’ (HK17). These narrative abilities, especially in digital storytelling, are key soft skills in supporting the creative digital economy.

Secondly, opinions were split regarding the linkage of smartness and sustainability. The initiatives of Hong Kong’s smart environment dimension include enhancing waste collection’s traceability and reducing pollution through smart metering. However, interviewee HK20 deemed these measures inefficient, using the example of plastic waste. Ambitious policies like waste charging, as adopted in other places, were suggested to control the consumption-based emissions, but they face obstacles due to entrenched political and economic resistance. Interviewee HK05 offered a compelling perspective, suggesting that Hong Kong’s identity as a free port sees all taxes, including new environmental tax, as undesirable, because its status is built on global trade and the preservation of free commerce. Regarding the smart and sustainable city of Hong Kong, several interviewees stated that Mainland China, particularly Shenzhen, outpaces Hong Kong in terms of smart and sustainable management. The small size of the territory also raises logistical problems: interviewee HK04 believed that it made no sense to develop a waste management facility in Hong Kong due to the limited population in a small territory, urging Greater Bay Area collaboration.

While acknowledging Hong Kong’s strengths, such as its advanced infrastructure, technological readiness and openness to innovation, the interviewees collectively painted a nuanced picture of the city’s smart city progress thus far. Their critiques and concerns suggest that, while Hong Kong possesses many advantages, realizing its smart city vision will require it to overcome systemic rigidities, foster cross-sector collaboration and strike a careful balance between embracing new technologies and upholding the principles of good governance, transparency and citizen empowerment.

Understanding smart city dynamics requires a mixed-methods approach, mapping statistical relationships, receiving data-rich interview accounts from relevant experts and engaging in more interactive forms of learning and exchange via focus groups. No single method can capture the full complexity of the case. There are strengths and limitations to all methods. Surveys provide aggregate data, elucidating powerful correlations between some sociodemographic and attitudinal data [8], but the findings are non-contextualized and lack granularity. Interviews are well suited to elicit certain types of responses and provide innovative perspectives on the dimensions of the smart city; however, in 2020–2021, the interviewees showed strong reluctance to discuss Hong Kong’s past, present and future amid the protests in 2019 against the extradition bill and in 2020 against the National Security Law, which shaped the context of our empirical data collection.

2.3. Step 3: Focus Groups

At the third stage, the focus groups allowed the research team to understand and interpret the complex relationship between trust and the smart city by introducing an interactive and deliberative dimension, rooted in a real-time comparative design. The focus groups addressed the three main research questions (trustworthiness, trust in data, trust in smart city as an epiphenomenon), while securing perspectives based on the place of origin (Hong Kong and China), profession (public and private sector) and sociodemographics (e.g., age, gender).

Focus groups can help in understanding this trust–technology nexus. The mixed-methods approach in our research project combined elements of concurrent and explanatory sequential designs, since the survey and interviews were conducted simultaneously within the same timeframe, while the focus groups were conducted at the later stage of the project in an attempt to explain the findings from the survey [31]. Particularly suitable for an in-depth understanding of social issues from a purposefully selected group of individuals [25], the focus group is a type of group interview in which the moderator/researcher asks a set of open-ended questions to obtain collective views about certain topics and encourage interactions among participants to elicit richer individual opinions [32,33].

The well-known flexibility of the focus group also extends to the data produced and their potential applications [34]. Focus groups are suitable for gathering different types of information, not only as a stand-alone method but also as part of a triangulation strategy [35]. They can yield results that combine responses to a pre-formulated questionnaire (basic information) with those of individual interviews (rich description); the focus group can also be used to explore findings from structured forms of investigation, such as surveys. Focus groups can uncover unexpected new perspectives and generate discussion about a research topic that requires collective views [25]. The results can be used to complement and explain statistical information obtained from other evaluation processes [36]. Thus, the focus group can be deemed as a mixed-methods research instrument that, when properly employed, can triangulate and transcend the findings of each individual research method [33,37]. The initial reason for implementing group interviews instead of traditional interviews was related to the excessive power of the interviewers. The focus group is a less directive research method to complement individual interviews, as the interviewers are less dominating and act more like moderators than investigators [25,36,38].

The general roles of the focus group as a research instrument are three-fold: as a qualitative method with quantitative elements contributing to the mixed-methods approach, serving to enhance the flexibility and credibility of the results (through triangulation) [34,39]; as a process of collective deliberation in a cross-sectional approach that uncovers the hidden relationships within complex domains (in this case, the trust dynamics in the smart city) [40]; and as a tool in a deliberative approach for the comparison of in-depth opinions between specific populations [35].

3. Presentation of the Focus Groups

Four specially convened focus groups, representing distinct audiences, accompanied the survey and purposive interview sample referred to in the introductory section. Focus groups operate well when the participants share characteristics in common, such as gender, age and ethnic and social class background, or even have similar shared experiences [25,26,37,41]. For Krueger and Casey [38], people tend to disclose more about themselves to people who are similar to them, particularly when addressing sensitive issues like trust, privacy and governance, to avoid a fear of judgement. As a result, one strategy for focus groups is to select largely homogeneous groups of individuals in participatory research to maximize the probability of the targeted outcome [25,37,41,42,43,44]. This selection reflects the fact that attitudes and opinions are socially formed and discussed in depth within the permissive social environment created by a focus group [36]. As participants are grouped with common qualities, focus groups can further enable researchers to examine and compare how the results are different across social groups such as age, gender and ethnicity [41]. While homogeneity helps to enhance the depth of insights within each group, it may limit the diversity of perspectives. As a common trade-off in focus group design, we thus attempted to avoid this limitation by also using survey and interview data to triangulate and interpret the findings.

The focus group encourages a deliberative dynamic, facilitated by the presence of moderators. An experienced moderator who stimulates debate and clarity [37,42], in combination with an expert co-moderator to provide participants with prior information on the topic, can effectively induce a deliberative discussion [35]. The focus group also builds on the group dynamic within a flexible and semi-structured framework to encourage interactions, and the participants are allowed to express their opinions in a spontaneous manner [25,27,34,37]. Focus group research offers two dominant virtues: they enable an understanding of a wide range of views that people have about a specific issue and how they interact (between participants and researchers or between participants) and discuss the issue [40,41].

Since the focus group played the explanatory role regarding the survey in our mixed-methods approach, the rationale for the formation of four groups was primarily derived from the survey’s findings, which showed that the region of birth and political inclination contribute substantively to trust dynamics [29]. The selection of four focus groups was intended to represent the most critical demographic and professional distinctions while avoiding redundancy. In fact, the thematic analysis conducted after each group indicated that most new themes emerged in the first and second groups, with a high degree of repetition in the third and fourth groups (as shown in Section 4), suggesting that data saturation was largely reached after four focus groups [42,45]. In the specific case, there were ten to eleven participants in each group, consisting of Hong Kong residents who had participated in the pilot survey: the participants were aged between 18 and 37, representing various political inclinations (pro-democracy, pro-establishment and localist) and diverse origins (from Hong Kong and Mainland China). Participants were recruited according to two of the five strategies suggested by Hennink [37], which are informal networking (reference by peers) and research-based (respondents from a survey). The selection of focus group members was not intended to yield a purely random sample but to reflect distinct professional groups with a mix of common experiences (ages and genders) and diverse professional and Chinese origins (Mainland and Hong Kong). Distinct group identities were identified for each group (see

Table 1. Characteristics of four focus groups.

Attribute	Focus Group 1 (FG1)	Focus Group 2 (FG2)	Focus Group 3 (FG3)	Focus Group 4 (FG4)
Age	18–24	22–31	22–37	22–32
Gender	Male: 3 Female: 7	Male: 3 Female: 7	Male: 3 Female: 8	Male: 3 Female: 8
Origin	Hong Kong: 0 Mainland China: 10	Hong Kong: 5 Mainland China: 5	Hong Kong: 8 Mainland China: 3	Hong Kong: 7 Mainland China: 4
Professional	Student: 10 Public sector: 0 Private sector: 0	Student: 10 Public sector: 0 Private sector: 0	Student: 0 Public sector: 11 Private sector: 0	Student: 0 Public sector: 0 Private sector: 11
Distinct group identity	Young Chinese	Mixed-origin group	Public sector	Private sector

), referring to the shared characteristics defining each focus group. In particular, Focus Group 1 (FG1) and Focus Group 2 (FG2) were grouped by common origins (Mainland or non-Mainland Chinese), while Focus Group 3 (FG3) and Focus Group 4 (FG4) were grouped by common professionals (civil servants or corporate employees), with the aim to compare their perspectives on trust in smart city initiatives as shaped by their professional roles [29]. Following the best practices in the literature [27,32,42,43], each group had one principal moderator (to assist the discussion of a particular topic) and two observers (to record the interactions between the participants and make notes of their facial expressions and gestures). Although the main language within the focus group discussions was English, the observers also provided for Mandarin–Cantonese–English simultaneous translation to ensure precise understanding and accurate expression among the participants. Information related to the specifications of the focus groups, substantiated by some of the leading sources in the literature, is shown in

Table 2. Design specifications of focus groups.

Design Specification	Operationalization	Literature
Number of focus groups	4 focus groups	3–4 focus groups [42,45]
Number of participants	42 participants in total	8–12 participants [32,44]
Duration	80–90 min each group	1–2 h [26,32,33,37,40,41]
Moderator	1 moderator, 1 notetaker, 1 interpreter	1 moderator, 1 notetaker [38,42]

.

The process involved careful preparation and protocols, with ethical approval granted by the Hong Kong Baptist University (HKBU) Research Ethics Committee under reference number [REC/19-20/0301]. Following advice from Lune and Berg [43], data collection was guided by compiling a schedule or an agenda. As a form of introduction, the moderator introduced the groups in a friendly and informal way, welcoming the participants and thanking them for attending the discussion. Each participant was then asked to introduce themselves (they were not pressed to specify other characteristics). The moderator asked for permission to record the discussion and then presented the ground rules for the group discussion: no personal attacks, no argument due to individual political inclinations and a permissive and non-threatening discussion environment. Participants signed informed consent forms and were assured that their participation would be anonymous and confidential. This procedure was repeated for each of the four groups.

The focus group vignette closely corresponded to the territory-wide survey (n. 808) and the interview schedule (n. 25) questions. Stewart and Shamdasani [44] suggest that the order of questions should range from the more general to the more specific. We began with an open question on the adjectives best describing Hong Kong and then sought to weigh the benefits and risks of the smart city in general and to establish the qualities of Hong Kong as a smart city in particular. The groups then deliberated on the questions of trust in providers and regulators of the smart city and issues of trust in data more generally. After a discussion of international and comparative (mainland) examples, the collective discussions ended with messages to the Hong Kong government. The structures of the questions are summarized in the Supplementary Materials (see Table S3).

4. Results: Data Analysis

According to Breen [36], a formal focus group data analysis should identify the most important themes, the most noteworthy quotes and all unexpected findings from the research. A purely thematic analysis of the whole corpus is interesting [46,47,48]. Equally important, if not more so, is the use of focus groups to illustrate distinct group perspectives (as well as converging ones across groups). To address the whole corpus, word frequency analysis, a lexical approach that has been validated to be useful for focus group interpretation, provides a thematic overview of all participants [49]. Group-specific word frequency analyses, which are shown in the latter part of this section, facilitate comparisons between the distinct groups, each representing distinct social realities. Emblematic citations bring to life each group. Unexpected findings can take various forms but are best elucidated by focus groups. Finally, focus groups are useful tools in terms of the triangulation of the validity of the research findings. Figure 2 shows the most frequent words (frequency higher than 30) across all four groups, analyzed by MAXQDA 24. The analysis presents only the top 10 most frequent words to ensure a focus on the most salient themes emerging from the discussions and to reflect the most visually accessible and analytically meaningful findings without overcomplicating the presentation [45,48].

The list is dominated by the words ‘Hong Kong’, ‘government’ and ‘citizens’, followed by variations of the city (‘city’ and ‘smart city’), trailed closely by the problem of trust. The data dimension is also present in the top ten list, with ‘information’, ‘application’ and ‘data’ central to discussions of smart city development in Hong Kong. Mainland China seems to be often mentioned as a rival or a target to be compared with when commenting on the digitalization progress in Hong Kong, since ‘China’ is within the top ten list. The simple word count summarizes well the project’s core, i.e., trust in the smart city in Hong Kong and China, with smartness understood mainly in terms of technical data applications.

There were many commonalities across all four groups (see

Table 3. Features of the smart city expressed in the four focus groups.

Element	FG1	FG2	FG3	FG4
City governance by algorithm		✓
Contactless/robot service	✓	✓		✓
e-health system				✓
e-payment	✓	✓	✓	✓
e-vehicle		✓		✓
Environmental protection		✓
Improvement in quality of life		✓
Internet of Things (IoT)	✓	✓	✓
Mobile applications	✓		✓
Next-generation energy		✓		✓
Online public services	✓		✓	✓
Smart identity cards	✓		✓
Traffic jam solution	✓		✓
Surveillance cameras		✓	✓
Usage of QR code				✓
Usage of urban space	✓

✓ = Discussed in the focus group

). In terms of definitions, that of ‘smart city’ is, above all, a technological one, centered on aspects of smart technology, such as e-payment, Internet of Things (IoT), smart identity cards, one-stop public services and police surveillance cameras. The focus groups demonstrated less of the rival framing (sustainability, e-governance) that was observed in the face-to-face interviews and drawn from the literature review [9].

All groups suggested convenience and efficiency as benefits of smart city development, primarily revolving around smart technology as the main reference (see

Table 4. Benefits and risks of the smart city.

Element	FG1	FG2	FG3	FG4
Benefits
Assistance in elderly and disabilities		✓
Civilizing citizens	✓
Convenience	✓	✓	✓	✓
Efficiency	✓	✓	✓	✓
Environmentally friendly				✓
New form of power for the authorities		✓
Transparent government		✓
Risks
Customized advertisement by private companies due to the misuse of personal information	✓		✓
Data loss by computer viruses				✓
Digital divide		✓		✓
Increase in unemployment rate		✓	✓
Inflexible decision made by digital system				✓
Invasion of privacy	✓	✓	✓
Leakage of personal information	✓		✓	✓
Rebellion of citizens due to implemented policies	✓
Unaccountable authority		✓
Vulnerable city due to relying on digital system		✓

✓ = Discussed in the focus group

). They were described in terms of the convenience of app-based public services, more efficient than traditional forms of service delivery in fields such as health and elderly assistance, consistent with the digitalization of public services to streamline bureaucratic processes and improve urban functionality, under the framework of smart governance [6,50]. Smart as sustainable was also mentioned. Several participants described the smart city as an environmentally friendly process, which aligns with some governments claiming to embrace the notion of smartness to target sustainable development [51]. Although specific elements such as e-vehicles and next-generation energy were mentioned, the connection between the environment and technology appeared weak, in contrast to the 25 individual interviews. Sustainability and smart city development raised conflicting priorities, such as the observation that electronic convenience and environmental sustainability can be in contradiction—for example, the explosion of takeaway food mobile applications, with a negative effect on carbon monoxide emissions. For most respondents, the benefits of the smart city outweighed the risks.

Although the participants overall tended to have a positive attitude towards smart city development, more types of risk than benefit were identified in the corpus, which also demonstrated substantial differences in understanding between the four groups (see Table 4). In general, privacy concerns, the safety of personal data and data misuse reflected the unease of the participants from the four focus groups, associated with the corporate-driven nature of many smart city projects that might prioritize surveillance and profit [22]. The limited participation of citizens, meaning that they have no control over how their personal data are used, exacerbates privacy risks [21]. The Hong Kong-centric private sector group (FG4) further noted risks of data mishandling by the government, the digital divide (excluding some citizens, especially the elderly, from digital public services) and the long-term consequences of non-human decisions being made by digital systems. In fact, the digital divide could also be understood in terms of the deepened inequalities caused by the outsourcing of smart city projects to private technology firms [22,52].

Classified as the group of young Chinese university students who had recently relocated to Hong Kong, FG1 illustrated well the variety of perspectives, understanding and support of the focus group participants regarding Hong Kong according to their place of origin and occupation (see Figure 3). Most FG1 participants compared Hong Kong unfavorably with Mainland Chinese cities in terms of digital services. They argued that the underdevelopment in accommodating electronic payments and transactions, which are widely utilized in Mainland China (e.g., Alipay and WeChat Pay), is the main reason that various cities in Mainland China, such as Beijing, Guangzhou, Hangzhou and Shanghai, appear to be ‘smarter’ than Hong Kong. Not surprisingly, considering their recent relocation to Hong Kong, the participants of FG1 were not aware of Hong Kong’s Smart City Blueprint, let alone the Smart City Blueprint 2.0. FG2, a mixed group of young Chinese and Hong Kong postgraduate students, was more balanced. Convenience was mainly discussed as the main benefit of digitalization. In fact, FG2, in comparison, raised the most benefits among all groups (see Table 4), logically implying the synergy of the mixed perspectives from a mixed-origin group. Several participants in FG2 (1, 4, 5) extended their observations that Hong Kong is a pioneer in neither e-payment nor the e-transportation system. Participant 9 primarily blamed the Hong Kong citizens’ concerns about privacy for the backwardness of digital Hong Kong when compared to Mainland China.

Participants in FG3 and FG4 were more concerned with questions of data privacy and digital trust. There were also differences according to the type of respondent. From the public administration group (FG3), issues were raised regarding the targeting of citizens by private companies, while the private sector group (FG4) expressed the view that data practices in multi-national firms were superior to those in the public sphere. For FG3, the privacy-related concern could be elaborated by the word frequency analysis in Figure 3, which shows that ‘trust’ and ‘data’ both stood out from the discussion. This observation not only substantiates the main direction of smart city development in Hong Kong, centered on big data utilization and management, but also highlights a dilemma within this group regarding trust in the government managing the data. In FG3, dominated by civil servants, most participants trusted the government but explicitly mentioned their distrust in the data management capacities of the outsourcing companies that managed the mobile applications ‘LeaveHomeSafe’ and ‘iAMSmart’. In FG4, in contrast, the discussion was centered around trust in the government. Participant 1 explicitly brought up the unfamiliarity of the elderly with smart city tools and devices and feared that this would discourage their usage of mobile applications, even though they trust the government. The fact that both the civil servants (in our focus groups) and the elderly cohorts (in the survey) expressed trust in the government but refrained from using these mobile applications for other reasons substantiates the existence of a form of ‘conditional trust’. Other participants in FG4 (5, 7, 10) indicated their equal measure of distrust in government and in private companies, because, in both cases, there was a lack of control or consent regarding how their data would be used by these parties. There was, however, the need for a trade-off between convenience and privacy. On balance, the opportunities provided by digital tools outweighed the dangers, as exemplified by the statements that the ‘leakage of personal information is an unavoidable risk in digitalization’ and ‘privacy is the trade-off of convenience’.

Significant differences in perception were also observed among the four groups regarding trust in providers and data processes and comparisons with Mainland China. In the words of one participant, people from Mainland China ‘don’t think (the government) will do something bad to our privacy’, but they are ‘concerned that the private company could steal our personal data, and thus the government needs to supervise them, or then we can’t trust them’ (FG1, participant 7). This shows the higher trust of Mainland Chinese citizens in the government than in private companies, which is exactly the opposite compared to Hong Kong citizens [29].

In summary, the consensus in the four focus groups is that the smart city is framed as a technological venture, based on the convenient delivery of public services by digital means, especially mobile applications. While the overall analysis identified core cross-cutting themes, there were also distinctive features.

5. Discussion

This paper studies focus groups as a research method across three dimensions. Firstly, focus groups provide a means to compare detailed discussions across distinct populations; secondly, focus groups enable collective deliberation to reveal hidden relationships; and, thirdly, focus groups serve as a mixed-methods approach to bridge qualitative and quantitative research and to validate findings.

For the first dimension, the case for using focus groups to elucidate shared understandings among distinct groups is clearly demonstrated. The focus group was closer to the social construction rather than the psychological perspective, seeking to elicit distinct narratives in relation to the central questions involved. Table S4, derived from Figure 3, in the Supplementary Materials presents the common themes and distinct traits of the four focus groups, as measured by the word frequency counts.

Figure 3 and Table S4 show that ‘Hong Kong’, ‘government’ and ‘information’ are the only three words that are commonly included in the top ten lists of the four focus groups. These three words crystallize the research topic: ‘Hong Kong’ and ‘government’ refer to the key player in Hong Kong’s smart city development, and ‘information’ is the key element of the smart city: information and communication technologies. The separate investigation of the word frequency analyses of the four focus groups clearly indicates the two classifications (FG1 and FG2 versus FG3 and FG4) regarding the origins of the participants, i.e., Mainland China and Hong Kong, and reveals two fundamentally different concepts towards smart city development. For Mainland Chinese citizens, convenience and safety come above all other values; thus, they are willing to have high trust in the government, believing that their data are rightfully and fairly protected and utilized by the government for their own sake. One participant in the mainland-centric focus group expressed one major difference between Hong Kong and Mainland Chinese citizens: ‘Unlike Hong Kong citizens, Chinese care about safety more than privacy’ (Participant 1, FG1).

Second, focus groups operationalize a process of collective deliberation that uncovers the hidden relationships within complex domains (in this case, the trust dynamics in the smart city). Granular findings are mediated and reinforced by the group of origin or current profession. Public trust is a decisive factor, and its demographic differentiation is especially demonstrated within the focus group, particularly regarding the origin of the participants. The underlying divergence of perspectives between Hong Kong and Mainland China respondents is the most striking finding. In terms of trustworthiness, questions of ‘trust’ in the government and service providers featured prominently in FG3 and FG4, where Hong Kong-based respondents were overweighed; they did not feature so prominently amongst the mainland participants. One participant in the Hong Kong group was of the view that ‘the biggest barrier for Hong Kong smart city development is the low public trust in the government’ (Participant 8, FG3). When pushed, in relation to trust in providers, the Mainland China respondents in FG1 and FG2 would always trust their government. They would trust local companies ‘if the government announces they are trustworthy’ (Participant 1, FG1), but foreign companies ‘should always be supervised by the government’ (Participant 7, FG1). Such trust in the central government went far, as the ‘Chinese government controlling the social media protects the citizens from fake news’ (Participant 9, FG2). In direct contrast, Hong Kong respondents were less likely to express blind faith in either the government or private firms.

Third, viewing this analysis from a broader sense, focus groups can assist in validation, defined by Hennink [37] as the process of identifying whether there is the recurrence of concepts and a consistent meaning in the data.

Table 5. Mixed-method validation.

Smart City Element	Survey	Interviews	Focus Groups	Validation
Understandings of smart city
Smart as technology	1	1	1	✓
Smart as sustainability	2	2	2	✓
Smart government	3	3	3	✓
Central findings of the smart city research project
Data trust paradox	✓	✓	x/✓	x/✓
Smart city as an empty signifier (difficult to understand what it intends to express)	✓	✓	x	x
Social impact of trust and the smart city (filtered by generation, locality, etc.)	✓	✓	✓	✓

✓ = Supported; x = Not supported; x/✓ = Partially supported

highlights the areas of mixed-method validation, where the focus groups strengthen the findings reported elsewhere in the survey and interviews.

The consistency and convergence across the territory-wide survey, expert interviews and focus groups provide support for focus groups as a valid and reliable research method within a mixed-methods approach. The ordinal scale (1 as the most important; 3 as the least important) presented in Table 5 represents the order of importance of the understandings of the smart city according to the type of narrative: in order, technology, environment and public administration. Even more firmly than in the survey or the interviews, the smart city was primarily related to the application of technology for urban-based public services. The central findings from the survey and interviews were expressed in terms of the data trust paradox, defined as high support for technology from the public in a low-trust environment; privacy concerns related to data management; the explicitly higher support of the government and subsequently the smart city development in Hong Kong from those born in Mainland China and, in contrast, the lower support from those younger generations who politically identified themselves as localists. The focus groups essentially confirmed and sharpened these distinctions, especially in relation to birth origins and political positions.

Despite the advantages of focus groups, several methodological limitations should be acknowledged. Adler et al. [42] question whether focus group participants can truly provide sincere answers due to the potential effects of social desirability, such as peer conformity and unwillingness to disagree with another participant. Additionally, the effectiveness of focus groups can be compromised by group dynamics, particularly the role of dominant individuals who may dominate the discussions, limiting contributions from other participants [38,40]. In larger groups, not all participants may actively take part in the discussions due to intimidation or the influence of dominant or aggressive participants [25,36]. Furthermore, the interpretative approach and subjectivity of this research method result in questionable objectivity and uncertain reliability [36,37]. The risk of sample bias also exists, as participant selection may lead to outcomes that are highly context-specific and not easily generalizable [36]. While we attempted to address these challenges through careful moderation, structured participant selection, and methodological triangulation, these limitations must necessarily be considered in this paper.

Another observation is the composition of the focus groups, which were dominated by females. While it was not our intention to recruit more female participants, research shows that females tend to emphasize trust and privacy in smart city discussions, due to their higher concern than males for privacy in digital contexts and caution in technology usage [53,54]. This not only aligns with the focus of this study, but we also observed the same phenomenon where female participants from the focus groups expressed comprehensive perspectives on trust and privacy issues, which provided us with valuable insights. Future research could explore whether a more gender-balanced composition would lead to different results.

6. Conclusions

This paper shows that the focus group adds value as part of a mixed-methods approach for social scientific projects to validate and substantiate the coherence of the results. The added value of the focus groups lay in their degree of granularity, the fruitfulness of interactions and the ability to present and compare distinct group perspectives (as well as converging ones across groups).

The focus group deliberations suggested major areas for improvement in the implementation of smart city initiatives in Hong Kong, to a greater extent than the results of the survey and interviews. In addition to emphasizing the critical role of trust in shaping public perceptions of smart city initiatives, the focus group results reconfirm the existence of the data trust paradox, i.e., that the high support for technology coexists with skepticism toward government- or corporate-driven smart city projects (in the case of Hong Kong). Although the participants were well aware of the promising opportunities for efficiency and innovation brought by technological advancements, they called for greater transparency and communication about smart city initiatives, urging for a more holistic and inclusive approach towards the smart city. The findings highlight the importance of transparency and citizen participation in regaining citizens’ trust in the government’s handling of smart city development. Improved communication by the HKSAR government might help to address public skepticism.

Beyond Hong Kong, these findings have broader global policy implications. The observed challenges, including distrust in government institutions and differing trust among demographic groups, are not unique to Hong Kong but also imply governance concerns in smart city development worldwide, supporting trust as a critical concept and key challenge in most cities’ smart city initiatives [55]. Singapore, a global city with an urban context similar to that of Hong Kong, empirically shows that citizens who feel that their concerns are heard in policymaking and who desire more public participation show higher trust in the government [56]. This supports the necessity of citizen engagement as part of the trust-building mechanism. While further studies are required to explore the trust framework in global smart city strategies, building public trust in not only technologies but also governing bodies is a fundamental governance imperative that determines long-term success.