Research on Users’ Willingness to Use the Urban Subway Wayfinding Signage System Based on the DeLone & McLean Model Theory: A Case Study of Wuxi Subway

Abstract

This study, which is grounded in the information systems success model (D&M model) proposed by Delone and Mclean, investigates user satisfaction and the intention to continue using the urban subway signage system; Wuxi subway is taken as a case study. Through a questionnaire survey, the research integrates elements from the D&M theory, such as information quality, system quality, service quality, user satisfaction, and intention to use. The data were collected using a combination of online and offline survey methods and were analyzed using IBM SPSS 23 and AMOS 23. The findings reveal that system quality serves as a crucial driver of user satisfaction, while service quality exerts the most significant influence on users’ intention to use. Additionally, information quality is equally important. The optimization suggestions encompass the need to ensure the relevance, completeness, timeliness, and accuracy of information; enhance the usability and reliability of the system; and bolster the responsiveness, empathy, and tangibles of the service. These discoveries provide a scientific basis and practical guidance for optimizing the signage systems of Wuxi subway and other urban public transportation systems, with the aim of elevating passengers’ intention to use and their satisfaction.

1. Introduction

Since the acceleration of industrialization and urbanization, cities have expanded significantly, populations have grown rapidly, and public transportation systems have become increasingly complex [1]. Without signage, especially systematic signage, many people would struggle to find their way and could even become lost in the city. To address this prominent issue in modern urban life, wayfinding systems were developed. These systems, comprising graphics, text, and symbols, are a critical aspect of modern graphic design [2]. Wayfinding refers to the process of determining and following a path or route between a starting point and a destination. It is a purposeful, guided, and motivated activity that can be observed as the traces of sensory and motor actions through an environment. Wayfinding design is based on human spatial cognition, helping people navigate from one location to another while maintaining a clear understanding of how to return to their starting point [3,4].

Urban subway wayfinding signage systems play a crucial role in modern urban rail transit. With the rapid development of subways in medium and large cities in economically advanced regions of China, subways have become one of the primary modes of transportation in these cities [5,6]. The design and optimization of subway wayfinding signage systems directly impact passengers’ travel experiences and are critical to the operational efficiency of subways. Researching urban subway wayfinding signage systems contributes to the improvement of passenger travel efficiency, the optimization of subway operations, and the enhancement of a city’s image. An excellent subway wayfinding signage system provides clear and accurate guidance, helping passengers to quickly understand their current location, choose the correct travel direction, and navigate transfer routes. This significantly reduces the time passengers spend searching for paths within stations and improves travel efficiency [7]. Through scientifically designed wayfinding signage systems, passenger flow can be better organized, reducing idle time within stations and increasing the transport capacity and operational efficiency of subways [8]. Furthermore, subway wayfinding signage systems are not merely functional facilities; they are also vital components of a city’s image. Well-designed, aesthetically pleasing wayfinding signage systems can elevate the overall image and status of a city, integrating into its cultural identity [9].

Over the past century, various design means have been employed to help people establish a clear spatial structure and to guide them through spaces. This approach has been applied frequently in architectural and landscape designs; however, it was only in the 1960s that it began to be called “wayfinding”, in studies spanning multiple research fields and related disciplines. An urban planner, Kevin A. Lynch, first introduced the term “wayfinding” in his 1960 book “Image of the City”, defining it as “a consistent use and organization of definite sensory cues from the external environment” designed for the external environment [10]. He summarized the five essential elements used by people to describe spaces: path, edge, node, landmark, and district. These five elements have subsequently become the main terminology in spatial research.

In the 1970s and 1980s, researchers began to focus on how people navigate and remember spaces in large and complex environments. A growing number of studies have revealed that human wayfinding behavior actually encompasses a series of multilayered actions, which are consistently influenced by environmental factors, such as spatial symmetry, user expectations, language, signage, and memory, among others. In 1984, the environmental psychologist Romedi Passini published his monograph titled “Wayfinding in Architecture” [11]. And in 1992, he co-authored “Wayfinding People, Signs, and Architecture” [12] with Paul Arthur. It clearly explains the human cognitive approach to space and explicitly proposes that “wayfinding” is a dynamic problem-solving behavior in spatial contexts, divisible into two steps: making a decision and implementing the plan. When individuals find themselves in unfamiliar spaces, they first need to determine their location, understand the spatial layout, and identify the destination, thereby formulating an action plan. For the first time, the design concept, which was previously confined solely to signage, was expanded and promoted to include graphic communication. Furthermore, it is suggested that wayfinding signage system design encompasses architectural space grammar, logical space planning, visual communication signage, audio-based communication systems, and map systems, as well as wayfinding systems tailored for individuals with special needs. It is evident that from this point on the design of modern wayfinding signage systems truly established a relatively stable development direction and system.

After entering the 21st century, scholars began conducting further research on wayfinding signage system design. Xiang Fan’s “Wayfinding Signage System Design” [2] and Craig M. Berger’s “Wayfinding: Designing and Implementing Graphic Navigational Systems” [7] are books that systematically organize the content of wayfinding signage, signage spatial composition, ontology structure, and signage interface design and comprehensively construct the knowledge system of wayfinding signage system design. In the past decade, scholars have begun to explore wayfinding behavior from various perspectives [13,14,15]. These studies have provided important references for optimizing the design of wayfinding signage systems.

The essential purpose of wayfinding system design is to integrate and organize information related to the spatial environment, thereby enabling people to reach their destinations quickly through information system design. A well-designed wayfinding system can enhance the efficiency of space utilization and fulfill or improve the fundamental functions of the space. Therefore, wayfinding system design is not merely about signage design. Many people misunderstand the wayfinding system as a series of signboards, but in reality, it is a comprehensive spatial information system. Signboards serve as a medium within the wayfinding system, which encompasses multiple integrated spatial information systems. Beyond signboards, maps, and graphic elements, the most fundamental component is the spatial information architecture system, including the spatial naming system, coding system, audio system, and more. Without a clear wayfinding system plan, signage alone cannot create a spatially coherent and clearly defined route.

Overall, the existing research primarily focuses on urban subway wayfinding signage itself, with limited systematic studies conducted from the perspective of user perception. To gain a more comprehensive understanding of this system, this paper innovatively employs a user perception model to investigate users’ willingness to use urban subway wayfinding signage systems.

Scholars have suggested utilizing different user perception models to elaborate on the factors influencing the success of information systems, such as the information systems success model (ISSM), also known as the D&M model. The ISSM, proposed by Delone and Mclean, has been widely adopted in the fields of information management and information systems [16]. The D&M model posits that users evaluate and measure the quality of a system based on their experience and perception of it, thereby determining its success. Subsequently, Delone and McLean revised the original D&M model, which resulted in a model that includes six variables: information quality, system quality, service quality, user satisfaction, usage intention, and net benefits. Among these, information quality, system quality, and service quality have an impact on user satisfaction and usage intention, which in turn influence the net benefits of the entire information system [17,18,19].

The Wuxi Metro is an urban rail transit system serving Wuxi City in Jiangsu Province. The first line, Wuxi Metro Line 1, opened on 1 July 2014, making Wuxi the 22nd city in China to have a metro system [20]. Currently, the Wuxi Metro operates five lines: Line 1, Line 2, Phase 1 of Line 3, Phase 1 of Line 4, and Line S1; these lines cover all five districts of the city and Jiangyin City. The total operational length is 143.93 km, with 91 stations. Additionally, Phase 2 of Line 4, Line 5, Line 6, and Line S2 are under construction, with a total length of approximately 120.5 km. In terms of passenger traffic [21], in 2016 the annual ridership of the Wuxi Metro was 81.468 million, with an average daily ridership of 223,200. In 2015, the ridership was 71 million, with an average daily ridership of 195,000. In the second half of 2014, the ridership was 17.7025 million, with an average daily ridership of 97,000.

This study selects the urban subway wayfinding signage system in Wuxi, China, as the subject of analysis due to its representativeness, innovation, and practicality. Firstly, as one of China’s economically developed cities, Wuxi has witnessed rapid development of its subway system, which plays a crucial role in urban transportation. The design and optimization of Wuxi’s subway wayfinding system not only reflect the city’s high standards for public transportation facilities but also serve as a representative example. Secondly, Wuxi’s subway has undertaken several innovative attempts in the design and optimization of its wayfinding system, such as by adopting a format of “enlarged layout, striking background colors, and combination of text and graphics”, as well as by utilizing intelligent guidance systems. These innovative measures provide valuable insights and references for other cities. The Wuxi subway wayfinding system emphasizes practicality and functionality, offering convenient and efficient travel guidance to passengers through scientific and reasonable layout and design. This passenger-oriented design concept holds significant research value.

The Wuxi Metro wayfinding signage system consists of various elements, including platform screen door panels, small station signs, linear route maps, platform directional signs, and external wayfinding signs, as well as electronic screens inside the metro platforms and carriages. These signs not only provide passengers with clear navigational information but also emphasize the integration of functionality and aesthetics in their design (Figure 1).

The platform screen door panels combine directional guidance, station names, and route map information, with a colored strip at the bottom and appropriate blank space to ensure simplicity and readability. The station name signs feature wider spacing between characters to highlight the current station, with the station name filled in the corresponding line color. They also clearly display information about the previous and next stations, allowing passengers to quickly identify their location. The route map distinguishes upcoming stations from those already passed through color coding and by using colored arrows for future sections and gray lines for completed sections, with station names also turning gray to enhance visual contrast. The transfer markings are relatively complex, utilizing a combination of nested arrows and directional colors to clearly indicate the current line and transfer options. The exit signs are arranged vertically by line number to further strengthen the guidance function.

The design of the small station signs aligns with the route map, with the main difference being the position of the colored strip, making them more distinctive. The linear route map, made of plastic material, is attached to the platform columns, with the graphic presented as a rounded rectangle. Each side represents different directions on the island platform. The design style is consistent with the platform screen door panels, reinforcing the sense of direction through continuous upward arrows.

The platform directional signs have more diverse combination designs, including station signs, route diagrams, and linear route maps. The external wayfinding signs include exit plaques and columns, featuring a light gray background with a metallic texture. The information is arranged from left to right and includes the universal metro symbol, Wuxi Metro logo, station name, station number, and exit number, ensuring that passengers can quickly obtain the necessary information.

In addition, electronic screens inside the metro platforms and carriages play a significant role. By dynamically displaying information, these screens update the metro station details in real time, enabling passengers to stay informed about the train’s status and station locations at any moment. Especially during emergencies, these screens can quickly convey urgent information, enhancing the passenger experience during transit.

Overall, the Wuxi Metro wayfinding signage system balances functionality with user experience. Through the thoughtful use of color, layout, and information hierarchy, passengers can quickly and accurately access the necessary navigational information, improving the efficiency and convenience of metro navigation. This systematic signage design not only meets the high-efficiency demands of modern urban transportation but also strikes a balance between visual appeal and practicality, enhancing the overall passenger experience.

Next, this paper will draw upon the D&M theory to explore user satisfaction and usage intention towards urban subway wayfinding signage systems from the perspective of user perception, aiming to provide valuable insights and directions for the development of urban subway wayfinding signage systems.

2. Theoretical Background and Research Hypotheses

Information systems (IS) play a crucial role in modern society, with applications across business management, public services, and personal life. The use of information technology is pervasive. How to evaluate the success and effectiveness of information systems has been one of the core issues in the field of IS research. The information systems success model (DeLone & McLean Information Systems Success Model, D&M model), proposed by DeLone and McLean in 1992, has become an important theoretical framework for assessing the success of information systems in this field. This model, based on a multidimensional perspective, identifies several key factors for measuring IS success and explains the interrelationships between these factors.

The original IS success model introduced by DeLone and McLean in 1992 focuses on six key components to measure the success of information systems. These factors include system quality, information quality, service quality, use, user satisfaction, and net benefits [16]. DeLone and McLean developed a causal relationship model, suggesting that system quality and information quality directly influence user satisfaction and usage, which, in turn, affect the net benefits of the information system.

With the rapid development of information technology and the changing needs of research, DeLone and McLean revised and extended their original model in 2003. The extended model not only retained the original six components but also introduced new variables such as intention to use, individual impact, and organizational impact [18]. The extended D&M model emphasizes the multidimensional benefits of information systems and highlights the long-term effects and potential value of IS. Through this expanded version, DeLone and McLean further demonstrated that the success of information systems is not only an immediate issue of user satisfaction but also involves long-term benefits at the organizational level and the degree of integration of the system in organizational operations [22].

This study is based on the D&M model and integrates the Wuxi Metro wayfinding signage system to construct research hypotheses from multiple factors, including information quality, system quality, and service quality. Through a systematic analysis of these key factors, the study aims to explore their impact on user satisfaction and intention to use. The findings will provide theoretical support and practical recommendations for optimizing and improving the Wuxi Metro wayfinding signage system.

2.1. Information Quality

Information quality refers to the output quality of an information system, reflecting the semantic success of the system [19,23]. Balog defines information quality in four dimensions: relevance, completeness, timeliness, and accuracy [24].

Relevance: Relevance refers to the degree of direct connection between information and the user’s needs, goals, or tasks. High-quality information should be able to directly answer the user’s questions, satisfy their needs, or support their decision-making process. When screening and providing information, it is crucial to ensure that the information is closely related to the user’s current or future needs. The issue of relevance in the field of information retrieval has been explored, and it has been suggested that the concept of relevance has not yet been fully understood. Cosijn, E., & Ingwersen, P. have constructed a systematic model of relevance attributes, demonstrating the role of different manifestations of relevance across various dimensions. They further propose a comprehensive model of relevance manifestations that incorporates socio-cognitive relevance [25]. Tanner, J., & Itti, L. introduced the theoretical concept of “goal relevance”, which is used to quantify the informational value of observed data for achieving a goal. This definition is universal, quantifiable, and explicit and enables the numerical representation of previously elusive correlations between observations and goals [26].

Completeness: Completeness requires that information be comprehensive and sufficient in content, with no omissions or missing critical details. This implies that information should encompass all the necessary components to enable users to fully understand and make accurate decisions. Information lacking completeness may lead to misunderstandings or incorrect judgments. Ballou, D.P., & Pazer, H.L. have examined the trade-offs between completeness and consistency faced by decision makers in an information-based decision-making environment. They propose a framework for systematically exploring the trade-offs between data completeness and consistency, taking into account the relative weights assigned by decision makers to both completeness and consistency [27]. Naumann, F., Freytag, J.C., & Leser, U. have delved into the issue of multisource data integration in information systems, exploring how to ensure the completeness of information and maintain its integrity within these systems [28]. Bardaki, C., Kourouthanassis, P., & Pramatari, K. have proposed an analytical model for assessing the information completeness of a system based on its size, scope, and depth [29].

Timeliness: Timeliness emphasizes the relationship between the speed of information provision and its value. In today’s rapidly evolving society, the timeliness of information is particularly crucial. Outdated information may no longer hold referential value and can even mislead users. Consequently, high-quality information should be provided promptly when needed to ensure its effectiveness. Ballou, D. P., Wang, R., Pazer, H. L., & Tayi, G. K. have explored the relationships between data quality, timely information, and information systems, proposing a series of ideas, concepts, models, and procedures that are applicable to information manufacturing systems. These aim to determine the quality of the information products delivered or transferred to information consumers [30]. Sampaio, S. D. M., Dong, C., & Sampaio, P. R. F. focus on data quality issues, particularly the timeliness of data, in internet query systems (IQS). They specifically introduce the quality model adopted, the design of the data source layer, and the algebraic query processing framework for quality awareness [31]. Edwards et al. proposed a quality of information (QoI) aware network framework aimed at quantifying the usefulness of information for specific applications. The QoI value is jointly determined by the intrinsic and contextual attributes of information. The former includes the freshness and accuracy of information, while the latter encompasses completeness and timeliness [32].

Accuracy: Accuracy is one of the most fundamental dimensions in assessing information quality. It requires that information be truthful and reliable in content and free from errors or misleading content. Accurate information serves as the foundation for making correct decisions. During the process of information gathering and transmission, appropriate measures must be taken to verify the authenticity of information, in order to avoid misleading users or causing negative consequences. Letzring, T. D., Wells, S. A., & Funder, D. C. have studied the effects of information quantity and quality on judgmental accuracy [33]. Ge, M., & Helfert, M. have explored the impact of information quality on decision making, adopting a multidimensional perspective of information quality to study the effects of information accuracy, completeness, and consistency on decision making. Their results indicate that information accuracy and completeness have significant effects on the quality of decision making, while the impact of information consistency on decision-making quality is not significant. However, information consistency may enhance the role of accuracy, suggesting that both information accuracy and consistency jointly influence the quality of decision making [34].

In urban subway wayfinding signage systems, high-quality information output implies that the signage information is highly relevant, complete, timely, and accurate and provides effective navigation and assistance to users. When the information quality of an urban subway wayfinding signage system is high, users are able to more easily access and understand the required information, thereby increasing their willingness to use the system. Furthermore, users can obtain the necessary information more accurately, making it easier for them to achieve their travel goals. Based on this, the following hypotheses are proposed:

H1: 

The information quality of urban subway wayfinding signage systems positively influences the willingness to use.

H2: 

The information quality of urban subway wayfinding signage systems positively influences user satisfaction.

2.2. System Quality

System quality refers to the performance or desired characteristics of an information system. It serves as the cornerstone of an information system, directly impacting the user experience and perception [16]. When evaluating system quality, multiple indicators, such as ease of use and reliability, are typically considered [19].

Usability: Usability refers to the degree of convenience and satisfaction users experience when using a system. A highly usable system can reduce user learning costs, improve operational efficiency, and enhance the overall user experience [35]. Zhou’s research indicates that system quality primarily influences perceived ease of use, while information quality primarily impacts perceived usefulness [36]. Furthermore, service quality has a significant impact on trust and perceived ease of use. Perceived usefulness, perceived ease of use, and trust jointly determine user satisfaction [37].

Reliability: Reliability refers to the ability of a system to perform its specified functions under specified conditions and within a specified time period. Chen and his team have explored reliability assessment methods for polymorphic manufacturing systems with quality–reliability dependencies [38].

A high-quality information system can reduce the frustrations and obstacles users encounter during the usage process, thereby enhancing their willingness to use it [39]. Moreover, as the cornerstone of an information system, system quality has a direct impact on user satisfaction. A high-quality information system can provide a better user experience, thereby increasing user satisfaction [40]. The urban subway wayfinding signage system is a specific application of an information system; thus, its quality also affects users’ willingness to use the system and their satisfaction. Based on this, the following hypotheses are proposed:

H3: 

The quality of urban subway wayfinding signage systems positively influences the willingness to use.

H4: 

The quality of urban subway wayfinding signage systems positively influences user satisfaction.

2.3. Service Quality

Service quality refers to the degree of discrepancy between users’ normative expectations of a service and their perceived service performance. Service quality is an important factor in evaluating the level of an information system [19]. Service quality is a multidimensional and comprehensive concept that is influenced by a combination of factors. The main influencing factors include responsiveness, empathy, and tangibility, among others [41].

Responsiveness: Responsiveness refers to the speed and manner in which a service system responds to a customer’s request or issue. Yusefi and his team have explored the relationship between response level and service quality, and they believe that there is a significant positive correlation between the two [42].

Empathy: Empathy refers to the ability of a service system to put itself in the user’s shoes, to understand, and to fulfill their individual needs and expectations. This factor plays a crucial role in enhancing service quality. Zhang and his team have investigated the impact of caring and personalized attention (i.e., empathy) within service quality on loyalty. Their research found that service quality can strengthen the perception of empathy, which in turn enhances loyalty [43].

Cadet and others argue that empathy is one of the five main dimensions of service quality and is a necessary condition for a successful service experience. They have explored the impact of service providers’ empathy on service quality and found a positive correlation between the two [44].

Tangibles: Tangibles refer to the quality of the physical environment, equipment, and other factors that customers encounter during their use of a service system. Although these factors do not directly constitute the service itself, they significantly impact the perception of service quality. Through analysis and empirical evidence, Park and others have shown that the two dimensions of tangibles and non-tangibles are more stable and representative of the psychometric dimensions of service quality than the five dimensions of SERVQUAL [45].

In a system, the level of service quality directly affects the user’s experience and perception [46]. When the service quality provided by the urban subway wayfinding signage system is high, meaning that there is a high degree of match between the user’s expectations and the actual service experience, users are more likely to use the system. Furthermore, service quality, as a crucial factor in measuring service levels, has a direct impact on user satisfaction [47].

When the service quality provided by the urban subway wayfinding signage system is high, meaning that it can meet or exceed users’ expectations, user satisfaction will increase accordingly. Based on this, the following hypotheses are proposed:

H5: 

The service quality of urban subway wayfinding signage systems positively influences the willingness to use.

H6: 

The service quality of urban subway wayfinding signage systems positively influences user satisfaction.

2.4. User Satisfaction

User satisfaction is a crucial factor influencing users’ willingness to use information systems [19,48]. When users are satisfied with a system, meaning that the system is able to meet or exceed their expectations and needs, their willingness to use the system becomes stronger [49]. Conversely, if users are dissatisfied with the system, whether due to unmet expectations or other issues, their willingness to continue using it will decrease, prompting them to seek alternative solutions [50]. Based on this, the following hypothesis is proposed:

H7: 

User satisfaction with the urban subway guidance signage system positively affects the willingness to use.

2.5. Intention to Use

Intention to use refers to the willingness of users to continue using an information system after their initial experience with it. It is one of the crucial indicators for evaluating the success of an information system [19,51]. Delone and McLean employed this concept in their model, which is similar to the empathy effect proposed by Kettinger et al. in their discussion of information systems. Specifically, whether users have the intention to continue using the information system after adopting it serves as a significant metric for assessing the system’s success [52,53].

In summary, this study constructs a model to explain the influencing factors of passengers’ intention to use the Wuxi urban subway guidance signage system, as shown in Figure 2.

3. Methodology

3.1. Research Framework

The research framework primarily comprises three stages: questionnaire design, data collection, and data analysis.

In the first stage, the questionnaire was meticulously designed based on the D&M theory, incorporating elements such as information quality, system quality, service quality, user satisfaction, and intention to use. The questionnaire consists of multiple sections, each tightly focused on a specific element, to facilitate the collection of information on users’ willingness to use the Wuxi Metro guidance signage system. Prior to its official release, a pilot test was conducted, and the questionnaire was adjusted based on participant feedback to ensure the reliability of its content.

In the second stage, this study collected survey data through both online and offline methods. The survey subjects were limited to people who had used the Wuxi Metro. For the online portion, the questionnaire was distributed via social media platforms and digital communities, including WeChat, TikTok (Douyin), and Xiaohongshu, to reach a wide range of user groups. Offline data collection involved posting survey posters at metro stations and their surrounding areas, key urban commercial districts, residential communities, and school campuses to attract participants. To ensure the validity and broad coverage of the data, the Tencent Survey platform was used to distribute the questionnaire, and respondents provided feedback based on their personal experiences with the Wuxi Metro wayfinding signage system.

In the third stage, the study conducted a comprehensive data analysis using IBM SPSS 23 and AMOS 23 to ensure the accuracy and scientific rigor of the results. Through an in-depth exploration of the data and a discussion of the findings, the study ultimately arrived at targeted conclusions and recommendations.

3.2. Questionnaire Design

Taking into account the influencing factors of passengers’ intention to use the Wuxi urban metro guidance signage system, this study designed a multidimensional questionnaire by setting and optimizing measurement items based on relevant research findings. The questionnaire covers various dimensions, such as information quality, system quality, service quality, user satisfaction, intention to use, and the respondents’ basic information. After completing the initial draft of the questionnaire, a pre-test was conducted, and feedback was sought from three experts in related fields. Based on their feedback, explanations of relevant concepts were added, and the language expression was improved and revised. The following table lists the research variables and measurement items for the study on the intention to use the urban metro guidance signage system (

Table 1. Research variables and measurement items for the study on intention to use the urban metro guidance signage system.

Second-Order Facet	First-Order Facet	Question Item	Question Item Content	Question Item Source
Information Quality	Relevance	REE1	I think the guide sign system of Wuxi urban subway provides useful navigation information.	Sun Shaowei [41]; Balong [24]; Zhou [36]
		REE2	I think the information provided by the guide sign system of Wuxi urban subway is relevant to my travel needs.
		REE3	I think the information provided by the guide sign system of Wuxi urban subway is helpful for the wayfinding process.
	Completeness	COM1	I think the guide sign system of Wuxi urban subway provides complete station information.	Sun Shaowei [41]; Balong [24]; Zhou [36]
		COM2	I think the guide sign system of Wuxi urban subway provides complete transfer information.
		COM3	I think the guide sign system of Wuxi urban subway provides complete information about facilities.
	Timeliness	TIM1	I think the operational status information of the guide sign system of Wuxi urban subway is updated in a timely manner.	Sun Shaowei [41]; Balong [24]; Zhou [36]
		TIM2	I think the guide sign system of Wuxi urban subway is capable of displaying real-time information to passengers in a dynamic manner. (Can the electronic screens within subway platforms and carriages dynamically display real-time information to passengers?)
		TIM3	I think that in special circumstances, the guide sign system of Wuxi urban subway is able to promptly convey emergency information through multiple channels.
	Accuracy	ACU1	I think the information provided by the guide sign system of Wuxi subway is accurate and error-free.	Sun Shaowei [41]; Balong [24]; Zhou [36]
		ACU2	I think the information provided by the guide sign system of Wuxi subway is unambiguous and does not lead to confusion or misdirection.
		ACU3	I think the layout and positioning of the guide sign system of Wuxi subway are scientific and reasonable.
System Quality	Usability	UAB1	I think the information provided by the guide sign system of Wuxi subway is simple and easy to understand.	YANG Yi-weng [54]; ZHANG Xing [55]; Barnes [56];
		UAB2	I think the symbols used in the guide sign system of Wuxi subway are clear and intuitive.
		UAB3	I think the language used in the guide sign system of Wuxi subway is simple and easy to comprehend.
	Reliability	REI1	I think the guide sign system of Wuxi subway is reliable.	YANG Yi-weng [54]; Negash [57]
		REI2	I think the information provided by the guide sign system of Wuxi subway is reliable.
		REI3	I believe the guide sign system of Wuxi subway maintains long-term stability and reliability in its operation.
Service Quality	Responsiveness	RES1	I think the guide sign system of Wuxi subway is able to provide an instant response to passengers’ needs.	YANG Yi-weng [54]; Negash [57]
		RES2	I believe the guide sign system of Wuxi subway can assist passengers in handling unexpected situations.
		RES3	I think the guide sign system of Wuxi subway can help solve issues related to wayfinding and navigation.
	Empathy	EMP1	I believe the guide sign system of Wuxi subway is capable of meeting the special needs of passengers with special requirements. (Can the Wuxi subway wayfinding system enable people with special needs, such as those with visual or hearing impairments, to easily and conveniently access information?)	YANG Yi-weng [54]; Negash [57]
		EMP2	I think the guide sign system of Wuxi subway can cater to the personalized needs of passengers. (Can the Wuxi subway wayfinding system provide personalized navigation information based on different passengers’ travel purposes, preferences, and habits?)
		EMP3	I believe the guide sign system of Wuxi subway can fulfill the emotional needs of passengers. (Can the Wuxi subway wayfinding system alleviate negative emotions, such as unfamiliarity, anxiety, or urgency, and enhance users’ sense of trust and security?)
	Tangibles	TAN1	I think the visibility and clarity of the guide signs in Wuxi subway are able to meet the needs of passengers.	YANG Yi-weng [54]; Negash [57]
		TAN2	I believe the guide sign system of Wuxi subway exhibits consistency and standardization. (For example, unity in design style, uniformity in sign specifications, and adherence to international standards, among others.)
		TAN3	I think the guide sign system of Wuxi subway aligns with the aesthetic preferences of passengers.
User Satisfaction		US	I am satisfied with the guide sign system of Wuxi subway.	Sun Shaowei [41]; Bhattacherjee [49]
			I think the experience of using the guide sign system of Wuxi subway is good.
			I feel at ease when using the guide sign system of Wuxi subway.
Intention to Use		ITU	I consider the guide sign system of Wuxi subway to be the first choice for navigation while taking the subway.	Sun Shaowei [41]; Bhattacherjee [49]
			I believe I will continue to use the guide sign system of Wuxi subway.
			I think I would recommend the guide sign system of Wuxi subway to others.

).

This study aims to explore the usage intention of the Wuxi Metro wayfinding signage system, with the questionnaire designed around several key dimensions to comprehensively understand users’ experiences and needs. In terms of information quality, we focus on relevance, completeness, timeliness, and accuracy. Through a series of questions, we seek to understand whether users find the information provided by the system helpful for their travel needs and how the timeliness and accuracy of this information affect their navigation experience.

The evaluation of system quality centers mainly on usability and reliability. We ask users about the ease of understanding the information and the system’s stability over time to assess whether it is user-friendly and effectively supports them. Service quality covers responsiveness, empathy, and tangibles. We are concerned with the system’s ability to meet users’ needs, handle unexpected situations, and provide personalized services, especially regarding its adaptability for groups with special needs, which is crucial for improving the overall user experience. We assess the visibility and design consistency of the signage system to determine whether it visually attracts users and meets passengers’ aesthetic preferences. By investigating the users’ satisfaction, we aim to understand their overall evaluation of the system, as well as the pleasure and convenience experienced during use. Additionally, we explore the users’ intention to use the system, evaluating their attitudes toward continued use and recommending the system to others.

Overall, this questionnaire provides quantitative data support for studying the usage intention of the Wuxi Metro wayfinding signage system, and it also offers important insights for future system optimization and improvement.

4. Data Analysis

4.1. Data Collection

The questionnaire was collected from March to August 2024 using both offline and online methods, combined with the “Tencent Questionnaire” online survey platform. A total of 709 valid samples were collected during the formal survey phase. To ensure the relevance and reliability of the data, strict screening criteria were applied, excluding invalid questionnaires with short completion times or responses exhibiting clear patterns. Ultimately, 605 high-quality valid questionnaires were obtained, resulting in a valid response rate of 85.33%. Additionally, this study strictly protected the confidentiality of the respondents’ personal information and answers, and informed consent was obtained from the participants before the survey.

According to the minimum sample size calculation formula, assuming that the total number of Wuxi Metro users is nearly infinite, the required minimum sample size is 385. A total of 709 valid samples were collected during the formal survey phase, and after screening, 605 high-quality questionnaires were obtained. This sample size not only exceeds the minimum sample size of 385 but is also nearly 20 times the number of analysis items, fully meeting the needs for statistical analysis and further enhancing the reliability of the research results [58].

The survey used a combination of online and offline methods to ensure broader audience coverage, thus increasing the diversity and representativeness of the data. In today’s information-driven society, online data collection offers great convenience and efficiency, allowing users to fill out the questionnaire at any time and from any location, breaking the time and space limitations of traditional offline surveys. Therefore, online data collection became one of the key methods for gathering the survey responses. We promoted the survey through various mainstream social media platforms and digital communities, including WeChat, Douyin (TikTok), and Xiaohongshu (Little Red Book). As the largest social platform in China, WeChat has a broad user base and powerful social functions, enabling rapid dissemination of the questionnaire link through Moments, groups, and other channels, thus increasing exposure and participation rates. Douyin and Xiaohongshu, with their strong interactivity and user engagement, are effective in attracting the attention and participation of more target groups through short videos and hashtags.

To ensure the diversity and representativeness of the sample, we also designed an offline questionnaire collection plan to complement the online channels. The offline collection mainly involved posting questionnaire posters at transportation hubs and high-footfall locations (Figure 3). Specific locations included Wuxi Metro stations and the surrounding areas, the city’s core business districts, major residential communities, and university campuses. As a key urban transportation hub, the metro stations provide access to a large number of commuters, who are the main users of the Wuxi Metro wayfinding signage system. By posting posters at metro stations and surrounding areas, we were able to effectively attract this target group and gather their real feedback on the system’s use. Moreover, the city’s business districts and residential communities allowed us to reach people of various ages and professional backgrounds. Business districts attract numerous consumers for shopping, entertainment, and leisure; their diverse travel needs provide a variety of feedback on their user experiences. In densely populated residential communities, particularly in areas with a high density of commuters, the posters reached more daily commuters and residents, further enhancing the diversity of the data. University campuses provided us with the opportunity to engage with younger groups. University students, as frequent public transport users, also offer important insights into the needs and experiences related to the metro wayfinding signage system. To increase the participation rate for the offline questionnaires, the poster design focused on clarity and visual appeal. Through concise explanations and prominent QR codes, the respondents could quickly access the online questionnaire by scanning the code with their mobile phones.

By combining both online and offline collection methods, we aimed to ensure a large sample size while also guaranteeing the diversity and representativeness of the respondents. The online channel helped us quickly and broadly reach potential users, while the offline channel ensured in-depth coverage of specific target groups, particularly those who were harder to reach through online means. This dual approach maximized the response rate and ensured the comprehensiveness and authenticity of the data. This comprehensive questionnaire collection strategy not only improved accessibility but also helped us obtain rich feedback from a diverse range of respondents.

4.2. Demographic Characteristics of the Sample

Overall, the sample primarily consists of users aged 25–40 who frequently use the metro wayfinding system. Male participants (56%) slightly outnumber female participants (44%), but the gender ratio is close to balanced, which helps reduce any gender bias in the study results. In terms of age distribution, the respondents are mainly concentrated in the 25-30 age range, accounting for 38.5%, followed by the 31–40 age group, which makes up 24.1%. This age structure reflects a relatively younger demographic, with older age groups (51 and above) accounting for only 7.8%. This distribution likely mirrors the characteristics of Wuxi as a vibrant, economically active city with a large influx of people [59], where metro users are primarily young individuals who rely more frequently on and have a greater demand for the metro wayfinding system.

In terms of education level, the respondents generally have a high level of education, with the majority holding a university degree or higher. Specifically, the highest proportion of respondents (47.3%) holds a bachelor’s degree, followed by those with a vocational college education (23.8%), and those with a master’s degree or higher account for 17%. This suggests that higher-educated individuals may be more attentive to or reliant on public transportation wayfinding systems and potentially more willing to participate in surveys.

Regarding the frequency of metro wayfinding system usage, nearly 80% of users rated it 5 or higher, with 35% rating it 5, and 30.7% and 14.2% giving ratings of 6 and 7, respectively. This indicates that the respondents generally use the metro wayfinding system frequently, further demonstrating the sample group’s dependence on and familiarity with the system. The data are highly representative and provide strong support for the reliability of this study, helping to analyze the actual needs and user behavior patterns regarding the metro wayfinding system in Wuxi (

Table 2. Demographic information statistics.

Demographic Information Statistics
	Options	Frequency	Percentage
Gender	male	339	56%
	female	266	44%
Age	18–24 years old	102	16.9%
	25–30 years old	233	38.5%
	31–40 years old	146	24.1%
	41–50 years old	77	12.7%
	51–60 years old	26	4.3%
	61 years and above	21	3.5%
Education	Junior high school and below	5	0.8%
	High school/technical secondary school/technical school	67	11.1%
	College	144	23.8%
	Undergraduate	286	47.3%
	Master’s degree and above	103	17%
Frequency of using subway sign system	1 (Not often)	5	0.8%
	2	3	0.5%
	3	29	4.8%
	4 (Generally)	84	13.9%
	5	212	35%
	6	186	30.7%
	7 (Often)	86	14.2%

).

4.3. Reliability Analysis

According to

Table 3. Cronbach’s reliability analysis table.

Cronbach’s Reliability Analysis
Scale	Dimensions	Number of Items	Cronbach’s α Coefficient	Overall Cronbach’s α Coefficient
Information Quality	Relevance	3	0.841	0.9
	Completeness	3	0.895
	Timeliness	3	0.897
	Accuracy	3	0.846
System Quality	Usability	3	0.869	0.842
	Reliability	3	0.877
Quality of Service	Responsiveness	3	0.833	0.873
	Empathy	3	0.863
	Tangibles	3	0.846
User Satisfaction	User Satisfaction	3	0.918	0.918
Intention to Use	Intention to Use	3	0.912	0.912

, the Cronbach’s alpha coefficients of the main research scales in this article are 0.9, 0.842, 0.873, 0.918 and 0.912, all of which are greater than 0.8. Furthermore, the Cronbach’s alpha coefficients within each variable’s dimensions are also all greater than 0.8, indicating the good internal consistency and high reliability of the scales. Consequently, the data can be used for subsequent analysis.

4.4. Validity Analysis

4.4.1. Exploratory Factor Analysis

To conduct information condensation research using factor analysis, it is first necessary to analyze whether the research data are suitable for factor analysis. As can be seen from the above table, the KMO value is 0.920, which is greater than 0.6, satisfying the prerequisite for factor analysis, indicating that the data can be used for factor analysis research. Additionally, the data pass Bartlett’s test of sphericity (p < 0.05), suggesting that the research data are suitable for factor analysis (

Table 4. KMO and Bartlett’s test.

KMO and Bartlett’s Test
KMO value		0.920
Bartlett’s test of sphericity	Approximate Chi-Square	13,480.129
	df	528
	p-value	0.000

).

Table 5. Variance explanation rate table.

Variance Explained Table
Factor Number	Characteristic Root			Rotational Front Variance Explained			Variance Explained After Rotation
	Characteristic Root	Variance Explained%	Accumulation%	Characteristic Root	Variance Explained%	Accumulation%	Characteristic Root	Variance Explained%	Accumulation%
1	12.018	36.418	36.418	12.018	36.418	36.418	2.635	7.986	7.986
2	2.484	7.528	43.946	2.484	7.528	43.946	2.564	7.769	15.755
3	1.982	6.005	49.950	1.982	6.005	49.950	2.503	7.586	23.341
4	1.676	5.078	55.028	1.676	5.078	55.028	2.447	7.416	30.757
5	1.514	4.589	59.617	1.514	4.589	59.617	2.424	7.344	38.101
6	1.359	4.118	63.735	1.359	4.118	63.735	2.416	7.321	45.423
7	1.249	3.784	67.520	1.249	3.784	67.520	2.376	7.200	52.623
8	1.175	3.560	71.080	1.175	3.560	71.080	2.375	7.197	59.820
9	1.120	3.393	74.473	1.120	3.393	74.473	2.326	7.050	66.870
10	1.072	3.249	77.722	1.072	3.249	77.722	2.311	7.004	73.873
11	1.013	3.069	80.791	1.013	3.069	80.791	2.283	6.918	80.791
12	0.647	1.962	82.753	-	-	-	-	-	-
13	0.431	1.305	84.058	-	-	-	-	-	-
14	0.397	1.205	85.262	-	-	-	-	-	-
15	0.370	1.122	86.384	-	-	-	-	-	-
16	0.351	1.062	87.446	-	-	-	-	-	-
17	0.337	1.021	88.467	-	-	-	-	-	-
18	0.332	1.005	89.472	-	-	-	-	-	-
19	0.308	0.933	90.405	-	-	-	-	-	-
20	0.300	0.909	91.314	-	-	-	-	-	-
21	0.282	0.853	92.167	-	-	-	-	-	-
22	0.266	0.807	92.974	-	-	-	-	-	-
23	0.256	0.776	93.750	-	-	-	-	-	-
24	0.248	0.751	94.501	-	-	-	-	-	-
25	0.243	0.736	95.237	-	-	-	-	-	-
26	0.236	0.714	95.950	-	-	-	-	-	-
27	0.226	0.686	96.636	-	-	-	-	-	-
28	0.215	0.651	97.288	-	-	-	-	-	-
29	0.193	0.584	97.872	-	-	-	-	-	-
30	0.189	0.572	98.443	-	-	-	-	-	-
31	0.180	0.545	98.988	-	-	-	-	-	-
32	0.175	0.529	99.517	-	-	-	-	-	-
33	0.159	0.483	100.000	-	-	-	-	-	-

analyzes the factor extraction and the amount of information extracted by the factors. From Table 5, it can be seen that a total of 11 factors were extracted through factor analysis, all with eigenvalues greater than 1. The variance explanation rates of these 11 factors after rotation are 7.986%, 7.769%, 7.586%, 7.416%, 7.344%, 7.321%, 7.200%, 7.197%, 7.050%, 7.004% and 6.918%, respectively. The cumulative variance explanation rate after rotation is 80.791%.

The data in this study were rotated using the varimax method of maximum variance rotation to identify the correspondence between factors and research items.

Table 6. Table of factor loading coefficients after rotation.

Table of Factor Loading Coefficients After Rotation
Name	Factor Loading Coefficient											Commonality
	Factor 1	Factor 2	Factor 3	Factor 4	Factor 5	Factor 6	Factor 7	Factor 8	Factor 9	Factor 10	Factor 11
REE1	0.104	0.101	0.145	0.073	0.068	0.203	0.127	0.148	0.063	0.795	0.156	0.792
REE2	0.095	0.164	0.172	0.152	0.087	0.068	0.113	0.045	0.112	0.781	0.150	0.760
REE3	0.186	0.183	0.179	0.138	0.075	0.080	0.109	0.117	0.091	0.759	0.162	0.767
COM1	0.102	0.124	0.824	0.103	0.086	0.124	0.159	0.109	0.107	0.142	0.136	0.826
COM2	0.105	0.169	0.803	0.147	0.066	0.078	0.124	0.142	0.071	0.202	0.148	0.820
COM3	0.098	0.143	0.828	0.111	0.124	0.145	0.146	0.075	0.092	0.153	0.140	0.843
TIM1	0.127	0.831	0.111	0.141	0.073	0.075	0.103	0.116	0.092	0.133	0.160	0.825
TIM2	0.090	0.841	0.148	0.164	0.099	0.079	0.088	0.086	0.060	0.154	0.157	0.848
TIM3	0.097	0.842	0.151	0.059	0.078	0.071	0.079	0.083	0.055	0.130	0.195	0.827
ACU1	0.095	0.186	0.107	0.136	0.100	0.171	0.107	0.097	0.071	0.119	0.786	0.771
ACU2	0.104	0.217	0.136	0.171	0.088	0.082	0.121	0.093	0.128	0.181	0.756	0.763
ACU3	0.148	0.157	0.191	0.108	0.079	0.116	0.094	0.122	0.059	0.180	0.783	0.787
UAB1	0.103	0.179	0.153	0.801	0.086	0.106	0.147	0.032	0.053	0.101	0.145	0.782
UAB2	0.127	0.102	0.086	0.799	0.084	0.252	0.117	0.103	0.125	0.125	0.102	0.808
UAB3	0.115	0.099	0.117	0.795	0.071	0.230	0.135	0.130	0.104	0.134	0.157	0.816
REI1	0.154	0.114	0.134	0.094	0.060	0.834	0.116	0.093	0.104	0.110	0.123	0.822
REI2	0.094	0.083	0.134	0.270	0.161	0.793	0.109	0.070	0.105	0.125	0.131	0.822
REI3	0.198	0.038	0.086	0.270	0.139	0.771	0.187	0.091	0.095	0.122	0.123	0.817
RES1	0.043	0.000	0.073	0.104	0.215	0.084	0.117	0.236	0.783	0.050	0.039	0.758
RES2	0.155	0.097	0.091	0.109	0.192	0.048	0.083	0.063	0.805	0.082	0.090	0.767
RES3	0.119	0.099	0.083	0.046	0.120	0.139	0.105	0.107	0.820	0.110	0.098	0.784
EMP1	0.131	0.080	0.109	0.038	0.764	0.089	0.121	0.180	0.197	0.060	0.087	0.725
EMP2	0.077	0.070	0.056	0.080	0.843	0.113	0.073	0.180	0.158	0.067	0.079	0.817
EMP3	0.059	0.100	0.096	0.114	0.824	0.111	0.146	0.186	0.181	0.091	0.082	0.832
TAN1	0.128	0.078	0.137	0.049	0.140	0.052	0.149	0.758	0.156	0.054	0.150	0.712
TAN2	0.112	0.113	0.053	0.100	0.214	0.082	0.165	0.801	0.133	0.101	0.084	0.795
TAN3	0.091	0.098	0.118	0.099	0.206	0.102	0.085	0.826	0.117	0.139	0.061	0.820
US1	0.844	0.133	0.104	0.097	0.076	0.158	0.166	0.123	0.094	0.095	0.135	0.860
US2	0.843	0.098	0.129	0.127	0.068	0.148	0.175	0.101	0.136	0.131	0.104	0.866
US3	0.843	0.099	0.071	0.119	0.138	0.115	0.170	0.117	0.117	0.146	0.099	0.860
ITU1	0.179	0.108	0.141	0.139	0.134	0.160	0.811	0.136	0.125	0.063	0.134	0.840
ITU2	0.238	0.124	0.164	0.151	0.132	0.163	0.776	0.182	0.134	0.176	0.100	0.859
ITU3	0.191	0.095	0.198	0.175	0.147	0.126	0.790	0.169	0.124	0.181	0.126	0.870

Note: The bold numbers in the table indicate that the absolute value of the load factor is greater than 0.4. Rotation method: Varimax.

presents the information extraction of the factors for the research items and the corresponding relationship between the factors and research items. From Table 6, it can be seen that the communalities corresponding to all the research items are higher than 0.4, indicating a strong correlation between the research items and factors and that the factors can effectively extract information. Meanwhile, the items corresponding to each variable are consistent with the expected division of the scale, and the corresponding factor loading values are all greater than 0.5, indicating that the scale has good validity.

4.4.2. Confirmatory Factor Analysis

Confirmatory factor analysis (CFA) is primarily utilized to examine the convergent validity of a scale. Convergent validity, also known as convergence validity, is evaluated primarily through the average variance extracted (AVE) value and composite reliability (CR) value. Fornell argues that the assessment of convergent validity must meet the following criteria: standardized factor loadings should be greater than 0.7; the composite reliability (CR) value should exceed 0.7; and the AVE should be higher than 0.5. Discriminant validity, on the other hand, is used to detect the absence of correlation between items representing different constructs where no correlation is expected. Fornell suggests that discriminant validity is primarily evaluated by comparing the AVE square root with the correlation coefficients between constructs. If the AVE square root of any construct is greater than the correlation coefficients of other constructs, it indicates good discriminant validity among dimensions [60].

According to

Table 7. Factor loading coefficients table.

Factor Loading Coefficients Table
Latent Variables	Explicit Variables	Coef.	SE	t	p	Factor Loading	SMC	AVE	CR
Relevance	REE1	1	-	-	-	0.804	0.646	0.646	0.846
	REE2	0.819	0.042	19.634	0	0.782	0.611
	REE3	0.845	0.041	20.637	0	0.826	0.681
Completeness	COM1	1	-	-	-	0.851	0.724	0.741	0.896
	COM2	0.947	0.037	25.406	0	0.854	0.729
	COM3	0.972	0.037	26.252	0	0.877	0.77
Timeliness	TIM1	1	-	-	-	0.855	0.731	0.747	0.899
	TIM2	0.957	0.035	27.018	0	0.889	0.791
	TIM3	0.923	0.036	25.527	0	0.848	0.72
Accuracy	ACU1	1	-	-	-	0.784	0.615	0.653	0.849
	ACU2	0.895	0.044	20.122	0	0.815	0.664
	ACU3	0.904	0.044	20.324	0	0.825	0.68
Usability	UAB1	1	-	-	-	0.777	0.603	0.696	0.873
	UAB2	0.965	0.045	21.665	0	0.85	0.723
	UAB3	1.002	0.045	22.145	0	0.873	0.763
Reliability	REI1	1	-	-	-	0.795	0.632	0.715	0.882
	REI2	0.899	0.039	22.95	0	0.859	0.738
	REI3	0.927	0.04	23.465	0	0.88	0.775
Responsiveness	RES1	1	-	-	-	0.774	0.598	0.634	0.838
	RES2	0.878	0.047	18.779	0	0.799	0.639
	RES3	0.884	0.046	19.015	0	0.814	0.663
Empathy	EMP1	1	-	-	-	0.75	0.562	0.688	0.868
	EMP2	1.01	0.049	20.462	0	0.841	0.707
	EMP3	1.095	0.052	21.248	0	0.892	0.796
Tangibles	TAN1	1	-	-	-	0.712	0.507	0.665	0.855
	TAN2	1.066	0.056	19.075	0	0.86	0.74
	TAN3	1.082	0.057	19.123	0	0.865	0.748
User Satisfaction	US1	1	-	-	-	0.882	0.779	0.79	0.919
	US2	0.999	0.032	30.774	0	0.901	0.811
	US3	0.963	0.032	29.84	0	0.883	0.78
Intention to Use	ITU1	1	-	-	-	0.834	0.696	0.779	0.913
	ITU2	1.018	0.036	28.008	0	0.904	0.818
	ITU3	1.003	0.036	28.106	0	0.907	0.822

, the factor loadings of the observed variables corresponding to the latent variables in the model are all greater than 0.5, indicating that these manifest variables can adequately explain their respective latent variables. Additionally, the average variance extracted (AVE) values for each latent variable are all greater than 0.5, and the composite reliability (CR) values are all above 0.7, suggesting that the scale exhibits good convergent validity.

According to

Table 8. Discriminant validity: Pearson correlation and AVE square root value.

Discriminant Validity: Pearson Correlation and AVE Square Root Value
	Factor1	Factor2	Factor3	Factor4	Factor5	Factor6	Factor7	Factor8	Factor9	Factor10	Factor11
Relevance	0.804
Completeness	0.491	0.861
Timeliness	0.439	0.424	0.864
Accuracy	0.493	0.459	0.502	0.808
Usability	0.417	0.406	0.395	0.448	0.834
Reliability	0.415	0.400	0.315	0.427	0.541	0.845
Responsiveness	0.316	0.315	0.263	0.312	0.323	0.341	0.796
Empathy	0.309	0.327	0.297	0.329	0.312	0.365	0.485	0.830
Tangibles	0.366	0.368	0.331	0.369	0.328	0.332	0.422	0.502	0.815
User Satisfaction	0.411	0.364	0.353	0.395	0.389	0.439	0.354	0.330	0.373	0.889
Intention to Use	0.449	0.482	0.371	0.430	0.467	0.473	0.397	0.419	0.466	0.526	0.882

Note: The diagonal numbers are the square root values of AVE.

, the AVE square root value of each variable in this study is greater than its corresponding correlation coefficient, which indicates that there is good discriminant validity between the variables.

4.5. Correlation Analysis

As shown in

Table 9. Pearson correlation analysis.

Pearson Correlation
	Average Value	Standard Deviation	Information Quality	System Quality	Service Quality	User Satisfaction	Intention to Use
Information Quality	4.554	1.253	1
System Quality	4.843	1.439	0.590 **	1
Service Quality	4.507	1.304	0.522 **	0.473 **	1
User Satisfaction	4.630	1.730	0.490 **	0.472 **	0.438 **	1
Intention to Use	4.569	1.765	0.558 **	0.536 **	0.532 **	0.526 **	1

** p < 0.01.

, the correlation coefficient between willingness to use and information quality is 0.558, with significance at the 0.01 level, indicating a significant positive correlation between willingness to use and information quality. Similarly, the correlation coefficient between willingness to use and system quality is 0.536, which is also significant at the 0.01 level, suggesting a significant positive correlation between willingness to use and system quality. Furthermore, the correlation coefficient between willingness to use and service quality is 0.532, which is significant at the 0.01 level, demonstrating a significant positive correlation between willingness to use and service quality. Additionally, the correlation coefficient between willingness to use and user satisfaction is 0.526, which is significant at the 0.01 level, indicating a significant positive correlation between willingness to use and user satisfaction.

Regarding user satisfaction, the correlation coefficient with information quality is 0.490, which is significant at the 0.01 level, revealing a significant positive correlation. Similarly, the correlation coefficient between user satisfaction and system quality is 0.472, which is significant at the 0.01 level, suggesting a significant positive correlation. Lastly, the correlation coefficient between user satisfaction and service quality is 0.438, which is also significant at the 0.01 level, demonstrating a significant positive correlation between user satisfaction and service quality.

4.6. Analysis of Structural Equation Modeling

4.6.1. Model Construction

This study employs AMOS 23 software for the construction and analysis of the structural equation model. The model comprises a total of three second-order latent variables, namely information quality, system quality, and service quality, along with two first-order latent variables, user satisfaction and willingness to use, and 33 observed variables. The specific model construction is illustrated in Figure 4.

4.6.2. Model Fit

The maximum likelihood method in the AMOS 23 software is selected to calculate whether the various fit indices of the model using the collected data are within the standard range, verifying whether the model’s structure aligns with the hypothetical structure proposed in the study. The fit indices mainly include absolute fit indices (chi-square degrees of freedom ratio X2/df, standardized root mean square residual (SRMR), root mean square error of approximation (RMSEA), goodness-of-fit index (GFI), and adjusted goodness-of-fit index (AGFI)); relative fit indices (normed fit index (NFI), incremental fit index (IFI), Tucker–Lewis index (TLI), and comparative fit index (CFI)); and parsimony fit indices (parsimony normed fit index (PNFI) and parsimony comparative fit index (PCFI)).

The data collected from the questionnaire were imported into the AMOS 23 software, and the model fit parameters obtained using the maximum likelihood method are shown in

Table 10. Model fit.

Model Fit
Fit Index	Judgment Criteria	Actual Value	Fitting Results
Absolute fit index
CMIN/DF	<3	1.619	Excellent
SRMR	<0.08	0.031	Excellent
GFI	>0.8	0.928	Excellent
AGFI	>0.8	0.915	Excellent
RMSEA	<0.08	0.032	Excellent
Relative fit index
NFI	>0.9	0.944	Excellent
IFI	>0.9	0.978	Excellent
TLI	>0.9	0.975	Excellent
CFI	>0.9	0.978	Excellent
Parsimonious fit index
PNFI	>0.5	0.851	Excellent
PCFI	>0.5	0.881	Excellent

. As can be seen from Table 10, the displayed values of the fit parameters all meet the standard requirements, indicating that the model fits very well. Therefore, this structural equation model has a good fit effect for the sample data obtained from the questionnaire. The model run results are shown in Figure 5.

4.6.3. Path Effect Analysis

According to

Table 11. Path effect analysis.

Path Effect Analysis
Path			B	β	S.E.	C.R.	P
Information Quality	→	User Satisfaction	0.353	0.254	0.131	2.701	0.007
System Quality	→	User Satisfaction	0.359	0.262	0.127	2.832	0.005
Service Quality	→	User Satisfaction	0.319	0.178	0.13	2.446	0.014
Information Quality	→	Intention to Use	0.285	0.206	0.116	2.467	0.014
System Quality	→	Intention to Use	0.28	0.205	0.113	2.475	0.013
Service Quality	→	Intention to Use	0.487	0.272	0.119	4.077	***
User Satisfaction	→	Intention to Use	0.193	0.193	0.045	4.29	***

Note: The arrow (→) in the table indicates the direction of the causal relationship in the path analysis. *** p < 0.001.

, information quality has a significant positive impact on intention to use (β = 0.206, p < 0.05), thus supporting Hypothesis H1. Information quality also has a significant positive impact on user satisfaction (β = 0.254, p < 0.05), thus supporting Hypothesis H2. System quality has a significant positive impact on intention to use (β = 0.205, p < 0.05), thus supporting Hypothesis H3. System quality has a significant positive impact on user satisfaction (β = 0.262, p < 0.05), thus supporting Hypothesis H4. Service quality has a significant positive impact on intention to use (β = 0.272, p < 0.05), thus supporting Hypothesis H5. Service quality has a significant positive impact on user satisfaction (β = 0.178, p < 0.05), thus supporting Hypothesis H6. User satisfaction has a significant positive impact on intention to use (β = 0.193, p < 0.05), thus supporting Hypothesis H7.

4.6.4. Robustness Test

The main approaches for robustness tests include altering research variables and estimation methods. If the signs and significance of the model coefficients remain largely unchanged after these adjustments, the results are considered robust. In this study, robustness is tested by changing the estimation method, with the model re-validated using multiple linear regression. The results, as shown in

Table 12. Results of mediation analysis.

Results of Mediation Analysis (n = 605)
	Intention to Use		User Satisfaction		Intention to Use
	B	t	B	t	B	t
Constant	−0.303	−1.284	0.527 *	2.087	−0.427	−1.862
Information Quality	0.379 **	6.623	0.345 **	5.632	0.298 **	5.237
System Quality	0.303 **	6.285	0.278 **	5.374	0.238 **	4.974
Service Quality	0.372 **	7.382	0.263 **	4.883	0.310 **	6.23
User Satisfaction					0.235 **	6.373
R-squared	0.429		0.318		0.465
Adjusted R-squared	0.426		0.315		0.461
F-value	150.378 **		93.563 **		130.368 **

* p < 0.05, ** p < 0.01.

, indicate that after modifying the variables, the signs and significance of the main research variables remain consistent, confirming the robustness of the findings.

5. Discussion

5.1. Analysis of Factors Influencing User Satisfaction with the Wayfinding Signage System of Wuxi Subway

Dominance of System Quality: The path analysis results indicate that system quality has the greatest impact on user satisfaction (path coefficient of 0.359), with a highly significant level (p = 0.005). This suggests that during the process of using the Wuxi subway wayfinding signage system, the system’s usability and reliability play crucial roles. A city subway wayfinding signage system that is easy to use, with clear and intuitive symbols, and operates stably over a long period can significantly enhance the overall satisfaction of users.

Crucial Role of Information Quality: Following system quality, information quality also has a notable impact on user satisfaction (path coefficient of 0.353, p = 0.007). Users have extremely high requirements for the accuracy, completeness, timeliness, and relevance of information provided by the Wuxi subway wayfinding signage system. High-quality information can help passengers make more informed navigation decisions and improve the efficiency of using the subway wayfinding system, thereby enhancing satisfaction with the system. Thus, optimizing information quality is an indispensable aspect of enhancing user satisfaction.

Complementary Effect of Service Quality: Although the path coefficient of service quality on user satisfaction is slightly lower than the previous two factors (0.319, p = 0.014), its importance cannot be overlooked. Service quality encompasses empathy, responsiveness, and tangibles during users’ interactions with the Wuxi subway wayfinding signage system. Quality service can alleviate users’ concerns, increase their sense of trust and belonging, and thereby positively contribute to user satisfaction.

5.2. Analysis of Factors Influencing Willingness to Use the Wuxi Subway Wayfinding Signage System

Decisive Impact of Service Quality: Among the various factors influencing users’ willingness to use the system, service quality emerges as the most prominent (path coefficient of 0.487, p = 0.001). This indicates that a superior service experience can significantly motivate users to continue using the Wuxi subway wayfinding signage system.

Auxiliary Roles of Information Quality and System Quality: Although the impact of information quality and system quality on willingness to use the system is less pronounced than that of service quality (with path coefficients of 0.285 and 0.28, respectively), they still play a positive role in enhancing users’ willingness to use the system. High-quality information and systems can improve users’ experience and efficiency during their interactions with the Wuxi subway wayfinding signage system, making them feel more satisfied and comfortable and thereby strengthening their intention to continue using it.

Indirect Influence of User Satisfaction: Although the impact of user satisfaction on willingness to use the system is relatively lower (path coefficient of 0.193, p = 0.001), it remains a non-negligible factor. User satisfaction can indirectly promote users’ willingness to use the system. When users are satisfied with the system or service, they are more likely to have the motivation to continue using it and potentially become long-term users.

6. Conclusions and Future Research

6.1. Conclusions

This study demonstrates that in Wuxi, a rapidly developing modern city in China, the design and optimization of the subway system’s wayfinding signage, as a critical component of urban public transportation, are directly linked to passenger satisfaction and their willingness to use the system. To enhance both satisfaction and usage, the providers of the Wuxi subway wayfinding signage system should focus on improving system quality and service quality. System quality is the most significant driver of user satisfaction, while service quality has the strongest influence on passengers’ willingness to use the system. Additionally, information quality plays a key role and should not be overlooked, as it significantly impacts both satisfaction and willingness to use the system.

From the perspective of information quality, the study clearly highlights the importance of relevance, completeness, timeliness, and accuracy of information in enhancing user satisfaction and willingness to use the system. This implies that during optimization, it is essential to ensure that the information provided by the wayfinding signage system is closely related to passengers’ travel needs, comprehensively covers station, transfer, and facility information, and remains up-to-date and accurate, reducing ambiguity and misdirection. By adopting a scientific and rational layout and positioning, the efficiency of wayfinding can be improved.

The optimization of system quality centers on usability and reliability. Enhancing the clarity and intuitiveness of symbols, simplifying language, and making information easier to understand can effectively reduce the users’ learning curve, thereby improving convenience. Additionally, the system’s reliability and long-term stable operation are vital for building user trust, which plays a key role in strengthening users’ willingness to continue using the system.

In terms of service quality, the study emphasizes the positive impact of responsiveness, empathy, and tangibles on the user experience. This suggests that the subway wayfinding signage system must not only respond quickly to passenger needs and handle emergencies but also attend to special passengers and personalized requirements, enhancing the emotional warmth of the service. Furthermore, maintaining the visibility, clarity, consistency, and standardization of signage, in line with passenger aesthetics, is also an essential means of improving user satisfaction.

6.2. Theoretical Contributions

This study is deeply rooted in the DeLone and McLean (D&M) information systems success model. Through a series of meticulously designed surveys and extensive data collection efforts, a comprehensive and detailed analysis of the multidimensional quality characteristics of the Wuxi subway wayfinding signage system was conducted. By employing advanced statistical methods and path analysis techniques, the study explores the impact of factors such as the information quality, system quality, and service quality of the Wuxi subway wayfinding signage system on users’ willingness to use the system and their satisfaction. This research enriches the application of the D&M theory in the field of public transportation wayfinding systems and provides invaluable theoretical guidance and empirical evidence for the subsequent optimization of wayfinding signage in the Wuxi subway, as well as in other urban public transportation systems.

6.3. Practical Significance

The practical significance of this study lies in its provision of a solid theoretical foundation and actionable guidance plan for the optimization and improvement of the Wuxi subway wayfinding signage system. Specifically, the following aspects highlight its profound practical application value:

Precise Guidance for System Optimization: By applying the D&M theoretical framework, the study systematically identifies the key factors influencing user satisfaction and willingness to use the system—information quality, system quality, and service quality. This discovery enables the Wuxi subway to target its system optimization efforts, ensuring that while enhancing the usefulness, completeness, timeliness, and accuracy of information, it also takes into account the ease of use and reliability of the system, as well as the responsiveness, empathy, and tangibility of the service. This comprehensive and precise optimization strategy will significantly enhance passengers’ travel experience.

Enhancing Passenger Satisfaction: The study delves into the specific manifestations of each quality characteristic in practical applications and the mechanisms of their influence on passengers’ decision-making processes, providing the Wuxi subway with effective ways to increase passenger satisfaction. By ensuring that the information provided by the wayfinding signage system is highly relevant and accurate to passengers’ travel needs, that the signage is user-friendly and easily recognizable, and that services are responsive and emotionally supportive, passenger satisfaction can be significantly boosted.

Promoting Sustainable Development of Public Transportation Systems: The findings of this study are not limited to the Wuxi subway but offer valuable insights for the optimization of public transportation systems in other cities as well. By drawing on the discoveries and recommendations of this study, other cities can formulate more scientific and rational optimization strategies tailored to their specific conditions, driving the continuous improvement and development of public transportation wayfinding signage systems. This holistic enhancement will contribute to the construction of a more convenient, efficient, and comfortable public transportation network, thereby promoting sustainable urban development.

Strengthening the Integration of Theory and Practice: This study successfully applies the D&M theoretical framework to the field of public transportation wayfinding signage systems, not only enriching the application of the theory in specific scenarios but also verifying its effectiveness through empirical research. This close integration of theory and practice provides the academic community with abundant research cases and data support, fostering further in-depth research and development in related fields.

6.4. Limitations and Future Research Directions

Despite the in-depth exploration of the impact of the Wuxi subway wayfinding signage system on user satisfaction and willingness to use the system across multiple dimensions, this study still has certain limitations that offer directions for future research.

Limitations in Sample: The data collection in this study may be limited by specific locations and sample groups, which may affect the generalizability of the results. Additionally, due to constraints in time and resources during the data collection process, this study currently focuses only on the Wuxi region. Future research could consider expanding the sample range to include passengers from different cities and cultural backgrounds, in order to enhance the broader applicability of the findings.

Insufficient Consideration of Dynamics: While this study primarily focuses on the impact of the current system state on user satisfaction and willingness to use the system, it does not fully account for the dynamic effects of system improvements or external environmental changes on passengers’ perceptions. Future research could adopt longitudinal methodologies to track changes in user feedback before and after system improvements, providing a more comprehensive understanding of the actual effects of system optimization.

Technological Limitations: As technology advances, new information system characteristics (such as intelligence and personalization) continue to emerge; however, this study may not have adequately explored the potential impacts of these emerging features on user satisfaction and willingness to use the system. Future research can focus on how these new technologies integrate with existing systems and their specific impacts on user experience.

In light of the aforementioned limitations, future research can expand and deepen its focus in the following directions:

Expanding Sample Scope and Diversity: Conducting cross-regional and cross-cultural comparative studies will help uncover differences in passenger satisfaction with subway wayfinding signage systems across various contexts, along with the underlying reasons. These insights will provide valuable references for broader system optimization [61,62].

Focusing on Dynamic Changes and Long-term Effects: It is necessary to conduct longitudinal research to track changes in user feedback after system improvements, evaluate the long-term effects of optimization measures, and provide empirical support for continuous optimization.

Exploring New Technologies and Emerging Features: It is necessary to pay attention to the application of new technologies, such as intelligence and personalization in subway wayfinding signage systems, investigate how these technologies enhance user experience and meet the diverse needs of passengers, and discuss their potential risks and challenges.

Deepening Theoretical Application and Expanding the Model: By building on the D&M theoretical framework, this study will integrate other relevant theories—such as user experience theory and service management theory—to construct a more comprehensive and systematic model. This enhanced framework will provide deeper insights and more effective guidance for optimizing subway wayfinding signage systems [63].