Evaluation of Urban Sustainability through Perceived Importance, Performance, Satisfaction and Loyalty: An Integrated IPA–SEM-Based Modelling Approach

Abstract

In recent years, there has been a surge in research pertaining to sustainable urban development. Importance–performance analysis (IPA) has emerged as one of the most widely used methods. However, few studies have combined IPA with structural equation modelling (SEM). This study introduces and evaluates an integrated IPA–SEM approach to assess the impact of perceived importance and performance of sustainable city dimensions on residents’ satisfaction and loyalty. The data for this study were collected from 425 survey respondents residing in Siliguri City of West Bengal, India. The results indicate that the ‘Keep Up the Good Work’ quadrant, consisting of social quality, connectivity conditions, and environmental quality, significantly and positively affected satisfaction. On the other hand, the ‘Concentrate Here’ quadrant concerning environmental pollution presented a negative yet insignificant relationship with satisfaction while both the ‘Low Priority’ and ‘Potential Overkill’ quadrants demonstrated no notable influence. Furthermore, a robust positive correlation between satisfaction and loyalty is confirmed. Overall, the findings offer valuable implications for urban planning, policy-making, and strategies aimed at enhancing Siliguri residents’ quality of life.

1. Introduction

Over the past few years, there has been a significant increase in research and investigations focused on developing and maintaining sustainable cities [1,2,3,4]. Currently, more than half of the global population resides in urban areas; this number is projected to rise to nearly 68% by 2050 [5,6,7]. Urbanization poses a challenge for many countries, as they struggle with limited resources to address the basic needs of their residents, such as housing, infrastructure development, safety measures, and employment opportunities, among others [8,9]. Therefore, it is crucial that we continue studying sustainable urban development so that cities can grow responsibly through a coordinated improvement of ecological financial cultural and social conditions [10,11].

Sustainable development principles are based on meeting present-day needs while ensuring that future generations’ ability to fulfil their own requirements will not be compromised at any level [12,13,14]. Sustainable cities that ensure every inhabitant can thrive without negatively impacting future generations’ prospects by providing secure healthy living environments [15,16,17]. The United Nations (UN) established the 2030 Agenda for Sustainable Development consisting of 17 Sustainable Development Goals (SDGs). These goals guide nations towards formulating policies promoting sustainable urban development initiatives [18,19]. The ninth World Urban Forum emphasizes ‘Cities 2030, Cities for all’, highlighting equitable and ecologically responsible urbanization’s importance by 2030 [20,21]. In essence, cities must strive for environmental, social, and economic sustainability when aiming to protect current and future residents’ well-being [22]. Evaluating ongoing urban development initiatives helps nations guarantee long-term strategic approaches towards achieving these objectives [23,24].

The analytical method known as importance–performance analysis (IPA) has been widely used in urban planning research to evaluate the relative worth and efficacy of various components [25,26,27,28]. It is considered that IPA’s ability to apply importance and performance scales to multiple sustainable city attributes, such as sustainable transport innovation [29], sustainable tourism [30], and sustainable urban green space [26], will lead to its widespread use in research on sustainable development indicators. By identifying areas most in need of improvement and those needing the least, IPA may help direct development where it will have the greatest impact [31,32]. The intended outcome of utilizing IPA is the collection of crucial information that can back up appropriate authorities’ policies towards promoting sustainability [33]. However, metrics like the four importance and performance quadrants used in research could be inadequate or faulty at revealing a city’s developmental circumstances. Urban researchers incorporate other methodologies with IPA, including structural equation modelling (SEM) [25], analytical hierarchy process (AHP) [34], and correspondence analysis [35] to strengthen their findings’ validity. Thus far, few studies have coupled IPA with SEM to investigate the links between importance, performance, satisfaction, and loyalty among local inhabitants. Deng and Pierskalla utilized SEM to examine the impacts of four sub-regions identified by IPA on visitor satisfaction and loyalty.

This research aims to assess Siliguri residents’ levels of contentment towards their city along with their perception regarding its advances made toward achieving sustained growth. A total of 425 adult residents who had lived within Siliguri City for over ten years participated via questionnaire survey. This study aimed to develop an integrated approach using both IPA–SEM methods to gauge correlation between perceived significance of social, economic, cultural, and environmental dimensions related to Sustainable Urban Development elements alongside overall performance across these domains, while measuring satisfaction level amongst urban residents. Therefore, the present study unveils a novel methodology that combines importance–performance analysis (IPA) and structural equation modelling (SEM) to evaluate the multifaceted dimensions of urban sustainability. After conducting a thorough analysis of relevant works, this study presents a comprehensive overview of Siliguri’s geographical and demographic characteristics. The purpose of this investigation is highlighted, emphasizing its novelty and significance. The work outlines the theoretical underpinnings, creating a conceptual structure encompassing perceived importance, performance, satisfaction, and loyalty. The following section delineates the research methodology and analytical rigorousness of integrating IPA and SEM. Subsequently, the results are elaborated upon, providing insight into the complex effects of sustainability dimensions on the satisfaction and loyalty of residents. The subsequent discussion places these findings in the context of existing literature and theoretical implications. The concluding section summarizes the significant findings and provides suggestions for urban planning stakeholders while recognizing the constraints of the study and proposing potential areas for future research.

2. Study Area

Siliguri, a rapidly growing urban hub, is situated in West Bengal, India and is renowned for its strategic geographic location as the ‘Gateway to North-East India’ [36]. This region serves as an important nexus between Nepal, Bhutan, and Bangladesh, driving demographic expansion along with economic development. The territory comprises portions of Darjeeling and Jalpaiguri districts within the state of West Bengal. Siliguri ranks as the third most populous urban agglomeration following Kolkata and Asansol. The area has undergone a significant transformation from a modest settlement into a thriving commercial centre [37,38]. Located in the Himalayan foothills alongside Mahananda River, Siliguri’s average elevation is approximately 122 m (400 feet). The jurisdiction of Siliguri Municipal Corporation covers around 41.9 square kilometres, and is divided into 47 wards (See Figure 1). Bhattacharyya and Mitra [36] noted that, since 1931, the population of this city has grown notably, with substantial growth recorded between 2001 and 2011 when it escalated from four lakhs to over seven lakhs. The urban configuration extends roughly up to 9 km in the north–south direction and slightly over five kilometres in the east–west direction.

Recent research [39,40,41] indicates robust economic growth; however, environmental sustainability remains one of several developmental challenges facing Siliguri. Recognizing pivotal roles played by social, economic, cultural, and environmental factors in advancing urban centres like Siliguri becomes imperative. This study aims to assess various crucial factors, particularly those pertaining to social, economic, cultural, and environmental dimensions concerning sustainable urban areas. This evaluation encompasses their significance, contentment, effectiveness, and allegiance within the context of a rapidly expanding environment. Implementing all-inclusive strategies to yielding significant results while steering policy formulation facilitating sustainable urban milieus is the principal objective of this study.

3. Theoretical Background and Hypotheses

3.1. Importance–Performance Analysis

The marketing research community developed the importance–performance Analysis (IPA) framework by incorporating consumer preferences into service quality evaluation [42,43]. The two-dimensional grid employed in IPA represents importance and performance ratings, dividing attributes into four categories: ‘Keep Up the Good Work’, ‘Concentrate Here’, ‘Low Priority’, and ‘Potential Overkill’ [44]. Although popularly used, debates continue regarding crosshair placement at intersecting lines [45,46]. Scale-centred crosshairs utilizing a midpoint scale are often used [47,48,49], as are data-centered crosshairs relying on an average of all importance–performance evaluations [50,51,52]. For this study, researchers utilized a data-driven approach using average mean differences (0,0), instead of the original ratings.

Another point of contention is the relative value between absolute importance, expressed directly by visitors, and derived importance, based on statistical methodologies including simple regression analysis [53,54,55], partial correlation analysis [56], and multiple regression analysis [57]. While certain scholars accept the indirect methods, it has been shown that declared or absolute importance is more reliable than derived importance when predicting overall pleasure and brand choice [57,58]. Therefore, this study’s researchers utilized an absolute importance criterion.

Due to its accessibility, intuitiveness, and apparent usefulness, the IPA has found widespread application in goods-and-services quality assessment [59]; however, major methodological issues affecting reliability and validity exist, with IPA being originally designed as a market approach for company management strategies. Despite these limitations, IPA has experienced broad use across service sectors, including hospitality and tourism, where it is used to examine various attributes’ importance and performance alongside visitor satisfaction levels.

3.2. Importance, Performance, and Satisfaction

A satisfied customer is one who has an affirmative experience with a product or service, overall, or with any of its features [55,60]. Satisfaction traditionally results from cognitive processes that evaluate services or products based on their performance against criteria. The performer’s level of happiness depends on whether they feel the performance exceeded or met anticipations [25]. O’Neill et al. [61] identified direct and inferred measures to assess perceptions: direct measurement explicitly asks consumers to rate their expectations using a Likert scale, while inferred measurement calculates differences between perception scores and measured standards.

Before purchase or use of a service, consumer expectations are polled. On-site evaluations determine how well these services meet needs [62]; it is impractical to inquire about anticipation before travel or recall after arrival at destination [63]. Most studies focus on the correlation between attribute importance–performance and attribute-level satisfaction and overall contentment [64,65,66], even though perceived performance and quality are fundamentally distinct. Some use them as proxies for measuring customer happiness [67,68].

Performance may be influenced by factors like task importance and people’s high expectation levels [69]. Proponents claim there exists a causal link between importance and performance [45], shown in prior research to significantly impact consumer happiness and satisfaction levels [70,71]. When assessing overall customer happiness, considering both importance and performance together becomes imperative, since they are inextricably linked [72]. Thus, tourist satisfaction does not change, regardless of attribute-performance, if tourists do not care/experience much regarding the characteristic; however, it would prove true for attributes with elevated amounts of significance and performance [73]. According to this presumption, the following four hypotheses can be derived (See Figure 2):

H1. 

The quadrant of attributes represented by ‘Keep Up the Good Work’ will have a large and positive impact on overall satisfaction.

H2. 

The attributes that fall into the ‘Concentrate Here’ quadrant will significantly and negatively influence overall satisfaction.

H3. 

The ‘Low Priority’ category of attributes will not have any effect on the overall satisfaction.

H4. 

The attributes that fall under the ‘Potential Overkill’ quadrant will not have any effect on the overall satisfaction.

3.3. Satisfaction and Loyalty

In business literature, loyalty is categorized into attitudes, intentions, and actions [74]. The term ‘attitude loyalty’ refers to an individual’s implicit preference for a specific place or event; on the other hand, ‘behavioural intention’ describes one’s desire to return while promoting and sharing positive feedback about their experience [26]. Some studies use behavioural intention measures like willingness to pay premium prices and brand-switching probability [75], while others employ behavioural loyalty measured by repeat purchases (e.g., nights/visits/amount of money spent on a specific brand and location, frequency of visits, and real patronage) [76,77]. Satisfaction is viewed as a precondition for developing loyalty [74]. In social science and hospitality- and tourism-related research, satisfaction has been quantified and correlated with loyalty [78]. Meta-analyses reveal robust positive correlations between satisfaction and loyalty regarding festivals [79,80], the hospitality industry, and individual tourist hotspots. Therefore, the hypothesis put forth supports this notion (See Figure 2).

H5. 

Loyalty will be significantly and positively influenced by overall satisfaction.

4. Database and Methodology

4.1. Comprehensive Overview of the Research Methodology

The present research utilizes a comprehensive and cautious methodology to evaluate the sustainable city indicators of Siliguri, taking into account diverse indicators spanning across social, economic, environmental, and cultural domains. The methodology involves a series of organized procedures that are combined to achieve the ultimate outcomes. After conducting a comprehensive examination of the available literature and seeking input from knowledgeable individuals, a collection of 26 indicators for sustainable cities (designated as F1 through F26) was selected (See Figure 3). The objective of the selection process was to guarantee the pertinence and practicality of the indicators in relation to the particular circumstances of Siliguri. Subsequently, a survey tool was created, comprising 26 inquiries that aligned with the chosen indicators. The objective of the survey was to gather the viewpoints of adult inhabitants who have been residing in Siliguri for over ten years, concerning the significance and efficacy of said indicators. This study employed a five-point Likert scale to assess the participants’ responses, thereby facilitating the collection of detailed feedback [81]. Subsequently, the raw data obtained from the survey underwent a process of data cleansing aimed at detecting and addressing incomplete or inconsistent responses (Figure 3). To ensure the data’s suitability for further analysis, a normality check was conducted by evaluating the skewness and kurtosis [82,83]. The subsequent step entailed performing a gap analysis to discern the discrepancies between the perceived significance and execution of each indicator (see Figure 4). The statistical significance of these gaps was evaluated using t-tests. Subsequently, an IPA [84] was performed to graphically depict the relative locations of the indicators within a two-dimensional plane delineated by their significance and effectiveness. The outcome of this approach was the creation of a four-quadrant matrix, consisting of the categories ‘Concentrate Here’, ‘Keep Up the Good Work’, ‘Low Priority’, and ‘possible overkill’ (Figure 3 and Figure 5). This matrix served to enhance comprehension and enable the prioritization of the indicators. Factor analysis was employed to gain insight into the fundamental constructs among the indicators within each quadrant of the IPA matrix. This study evaluated the degree of contentment among inhabitants with regard to the efficacy and significance of the sustainable urban measurements. The investigation examined the correlation between satisfaction levels and residents’ loyalty towards the city, acknowledging that the latter is dependent on the former. Structural equation modelling (SEM) [85] is a sophisticated statistical methodology that can constructs a model that illustrates the interconnections between satisfaction, loyalty, and sustainable city indicators. The utilization of structural equation modelling (SEM) [85] facilitated a more profound comprehension of the interrelatedness and impact of these variables (Figure 3 and Figure 6).

The methodology employed in this study ensures a thorough, organized, and all-encompassing evaluation of sustainable urban development in Siliguri. This is achieved through the use of primary data and the application of rigorous analytical techniques. The analysis and results in the following sections are based on the research methodology overview, which serves as the fundamental foundation of this research.

4.2. Development of Survey Instrument

To assess views on importance, performance and satisfaction of various attributes, 26 questions were asked to city residents with answers rated on a five-point Likert scale. Performance ratings ranged from 1 (strongly disagree) to 5 (strongly agree); importance ratings from 1 (not important at all) to 5 (very important); satisfaction ratings from 1 (not satisfied at all) to 5 (extremely satisfied) and loyalty ratings from 1 (strongly disagree) to 5 (strongly agree). Additionally, data on socio-economics and demography (city duration, genders, education levels, and employment status, etc.) and town features were requested from city residents. The final survey section consisted of open-ended question regarding the level of satisfaction and loyalty experienced towards the city by its citizens.

4.3. Data Collection and Analysis

This research used a convenient sample of people strolling through crucial locations, including a retail centre and a well-known street in the city. Convenience sampling is standard among numerous prominent public opinion polling organizations, political polling groups, and market research firms, exemplifying a nonprobability sample [86]. Participants in the study who were at least 18 years old were asked to fill out a questionnaire about the characteristics of the city, their perceptions of the city’s attributes (including their importance, performance, and level of satisfaction), their overall level of satisfaction, and their loyalty to the town. The sampling was carried out throughout January and April 2023.

In order to address the issue of missing data, the expectation maximization algorithm was employed prior to conducting data analyses [81]. Out of the 470 returned questionnaires, 45 were disqualified due to repeated inadequate responses or suspicious patterns, leaving a total of 425 samples for further investigation. Following the initial screening, less than 5% of data were missing entirely, which is considered low [87,88]. Moreover, the skewness and kurtosis of the observed variables were examined, and it was determined that the data did not deviate significantly from a normal distribution as all endogenous variables had skewness values below 2 and kurtosis values below 3, with their absolute values falling within acceptable limits [89,90]. The data analysis procedure consisted of four steps:

• Pairwise t-tests were conducted for a gap analysis to compare individual perceptions of importance and performance.
• IPA was performed using graphical representations of each attribute on Importance and Performance grids.
• The five-question loyalty survey and the four-quadrant characteristics were subjected to factor analysis.
• The five proposed hypotheses were tested using SEM.

Principal components analysis was utilised to identify unobserved variables in the factor analysis. Potential factors were identified using varimax rotation and eigenvalues larger than 1.00. To evaluate the suitability of the data for factor analysis, statistical tests such as Kaiser–Meyer–Olkin (KMO) and Bartlett’s Sphericity tests were conducted. Items were selected for a factor if they had a threshold of 0.45 [82]. Cronbach’s alpha coefficients determined factor consistency. A scale’s reliability is usually assessed using a 0.70 Cronbach’s alpha score [83]; a value below 0.60 is also acceptable in exploratory research in social science and psychological studies [91]. Exploratory factor analysis (EFA) and the calculation behind the structural equation modelling (SEM) was carried out using SPSS [92].

The fit of the structural models was tested using Smart-PLS, while the structural relationships among the variables were established using SEM. The bootstrapping technique in Smart-PLS, including user-defined settings, was used to determine the significance of the relationships between variables and to test the hypotheses (See Figure 3).

4.4. Model Specification

Importance–performance analysis employs a matrix-style graph, with ‘importance’ represented on the y-axis while ‘performance’ is depicted along the x-axis. The matrix plots items according to significance and performance averages. Survey data, usually from Likert-scale questions, is used to calculate IPA grid attribute coordinates [84]. The simple calculations include:

Mean Importance (I): This is the average score of respondents’ ratings of the importance of a specific attribute or factor.

I = Σi/n

(1)

where:

i = importance rating of an attribute;

n = number of respondents.

Mean Performance (P): This is the average score of respondents’ ratings of the performance of a specific attribute or factor.

P = Σp/n

(2)

where:

p = performance rating of an attribute;

n = number of respondents.

4.5. Structural Equation Modelling (SEM)

Multivariate Structural Equation Modelling is more complicated. The SEM equations express variable connections, but its mathematics is complex [85]. For example:

η = Bη + Γξ + ζ

(3)

y = Λyη + ε

(4)

where:

η represents the dependent variables vector;

ξ denotes the independent variables vector;

B represents the coefficients matrix used for the dependent variables;

Γ indicates the coefficients matrix used for the independent variables;

Λy is the coefficients matrix used to measure the model for the dependent variables;

ζ represents the error vectors for the dependent variables; and

ε indicated error vectors for the observed variables.

5. Results

5.1. Respondents Characteristics

A total of 659 people were sent an invitation to take part in the survey; of them, 425 accepted the request to participate, which is a response rate of 64.49 percent. Most of the people who answered were women (53.45%), between the ages of 26 and 54 (45.9%), had a high level of education, and were not in need of financial assistance. The respondents who were between the ages of 18 and 30 constitute the most significant proportion of the city’s population (29.41%), followed by those who were between the ages of 30–45 (27.06%) and those who were between 45 and 60 (22.82%), who are also significant. Approximately 24% of residents have a bachelor’s degree, 27.29% are postgraduate, and 6.35 are doctorate. Most respondents were engaged in private jobs (36.71%), followed by business (25.65%), government jobs (11.29%), and housewives (9.18%). Also, the majority of the respondents (35.05%) had resided in the city for more than 10 to 15 years, followed by 22.35% for 15 to 20 years, 20.94% for 5 to 10 years, 11.53% for less than five years, and 10.11% for more than 20 years. Thus, most respondents had resided in this city for more than 10 to 15 years and had a good experience and perception of the city. The vast majority of respondents, 56.4%, called the town their permanent home or were the city’s permanent residents.

5.2. Gap Analysis

The findings of the paired-sample t-tests that were carried out in order to establish the significance of the mean differences between importance and performance are presented in

Table 1. Socio-demographic characteristics of the respondents.

Attribute	Category	Frequency	Percentage
Gender	Male	198	46.60
	Female	227	53.45
Age Group	<18	52	12.24
	18–30	125	29.41
	30–45	115	27.06
	45–60	97	22.82
	>60	36	8.47
Marital Status	Married	278	65.41
	Single	108	25.41
	Divorced	39	9.18
Education	Primary	34	8.00
	Secondary	54	12.71
	Higher Secondary	92	21.65
	Graduate	102	24.00
	Post Graduate	116	27.29
	Doctorate	27	6.35
Occupation	House Wife	39	9.18
	Student	73	17.18
	Business	109	25.65
	Private Job	156	36.71
	Govt. Job	48	11.29
Duration of Residence	<5 years	49	11.53
	5–10	89	20.94
	10–15	149	35.05
	15–20	95	22.35
	>20	43	10.11

. According to the data presented, the performance and significance characteristics received mean ratings of 5.692 and 5.756, respectively. On the significance scale, the item ‘control of traffic congestion’ obtained the highest rating, while on the performance scale, the item ‘preservation of natural areas’ received the highest rating (M = 6.28).

The results of the t-test indicated that 23 out of 26 pairs had considerably reduced levels of performance than relevance in the following nine items: Access to many locations inside the city and the public transportation system, the control of traffic congestion, the control of the crowdedness of people, pollution, the maintenance of streets and buildings, the cost of living affordability, and community identity and belongingness.

On the other hand, 17 indicators assessing aspects such as social security, connectedness, and environmental quality scored better in performance than relevance. An examination of correlations revealed that performance, importance, and the degree to which an attribute is satisfied have a strong relationship. It is essential to point out that attribute performance and attribute satisfaction have a strong correlation, with a coefficient of 0.970, which is very near to 1.0 and suggests that there is a perfect match between the two (See Figure 4).

5.3. IPA Matrix Analysis of Resident’s Satisfaction towards City Loyalty

IPA was used to assess and analyse each characteristic’s importance and performance through the resident’s satisfaction and city dwellers’ loyalty. The median of the highest and lowest values of the entire averages was used to calculate the IPA intersection points. Importance was on the horizontal axis, and performance was on the vertical axis [93]. The findings are shown in Figure 5 and

Table 2. Results of paired-sample t-test.

Item	Average Performance	Average Importance	Gap	t	p
The assurance of safety and serenity in the city (F1)	6.13	5.98	0.15	6.090	0.001
Caliber and comfort of residential accommodations (F2)	6.15	6.01	0.14	0.693	0.489
Superiority and accessibility of health-related services (F3)	6.09	5.79	0.30	2.698	0.007
The standard of educational facilities and offerings (F4)	6.16	6.11	0.05	5.277	0.001
Availability of sports and recreational amenities (F5)	5.83	5.43	0.40	19.887	0.001
Proximity and accessibility to shopping centres and dining establishments (F6)	5.67	5.26	0.50	11.236	0.001
Integrated urban mobility and public transit systems (F7)	5.68	5.95	−0.27	20.327	0.001
Effective management of vehicular traffic (F8)	5.61	6.31	−0.70	10.494	0.001
Strategies to regulate population density (F9)	5.80	5.91	−0.11	11.933	0.001
Efforts to mitigate air pollution (F10)	5.38	6.02	−0.64	4.200	0.001
Strategies for noise disturbance abatement (F11)	5.36	5.89	−0.53	3.537	0.001
Measures to combat water pollution (F12)	5.36	5.76	−0.40	2.942	0.003
Effective systems for waste disposal and sewage treatment (F13)	6.29	6.10	0.19	9.248	0.001
Protection and maintenance of natural areas (F14)	6.26	6.02	0.24	9.254	0.001
Presence of lush greenery and open spaces like parks (F15)	6.27	6.10	0.17	14.703	0.001
Upkeep of civic infrastructure and architectural aesthetics (F16)	5.22	5.62	−0.40	16.533	0.001
Sufficiency of the average monthly earnings (F17)	5.63	5.65	−0.02	−3.047	0.002
Balance between the cost of living and affordability (F18)	4.12	5.47	−10.35	−4.125	0.001
Availability of housing options within economic reach (F19)	5.61	5.34	0.27	4.741	0.001
Economic vibrancy and robustness of the city (F20)	6.02	5.91	0.11	5.397	0.001
Diversity within the city’s economic fabric (F21)	6.07	5.97	0.10	1.708	0.088
Plenitude of professional prospects (F22)	6.10	6.08	0.02	3.898	0.001
Safeguarding of cultural landmarks and traditions (F23)	5.24	4.83	0.41	14.445	0.001
The sense of camaraderie and identification with the community (F24)	5.19	5.65	−0.46	19.133	0.001
Presence of cultural hubs such as museums and theatres (F25)	5.56	5.52	0.04	0.954	0.341
Plenty of opportunities for cultural engagement and amusement (F26)	5.40	5.24	0.16	−7.623	0.001
Average	5.700 *	5.756 *

Note: * performance and importance scores are based on a 1 to 5 Likert scale.

.

5.4. Results of Exploratory Factor Analysis (EFA)

The Kaiser–Meyer–Olkin (KMO) value achieved an acceptable level for each quadrant’s dataset, and Bartlett’s test of sphericity indicated that the data were suitable for factor analysis with significance at p < 0.001. Four quadrants were identified in the table, and EFA was performed on each quadrant to identify several components, with loadings exceeding the 0.50 threshold considered acceptable. Separate factor analyses were carried out for resident loyalty and satisfaction while excluding factors with low loadings. Findings show that the ‘Keep Up the Good Work’ quadrant had three components: social quality, connectivity conditions, and environmental quality; accounting cumulatively for 29.89%, 21.41% and 14.44%, with a cumulative variance of 65.74% (

Table 3. IPA matrix results.

Quadrant	Item
Keep up the Good Work	The assurance of safety and serenity in the city; Caliber and comfort of residential accommodations; Superiority and accessibility of health-related services; The standard of educational facilities and offerings; Integrated urban mobility and public transit systems; Effective management of vehicular traffic; Effective systems for waste disposal and sewage treatment; Protection and maintenance of natural areas; Presence of lush greenery and open spaces like parks.
Concentrate Here	Efforts to mitigate air pollution; Strategies for noise disturbance abatement; Measures to combat water pollution.
Low Priority	Upkeep of civic infrastructure and architectural aesthetics; Balance between the cost of living and affordability; Safeguarding of cultural landmarks and traditions; The sense of camaraderie and identification with the community; Presence of cultural hubs such as museums and theatres.
Potential Overkill	Sufficiency of the average monthly earnings; Availability of housing options within economic reach; Availability of sports and recreational amenities; Proximity and accessibility to shopping centres and dining establishments.

). In the ‘concentrate here’ quadrant, only one component, named ‘Environmental Pollution’, explained a cumulative variance of 66.73%. Two-components were found in the ‘Low Priority’ quadrant: Component I was renamed as Variety, and Component II as Heritage. These accounted cumulatively for 69.112% variance. In addition, the ‘Potential Overkill’ quadrant contained two components; Component I, renamed Economic Quality, and Component II, renamed Recreational Facility. They both explained a total of 67.238% cumulative variance.

This study assessed loyalty using four items, while satisfaction was evaluated with two items. These items underwent factor analysis to identify the underlying factors. For loyalty, one factor emerged (KMO = 0.755, p < 0.001), accounting for 78.14 per cent of the variance with an α of 0.899 (refer to Table 3). Similarly, analysis of satisfaction also revealed one factor (KMO = 0.500, p < 0.00), explaining 88.22% of the variance with an α of 0.866 (See Table 3). The structural equation modelling (SEM) study used these factors as observable variables to quantify loyalty and satisfaction.

The existence of subscales in both quadrants justifies the incorporation of a second-order component in the structural equation model.

5.5. Structural Equation Modelling in Addition to the Importance–Performance Quadrants

The structural model’s paths establish hypothesised relationships between latent variables. Path coefficients are a statistical tool that the researcher can use to validate or invalidate their hypotheses and get insight into the dynamic nature of the relationship between the dependent and independent variables. Path coefficients, also known as standardised beta coefficients, are a kind of regression coefficient that may be determined using the ordinary least squares method. In conjunction with t-statistics, the bootstrapping approach is used to ascertain whether or not route coefficients are statistically significant (Figure 6).

Table 4. Exploratory factor analysis.

Domain	Indicator	Component I	Component II	Component III	Cronbach Alpha(α)
‘Keep up the Good Work’ quadrant:	IP1	0.818			0.786
	IP2	0.798
	IP3	0.864
	IP4	0.754
	IP7		0.757		0.689
	IP8		0.831
	IP13			0.891	0.725
	IP14			0.864
	IP15			0.592
Eigenvalues		2.974	1.757	1.186
% of Variance		29.890	21.415	14.441
Cumulative Percentage		29.890	51.305	65.746
KMO = 0.738, p ≤ 0.001 ** Items 9, 20, 21, and 22 were omitted from the factor analysis because their loading on the factor was less than 0.45.
‘Concentrate Here’ quadrant:	IP10	0.810			0.749
	IP11	0.808
	IP12	0.832
Eigenvalues		2.002
% of Variance		66.731
Cumulative Percentage		66.731
KMO = 0.738, p ≤ 0.001
‘Low Priority’ quadrant	IP16	0.743			0.610
	IP18	0.814
	IP23		0.813		0.760
	IP24		0.853
	IP25		0.888
Eigenvalues		2.366	1.090
% of Variance		44.112	25.000
Cumulative Percentage		44.112	69.112
KMO = 0.718, p ≤ 0.001 ** Item 26 was eliminated from consideration as a factor because its loading on factors was less than 0.45
‘Potential Overkill’ quadrant	IP17	0.837			0.672
	IP19	0.819
	IP5		0.799		0.785
	IP6		0.808
Eigenvalues		1.600	1.089
% of Variance		34.569	32.669
Cumulative Percentage		34.569	67.238
KMO = 0.60 p ≤ 0.001
Resident’s Loyalty	RL 1	0.896			0.899
	RL 2	0.915
	RL 3	0.769
	RL 4	0.946
Eigenvalues		3.126
% of Variance		78.148
Cumulative Percentage		78.148
KMO = 0.755, p < 0.001
Resident’s Loyalty	RS1	0.939			0.866
	RS2	0.829
Eigenvalues		88.226
% of Variance		88.226
Cumulative Percentage		0.866
KMO = 0.720, p < 0.00

** IP importance Performance. Detailed indicator refer to the Supplementary Materials.

provides extensive information on the path coefficients, t-statistics, and degrees of significance for each of the hypothesised correlations. Using the findings from the pathway analysis, each submitted hypothesis is either accepted or rejected. The following section will go through these findings in more detail.

5.6. Hypotheses Testing

In order to verify the proposed hypotheses and the structural model, we analysed the path coefficients between latent variables. The value of the path coefficient has to be at least 0.1 for a specific impact to be considered when the model is being developed [94,95]. There is evidence to support five of the hypotheses based on the path coefficients in this model (Table 4). The supported hypotheses were statistically significant at the level of 0.05.

It indicates that green space, safety, and heritage, which are all first-order constructs, are all linked to the second-order constructs in a way that is significant and positively related to ‘Keep Up the Good Work’, which has a significant, positive relationship with overall satisfaction (β = 0.067, t = 2.055, p < 0.05). Therefore, hypothesis one is supported.

This study found no significant connections between the other three concepts, i.e., ‘Concentrate Here’, ‘Low Priority’, and ‘Potential Overkill’, and overall satisfaction. Notably, the ‘Concentrate Here’, concept was associated with lower overall satisfaction (β = −0.008, t = 2.168, p < 0.05), providing partial support for H2. The ‘Low Priority’ and ‘Potential Overkill’ concepts had coefficients of (β = −0.009, t = 2.168, p > 0.292) and (β = −0.034, t = 0.628, p > 0.053), respectively, indicating that these two quadrants had minimal or no influence on overall satisfaction. Consequently, hypotheses three and four were fully supported. The positive relationship between overall satisfaction and loyalty (r = 0.743, t = 29.265, p < 0.05) confirmed hypothesis five. A summary of the results for all tested hypotheses can be found in

Table 5. Results of hypothesis testing.

Hypothesized Relationships	β	SD (σ)	t Value	p-Value	Decision
H1 Keep up the Good Work—Overall Satisfaction	0.067	0.071	2.055	0.000 **	Accepted
H2 Concentrate Here—Overall Satisfaction	−0.008	0.050	2.168	0.000 **	Accepted
H3 Low Priority—Overall Satisfaction	−0.009	0.063	0.161	0.292	Accepted
H4 Potential Overkill—Overall Satisfaction	−0.034	0.053	0.628	0.530	Accepted
H5 Overall Satisfaction—City Loyalty	0.743	0.025	29.265	0.000 **	Accepted

** p < 0.001.

.

6. Discussion and Conclusions

In order to evaluate the sustainability of the city, the current research sought to examine the connections between importance–performance, satisfaction, and loyalty in the setting of Siliguri City. This study provides a thorough insight into the inhabitants’ impressions of their urban environment and its influence on overall satisfaction and loyalty by assessing the significance and performance of all attributes. The results have significant ramifications for urban planning, policy development, and the formulation of plans to enhance citizens’ quality of life.

The ‘Keep Up the Good Work’ quadrant, which includes social quality, connectivity conditions, and environmental quality, significantly and favourably influenced overall satisfaction, according to the findings, which supported hypothesis 1. This result is consistent with other studies, e.g., ref. [96,97,98], highlighting the critical importance of green spaces, safety, and the preservation of natural areas in raising inhabitants’ contentment and overall quality of life. It highlights the need to maintain and improve these qualities to maintain a high level of pleasure among the residents of Siliguri. To preserve a good quality of life for people, urban planners and legislators should prioritise the preservation and growth of open areas, invest in safety measures, and safeguard the city’s cultural legacy.

The ‘Concentrate Here’ quadrant, which includes characteristics connected to environmental pollution, revealed a negative but non-significant connection with overall satisfaction, contrary to expectations, which partly supported hypothesis 2. The result suggests that even if people value these elements, they may not significantly impact their overall satisfaction levels. Residents may have become acclimated to the city’s water, air, and noise pollution, or consider these problems unavoidable in an urban environment, which is one explanation for this conclusion. This is consistent with earlier studies that identified environmental pollution as a significant issue in Siliguri [39,99,100,101]. However, addressing environmental pollution is essential for ensuring sustainable growth and raising the standard of living for locals. To reduce pollution levels in the city, policymakers should put strong pollution control measures into place, encourage ecologically friendly behaviours, and invest in renewable energy sources.

Contrary to the expectations set out by hypotheses 3 and 4, the ‘Low Priority’ and ‘Potential Overkill’ categories did not significantly impact respondents’ level of contentment. These results show that inhabitants do not value these characteristics highly when deciding how satisfied they are with their current living situation. This finding has real-world implications for helping city planners and legislators choose initiatives that significantly impact residents’ happiness. Instead of putting too much money into leisure centres, governments might invest in cleaner air, safer streets, and more environmentally friendly parks.

Finally, the research proved hypothesis 5 by showing a significant correlation between complete satisfaction and continued loyalty. In keeping with prior research on satisfaction and loyalty in various circumstances, e.g., [102,103], this conclusion shows that contented inhabitants are more likely to display allegiance towards their city. An increase in loyalty from satisfied customers might positively affect civic participation, social harmony, and even economic activity in the area.

A significant discovery of this study is that social quality, connectivity, and environmental quality are essential factors that contribute significantly to the overall satisfaction of residents. This highlights the significance of urban planners acknowledging the importance that inhabitants attribute to social connections, convenient transportation, and environmental excellence. Particular interventions such as community development initiatives, improvements to public transportation, and the implementation of sustainable measures can yield favourable outcomes. The aforementioned discovery holds significant importance for comprehending how inhabitants perceive the amalgamation of said characteristics as factors that determine their standard of living. Any developmental initiatives that disregard these facets may not yield favourable outcomes in terms of their overall contentment.

Conversely, the outcomes pertaining to environmental contamination exhibit a degree of unexpectedness [40]. The absence of a noteworthy correlation between pollution and contentment may be attributed to diverse factors, including the residents’ acquiescence to specific degrees of pollution as a compromise for inhabiting an urban setting [37,100,101]. This offers a prospect to explore the psychological and social factors that could potentially impact the perceptions and attitudes of inhabitants towards pollution. Additionally, this research validated the correlation between contentment and allegiance, which holds significant ramifications for the enduring viability of urban areas. Cities that possess a population that exhibits strong loyalty are expected to encounter reduced levels of emigration, heightened degrees of civic involvement, and greater consistency in economic advancement [37,41]. This argument advocates for the establishment of socially sustainable communities in addition to physically sustainable cities.

The IPA–SEM [44,45,46,84,93,100] methodology has the potential to function as a valuable instrument for policymakers and urban planners, not only in India but also worldwide, in facilitating evidence-based decision making. Furthermore, through the identification of the attributes of urban living most highly valued by its inhabitants, this methodology can more efficiently direct investment and policy priorities. The present study holds pedagogical significance as it can serve as a valuable resource in academic contexts for educating students and practitioners in the domains of urban planning, public policy, environmental studies, and allied disciplines regarding the comprehensive assessment techniques. Additionally, it contributes to the academic conversation surrounding urban sustainability, particularly in burgeoning urban areas within developing nations.

Although this study contributes substantially to the practical and theoretical understanding of urban sustainability, it is not exempt from certain limitations. The generalizability of the study is limited due to its exclusive focus on Siliguri City [38,100,101]. Furthermore, there can be substantial variations in cultural, socio-economic, and governance contexts, even within a single nation. Hence, prudence must be exercised while extrapolating these results to other urban areas. Furthermore, the cross-sectional design is constraining as it only captures the perceptions and attitudes at a singular moment in time. Longitudinal investigations could yield a greater understanding regarding the development of these perceptions and attitudes over time, particularly in reaction to policy modifications or noteworthy occurrences. Despite its few limitations, the present study has the potential to measure urban sustainability by considering residents’ perceptions regarding satisfaction and loyalty towards the city.

In conclusion, this research sheds light on the complex interplay between relevance, efficiency, happiness, and commitment in the urban setting of Siliguri. Siliguri is not just a major city in West Bengal, but also a gateway to Northeast India; therefore, these results have crucial implications for urban planning, policymaking, and the creation of initiatives to enhance the quality of life for citizens there. It emphasizes the need of preserving and improving variables in the ‘Keep Up the Good Work’ quadrant for urban planners and politicians to guarantee high levels of overall happiness and loyalty among people. For sustainable urban growth and higher standards of living, reducing environmental pollution must also be a top priority. Moreover, resources should be allocated more effectively by focusing in on the factors that have a major impact on residents’ happiness as opposed to pouring money into areas that have a negligible effect. The findings suggest that urban sustainability is not merely a product of infrastructure and services but is deeply interconnected with social factors and residents’ perceptions. This study adds to the expanding body of literature on urban planning and policymaking by providing evidence-based guidelines for promoting well-being in urban settings and highlighting the relevance of understanding people’s opinions when designing successful policies and initiatives. The implications of these findings are far-reaching for urban planners, policymakers, environmentalists, and social scientists. They collectively emphasize the importance of a balanced approach to urban development that not only focuses on physical infrastructure but also on building strong, sustainable communities. This study paves the way for future research in Siliguri and other urban areas by highlighting the need for longitudinal studies, comparative analysis with other cities, exploration of additional attributes like digital infrastructure and social integration, and investigation of demographic factors’ influence on residents’ perceptions, satisfaction, and loyalty. These methods may help us achieve a deeper comprehension of urban dynamics, which in turn can guide better policy formulation and planning.