In the rapidly evolving arena of sustainable tourism development, researchers have embarked on a meticulous journey to refine the tools used for assessing quality management frameworks. A recent pivotal study has illuminated the rigorous process of scale development and validation, incorporating robust methodologies to optimize measurement accuracy and theoretical coherence. This analytical deep dive utilized a comprehensive item analysis followed by exploratory and confirmatory factor analyses, ultimately sculpting a high-fidelity instrument capable of capturing nuanced dimensions critical to the sustainable management of tourism destinations.
Item construction, a foundational phase in scale development, was approached with scientific rigor through an initial pool of items subjected to intense scrutiny. Utilizing key techniques such as the Index of Item-Objective Congruence (IOC) and collinearity diagnostics, researchers ensured that only items with distinctive contributions to the construct were retained. This screening was supplemented by data collected from a targeted survey at a well-known tourist attraction, yielding 165 valid questionnaires after rigorous preprocessing. The effective response rate of 91.7% lent statistical confidence to subsequent analyses.
The study employed a discriminant analysis method to refine the scale further. By segmenting respondents into high and low-scoring groups based on cumulative scores, the team calculated discriminant coefficients to evaluate each item’s ability to differentiate between these cohorts. Items with insignificant t-values—specifically WPM3, OP3, and OP9—were statistically demonstrated to lack discriminative power and were thus excised from the pool. This pruning reduced the initial item set to 33—ensuring that subsequent analyses were rooted in more psychometrically sound data.
Beyond discrimination, reliability formed a crucial lens through which items were assessed. Cronbach’s alpha, a widely-accepted measure of internal consistency, was analyzed in detail. The removal of items FP1, WPM5, and OP6 emerged as necessary to bolster the reliability coefficients of their respective dimensions—FP, WPM, and OP—a process justified by both empirical improvements and theoretical alignment. Post-deletion, alpha values surged above the 0.80 threshold, indicating excellent internal consistency and reinforcing the conceptual integrity of each dimension within the scale.
The suitability of the data for exploratory factor analysis was confirmed through the Kaiser-Meyer-Olkin (KMO) measure and Bartlett’s Test of Sphericity. A KMO value of 0.85 signaled strong sampling adequacy, while a highly significant Bartlett’s test (p < 0.001) established the presence of robust inter-item correlations amenable to factor analysis. These statistical validations fortified the foundation for identifying latent constructs underpinning the data, promising a faithful representation of the theoretical framework in empirical form.
Item communalities and factor loadings further underscored the retention of all items within the item pool. Each variable reflected communalities above 0.2 and loadings over 0.45, benchmarks that advocate for their relevance in the factor solution. This statistical backdrop led to the establishment of five distinct factors: full participation (FP), whole process management (WPM), comprehensive management (CM), tourism benefits (TB), and overall perception (OP), with respective item counts aligning to 5, 5, 5, 9, and 6.
Transitioning to a broader participant base, the study administered 700 surveys, achieving a robust 90.2% effective response rate with 632 valid questionnaires ultimately analyzed. Deploying SPSS for exploratory factor analysis (EFA), the research team capitalized on the enhanced sample size to verify the scale’s dimensionality. An impressive KMO of 0.941 coupled with a Bartlett’s test chi-square of over 11,000 confirmed the dataset’s factorability. The extraction of five factors with eigenvalues exceeding one accounted for nearly 67% of total variance, a solid illustration of the scale’s explanatory power.
The factor structure revealed through principal component analysis with Varimax rotation resonated well with theoretical expectations, as each dimension captured conceptually consistent items. FP items loaded strongly on the fourth factor, WPM on the fifth, CM on the third, TB on the first, and OP on the second. The scree plot corroborated the five-factor solution, signaling the robustness of the construct delineation. This alignment between empirical results and theoretical constructs elevated the credibility of the newly crafted scale.
To ensure structural stability beyond EFA, confirmatory factor analysis (CFA) was implemented using AMOS software on an independent subsample of 290 completed questionnaires. The measurement model exhibited exemplary goodness-of-fit indices: chi-square/degrees of freedom at 1.377; Comparative Fit Index (CFI) and Incremental Fit Index (IFI) both near 0.99; and Root Mean Square Error of Approximation (RMSEA) as low as 0.02. These values far surpass conventional cutoff criteria, substantiating the model’s fitness and confirming the scale’s underlying structure.
Construct validity received rigorous evaluation through convergence and discrimination assessments. Factor loadings for the finalized 30-item scale ranged between 0.69 and 0.87, signaling strong convergent validity. Composite Reliability (CR) values ranged from 0.88 to 0.93, comfortably exceeding the 0.70 benchmark, while Average Variance Extracted (AVE) values were all above 0.50, further validating the convergent credentials of the instrument. These indicators collectively reinforce confidence in the scale’s ability to precisely measure the intended dimensions.
Discriminant validity was gauged through an elegant comparison of the square root of AVEs against inter-factor correlation coefficients. For all five dimensions, the square roots of AVEs surpassed the correlation values between other latent variables, signifying clear conceptual distinctions among factors. Moreover, the observed correlations remained below the 0.75 threshold, as advocated in methodological literature, thereby negating concerns of multicollinearity and underscoring the instrument’s discriminative sharpness.
This methodologically robust research journey—from item generation through discriminant and reliability analyses, onward to factor extraction and measurement modeling—exemplifies the scientific precision needed for scale development in complex applied contexts. The final instrument, possessing both solid psychometric attributes and theoretical fidelity, equips researchers and practitioners with a meaningful tool for assessing sustainable tourism development through the lens of total quality management frameworks.
In an era where sustainable tourism endeavors demand sophisticated evaluative instruments, this study’s contributions hold tremendous promise. Its methodological transparency and empirical thoroughness set a new standard, fostering greater understanding of multidimensional management practices and their perceived outcomes. By ensuring that measurement scales are both reliable and valid, future investigations can build upon this foundation to advance policy-making, operational improvements, and scholarly knowledge alike.
Ultimately, the integration of advanced statistical validations, including discriminant and confirmatory factor analyses, with practical scale refinement processes exemplifies cutting-edge research in social science measurement. This work not only enriches the tourism management literature but also illustrates the critical role of rigorous scale development in driving impactful empirical insights for sustainable development strategies worldwide.
Article Title: Innovative application of total quality management in sustainable tourism development: framework development and empirical analysis
Article References:
Kong, X., Wang, H., Chen, Y. et al. Innovative application of total quality management in sustainable tourism development: framework development and empirical analysis. Humanit Soc Sci Commun 12, 1055 (2025). https://doi.org/10.1057/s41599-025-05139-6
Image Credits: AI Generated
