The influence of rural tourism landscape perception on tourists’ revisit intentions—a case study in Nangou village, China

The study situates rural tourism within China’s Rural Revitalization Strategy, emphasizing its role in optimizing rural industrial structures, integrating primary–tertiary sectors, and enhancing rural incomes. It notes China’s evolution from early agritainment to policy-driven, top-down rural tourism practices that stress local culture, ecological concepts, and industrial integration, while facing challenges of homogenization and coordination among culture, ecology, and economy. The research gap identified is the predominance of supply-side planning studies and lack of demand-side, quantitatively grounded analyses of tourists’ experiences and values. The research question explores how landscape perception influences tourist satisfaction and revisit intention in rural tourism contexts, using Nangou Village (Yan’an, Shaanxi) as a case. The study aims to test a Landscape Perception→Satisfaction→Revisit Intention framework and provide evidence-based optimization strategies for rural tourism landscape planning and product design in line with Chinese cultural characteristics.

Landscape perception theory, rooted in environmental psychology and human geography, examines how individuals cognitively and affectively process environmental stimuli, shaping emotions (e.g., satisfaction, place identity) and behaviors (e.g., approach, revisit). Prior work outlines that destination image and physical landscape environment are key determinants of satisfaction and behavioral intentions. The expectation-discrepancy model suggests satisfaction arises when experiences meet or exceed expectations. In tourism, revisit intention includes intentions to return and recommend a destination. Empirical studies have linked ecological quality, cultural authenticity, recreational activities, and route design to satisfaction and revisit; several demonstrate that satisfaction mediates the effect of image/landscape perception on revisit intention. Gaps remain in fully modeling the path from landscape perception to behavior and translating quantitative findings into landscape optimization strategies. This study posits: H1: Landscape perception positively affects satisfaction; H2: Landscape perception positively affects revisit intention; H3: Satisfaction positively affects revisit intention; H4: Satisfaction mediates the relationship between landscape perception and revisit intention.

Study area: Nangou Village, Gaoqiao Town, Ansai District, Yan’an City, Shaanxi Province, China (~1716 ha), a key rural revitalization model with resources spanning natural ecology, Ansai folk culture, and revolutionary (red) culture, and facilities including sightseeing parks, soil and water conservation demonstration park, agricultural picking park, red culture camps, and themed landscapes. Measures: Three latent variables—Landscape Perception (LP), Satisfaction (SA), Revisit Intention (RI). LP is modeled as a second-order construct with five first-order dimensions and items: Natural Ecology (NE1–NE3), Historical Culture (HC1–HC3), Leisure Recreation (LR1–LR3), Research Experience (RE1–RE4), and Integral Route (IR1–IR3). SA (SA1–SA3) gauges overall quality, expectations, and competitiveness. RI (RI1–RI3) captures loyalty, revisit willingness, and recommendation. All items used a 5-point Likert scale. Questionnaire: Four parts—demographics; cultural image perception; environmental design evaluation; place perception—mapped to the measurement indices. Field survey conducted November 2022; online survey ran Nov 15, 2022–Sep 12, 2023. A total of 344 valid responses were obtained (meeting SEM sample recommendations). Demographic profiles showed balanced gender, broad age distribution dominated by young/middle-aged, varied occupations, mostly middle-to-high education and middle-income respondents. Analysis: SPSS 27.0 was used for descriptive statistics and reliability (Cronbach’s alpha). AMOS 27.0 performed confirmatory factor analysis (CFA) to assess convergent/discriminant validity, and structural equation modeling (SEM) to test hypothesized paths among LP, SA, and RI. Model fit employed indices CMIN/DF, GFI, AGFI, CFI, TLI, RMSEA, and SRMR. Mediation was tested with bias-corrected bootstrap (n=5000) and 95% CIs for indirect, direct, and total effects.

- Reliability: Cronbach’s α for LP subdimensions: NE=0.861; HC=0.831; LR=0.805; RE=0.863; IR=0.822; LP total (16 items)=0.898; SA=0.803; RI=0.845; overall scale (22 items)=0.913, indicating good internal consistency. - Descriptive scores (means): Overall LP=3.748 (near good). Subdimensions ranking: NE (3.976) > HC (3.906) > RE (3.889) > LR (3.836) > IR (3.826). SA=3.625 (between average and satisfactory). RI=3.452 (between average and willing). - CFA: Standardized factor loadings ranged 0.686–0.891; CR>0.8 for all constructs; AVE>0.5, supporting convergent validity. Discriminant validity largely supported as the square roots of AVE exceeded inter-construct correlations (with LP and some subdimensions slightly close but acceptable overall). - Model fit (SEM): CMIN/DF=1.097; GFI=0.949; AGFI=0.936; CFI=0.995; TLI=0.994; RMSEA=0.017; SRMR=0.037, indicating excellent fit. - Structural paths: LP→SA β=0.559 (p<0.001); LP→RI β=0.434 (p<0.001); SA→RI β=0.377 (p<0.001). H1–H3 supported. - Mediation (bootstrap 5000): Indirect effect (LP→SA→RI)=0.221, 95% CI [0.141, 0.314]; direct effect (LP→RI)=0.603, 95% CI [0.456, 0.755]; total effect=0.823, 95% CI [0.682, 0.961]; relative indirect effect 26.79%. H4 supported. - LP subdimensions: All five subdimensions significantly positively loaded on LP (p<0.01). Influence on LP followed IR > HC > LR > NE > RE. All subdimensions correlated positively with RI; reported order of correlations with RI: NE > IR > HC > LR > RE. Historical culture and integral route emerged as having the greatest influence on LP and strong associations with revisit intention. - Preference insights: Tourists showed greater interest in historical culture and research experience projects; one- to two-day and small group tours were common; most visitors were first-time visitors with relatively low current revisit intention levels.

The findings confirm that landscape perception significantly enhances both satisfaction and revisit intention, and that satisfaction partially mediates the perception–behavior link. This validates the theoretical model for rural tourism contexts: high-quality, culturally resonant, and well-integrated landscape environments elevate satisfaction, which in turn strengthens loyalty and intention to return. Decomposing landscape perception reveals that integral route design and historical-cultural expression are pivotal drivers of perceived quality and downstream behavioral intentions. Practically, these results emphasize that optimizing route coherence and thematic integration, and deepening authentic historical-cultural elements, can meaningfully improve satisfaction and foster repeat visitation in rural destinations like Nangou Village.

- Landscape perception exerts significant positive effects on tourist satisfaction and revisit intention, with satisfaction acting as a mediator. - Five LP dimensions—natural ecology, historical culture, leisure recreation, research experience, and integral route—each contribute significantly to satisfaction and revisit intention, with historical culture and integral route most influential. - Average satisfaction and revisit intention are moderate, indicating room for improvement. - Strategy recommendations derived from empirical results: (1) Deeply mine and present vernacular historical-cultural resources to enhance authenticity and distinctiveness; (2) Shape and expand a compelling red culture brand integrated with ecological and folk resources; (3) Create boutique, theme-coherent rural tourism routes that connect and upgrade dispersed attractions and foster integrated industries (e.g., agricultural science, folklore, parent–child leisure, and leisure agriculture). Future research should refine the dimensionality of landscape perception scales, extend data collection across seasons for generalizability, and integrate landscape optimization with broader destination factors such as industrial transformation, service quality, and marketing.

- The division of landscape perception dimensions is partly subjective and exploratory, suggesting potential construct refinement in future work. - Data collection spanned a relatively short period and may not capture seasonal or annual variability, limiting generalizability. - Recommendations focus on landscape-level optimization and should be integrated with other destination factors (industrial structure, planning, branding, service quality) in comprehensive planning.