TL;DR
This paper introduces "Hierarchical Variance Analysis," a novel quantitative method to understand how residents perceive tourism impacts. By combining ANOVA and regression, it reveals that socio-demographic factors like education and type of work significantly influence these perceptions, offering a data-driven approach for sustainable tourism planning.
What Problem Does This Paper Address?
Tourism development, while economically beneficial, often brings complex social, cultural, and environmental impacts. Understanding how local residents perceive these impacts is crucial for sustainable tourism planning and policy-making. However, existing research often struggles to quantitatively disentangle the contributions of various socio-demographic factors and their interactions on these perceptions. This paper aims to fill this gap by providing a robust, hierarchical analytical framework to explore and confirm these relationships.
Methods
The core of this study is the development and application of Hierarchical Variance Analysis. This approach integrates three key statistical procedures:
- Cronbach's Alpha Tests: Used to assess the reliability and internal consistency of the survey questions measuring different tourism impacts (economic, socio-cultural, environmental). This ensures the survey instruments are sound.
- One-Factor ANOVA: Employed to explore the relationship between individual socio-demographic variables (e.g., gender, education level, type of work) and the perceived tourism impacts. This helps identify which factors have a significant effect.
- Linear Regression Analysis: Used to analyze the simultaneous effects of multiple socio-demographic variables on the perceived impacts, providing a more comprehensive understanding of their combined influence.
- Two-Factor ANOVA: Specifically used to confirm and analyze the interaction effects between key socio-demographic variables, such as "Education level" and "Type of work," on tourism perceptions. This is a novel aspect, revealing how these factors jointly shape resident attitudes.
The methodology is designed to be hierarchical, moving from assessing individual factor reliability to exploring their main effects, then their combined influence, and finally their interactive effects.
Datasets
The study utilizes data collected from a structured questionnaire administered to 452 residents of Ly Son Island, Vietnam. This dataset is specific to the research context and was collected through a field trip in September 2018. The questionnaire captured socio-demographic information (gender, marital status, education level, age, type of work, social network) and residents' perceptions of economic, socio-cultural, and environmental tourism impacts using a 5-point Likert scale.
Key Insights and Ideas
- Socio-demographics Matter: The study strongly confirms that who you are (your socio-demographic profile) significantly shapes how you perceive tourism's effects. It's not a one-size-fits-all perception.
- Interactions are Key: The most compelling insight is the significant interaction between "Education level" and "Type of work." This means the impact of education on tourism perception is not uniform; it's heavily influenced by what job you do. For example, the same level of education might lead to very different views on tourism depending on whether you're a farmer or work in the tourism service industry.
- Nuanced Perceptions: Residents don't just see tourism as "good" or "bad." They have distinct views on economic, socio-cultural, and environmental impacts, and these views are differentially affected by their backgrounds.
- A Practical Framework: The Hierarchical Variance Analysis offers a structured, quantitative way to break down complex perceptions, making it easier for policymakers to identify specific groups and issues.
Key Results
- Gender: Had a less marked effect on economic impacts, but males were more optimistic about socio-cultural impacts and less worried about environmental ones compared to females.
- Marital Status: Married residents generally reported more satisfaction with positive economic and environmental impacts and overall assessments.
- Education Level: Higher education levels were associated with greater worry about negative economic and socio-cultural impacts, but also increased satisfaction with positive environmental aspects.
- Type of Work: Farmers and fishermen expressed more concern about negative socio-cultural and environmental impacts compared to those in trade and tourism services. Free laborers showed the most pessimism regarding negative economic impacts.
- Interaction Effects: The interaction between "Education level" and "Type of work" significantly predicted perceptions across all three impact categories (economic, socio-cultural, environmental). For instance, in economic impacts, the "Beyond high-school level" group showed dramatic shifts in perception based on their "Type of work."
Limitations and Future Work
- Questionnaire Design: The authors acknowledge that if the questionnaire is not well-designed, it can violate factor independence, potentially invalidating the regression model.
- Alternative Models: In cases where the Fisher test doesn't yield significant results, suggesting an invalid linear regression model, the authors suggest exploring other advanced models like Structural Equation Models (SEM) or Bayesian networks.
- Generalizability: While the study provides valuable insights for Ly Son Island, a small island destination, future research could explore the applicability of the Hierarchical Variance Analysis to other types of destinations and cultural contexts.
- Deeper Dive into Interactions: Further research could delve deeper into the mechanisms behind the observed interaction effects, exploring the underlying reasons why education and occupation combine to shape perceptions so profoundly.
Summary updated