Factors influencing farming households’ climate change adaptation strategies in Central Vietnam

PLOS One | https://doi.org/10.1371/journal.pone.0328058 July 10, 2025 13 / 22 sample mean were coded as ‘1’ (i.e., indicating active application of adaptive strategies), while those adopting strategies at or below the average level were coded as ‘0’ (i.e., no or limited adaptation). This dichotomization facilitated the identifi- cation of significant factors distinguishing adaptive versus non-adaptive households. Prior to estimating the MNL model, the potential multicollinearity among independent variables was assessed using Variance Inflation Factor (VIF) [65]. All VIF values were found to be below the threshold value of 2.0, indicating an absence of serious multicollinearity concerns. This confirms that the explanatory variables included in the model were sufficiently independent of one another, ensuring the reliability of coefficient estimates. Model adequacy and overall goodness-of-fit were assessed through the Likelihood Ratio (LR) Chi-square test and Pearson Chi-square statistics. The LR Chi-square value was 81.032, statistically significant at the 1% level (p < 0.01), suggesting that the explanatory vari- ables jointly contribute to the variation in adaptation strategy choices and that the model significantly improves upon the null model with no predictors. Furthermore, the Pearson Chi-square statistic for goodness-of-fit was 182.342, also highly significant (p < 0.01), providing strong evidence that the model has a good fit to the observed data and demonstrates ade- quate explanatory power.

The empirical results from the MNL model indicate that the selection of climate change adaptation strategies among farming households is not random but is systematically influenced by a range of factors. These include household-level socio-economic characteristics (such as education, income, household size, and farming experience), institutional factors (such as access to extension services, information on climate change, and participation in civic organizations), organiza- tional capacity (such as access to credit and irrigation infrastructure), and environmental characteristics (including land- holding size, frequency of extreme weather events, and water availability). Each of these variables exerts a differentiated impact on the probability of selecting a specific adaptive strategy, reflecting the heterogeneity in household capacities and climate exposures. These findings underscore the importance of designing context-specific adaptation policies that are sensitive to the underlying socio-economic and environmental drivers influencing household behavior. In particular, strengthening institutional support systems, enhancing access to agricultural services and information, and addressing resource-related constraints (e.g., water, land, and capital) are crucial for promoting the adoption of effective and sustain- able adaptation practices at the grassroots level (Table 5). Specifically, the findings in Table 6 indicated that gender (GEN) significantly influenced adaptation choices. Male-

headed households were more likely to adopt shrimp farming (+31.6%), lotus-fish farming (+41.2%), and salt-tolerant rice Fig 2. Challenges for climate change adaptation of farmers. Source: Study results (2024). https://doi.org/10.1371/journal.pone.0328058.g002