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

PLOS One | https://doi.org/10.1371/journal.pone.0328058 July 10, 2025 8 / 22 and validate the primary findings. These included socio-economic and climatic data Quang Binh Statistics Department, and the Department of Climate Change under the Ministry of Natural Resources and Environment (MONRE). These sources provided critical background information on agricultural practices, historical climate variability, and household demographics in the study area. The sample size for the household survey was determined using a simplified formula recommended by Greene (2000) [65] expressed as: n = N 1 + N ∗ e2 (5) where n represents the required sample size, N is the total population, and e is the allowable margin of error, set at 5% in this study. Based on official statistics from the Quang Binh Statistics Office (2021) [48], the combined population of the three selected communes—Chau Hoa, Phong Hoa, and Duc Hoa— were approximately 14,521 individuals. With an average household size of 4.5 persons per household in rural areas [46], the estimated total number of households was 3,226. Applying the above formula yielded a required minimum sample size of 356 households to ensure statistical reli- ability and representativeness. The three communes were purposefully selected due to their high concentration of farming households and significant exposure to climate risks [46,48]. Given the relatively even distribution of households across the selected sites, a proportional random sampling approach was adopted. Accordingly, 119 households were surveyed

in Phong Hoa, 119 in Duc Hoa, and 118 in Chau Hoa. Lists of households were obtained from communal authorities, from which respondents were selected using simple random sampling techniques to minimize selection bias. To ensure ethical compliance and participant engagement, surveys were conducted in the evenings when household members were more likely to be present. Prior to the interviews, each participant was provided with a clear explanation of the study’s objectives and procedures. Informed consent was obtained in writing through a consent form, in which partici- pants voluntarily indicated their willingness to participate and signed their names. All selected households agreed to take part in the study. The official survey was conducted over a two-month period, from November to December 2023. Upon completion of data collection, the survey responses were screened for completeness and consistency. Cleaned data were then coded and entered into the Statistical Package for the Social Sciences (SPSS) version 23.0 for quanti- tative analysis. The initial phase of analysis included descriptive statistics, frequency distributions, and cross-tabulations to summarize household characteristics and assess general attitudes and perceptions regarding climate change and its effects on agricultural livelihoods. ANOVA tests were also conducted to examine differences in awareness and adaptive behavior across socio-economic subgroups. To identify the key determinants of adaptation strategy choices, the study

employed the MNL model. This model was appropriate given the categorical nature of the dependent variable—house- holds’ choice of specific adaptation strategies—and allowed for the estimation of the likelihood that a household selects a particular strategy given a set of independent socio-economic, demographic, and institutional variables. The MNL results provided empirical evidence on the significance and direction of each factor’s influence on adaptive decision-making in the context of climate-related stressors. 3. Results 3.1 Socio-economic characteristics of the sample Table 2 presents the socio-economic characteristics of the sampled farming households. The gender distribution of house- hold heads indicated that 59.78% were male and 40.22% were female, suggesting that male-headed households consti- tuted the majority of the sample. The average household size was 5.23 members. Larger households were observed to utilize a broader range of livelihood resources, derive greater income from environmental activities, and exhibit a higher likelihood of adopting climate change adaptation strategies compared to smaller households. The age of the household