Socio-Economic Determinants of Climate-Smart Agriculture Adoption: A Novel Perspective from Agritourism Farmers in Nigeria

Abstract

The existing body of scholarly work on the adoption of Climate-Smart Agriculture (CSA) in Africa and Nigeria has predominantly concentrated on the experiences and practices of smallholder farmers. While these studies offer valuable insights into the general factors that influence the adoption of CSA practices, their findings may not be fully applicable to the burgeoning agritourism farmers in Nigeria. This study presents a novel perspective on the socio-economic determinants of CSA adoption among the nascent agritourism farmers in Nigeria. The data were collected through a well-structured questionnaire administered to 436 agritourism farmers in Nigeria. The five mutually inclusive endogenous variables that capture the impact of CSA practices were agroforestry system, improved livestock management, organic farming, crop rotation/intercropping, and farmer field schools. While the agritourism farmers possess moderate experience and education, significant gaps exist in access to critical resources like credit, climate information, extension services, and membership in agritourism cooperatives/associations. The multivariate probit (MVP) model revealed that agritourism farming experience significantly boosts crop rotation/intercropping adoption. Education enhances organic farming uptake but negatively impacts improved livestock management. Similarly, extension services access promotes farmer field schools while discouraging organic farming. Significant negative covariance matrix between CSA practices suggests overlapping demands for limited farm resources.

1. Introduction

Climate change presents an imminent threat to the long-term sustainability and viability of the burgeoning agritourism sector in Sub-Saharan Africa. The increasing frequency and intensity of extreme weather events, including severe floods and prolonged droughts, inflict substantial damage on farm and agritourism infrastructure. These disruptions interfere with planned agricultural activities that form part of the tourism experience, leading to a decline in tourist engagement. Climate change also leads to land degradation and loss of soil biodiversity, which diminishes the natural aesthetic appeal of agritourism farms, resulting in diminished visitor engagement and a downturn in revenue streams for the agritourism farmers [1,2,3].

The agritourism sector in Sub-Saharan Africa encompasses a wide spectrum of agriculture-related tourism activities. These activities include farm stays: offering immersive rural experiences, the exploration of hills and valleys through climbing and hiking, and guided agricultural tours showcasing farming techniques and landscapes. Farm-gate markets provide fresh local produce, while culinary experiences highlight traditional delicacies. Enriching educational programs emphasise agricultural, sociocultural, and environmental sustainability. Recreational farm activities such as fruit picking, animal feeding and petting, horseback riding, harvest festivals, and other indigenous cultural events tied to agriculture also form an integral part of the agritourism activities in Africa. Agritourism, by its very nature, depends on the quality and attractiveness of the agricultural landscape and the broader environment. Typically, agritourism integrates tourism with agricultural activities, attracting tourists to these activities while generating supplementary income for the farmers [4]. Climate change poses a direct threat to the growth and viability of agritourism farms. As a result, agritourism farmers have an inherent drive to adopt climate-smart practices that can help mitigate these risks and ensure the long-term sustainability of their operations.

Climate-Smart Agriculture (CSA) represents a sustainable approach to increasing agricultural productivity and income, reducing greenhouse gas emissions from agriculture, and enhancing the resilience of agricultural systems to respond to the capricious effects of climate change through effective adaptation strategies [5,6]. Ref. [7] noted that CSA offers a transformative framework that addresses the interconnected challenges of food insecurity, climate crisis, and rural underdevelopment. This transformative framework rests on three core objectives: (i) sustainably increasing agricultural productivity and incomes; (ii) adapting to and building resilience against climate change; and (iii) reducing greenhouse gas emissions where possible. In 2010, the Food and Agriculture Organisation proposed CSA as a solution to transform agricultural systems and support food security in a changing climate [6]. CSA is not a completely new strategy, but rather a production philosophy that integrates existing agricultural practices with innovative techniques to optimise resource use, increase agricultural yield, and enhance farmers’ capacity to adapt to climate change, thereby forging a resilient pathway towards sustainable rural development [8].

The novelty of this research lies in its focus on agritourism farmers and the socio-demographic factors that influence their adoption of Climate-Smart Agriculture in Nigeria. Agritourism farmers represent a key demographic for CSA implementation, given their dual roles in agriculture and rural tourism development. Through their efforts, the agritourism sector contributes to the economic sustainability of rural areas by preserving traditional farming practices and maintaining the natural beauty of agricultural landscapes [9]. The adoption of CSA practices by agritourism farmers strengthens their resilience to climate change, enhances the long-term viability and attractiveness of their farms, delivers high-quality, sustainably produced farm products, and provides unique tourism experiences [10]. Examples of CSA practices include the adoption of crop varieties resistant to climate change, effective water management strategies, agroforestry systems, crop rotation, and soil conservation measures. Additional practices include crop diversification, organic farming, the use of renewable energy, waste recycling, etc.

There is a notable research gap regarding CSA adoption within the agritourism sector, both in Nigeria and across Sub-Saharan Africa. A cursory review of the academic literature reveals a limited number of studies examining the confluence of agritourism and CSA practices, with a near total absence of research into the factors influencing CSA adoption among agritourism farmers in Nigeria. The prior-existing scholarly works on the adoption of CSA practices or technologies in Africa and Nigeria have predominantly concentrated on the experiences of smallholder farmers engaged in traditional crops and livestock production [11,12,13,14,15]. This notable scarcity of research highlights a critical gap in our understanding of this emerging and increasingly important sector.

The central issue this research aims to address is the lack of comprehensive empirical evidence on the socio-economic factors influencing the adoption of CSA practices among agritourism farmers in Nigeria. To effectively ascertain this, the researchers conducted a field survey in South–Southern and South–Eastern geopolitical zones of Nigeria to identify the specific socio-economic or institutional factors that either facilitate or impede agritourism farmers’ decisions to adopt CSA practices.

2. Materials and Methods

2.1. Study Area

The research was carried out in South–Southern and South–Eastern geopolitical zones of Nigeria. The zones are prone to the capricious effects of climate change, yet possess unique culture, diverse languages, numerous agricultural festivals/carnivals, and beautiful topography (deltas, waterfalls, valleys, caves, etc.), which makes it well suited for agritourism development. Due to the nascent nature of agritourism in the South–South and South–East geopolitical zones of Nigeria, Section 2.1.1 and Section 2.1.2 explore agritourism growth and development in the study area, amidst the deleterious effects of climate change in the regions.

2.1.1. Exploring the Potential of Agritourism Development in South—South, Nigeria

The South–South geopolitical zone of Nigeria comprises six states, namely Akwa Ibom, Bayelsa, Cross River, Delta, Edo, and Rivers, with a total of 123 Local Government Areas (LGAs) or administrative units [Figure 1]. The South–South is geographically located within the Niger Delta region of Nigeria. The Niger Delta is renowned for its vast oil reserves and intricate network of rivers, creeks, and mangrove forests. The results summarized in Table 1 show that the total land area of the South–South geopolitical zone is approximately 84,600 km², making it slightly larger in size to Austria (83,879 km²) or Serbia (88,361 km²) [Table 1]. The South–South geopolitical Zone of Nigeria has a population of 26 million people, which is around 12% of the total population of Nigeria [16]. The zone is home to over 30 distinct ethnic groups, with each contributing unique customs, cuisines, and artistic expressions [17,18]. Festivals such as the Calabar Carnival in Cross River State and the annual fishing festivals in Delta State highlight the region’s dynamic cultural scene and draw tourists from across Nigeria and beyond. Likewise, historical sites like the ancient city of Benin in Edo State offer glimpses into the rich history of the region, attracting scholars and enthusiasts alike [18,19]. Agritourism serves as a platform to showcase these cultural assets, educating visitors about traditional farming techniques, indigenous crops, as well as traditional dances and festivals, among other activities.

The South–South region holds substantial potential for further growth in agritourism, primarily due to its rich and diverse agricultural heritage, encompassing a wide array of natural resources, animal husbandry, and indigenous crop species [19,20]. This potential for leveraging agriculture for tourism remains largely untapped, offering significant prospects for economic diversification and the broader promotion of the tourism sector within the region. The Nigerian government’s growing recognition of agritourism’s importance, as demonstrated by national initiatives to establish agritourism villages across the country [21], suggests an increasingly supportive environment for the development of this sector in the region.

Meanwhile, climate change poses a critical threat to the South–South geopolitical zone, exacerbating existing vulnerabilities such as coastal erosion, flooding, and oil pollution. Rising sea levels endanger communities along the Atlantic coast, while erratic rainfall patterns disrupt agricultural activities, which are vital to the region’s economy. Other challenges in the region include armed militants, oil theft, gas flaring, and herdsmen attacks. Agriculture remains a cornerstone of livelihoods in the South–South region, with crops like cassava, plantain, yam, and palm oil thriving in its fertile soils. However, environmental degradation and inadequate infrastructure hinder productivity, creating an urgent need for innovative solutions that balance economic growth with ecological preservation. Agritourism stands out as a promising way to foster sustainable rural development, which encourages the adoption of environmentally friendly practices, aligning with global trends toward sustainability [22].

Figure 1. Map of Nigeria, showing South–South Geopolitical Zone. Source: Reprinted from ref. [16,17,23,24].

Figure 1. Map of Nigeria, showing South–South Geopolitical Zone. Source: Reprinted from ref. [16,17,23,24].

2.1.2. Potential of Agritourism Development in South—East Zone of Nigeria

Figure 2 shows the map of South–East geopolitical zone of Nigeria with its constituent provinces/states, namely Abia, Anambra, Ebonyi, Enugu, and Imo, with a total of 95 Local Government Areas (LGAs) [25]. It has a population of about 36 million people (18% of the total population of Nigeria). The major language spoken in the region is Igbo, hence, they are sometimes referred to as the Igbo people. The South–East geopolitical zone covers an area of approximately 30,000 km² [25,26], which is comparable to the size of Albania [Table 1]. Although the South–East is the smallest geopolitical zone in Nigeria, it contributes greatly to the Nigerian economy due to its oil and natural gas reserves, along with a growing commercial agricultural economy. Apart from agriculture, which is the major economic activity, the zone is also known for its commercial and trading activities with numerous small and medium indigenous enterprises [27]. As illustrated in Table 1, the major agricultural produce in the zone are yam, cassava, rice, oil palm, cocoyam, etc. The zone also has solid minerals and natural resources such as crude oil, natural gas, bauxite, iron ore, sandstone, lignite, clay, coal, tin, and columbite. The zone has distinct music referred to as Ogene music, Igbo highlife, Odumodu, Egwu Ekpili, Ikorodo, Ikwokirikwo, and others [25].

The region’s predominantly rural character, coupled with its reliance on agriculture as a primary livelihood source, positions it as an ideal region for agritourism growth and development. Traditional festivals, such as the New Yam Festival, Ekpe, and Ntaka masquerade, celebrated across various communities in Igbo land, highlight the deep-rooted connection between agriculture and cultural heritage. Such events provide an excellent platform for promoting agritourism by attracting visitors eager to experience authentic Igbo customs while supporting local economies [28]. Historical landmarks, including the Ogbunike Caves in Anambra State and the National War Museum in Umuahia, Abia State, further enhance the region’s appeal as a destination for culturally immersive tourism experiences.

Potential agritourism activities in South–Eastern Nigeria are diverse. Farm stays provide accommodations on agricultural farms, offering tourists immersive experiences of rural life in the region [20]. Guided farm tours highlight processes such as crop cultivation, animal husbandry, and agro-processing, delivering educational and interactive experiences to visitors [29]. Additionally, the direct sale of farm-fresh produce, processed produce, and locally handcrafted items through farm stands in community markets strengthens connections between producers and consumers [30,31]. The region’s natural beauty also lends itself to outdoor recreational activities such as fishing in farm ponds, bird watching in agricultural landscapes, nature walks through farmlands, and horseback riding in suitable terrains [32]. Educational experiences can be curated through workshops and classes focusing on traditional farming techniques, the preparation of local cuisine using farm-sourced ingredients, and insights into the cultural practices intertwined with agriculture [33]. Furthermore, agritainment opportunities, such as hosting harvest festivals that celebrate the agricultural calendar, attract a wide range of visitors.

The South–East geopolitical zone faces significant threats from climate change, including erratic rainfall patterns, prolonged dry seasons, and floods, which collectively disrupt traditional farming practices [34]. These climatic challenges exacerbate soil degradation and reduce crop yields, thereby threatening food security and rural livelihoods [35]. In response, agritourism emerges as a viable solution by diversifying income streams for farmers and encouraging the adoption of climate-smart agricultural practices. For instance, farms practicing organic cultivation or engaging in livestock rearing could host tourists interested in learning about sustainable farming techniques, thus creating additional revenue sources while raising awareness about environmental conservation [19].

Figure 2. Map of Nigeria, showing South–East Zone. Source: Reprinted from ref. [25,26,36].

Figure 2. Map of Nigeria, showing South–East Zone. Source: Reprinted from ref. [25,26,36].

Table 1. Summarised description of the study area.

Characteristics	South—South	South—East
Total estimated land area (km2)	84,587	30,000
Coordinates	Longitude 5.05° E to 7.35° E Latitude 4.15° N to 6.2059° N	Longitude 7° E to 9° E Latitude 4° N to 7° N
Population	26 million people	36 million people
Mean household size	6	7
Vegetation	Mangrove Swamps, Freshwater Swamps, Rainforests, Derived Savannas	Rainforests, Highland Vegetation, Freshwater Swamps
Topography	Predominantly low-lying coastal plains and riverine areas	Undulating hills and valleys with some highland areas
Average Temperature Range (°C)	~21–36	~21–34
Average annual rainfall (mm/m2)	1900.71 ± 280.88	2600 ± 500
Common crops	Cassava, yam, plantain, oil palm, rubber, cocoa	Cassava, yam, cocoyam, maize, oil palm, rice
No. of States	6 States (Akwa Ibom, Bayelsa, Cross River, Delta, Edo Rivers)	5 States (Abia, Anambra, Ebonyi, Enugu, Imo)
No. of LGA (No. of respondents)	123	95
Main Soil Types	Sandy loam and alluvial soils, with acidic tendencies in coastal zones	Ferrallitic/ferralsols (red and deep) with patches of loamy soil
Major Livestock/Animal Husbandry	Poultry, goats, and fish farming in riverine communities	Poultry, goats, pigs, and minimal cattle rearing
Dominant Languages/Dialects	Edo, Ijaw, Ibibio, Efik, Urhobo, Itsekiri, and others	Igbo (and multiple Igbo dialects)
Marketing Channels for Agritourism	Heavy reliance on word-of-mouth, social media, radio, and television	Church groups, community associations, diaspora visits, radio, social media, and television,
Peak Agritourism period	Peak agritourism visits often coincide with dry season festivals (e.g., Christmas–New Year) and mild weather from Nov. to Mar.	Agritourist influx spikes during local cultural events (e.g., New Yam festivals, Christmas), typically in Aug./Sep. and Dec./Jan.

Source: Researcher field survey, 2025; with contributions from [16,17,20,25].

Table 1. Summarised description of the study area.

Characteristics	South—South	South—East
Total estimated land area (km2)	84,587	30,000
Coordinates	Longitude 5.05° E to 7.35° E Latitude 4.15° N to 6.2059° N	Longitude 7° E to 9° E Latitude 4° N to 7° N
Population	26 million people	36 million people
Mean household size	6	7
Vegetation	Mangrove Swamps, Freshwater Swamps, Rainforests, Derived Savannas	Rainforests, Highland Vegetation, Freshwater Swamps
Topography	Predominantly low-lying coastal plains and riverine areas	Undulating hills and valleys with some highland areas
Average Temperature Range (°C)	~21–36	~21–34
Average annual rainfall (mm/m2)	1900.71 ± 280.88	2600 ± 500
Common crops	Cassava, yam, plantain, oil palm, rubber, cocoa	Cassava, yam, cocoyam, maize, oil palm, rice
No. of States	6 States (Akwa Ibom, Bayelsa, Cross River, Delta, Edo Rivers)	5 States (Abia, Anambra, Ebonyi, Enugu, Imo)
No. of LGA (No. of respondents)	123	95
Main Soil Types	Sandy loam and alluvial soils, with acidic tendencies in coastal zones	Ferrallitic/ferralsols (red and deep) with patches of loamy soil
Major Livestock/Animal Husbandry	Poultry, goats, and fish farming in riverine communities	Poultry, goats, pigs, and minimal cattle rearing
Dominant Languages/Dialects	Edo, Ijaw, Ibibio, Efik, Urhobo, Itsekiri, and others	Igbo (and multiple Igbo dialects)
Marketing Channels for Agritourism	Heavy reliance on word-of-mouth, social media, radio, and television	Church groups, community associations, diaspora visits, radio, social media, and television,
Peak Agritourism period	Peak agritourism visits often coincide with dry season festivals (e.g., Christmas–New Year) and mild weather from Nov. to Mar.	Agritourist influx spikes during local cultural events (e.g., New Yam festivals, Christmas), typically in Aug./Sep. and Dec./Jan.

Source: Researcher field survey, 2025; with contributions from [16,17,20,25].

2.2. Method of Data Collection

The research data were collected using a well-structured questionnaire administered to agritourism farmers. This questionnaire was thoughtfully designed to obtain detailed quantitative insights into the prevalence of CSA practices among these farmers. Information was gathered on various CSA practices, including agroforestry systems, cover cropping, crop diversification, crop rotation and intercropping, drought-resistant crop varieties, farmer field schools, and greenhouse farming, among others. These practices were identified through both a comprehensive literature review [37,38,39] and a field survey. The field survey revealed that all agritourism farmers in the South–South and South–East geopolitical zones of Nigeria are engaged in two or more climate-smart practices. Additional data collected focused on the socioeconomic characteristics of the agritourism farmers, such as annual income, access to credit, availability of climate-related information, and extension services [Figure 3].

The researchers identified five key CSA practices adopted by agritourism farmers, which were treated as the dependent variables [Figure 3]. These practices include Agroforestry systems (AgrFSys), improved livestock management (ImpLiMgt), organic farming (OrgFarm), crop rotation and intercropping (CropRot), and participation in farmer field schools (FarSch) [40,41]. The adoption status of each practice was analysed as a binary outcome (1 if adopted, 0 otherwise), reflecting the discrete choices made by the agritourism farmers regarding adopting a CSA practice. Similarly, the selection of exogenous variables was informed by extensive CSA adoption literature [8,40,42,43], which highlights key socioeconomic factors influencing farmers’ decisions to adopt these practices. Ethical considerations were thoroughly addressed throughout the data collection process; informed consent was obtained from all agritourism farmers, who were assured of the confidentiality and anonymity of their responses.

2.3. Sample Size Determination

The researchers adopted a two-phase survey approach to gather socioeconomic and CSA data from agritourism farmers in the South–South and South–East geopolitical zones of Nigeria. The first phase was conducted from 29 August to 30 September 2024. The period aligned with peak agritourism activities in the South–East region, including agricultural harvesting ceremonies, cultural events, and numerous festivals and social meetings. During this period, the researcher visited numerous agritourism farms to observe and document essential socioeconomics and CSA data. The second phase involved the use of trained enumerators to collect vital socio-economic and CSA data from the agritourism farmers in the South–South geopolitical zone. This phase, conducted from 15 December 2024 to 15 January 2025, coincided with the yuletide season, which is characterised by heightened agritourism and vibrant cultural festivities and a high influx of visitors/tourists into the region, which facilitated a comprehensive and efficient data collection process.

To ensure that the sample of participants in the survey was representative of the agritourism farmers in the two geopolitical zones, a stratified random sampling technique was employed. The first stage involved the selection of two geopolitical zones out of the six geopolitical zones in Nigeria. The selected geopolitical zones were the South–South and South–East zones. Since the geopolitical zones in Nigeria are usually demarcated into States, of which a subset of the States comprises Local Government Areas (LGAs), thus, the South–South geopolitical zone comprises six states, namely Akwa Ibom (with 31 LGAs), Bayelsa (8 LGAs), Cross River (18 LGAs), Delta (25 LGAs), Edo (18 LGAs), and Rivers (23 LGAs), making it a total of 123 LGAs. Similarly, the South–Eastern geopolitical zone encompasses the following states: Abia (17 LGAs), Anambra (21 LGAs), Ebonyi (13 LGAs), Enugu (17 LGAs), and Imo (27 LGAs), with a total of 95 LGAs. Altogether, there are 218 LGAs (123 in South–South and 95 South–East).

With the assistance of trained enumerators, the researchers randomly selected two (2) agritourism farmers from each LGA in both the South–East and South–South geopolitical zones of Nigeria. In the 2nd stage, which was carried out between 29 August and 30 September 2024 in the South–East zone, two (2) agritourism farmers were randomly selected from each of the 95 LGAs in the region, yielding 190 agritourism farmers. Due to the emerging state of agritourism in Nigeria, farmers across different LGAs tend to share dissimilar characteristics in terms of scale, operations, and market engagement, making the random selection an effective method of ensuring even representativeness. In the 3rd stage, which was carried out from 1 December 2024 to 15 January 2025, two (2) agritourism farmers were also randomly selected from each of the 123 LGAs in the South–South zone, yielding 246 agritourism farmers. Together, the total sample size for this study comprised 436 agritourism farmers (190 in the South–East and 246 in the South–South geopolitical zones of Nigeria). This sampling strategy was employed to account for the geographical dispersion and relatively limited number of agritourism farmers across the geopolitical zones, states, and LGAs. Each LGA typically hosts between eight (8) and twenty (20) agritourism farmers or more, with variations influenced by factors such as proximity to state capitals, vicinity within game reserve or forest, propinquity within higher education institutions (most especially when it is located in the rural area), natural features like mountains, rivers/streams, and caves, proximity to airports, seaports, and other key infrastructure.

The selection of two (2) agritourism farmers per LGA was primarily driven by the need for broader geographical coverage. Due to the nascent and dispersed nature of agritourism operations in the study area, the random selection processes within each LGA were crucial. The random approach aimed to reduce selection bias by increasing the likelihood of selecting farmers with divergent scales of operations, different levels of agritourism activities, and varying engagement levels with CSA practices.

The major reasons for undertaking the research in South–South and South–Eastern geopolitical zones of Nigeria are that both zones experience high vulnerability to climate change [39], yet with very rich cultural heritage, diversified indigenous crop species, and noteworthy opportunity for agritourism development. The zones are characterised by diverse ecosystems, ranging from the wet, low-lying areas of the Niger Delta in the South–South to the more undulating terrains of the South–East, which experiences similar climatic challenges like flooding, irregular rainfall patterns, landslides, etc. Moreover, the South–South and South–East are historically rich in cultural heritage, indigenous crop species, traditional farming practices, etc., which provide a significant avenue for the growth and development of agritourism in Nigeria. This underscores the rapid rise of agritourism farms and enterprises in the regions. Basically, the two regions show an untapped market potential and the opportunity for substantial growth and development of agritourism in the country. The choice of these regions also addresses an existing gap in academic research. While there is considerable research on CSA in conventional farming systems, no attention has been paid to its application within the niche of agritourism.

2.4. Analytical Techniques

2.4.1. Determinants of CSA Adoption Among Agritourism Farmers in Nigeria: The Use of Multivariate Probit (MVP) Model

The multivariate probit (MVP) model was employed to analyse the socio-economic determinants of Climate-Smart Agriculture adoption among the agritourism farmers in the study area. Traditional univariate models, such as the binary logit or probit models, typically examine adoption decisions in isolation. However, in reality, farmers’ choices are often correlated. For instance, the decision to adopt improved seed varieties might be linked to the decision to use specific types of fertilisers or to implement certain soil conservation techniques. These correlations arise due to the complementarity of practices, where the effectiveness of one practice is enhanced by the adoption of another, or due to substitutability, or where farmers might choose one practice over another to achieve a similar goal [44,45]. The multivariate probit model addresses these limitations by simultaneously estimating a system of binary outcome equations, one for each adoption decision.

Unlike the univariate probit models, the MVP model acknowledges that the error terms associated with the adoption of different practices may be correlated, reflecting the influence of shared unobserved factors [44,46]. By jointly modelling these adoption decisions, the MVP model provides a more accurate and nuanced understanding of the factors driving the adoption of multiple technologies and/or practices. This is because farmers’ decisions to adopt different practices are often interdependent, influenced by a variety of shared unobserved factors, such as risk preferences, access to information networks, or managerial skills. Univariate models treat each adoption decision in isolation, failing to account for the potential correlations in these unobserved factors across different adoption choices.

Since the agritourism farmers employ more than two interdependent CSA practices, the MVP model is the best fit to analyse the multiple adoption decision. Previous studies also confirm that the MVP model is suitable for jointly estimating the adoption decisions of more than two correlated CSA practices. This approach effectively captures a range of determining independent variables while accounting for potential correlations among unobserved disturbances [7,47,48].

To analyse the determinants of CSA adoption among agritourism farmers in the study area, the researcher chose five CSA practices adopted by the agritourism farmers in the study area (based on field survey and literature review), as measurable effects or dependent variables. R-studio was employed to analyse the data. The exogenous variables were the associated socioeconomics/institutional data of the agritourism farmers.

2.4.2. Model Specifications

The various exogenous/independent variables are categorised as follows:

X₁ (AgExp) = Years of experience in agritourism (years);

X₂ (Edu) = Highest level of education (years);

X₃ (HHZ) = Household size (no. of persons);

X₄ (FMZ) = Farmland size (hectares);

X₅ (Cred) = Access to credit: It takes the value 1, if the agritourism farmer indicated access to formal financial products such as loans from banks or microfinance institutions, or formal credit facilities from cooperatives, it does not include subsidies, grants, or general financial aid programs; thus, it is 0 otherwise;

X₆ (AgY) = Agritourism annual income: It is the amount of money the agritourism farmers earn per annum in Nigerian Naira (NGN). It was converted to the United States Dollar $ (USD) for universality, at the average exchange rate in January 2025, which was 1 USD = NGN 1533.79;

X₇ (InfoC) = Access to climate information (Yes = 1, No = 0);

X₈ (Coop) = Membership in agritourism network/cooperative. It takes the value 1, if the agritourism farmers indicate they are members of an agricultural cooperative/social group, 0 otherwise;

X₉ (Teno) = Land Tenure Status. It takes the value 1 if the agritourism farmers are the landlord or own the land in which they engage in agritourism, 0 implies that the land is leased/communal land/otherwise;

X₁₀ (Ext) = Access to Agric extension services (Yes = 1, No = 0).

Model Components:

The outcome of CSA adoption is modelled using a random utility framework adapted from Kassie and Teklewold’s study on multiple sustainable agricultural practices in rural Ethiopia and Tanzania [44,45] (with modifications presented here). Consider the ith agritourism farmer:

$$\begin{gathered} y_1=1(\alpha y_2+x_1{\beta }_1+{\epsilon }_1>0), \\ {y}_2=1(x_2{\beta }_2+{\epsilon }_2>0), \\ {y}_n=1(x_n{\beta }_n+{\epsilon }_n>0), \\ ({\epsilon }_1,{\epsilon }_{2\dots .}{\epsilon }_n|x_1,x_{2,\dots .}{x}_n)~N(\mathrm{0,0},\mathrm{1,1},\dots n\rho ), \end{gathered}$$

(1)

where $1(.)$ is the indicator function taking the value one if the statement in the brackets is true and zero otherwise. $\alpha ,{\beta }_1,{\beta }_2$ are regression coefficients, and $N(.,.,.,.,\rho )$ indicates the standard bivariate normal distribution with correlation coefficients $\rho$, where variances have been normalised to one without loss of generalization.

There are five latent variables (Y_i1^∗,Y_i2^∗,Y_i3^∗,Y_i4^∗,Y_i5^∗), one for each CSA practice (k = 1 to 5) corresponding to each (ith) agritourism farmer. The latent variables serve as the explained outcome or predicted variable. They are expressed as follows:

Y₁ (AgrFSys): Agroforestry system. It takes the value 1 if the ith agritourism farmers practice agroforestry, 0 otherwise.

Y_i1^∗ = β₁₀ + β₁₁AgExp_i + β₁₂Edu_i + β₁₃HHZ_i + β₁₄FMZ_i + β₁₅Credi + β₁₆AgY_i +
β₁₇InfoC_i + β₁₈Coop_i + β₁₉Teno_i + β_1,10Ext_i + ϵ_i1.

(2)

Y₂ (ImpLiMgt): Improved livestock management. Value of 1 if the sampled agritourism farmers adopt an improved livestock management system, 0 otherwise.

Y_i2^∗ = β₂₀ + β₂₁AgExp_i + β₂₂Edu_i + β₂₃HHZ_i + β₂₄FMZ_i + β₂₅Cred_i + β₂₆AgY_i +
β₂₇InfoC_i + β₂₈Coop_i + β₂₉Teno_i + β_2,10Ext_i + ϵ_i2

(3)

Y₃ (OrgFarm): Organic farming. It takes the value 1 if the agritourism farmers engage in organic farming, 0 otherwise.

Y_i3^∗ = β₃₀ + β₃₁AgExpi + β₃₂Edui + β₃₃HHZ_i + β₃₄FMZ_i + β₃₅Cred_i + β₃₆AgY_i +
β₃₇InfoC_i + β₃₈Coop_i + β₃₉Teno_i + β_3,10Ext_i + ϵ_i3

(4)

Y₄ (CropRot): Crop rotation and intercropping. Takes the value of 1 if agritourism farmers practice crop rotation, 0 otherwise.

Y_i4^∗ = β₄₀ + β₄₁AgExp_i + β₄₂Edu_i + β₄₃HHZ_i + β₄₄FMZ_i + β₄₅Cred_i + β₄₆AgY_i +
β₄₇InfoC_i + β₄₈Coop_i + β₄₉Teno_i + β₄,₁₀Ext_i + ϵ_i4

(5)

Y₅ (FarSch): Farmer field schools. It has the value of 1 if agritourism farmers are involved in farmer field school, 0 otherwise.

Y_i5^∗ = β₅₀ + β₅₁AgExp_i + β₅₂Edu_i + β₅₃HHZ_i + β₅₄FMZ_i + β₅₅Cred_i + β₅₆AgY_i +
β₅₇InfoC_i + β₅₈Coop_i + β₅₉Teno_i + β_5,10Ext_i + ϵ_i5

(6)

Y_ik^∗ = Latent (unobserved) variable for agritourism farmer i and practice k.

AgExp_i, Edu_i, …, Exti represent the observed values of the 10 independent variables (X₁ to X₁₀) for the ith agritourism farmer.

β_k0 = Intercept for equation k.

β_k1 to β_k,10 = Parameters (coefficients) to be estimated for equation k, representing the effect of each independent variable on the adoption decision of practice k.

ϵ_ik = Unobserved error term for agritourism farmer i and practice k.

Y_ik = 1 if Y_ik^∗ > 0 (if the sampled agritourism farmer i adopts practice k)

Y_ik = 0 if Y_ik^∗ ≤ 0 (if the agritourism farmer i does not adopt practice k)

The error terms (ϵi₁, ϵi₂, ϵi₃, ϵi₄, ϵi₅) jointly follow a multivariate normal (MVN) distribution with a mean vector of zeros and a 5 × 5 variance-covariance matrix Ω.

Error Vector: ϵ_i = (ϵ_i1,ϵ_i2,ϵ_i3,ϵ_i4,ϵ_i5), Distribution: ϵ_i∼MVN (0, Ω)

Variance–Covariance Matrix (Ω):

Ω =

1	ρ12	ρ13	ρ14	ρ15
ρ21	1	ρ23	ρ24	ρ25
ρ31	ρ32	1	ρ34	ρ35
ρ41	ρ42	ρ43	1	ρ45
ρ51	ρ52	ρ53	ρ54	1

Diagonal elements = 1 (variances are normalised to 1).

The off-diagonal elements (ρ_jk) represent the correlation coefficient between the error terms of equation j and equation k (ρ_jk = ρ_kj). These measure the interdependence between the adoption decisions for practices j and k.

2.5. Diagnostics Test

2.5.1. Check for Collinearity

Before the analysis of Multivariate Probit (MVP), an essential diagnostic procedure involves assessing collinearity among explanatory variables. Collinearity poses a severe concern in econometric modelling as it can inflate the standard errors and obscure the true effects of independent variables on dependent outcomes. To mitigate these potential distortions, the Variance Inflation Factor (VIF) was employed as a diagnostic tool, providing a quantifiable measure of multicollinearity. A commonly accepted threshold for VIF is 5, beyond which multicollinearity is considered problematic and may necessitate corrective actions such as variable transformation or exclusion.

Variable	VIF Value
X1	1.4852
X2	1.0301
X3	1.0102
X4	2.9655
X6	2.9328

The VIF is intended for continuous independent variables (i.e., X₁, X₂, X₃, X_4, and X₆). The computed VIF values indicate that multicollinearity is not a substantial issue in this dataset. Agritourism experience (X₁) exhibits a VIF of approximately 1.49, suggesting negligible collinearity. Education level (X₂) and household size (X₃) report even lower VIF values of 1.03 and 1.01, respectively, affirming their independence from other predictors. Farmland size (X₄) and agritourism annual income (X₆) present slightly higher VIF values of 2.97 and 2.93, respectively, though these remain well within acceptable limits.

Test	BP Stat
Breusch–Pagan	1.4786
df	5
p_value	0.9155

2.5.2. Test for Heteroskedasticity

The Breusch–Pagan (BP) test is employed to detect heteroskedasticity in a regression model. Heteroskedasticity occurs when the variance of the residuals is not constant across the observations. This leads to inefficient standard errors, misleading statistical inference. The BP test helps address this issue by examining whether the residual variance systematically depends on the explanatory variables.

In the MVP analysis, verifying homoskedasticity is essential to ensure the accuracy of estimated coefficients and maintain statistical efficiency. Since MVP models involve interdependent binary choices, unexplained variability in residuals could distort inference and reduce the reliability of parameter estimates.

The results of the conducted BP test indicate that heteroskedasticity is not a concern in the model. The BP statistic (1.4786) is relatively low, and with 5 degrees of freedom, the p-value (0.9155) is well above the standard significance threshold of 0.05. This leads to the failure to reject the null hypothesis, confirming that the residual variance remains constant across different levels of the explanatory variables.

2.5.3. Standardization

Before proceeding with the MVP analysis, we standardised the continuous independent variables AgExp (X₁: years of experience in agritourism), Edu (X₂: highest level of education), HHZ (X₃: household size), FMZ (X₄: farmland size), and AgY (X₆: agritourism annual income) to ensure consistency in scale. Standardization is a statistical transformation technique that adjusts the values of a variable to have a mean of zero and a standard deviation of one. Given the heterogeneity in measurement units and value ranges across these variables, their inclusion in the model without standardization could lead to disproportionately weighted coefficient estimates, affecting the validity and robustness of the inferential statistics. Following the standardization process, we proceeded with the MVP analysis.

3. Results

3.1. Socioeconomics and CSA Characteristics of Agritourism Farmers in Nigeria

This section portrays selected socio-economic characteristics of the agritourism farmers, as well as the statistics of the adoption of CSA practices in the study area.

Table 2. Socio-economic and CSA characteristics of agritourism farmers in Nigeria.

Socioeconomic Variables	Mean	Std. Dev	Min	Max
Agritourism experience (years)	7.11	2.85	1	12
Highest level of education (No formal education = 0, primary = 6, secondary = 12, tertiary = 16, postgraduate = 20 years)	13.60	4.27	0	20
Household size (number of persons)	5.90	2.19	2	14
Farmland size (hectares)	2.09	1.08	0.50	6.00
Access to credit facilities (Yes = 1, No = 0)	0.47	0.50	0	1
Annual income from agritourism (USD $)	3730.69	857.79	1950	6250
Access to climate information (Yes = 1, No = 0)	0.48	0.50	0	1
Membership to agritourism network/cooperative (Yes = 1, No = 0)	0.22	0.42	0	1
Land tenure status (if the agritourism farmer is the landlord or owns the land = 1; if the land is leased/communal/otherwise = 0)	0.67	0.47	0	1
Access to agric extension services (Yes = 1, No = 0)	0.33	0.47	0	1
Climate Smart Agricultural Practices
Agroforestry system (Yes = 1, No = 0)	0.78	0.41	0	1
Improved Livestock Management (Yes = 1, No = 0)	0.59	0.49	0	1
Organic Farming (Yes = 1, No = 0)	0.80	0.40	0	1
Crop Rotation and Intercropping (Yes = 1, No = 0)	0.82	0.38	0	1
Farmer Field Schools (Yes = 1, No = 0)	0.55	0.50	0	1

Source: Researchers’ field survey data, 2025.

shows the socioeconomic characteristics of the agritourism farmers in the study area, alongside their CSA adoption statistics. The sampled agritourism farmers reported an average of 7.11 years of experience in agritourism farming (SD = 2.85), with individual experience ranging from 1 to 12 years. Educational attainment within the sample averaged 13.6 years (SD = 4.27), signifying that the agritourism farmers had completed secondary education and potentially some form of subsequent training or tertiary education. The educational spectrum spanned from farmers with no formal schooling to those possessing postgraduate qualifications. Household composition of the agritourism farmers averaged 5.90 or ≈ 6 persons (SD = 2.19), with a minimum of 2 and a maximum of 14 members per household. Regarding farm structure, the mean farmland size managed by the agritourism farmers was 2.09 hectares (SD = 1.08), varying from 0.50 to 6.00 hectares. According to Figure 3, the average annual income generated from agritourism activities stood at USD 3,730.69 (SD = 857.79), with reported incomes between USD 1,950 and USD 6,250. Land tenure analysis indicated that 67% of the agritourism farmers owned the land utilised for their operations.

Access to institutional support systems varied. Access to credit was available to only 47% of the agritourism farmers, while access to climate information was available to 48% of the agritourism farmers. Membership in agritourism networks or cooperatives was relatively low, at 22%. Furthermore, barely 33% of the randomly sampled agritourism farmers indicated they had access to agricultural extension services. The adoption levels of the assessed CSA practices demonstrated considerable engagement among the surveyed farmers. Crop rotation and intercropping showed the highest adoption rate at 82%, followed by organic farming at 80%, and agroforestry systems at 78%. Improved livestock management was practised by 59% of the agritourism farmers, while participation in farmer field schools was adopted by 55%.

The correlation matrix heatmap presented in Figure 4 visually represents the relationships between socio-economic determinants and the adoption of CSA practices among agritourism farmers in Nigeria. The heatmap matrices depict pairwise associations between variables but do not account for multivariate interactions. The colour gradient reflects the strength and direction of relationships, where dark blue indicates strong positive correlations and dark red represents strong negative correlations.

From the heatmap matrix in Figure 4, factors such as agritourism farmers education (X₂), access to credit (X₅), and access to climate information (X₇) show strong positive correlations with multiple CSA practices, suggesting that farmers with higher education levels, financial support, and climate awareness are more inclined to adopt sustainable agricultural methods. Similarly, membership in agritourism cooperatives (X₈) and access to agricultural extension services (X₁₀) are also positively associated with the adoption of CSA practices, indicating the importance of social networks and professional guidance in influencing farming decisions. On the other hand, variables like household size (X₃) and land tenure status (X₉) display weaker correlations, implying a less direct impact on farmers’ adoption behaviours.

Figure 5 shows a boxplot that delineates the disparities in annual income from agritourism among the farmers based on their land tenure status. It contrasts those farmers who own land with those operating under leased, communal, or other landforms of agreement. The two dots above the right-hand box are outliers. The colour difference between the two boxplots serves a categorical purpose: it visually distinguishes the two land tenure groups being compared. The figure shows that agritourism farmers who own land exhibit a higher median income compared to their non-owning counterparts, with the upper quartile extending further, indicating greater income potential. This pattern reflects the structural advantage that landowners hold in terms of capital stability and the ability to plan long-term agritourism investments that could yield substantial financial returns.

3.2. Determinants of CSA Adoption Among Agritourism Farmers in Nigeria

A Multivariate Probit (MVP) model was employed to analyse the factors influencing the adoption decisions of Climate-Smart Agriculture (CSA) practices among the agritourism farmers in Nigeria. Based on the field survey, the CSA practices examined were agroforestry system (Y₁), improved livestock management (Y₂), organic farming (Y₃), crop rotation/intercropping (Y₄), and farmer field schools (Y₅). The MVP framework was chosen to account for the potential interdependence in agritourism farmers’ adoption decisions, recognising that the choice to adopt one practice may be correlated with the decision to adopt another.

Table 3. Results of multivariate probit model for determinants of CSA adoption among agritourism farmers in Nigeria and the covariance matrix of the error terms associated with different CSA practices.

Variables	Various CSA Practices Adopted by the Agritourism Farmers
	AgrFSys (Y1) Coef. (Std. Error)	ImpLiMgt (Y2) Coef. (Std. Error)	OrgFarm (Y3) Coef. (Std. Error)	CropRot (Y4) Coef. (Std. Error)	FarSch (Y5) Coef. (Std. Error)
X1 (AgExp)	−0.0934 (0.0991)	−0.0112 (0.1021)	−0.1104 (0.1161)	0.3532 (0.1097) ***	−0.0791 (0.0885)
X2 (Edu)	−3.39E-03 (0.0787)	−0.1924 (0.0863) **	0.2579 (0.0879) ***	0.0613 (0.0960)	−0.0471 (0.0734)
X3 (HHZ)	0.0182 (0.0878)	0.0979 (0.0751)	0.0286 (0.0873)	3.30E-03 (0.0942)	0.0539 (0.0737)
X4 (FMZ)	0.0912 (0.1786)	−0.0262 (0.1398)	0.0439 (0.1932)	−0.1690 (0.1775)	−1.08E-03 (0.1419)
X5 (Cred)	−0.1282 (0.2333)	−0.2274 (0.1949)	−0.0551 (0.2250)	0.0609 (0.2361)	−0.2612 (0.1908)
X6 (AgY)	0.0903 (0.1246)	0.0251 (0.1326)	0.0421 (0.1537)	−0.0119 (0.1472)	−4.38E-03 (0.1241)
X7 (InfoC)	−0.0617 (0.2034)	−0.1273 (0.1676)	0.0376 (0.2007)	−0.0120 (0.2020)	0.1693 (0.1689)
X8 (Coop)	0.1983 (0.2385)	0.1594 (0.2154)	−0.0840 (0.2391)	−0.0514 (0.2418)	−0.0871 (0.2086)
X9 (Teno)	−0.2051 (0.1782)	−0.2641 (0.1537) *	0.1110 (0.1814)	−0.0413 (0.1939)	0.0746 (0.1507)
X10 (Ext)	0.3002 (0.1901)	2.22E-04 (0.1544)	−0.3196 (0.1753) *	0.2687 (0.2029)	0.4474 (0.1503) ***
(β0) Constant	0.8679 (0.2116) ***	0.5498 (0.1764) ***	0.9393 (0.1933) ***	0.9022 (0.2183) ***	−0.0079 (0.1663)
(Y1) AgrFSys	1	−0.5265 (0.0919) ***	−0.2560 (0.1201) **	−0.2724 (0.1330) **	0.0171 (0.1022)
(Y2) ImpLiMgt	−0.5265 (0.0919) ***	1	−0.1762 (0.1141)	−0.1437 (0.1147)	−0.1589 (0.0926) *
(Y3) OrgFarm	−0.2560 (0.1201) **	−0.1762 (0.1141)	1	−0.1756 (0.1489)	−0.1635 (0.1015)
(Y4) CropRot	−0.2724 (0.1330) **	−0.1437 (0.1147)	−0.1756 (0.1489)	1	−0.1290 (0.1138)
(Y5) FarSch	0.0171 (0.1022)	−0.1589 (0.0926) *	−0.1635 (0.1015)	−0.1290 (0.1138)	1

Coefficients are reported with standard errors in parentheses. Significance levels: * p < 0.10, ** p < 0.05, *** p < 0.01. R-Studio MVP model output 2025 (own calculation). Source: computed from field survey data, 2025.

presents the coefficients and standard errors of the matrix from the estimated MVP model. The MVP model provides a more robust statistical analysis by evaluating how multiple independent variables jointly influence the likelihood of adopting various CSA practices, unlike the heatmap matrix, which does not account for multivariate interactions. The estimation results in Table 3 reveal several statistically significant determinants influencing the adoption of CSA practices. For crop rotation and intercropping (Y₄), years of experience in agritourism (X₁) showed a significant positive effect (coefficient = 0.3532, p < 0.01), suggesting that more experienced agritourism farmers were more likely to adopt this practice. Education level (X₂) revealed contrasting effects: it significantly decreased the likelihood of adopting improved livestock management (Y₂) (coefficient = −0.1924, p < 0.05) but significantly increased the probability of adopting organic farming (Y₃) (Coef = 0.2579, p < 0.01). Land tenure status (X₉), where ownership was coded as 1, was found to negatively influence the adoption of improved livestock management (Y₂) at the 10% level of significance (Coef = −0.2641, p < 0.10), indicating that agritourism farmers owning their land were less likely to adopt this practice compared to those with other tenure arrangements. Access to agricultural extension services (X₁₀) significantly reduced the likelihood of adopting organic farming (Y₃) (Coef = −0.3196, p < 0.10), while strongly promoting participation in farmer field schools (Y₅) (Coef = 0.4474, p < 0.01). Other variables, including household size, farmland size, access to credit, agritourism annual income, access to climate information, and cooperative membership, did not show a statistically significant relationship with the adoption of any of the five CSA practices at conventional significance levels.

Analysing the correlations of the error terms offers valuable insights into the interdependencies among the various adoption decisions. Significant negative correlations were found between the adoption of the agroforestry system (Y₁) and improved livestock management (Y₂) (ρ12 = −0.527, p < 0.01) and the agroforestry system (Y₁) and organic farming (Y₃) (ρ13 = −0.256, p < 0.05). Agroforestry system (Y₁), crop rotation, and intercropping system (Y₄) (ρ14 = −0.272, p < 0.05) were also negatively correlated. This result portrays a substitutive relationship or competition for resources, mostly between the agroforestry system and other CSA practices, based on an unobserved or latent factor. Agroforestry may indeed require substantial land and labour, which restricts the adoption of other CSA practices like livestock management or organic farming. Similarly, a negative correlation was observed between improved livestock management (Y₂) and farmer field schools (Y₅) (ρ25 = −0.159, p < 0.10). This observation may stem from time and resource constraints, as agritourism farmers adopting the CSA practice of improved livestock management may prioritise hands-on, farm-based activities over attending farmer field schools. Other pairwise correlations were not statistically significant, implying independence between those adoption decisions once observed factors are accounted for.

4. Discussion

The findings from the MVP model offer novel insights into the socio-economic factors influencing the adoption of various Climate-Smart Agriculture practices among agritourism farmers in Nigeria. The result of the multivariate probit model indicates a significant negative relationship between educational attainment and the adoption of CSA practice of improved livestock management among the randomly sampled agritourism farmers in Nigeria. Although education is generally associated with increased technology adoption, several studies [49,50,51] have also documented an inverse relationship. This paradox may stem from the educated agritourism farmers prioritising tourism-related activities over direct engagement in livestock management, potentially perceiving the latter as labour-intensive, malodorous, or less economically viable compared to alternative practices. As [52] observed in South–South Nigeria, farmers with higher education (up to university degree) often diversify beyond traditional farming activities, potentially reducing their focus on livestock operations. Education broadens career opportunities, drawing attention away from labour-intensive tasks such as livestock management. The rationale for this is that higher education widens career prospects, thereby shifting focus away from labour-intensive activities like livestock management. Security challenges in the study area (South–Eastern and South–Southern Nigeria) could also contribute to the explanation of this negative relationship, as the educated agritourism farmers may exhibit greater risk aversion toward livestock investments due to persistent herdsmen–farmer conflicts, fearing loss of their assets. The authors of [53] reported that security concerns significantly influence agricultural investment decisions, particularly in the livestock sector, which is prone to theft. The historical prevalence of pastoralist conflicts in the region may amplify this deterrent effect, particularly among educated agritourism farmers who, due to their exposure, may have access to alternative livelihood options within the spheres of agritourism. Additionally, institutional and infrastructural factors further explain this negative relationship. In Nigeria, access to credit, training, and market opportunities for improved livestock management is limited, particularly for emerging agritourism farmers. This scarcity reduces the appeal of such practices for those with higher education, who may pursue other ventures with stronger institutional or financial support. In addition, the capital-intensive nature of improved livestock management, which demands high inputs (e.g., veterinary services, quality feed, adequate infrastructure), makes it less accessible to the average agritourism farmer. The authors of [54,55] reported that poultry farmers, despite their education, cited high feed costs, veterinary services, and financial constraints as barriers to technology adoption. Consequently, the educated agritourism farmers in the study area, recognising these systemic barriers, might avoid investing in the improved livestock management CSA practice, viewing it as structurally and economically challenging.

Land tenure status also exhibits a negative relation with the adoption of improved livestock management practices. In the context of this research, land tenure status is defined as whether an agritourism farmer owns, leases, or utilises communal land for their agritourism operation. Land tenure status substantially influences the decision-making processes of agritourism farmers by indicating the status of the security of their land ownership. The authors of [56] noted that insecure or unfavourable land tenure can significantly impede the adoption of CSA practices such as improved livestock management. This result implies that agritourism farmers who lack secure land rights may be less inclined to invest in improved livestock management systems. Those operating on communal or leased land may face added uncertainty regarding their rights to develop the necessary improved livestock infrastructure. This negative association can also be attributed to limited access to credit for agritourism farmers with insecure tenure. In Nigeria, land ownership often serves as collateral for loans, facilitating agricultural investments [57]. However, agritourism farmers who rent or rely on communal land may struggle to secure such financial support, thereby limiting their capacity to adopt improved livestock practices. The authors of [58] observed that insecure land tenure in South–Eastern Nigeria discourages long-term investments in livestock infrastructure, a trend that is similarly observed in the whole of Nigeria. The authors of [59,60] also noted that agritourism operations in Nigeria emphasise scenic landscapes and experiential crop-based activities, which generate direct tourist revenue, making livestock management less compatible with the desired tourism experience.

According to Table 3, the multivariate Probit model shows a strong positive relationship between educational attainment and the adoption of organic farming among the sampled agritourism farmers in Nigeria. This result implies that more educated agritourism farmers are more likely to adopt organic farming methods. The authors of [61,62] posit that highly educated farmers tend to develop critical thinking and gain exposure to global environmental challenges, fostering an appreciation for sustainable practices. On top of that, [63] opined that tourists are increasingly attracted to agritourism farms, which portray environmentally sustainable practices, making organic farming a strategic choice for Nigerian agritourism farmers aiming to attract more visitors and obtain higher returns on investment. According to [64,65], agritourism attracts environmentally conscious and health-aware consumers willing to pay a premium for organically produced food. Educated agritourism farmers, attuned to these market preferences, recognise the potential for value addition and profitability, reinforcing the positive link between education and the adoption of organic farming. Equally, [61] reported that formal education exposes farmers to knowledge about environmental sustainability, the risks of synthetic pesticides and fertilisers, and the advantages of organic farming for human health and ecosystems. Consequently, this study infers that educated agritourism farmers in the study area are more likely to embrace organic farming methods as a sustainable and health-conscious approach to agriculture.

Also, the multivariate probit model reveals a negative relationship between access to agricultural extension services and the adoption of Climate-Smart Agriculture (CSA) practice of organic farming among the sampled agritourism farmers in Nigeria. The findings indicate that agritourism farmers with access to extension services are less likely to adopt the organic farming method. A key reason for this peculiar relationship may be because of the historical focus of Nigeria’s agricultural extension services on promoting conventional, high-yield farming methods over sustainable organic practices. The authors of [66] posited that extension services in Nigeria have frequently prioritised yield maximization through synthetic fertilisers, pesticides, and genetically modified crop species over sustainable practices. This occurrence could be attributable to Nigeria’s persistent food insecurity crisis, as national efforts are primarily concentrated on enhancing agricultural yields while placing less emphasis on adopting the “perceived low-yielding” organic farming system [67]. Similarly, extension personnel might lack the specialised training, in-depth knowledge, or appropriate resources required to effectively advise agritourism farmers on the best principles and practices of implementing organic agriculture. Additionally, the limited number of agricultural extension agents in Nigeria exacerbates the situation. The authors of [68] reported that a general guideline recommends 1 agricultural extension agent per 1000 farmers, with higher extension-to-farmer ratios required in countries facing severe poverty and food insecurity. However, Nigeria has the lowest extension agent to farmer ratio in Africa, with 1 agricultural extension agent per 7500 farmers [69]. This severe shortage suggests that there are few extension agents available to instruct agritourism farmers in sustainable organic farming techniques, thereby corroborating this study’s finding of an inverse relationship.

The multivariate probit model identifies a statistically significant positive association between years of agritourism experience and the adoption of CSA practice of crop rotation and intercropping among the sampled agritourism farmers in the study area. The result indicates that agritourism farmers with longer years of experience in agritourism are more likely to adopt crop rotation and intercropping techniques. Experienced agritourism farmers may recognise that crop rotation and intercropping techniques enhance agricultural yield and output, as well as improve the visual and experiential appeal of their farms, thereby potentially attracting more tourists and increasing their revenue streams. The authors of [70] noted that the diverse and visually dynamic farm landscapes facilitated by crop rotation and intercropping often serve as a motivation for adoption due to their attractiveness to tourists. A study in Nigeria found that farmers using crop rotation and intercropping reported a higher visitor retention rate [71]. Another contributing factor to this positive relationship could be the low financial investment required for adopting crop rotation and intercropping techniques (compared to other CSA practices), making these practices accessible to resource-limited farmers. The authors of [72] ascertained that experienced farmers typically possess a better understanding of soil conditions, microclimates, and crop combination and spacing techniques, enabling them to choose effective practices that improve the aesthetic value of their farms while maximising yield. According to the findings from the analysis, it can be inferred that experienced agritourism farmers in the study area were more inclined towards adopting crop rotation and intercropping CSA practices.

Based on the last significant result, the multivariate probit model portrays a positive relationship between participation in farmer field schools and access to agricultural extension services. As [73] highlighted, farmer field schools are designed to empower farmers by providing hands-on training, fostering collaboration, and encouraging the adoption of sustainable and climate-smart practices. Therefore, access to agricultural extension services could enhance this process by bridging the gap between farmers and their resources, knowledge, and support systems necessary for effective technology transfer. In Nigeria, where farmer field schools are often supported by NGOs (Non-Governmental Organizations) or government initiatives, agricultural extension agents play a vital role in identifying motivated farmers and customising training to local agroecological conditions. This finding of a positive relationship between farmer field school and access to agricultural extension agents is consistent with the work of [74], who found that extension-supported farmer field schools significantly improve knowledge retention when compared to other top-down training models. This result is consistent with the findings of [75], who noted that Nigeria’s National Agricultural Extension and Research Liaison Services (NAERLS) regularly partners with World Bank-sponsored farmer field school programs, providing improved seedlings, farm tools, technical manuals, and various farm inputs to the farming community. The result implies that agricultural extension services and participation in farmer field schools reinforce one another. Agricultural extension services facilitate initial engagement by disseminating relevant information and connecting farmers to farmer field school opportunities, while the schools sustain learning through collaborative, hands-on skill development.

5. Summary and Conclusions

This study investigated the socio-economic determinants of Climate-Smart Agriculture (CSA) adoption among agritourism farmers in Nigeria. The research shows that the agritourism farmers possess considerable experience in their agritourism ventures (averaging over 7 years) with relatively high levels of education (averaging post-secondary levels). They manage small farms (average of two hectares), often under a lease agreement and direct ownership. Institutional support, however, remains a critical limitation affecting the broader diffusion of CSA practices among the respondents. Less than half of the sampled agritourism farmers could access credit or climate information, and far fewer participated in cooperative activities or received agricultural extension support services.

The adoption rates of CSA practices varied among the sampled agritourism farmers. High adoption was observed for crop rotation/intercropping, organic farming, and agroforestry systems. These practices often enhance farm ecosystems and may offer marketing advantages for tourism. Moderate adoption levels were found for improved livestock management and participation in farmer field schools. This pattern suggests that the agritourism farmers prioritise practices that align well with land management and potentially enhance the appeal of their farms to visitors, while practices requiring different types of investment or collective action are adopted less frequently.

The findings from the Multivariate Probit (MVP) model reveal several significant factors influencing the adoption of the CSA practices among the agritourism farmers. Notably, years of experience in agritourism positively and significantly impacted the adoption of crop rotation and intercropping. Education level exhibited contrasting effects, negatively influencing the adoption of improved livestock management while positively affecting the adoption of organic farming. Land ownership negatively influenced the adoption of improved livestock management. Access to agricultural extension services showed a negative association with the adoption of organic farming but a strong positive association with participation in farmer field schools. Furthermore, the significant negative correlations observed in the error terms, such as between agroforestry and improved livestock management, agroforestry and organic farming, and agroforestry and crop rotation, highlight potential substitutive relationships or resource competition among these practices. This indicates that adopting one practice can influence the decision regarding another, likely due to competition for resources like land, labour, or capital, or because of perceived incompatibilities within the farming system.

Intriguingly, certain conventional determinants showed limited statistical influence in this specific context. Factors such as household size, overall farmland size, annual income from agritourism, access to credit, and access to climate information did not emerge as significant drivers for the adoption of the assessed CSA practices at standard significance levels. While these factors are generally considered important in agricultural decision-making, their lack of significance here could point towards the unique characteristics of the burgeoning agritourism sector in Nigeria.

6. Policy Recommendations

Based on the intricate findings regarding the socio-economic determinants of Climate-Smart Agriculture (CSA) adoption among agritourism farmers in Nigeria, several targeted recommendations emerge. These recommendations, while rooted in the Nigerian context, offer valuable insights and replicable strategies for policymakers, extension agencies, and research institutions in other regions with comparable agricultural landscapes and the emerging agritourism sector.

Promoting Knowledge Sharing and Mentorship: Given the positive influence of years of agritourism experience on the adoption of crop rotation and intercropping techniques, efforts should be made to facilitate knowledge sharing and peer-to-peer learning among agritourism farmers, particularly between experienced and newer entrants in the agritourism sector. This can be achieved through organising workshops, establishing farmer exchange programs, and supporting mentorship initiatives where the experienced agritourism farmers can guide and advise their less experienced counterparts.

Tailored Educational Programs: In light of the contrasting effects of educational attainment on different CSA practices, there is a clear need for targeted educational and training programs tailored to the specific requirements of practices like improved livestock management and organic farming. Developing specialised curricula and training modules that address the specific techniques, benefits, and potential challenges of each practice is essential. For instance, programs on improved livestock management could focus on sustainable grazing, animal health, and efficient feeding practices, while those on organic farming could cover soil fertility management, natural pest control, and organic certification processes.

Financial Limitations, Agricultural Cooperatives and Access to Credit: Despite their lack of statistical significance as direct adoption drivers in the MVP model for the specific practices studied, the generally low levels of access to credit (47%), climate information (48%), and cooperative membership (22%) represent significant systemic weaknesses that likely act as underlying constraints. Therefore, broader efforts remain crucial to enhance access to appropriate financial products that accommodate the investment profiles of agritourism farmers. Similarly, improving the delivery and usability of climate information systems, ensuring forecasts and advisories are relevant for farm-level planning for agritourism, is essential. Promoting and strengthening agritourism cooperatives could also provide platforms for collective input sourcing, marketing, knowledge exchange, and advocacy, enhancing the overall enabling environment.

Addressing Barriers to Land Tenure: Given the negative influence of land ownership on the adoption of improved livestock management, policymakers should explore potential barriers faced by landowners and consider implementing incentives or support mechanisms to encourage the adoption of this practice. This could involve providing financial assistance for investments in livestock infrastructure, offering technical support tailored to landowners’ specific needs, or ensuring clear and secure land tenure arrangements to encourage long-term investments in sustainable livestock practices.

Reorienting Extension Services: Considering the negative impact of extension services on organic farming but positive impact on farmer field schools, a review and potential reorientation of extension service delivery is recommended to better support organic farming practices, while continuing to leverage the effectiveness of farmer field schools. This may involve training extension agents in organic farming techniques, establishing demonstration organic farms, and fostering stronger linkages between extension services and organic farmer organizations. The successful model of farmer field schools should continue to be supported and potentially expanded to include specific modules on organic farming practices.

Broader Applicability for Similar Regions: While this study is rooted in the Nigerian context, its findings and the subsequent policy implications hold relevance for other developing countries that are witnessing a growth in agritourism amidst pressing climate change concerns. Therefore, strategies focusing on tailored educational programs, fostering peer-to-peer learning, reorienting extension services towards sustainable and agritourism-relevant CSA practices, and improving the enabling institutional environment could be adapted and prove beneficial in promoting resilient agritourism sectors elsewhere.

7. Limitation of the Study

While this study offers significant insights into the socio-economic factors influencing CSA adoption among Nigeria’s emerging agritourism sector, it is important to acknowledge several limitations inherent in its design and execution, which temper the interpretation and generalizability of the findings. These constraints provide context for the results and highlight areas where further research could yield a more comprehensive understanding. The focus on a specific, nascent group (agritourism farmers), while novel and valuable for understanding this sub-sector, inherently limits the direct applicability of the findings to the broader agricultural community in Nigeria, whose circumstances, motivations, and constraints may differ significantly. Furthermore, this study was conducted within a specific geographical context in Nigeria (South–Southern and South–Eastern geopolitical zones of Nigeria), and regional variations in agro-ecological conditions, market access, cultural norms, and institutional support systems could lead to different adoption patterns elsewhere in the country.

The research relied on cross-sectional data collected through a field survey at two points in time—August/September 2024 and December/January 2025—due to heightened agritourism activities during these times. This methodological approach, while effective for identifying associations between variables, restricts the ability to establish causal relationships definitively, mainly because it precludes the analysis of adoption dynamics over time. This is because the cross-sectional design does not allow for the observation of how socio-economic determinants and agritourism engagement evolve over time.