The study was conducted in Kambia District, located in the North-Western region of Sierra Leone. Kambia District is recognized as one of the major rice-producing areas in the country due to its favorable agro-ecological conditions, including fertile lowlands, abundant rainfall, and extensive inland valley swamp systems suitable for rice cultivation. Rice farming is the dominant agricultural activity and serves as the major source of livelihood for most rural households within the district. The study specifically focused on Robana, in Mambolo Chiefdom (8.91˚N, 13.04˚W), and Rokupr in Magbema Chiefdom (9.01˚N, −12.95˚W) and Samu Chiefdom (Kasirie and Kychom), (8.94˚N, −13.11˚W, 8.93˚N, −13.14˚W) respectively. These chiefdoms were purposively selected because of their active involvement in rice production and the observable differences in access to agricultural extension services, irrigation facilities, farmer training opportunities, and modern agricultural technologies. In addition, the selected chiefdoms exhibit varying biophysical characteristics such as soil fertility, flooding patterns, and drought conditions, making them suitable for examining the influence of extension services, biophysical conditions, and farmers’ perceptions on the adoption of modern rice technologies.

The study adopted a cross-sectional survey research design using a mixed-methods approach that combined both quantitative and qualitative research techniques. The cross-sectional design was considered appropriate because it enabled the researcher to collect information from respondents at a single point in time while examining the relationships between extension services, biophysical conditions, farmers’ perceptions, and adoption of modern rice technologies. The quantitative component of the study focused on generating numerical data that could be statistically analyzed to determine patterns, associations, and determinants of adoption. Structured questionnaires were administered to rice farmers to obtain measurable information regarding socio-demographic characteristics, farm characteristics, access to extension services, and adoption behaviors.

The qualitative component involved key informant interviews and field observations aimed at obtaining in-depth explanations regarding farmers’ experiences, institutional support systems, challenges associated with technology adoption, and perceptions of modern rice technologies. The combination of quantitative and qualitative approaches enhanced the reliability, validity, and comprehensiveness of the findings.

The study targeted all rice farmers engaged in rice production activities in Magbema, Mambolo, and Samu Chiefdoms in Kambia District, including both male and female farmers practicing rain-fed lowland, upland, and irrigated rice farming systems. Rice farmers constituted the main unit of analysis because they are directly involved in the adoption and utilization of modern rice technologies. The sampling frame consisted of 1,718 rice farmers distributed across the four selected communities: Kychum (391), Kasiri (391), Rokupr (440), and Robana (496). From these sampling frames, 312 farmers were proportionately selected, comprising 71 from Kychum, 71 from Kasiri, 80 from Rokupr, and 90 from Robana. Farmer lists were compiled with the assistance of agricultural extension officers, local authorities, farmer organizations, and community leaders and were subsequently verified to ensure that only active rice farmers were included. Individual farmers were then selected using simple random sampling to provide each eligible farmer with an equal probability of selection. Where a selected farmer was unavailable after the specified contact attempts or declined to participate, the farmer was replaced by another eligible farmer randomly selected from the same verified sampling frame. Participation was voluntary, and only farmers who provided informed consent were included in the study. The study employed a multistage sampling procedure. First, purposive sampling was used to select the three chiefdoms because of their importance in rice production and differences in extension accessibility, farming practices, and environmental conditions. Second, rice-producing communities were identified with the assistance of extension officers, local authorities, and community leaders. Finally, simple random sampling was used to select individual farmers from farmer lists to ensure equal chances of participation and reduce selection bias (Table 1). The sample size for the study was determined using the Yamane (1967) formula for sample size determination. The formula is commonly used in quantitative research to determine a representative sample from a known population.

where:

n = required sample size;

N = total population size;

e = margin of error (0.05).

Table 1. Proportionate distribution of sample size.

The study utilized both primary and secondary sources of data. Primary data were collected directly from respondents through structured questionnaires, key informant interviews, and field observations. These data focused on socio-demographic characteristics, farm characteristics, access to extension services, training opportunities, membership in farmer organizations, biophysical farming conditions, farmers’ perceptions, and the adoption of modern rice technologies. The structured questionnaire served as the main instrument for collecting quantitative data from rice farmers. It consisted mainly of closed-ended questions with a few open-ended items and was administered through face-to-face interviews to accommodate respondents with low levels of formal education. In addition, key informant interviews were conducted with agricultural extension officers, farmer organization leaders, community leaders, and Ministry of Agriculture personnel to obtain qualitative information on extension delivery systems, institutional support, challenges facing rice farmers, and factors influencing technology adoption. Field observations were also conducted using an observation checklist to assess farming conditions and technology utilization practices, including irrigation facilities, soil conditions, mechanization equipment, demonstration plots, and rice farming systems. These observations helped verify information obtained from respondents. Secondary data were sourced from government reports, Ministry of Agriculture publications, agricultural policy documents, academic journals, books, previous studies, and reports from NGOs and development agencies involved in agricultural development and rice production in Sierra Leone. These sources provided background information and supported the interpretation of the study findings.

To ensure validity, the research instruments were reviewed by agricultural experts, research supervisors, and specialists in agricultural extension and rural development. Their comments and suggestions helped improve the clarity, relevance, and appropriateness of the questionnaire items and interview guides in relation to the study objectives. A pilot study was conducted among rice farmers outside the selected study communities to test the suitability, consistency, and clarity of the research instruments. Based on the pilot results, ambiguous questions were revised and necessary adjustments were made before the actual data collection exercise. Reliability of the instruments was ensured through careful wording of questions, proper training of research assistants, and standardization of interview procedures. Consistency in data collection procedures helped minimize errors and improve the accuracy of responses.

The collected data were coded, cleaned, and analyzed using statistical software through both descriptive and inferential statistical methods. Descriptive statistics such as frequencies, percentages, means, standard deviations, and tables were used to summarize respondents’ socio-demographic characteristics, extension service variables, biophysical conditions, and farmers’ perceptions of modern rice technologies. Chi-square tests examined associations between categorical variables, including training participation, household income sources, and farmer organization membership. Kendall’s W ranking analysis was used to assess the level of agreement among respondents on the perceived benefits, risks, constraints, and determinants of technology adoption, while post-hoc analysis identified specific group differences where significant Chi-square relationships existed.

Ethical principles were strictly observed throughout the study. Respondents were informed about the purpose and objectives of the research before participation. Participation in the study was voluntary, and respondents were assured of confidentiality and anonymity regarding the information they provided. Informed consent was obtained from all respondents prior to data collection, and participants were allowed to withdraw from the study at any stage without penalty.

The socio-demographic characteristics of the 312 farmers across the three chiefdoms demonstrated notable variation in gender composition, age, and educational attainment. In Magbema, females predominated (62.5%), whereas Mambolo and Samu were largely male, representing 78.9% and 71.8% of farmers, respectively (Table 2). The farming population was predominantly mature, with mean ages of 46.3 ± 9.8 years in Magbema, 43.8 ± 8.7 years in Mambolo, and 42.1 ± 11.2 years in Samu, indicating that rice farming activities were mainly undertaken by individuals in mid-to-late adulthood. Educational attainment varied across the chiefdoms; a substantial proportion of farmers had no formal education, particularly in Mambolo (41.1%), followed by Samu (33.8%) and Magbema (32.5%). Secondary education was attained by 26.3% of farmers in Magbema, 20.0% in Mambolo, and 28.9% in Samu, while tertiary education was relatively low in all chiefdoms. Farming characteristics further highlighted differences in experience and farm size. Mean farming experience was 14.8 ± 8.9 years in Magbema, 17.6 ± 9.7 years in Mambolo, and 13.2 ± 8.5 years in Samu, indicating a generally experienced workforce. Farm sizes also differed markedly, with Magbema having the largest mean farm size (19.6 ± 14.2 acres), followed by Mambolo (12.8 ± 10.5 acres) and Samu (6.3 ± 5.7 acres).

Economic characteristics showed that rice farming was the predominant source of household income across all chiefdoms, accounting for 92.5% of farmers in Magbema, 98.9% in Mambolo, and 90.8% in Samu. Alternative income sources, including livestock, non-farm activities, and other crops, were minimally represented. Participation in farmer organizations or cooperatives varied substantially; most farmers in Magbema were cooperative members (93.8%), while only 25.4% of Samu farmers reported membership. Mambolo had an intermediate level of cooperative participation (76.7%).

Table 2. Socio-demographic characteristics of farmers.

Adoption outcome: A farmer was classified as an adopter (1) if he or she had used at least one of the specified modern rice technologies during the study/reference period, and as a non-adopter (0) if none of the technologies had been used. The modern rice technologies considered included improved seed varieties, mechanized land preparation, improved water management/irrigation, integrated soil fertility management, improved planting methods, and climate-smart rice production practices. Extension services in Kambia District strongly influenced farmers’ adoption of modern rice technologies, though access varied sharply across chiefdoms. Magbema (71.3%) and Mambolo (83.3%) farmers reported regular access, whereas 79.6% of Samu farmers had no access (χ² = 254, df = 6, p < .001) (Table 3) ([31]). Extension agents visited very regularly for 65% of Magbema and 68.9% of Mambolo farmers, but 79.6% of Samu farmers never received visits (χ² = 234, df = 6, p < .001) ([34]).

Table 3. Role of extension services in shaping farmers’ adoption decisions.

Detailed information reached 76.3% of Magbema and 80% of Mambolo farmers, while 84.5% of Samu farmers received none (χ² = 237.7, df = 6, p < .001). Training attendance was high in Magbema (68.8%) and Mambolo (75.6%), versus 86.6% of Samu farmers who never attended (χ² = 226.3, df = 8, p < .001) ([25]). Demonstrations were most effective in Magbema (56.3%) and Mambolo (75.6%), whereas Samu farmers relied on group meetings (49.3%) and radio (21.1%) (χ² = 182.1, df = 8, p < .001) ([40]; [23]). Accessibility, follow-up support, and satisfaction followed similar patterns, consistently favoring Magbema and Mambolo (all p < .001).

The Chi-square results in Table 3 show that there is no statistically significant association between farmers’ participation in training on modern rice technologies and their main source of household income (χ² = 12.34, p = 0.195) (Table 4). A test of association was conducted to determine whether farmers’ main source of household income was associated with their participation in training on modern rice technologies. Although the Pearson chi-square statistic was χ² = 12.34 (df = 9, p = 0.195), the Pearson chi-square assumptions were not satisfied because 8 of the 16 expected cell frequencies were below 5, with particularly small expected counts for the livestock and other-crops categories. Therefore, a Monte Carlo Fisher-Freeman-Halton exact test was used to obtain a more reliable significance estimate. The Monte Carlo exact test produced a p-value of 0.122 based on 200,000 simulated tables. The result was not statistically significant at the 5% level, indicating that there was no sufficient evidence of an association between farmers’ main source of household income and their participation in training on modern rice technologies (Monte Carlo exact p = 0.122). This implies that access to training whether formal, informal, both, or none is not determined by the type of livelihood a household depends on ([29]; [8]). Although rice farmers dominate all training categories, particularly among those with informal training and those who have not received training, this pattern largely reflects the fact that rice farming is the primary income source for most respondents in the district, rather than indicating any targeted training differences. Similarly, smaller groups such as those engaged in non-farm activities, livestock, or other crops show limited and scattered participation in training, but not enough to create meaningful variation across income groups ([15]). Overall, the findings suggest that training opportunities are distributed broadly but not strategically targeted based on household income sources. This highlights the need for more inclusive and better-targeted training programs that ensure equitable access for all farmer categories while also improving the relevance and effectiveness of training interventions to enhance adoption of modern rice technologies ([19]; [48]).

Table 4. Chi-square test on the training of modern rice technologies associated with main source of household income.

The Chi-square results in Table 5 reveal a highly statistically significant association between farmers’ participation in training on modern rice technologies and their membership in farmer organizations (χ² = 131.28, p < 0.0001) (Table 5). This indicates that membership in farmer organizations strongly influences access to training opportunities ([33]; [55]). Farmers who belong to such organizations are far more likely to receive training, particularly formal training (92 cases) and combined formal and informal training (35 cases) ([50]; [56]). In contrast, non-members are largely excluded, with the majority (101 cases) reporting that they have not received any training. The large differences between observed and expected values further confirm the strength of this relationship.

Table 5. Chi-square test on the training of modern rice technologies associated with member of farmer organization.

These findings demonstrate that farmer organizations serve as critical platforms for knowledge dissemination, capacity building, and access to agricultural innovations ([35]; [38]). Membership enhances farmers’ exposure to extension services, training programs, and improved technologies, thereby increasing their likelihood of adopting modern rice practices ([56]). Conversely, farmers who are not part of such groups remain significantly disadvantaged in accessing training opportunities ([32]). This underscores the importance of promoting and strengthening farmer organizations as an effective strategy for expanding training coverage, improving information flow, and ultimately enhancing the adoption of modern rice technologies in Kambia District.

Key informants highlighted that agricultural extension services are instrumental in promoting the adoption of modern rice technologies in Kambia District. The Farmer Field School Approach (FFSA) and on-farm demonstrations were identified as the most effective extension strategies because they enhance farmers’ practical knowledge, skills, and confidence in using improved technologies. However, extension service delivery is constrained by limited mechanization, inadequate farming tools, and insufficient financial resources, which restrict training and farmer support activities. The informants emphasized that strengthening extension services through increased funding, improved access to mechanization, and expanded farmer training would significantly enhance the adoption of modern rice technologies and improve rice productivity.

The analysis of biophysical conditions for rice technology adoption in Kambia District revealed significant differences across chiefdoms. The type of rice farming system varied substantially (χ² = 50.54, df = 6, p < .001) (Table 6), with Magbema and Mambolo farmers predominantly cultivating rain-fed lowland (32.5% and 53.3%, respectively) and combined irrigated and rain-fed systems (30% and 28.9%), whereas Samu farmers relied mainly on rain-fed lowland (57.7%) and rain-fed upland systems (18.3%), and no fully irrigated systems were reported in Samu ([30]).

Table 6. Biophysical conditions for modern rice technology adoption.

Dominant soil types also differed significantly (χ² = 134.1, df = 6, p < .001); clay soils were prevalent in Samu (47.2%), loam soils in Mambolo (67.8%) and Magbema (45%), and mixed soils were most common in Magbema (52.5%) ([51]). Soil fertility ratings varied across chiefdoms (χ² = 81.48, df = 8, p < .001), with very high fertility recorded in Magbema (43.8%) and Mambolo (35.6%), while Samu fields were primarily medium fertility (58.5%) ([3]). Flooding frequency differed significantly (χ² = 63.27, df = 6, p < .001), with occasional flooding reported by 41.3% of Magbema, 43.3% of Mambolo, and 54.9% of Samu farmers, while rare flooding was reported by 41.3%, 35.6%, and 4.9%, respectively ([42]). Drought stress frequency was also significantly associated with chiefdom (χ² = 17.06, df = 6, p = 0.009), with most Magbema (36.3%) and Mambolo (44.4%) farmers experiencing drought rarely, and 44.4% of Samu farmers reporting no drought stress ([13]).

The key informants identified soil type and drought as the major biophysical factors influencing the adoption of modern rice technologies in Kambia District. They noted that drought reduces water availability, limiting farmers’ ability to adopt improved technologies. Farmers largely depend on indigenous knowledge and local farming experience to cope with these environmental challenges. The findings suggest that promoting climate-resilient, location-specific rice technologies that complement local biophysical conditions and indigenous practices would enhance technology adoption.

Farmers’ perceptions of modern rice technology indicate a cautious attitude toward adoption, balancing recognized benefits against perceived risks and constraints. Regarding benefits, farmers expressed only slight agreement that modern rice technologies increase rice yield (Mean = 2.04, SD = 1.30), improve rice quality (Mean = 1.92, SD = 1.26), and enhance farm income (Mean = 2.02, SD = 1.32) (Table 7), suggesting that while some advantages are acknowledged, their impact is not strongly endorsed ([8]; [52]). In contrast, farmers perceived the risks associated with adoption to be moderate to high. The risk of financial loss was considered moderate (Mean = 3.10, SD = 1.67), whereas the likelihood of crop failure (Mean = 4.16, SD = 1.18) and potential environmental harm (Mean = 4.35, SD = 1.06) were viewed as serious concerns ([14]). Constraints were generally rated in the moderate range, with the high cost of inputs being the most notable limitation (Mean = 2.94, SD = 1.59), followed by lack of technical knowledge (Mean = 2.46, SD = 1.44), limited access to credit (Mean = 2.43, SD = 1.49), and inadequate extension support (Mean = 2.44, SD = 1.46) ([1]).

Table 7. Farmers’ perception associated with adopting modern rice technology.

Key informants reported that farmers generally perceive modern rice technologies positively, particularly for their ability to increase rice yields, productivity, and household income. They explained that the adoption of mechanization and improved farming practices in recent farming seasons has resulted in higher rice production and improved harvests. These experiences have strengthened farmers’ confidence in the benefits of modern rice technologies and encouraged greater interest in their adoption.