A Study on Information Communication Technology in Ba Province, Fiji

Abstract

This study examined the socioeconomic factors that influence the adoption of information and communication technologies (ICTs) in enhancing farm productivity among farmers in Ba Province, Fiji. A structured questionnaire survey was administered to a sample of 320 randomly selected farmers across the province’s 16 mainland districts. The analysis demonstrated that, although farmers possessed conventional ICTs, there was no direct correlation between ownership and actual utilisation. Significant determinants affecting ICT use were identified as education, experience, type of farming, and business model. These findings underscore critical implications for both policy and theoretical frameworks, emphasising essential factors to consider in the implementation of ICT solutions for agricultural practitioners.

1. Introduction

The swift dissemination of information and communication technology (ICT) in developing countries has become a pillar of the agricultural revolution. ICTs, both traditional (e.g., radio, television) and emerging (e.g., mobile phones, e-platforms, online apps), have created opportunities for addressing the difficulties in transferring information and knowledge. Nevertheless, notable gaps remain, including in sectors such as agricultural extension. For example, because there are more farmers than extension officers available worldwide, many farmers cannot access extension services through local officers [1,2]. This is problematic, as extension services serve a pivotal role in the accurate flow of information for solving everyday farming problems and achieving desirable changes [3]. ICT has a high potential for addressing this gap in information transfer [4].

Because agriculture is a complex socioeconomic activity, it necessitates collecting a large amount of information using multiple sources and channels [5]. The timely dissemination and utilisation of essential information with ICT can aid in decreasing the vulnerability of farming households, promoting food security, and improving farmers’ livelihoods [6]. The adoption of ICT is closely linked to enhancing interaction, information delivery, and decision-making [7,8]. Through understanding the collective benefits, ICTs become particularly valuable for a nation like Fiji, which is severely affected by climate change and relies closely on its agricultural sector for food security. Thus, the need to study farmer ICT use and perspectives becomes essential. A clearer insight will help the Fijian agriculture sector to digitalise and integrate ICT for farmers better.

Moreover, since the Fijian agriculture sector is primarily serviced through its Ministry of Agriculture (MoA) Crop and Animal Extension Services, there is a heavy reliance on traditional extension services (agricultural officer and farmer direct contact). Nonetheless, COVID-19 restrictions have highlighted the necessity for a move from conventional information gathering and dissemination. Shifting farmers’ dependence on information from understaffed extension services to ICT presents an important solution to this problem, as well as that of gathering farming information.

Hence, the focus of this research was to explore the role of ICT in enhancing agricultural productivity and addressing information transfer gaps in Fiji. Specifically, the research questions emphasised (1) identification of the types of ICTs currently used by farmers in Fiji and their preferences for these technologies; (2) examining the factors influencing the adoption of ICTs among farmers, particularly in relation to farm productivity; and (3) evaluating the impact of ICT use on farm productivity and identifying key determinants that contribute to this relationship. By providing insights into ICT adoption and usage patterns among Fijian farmers, this study aims to fill a critical gap in knowledge regarding the integration of ICT in agricultural practices. The findings of this research are valuable for policy development aimed at promoting the use of ICT to enhance agricultural productivity, improve information dissemination, and support the digital transformation of Fiji’s agricultural sector.

ICT in the Fijian Agricultural Context

Fiji is well known for its tourism and diversity. With a population of just under a million people of diverse languages and cultures, sustainable agricultural development and food security relies on stable connections and effective communication. Different communication channels have served as conduits for sharing information, from conventional mediums like in-person conversation and the press to digital messaging applications and social media. However, each method bears its unique strengths and limitations, determining its use and applications in exchanging information.

In this context, voice-based information services represent a longstanding means of agricultural communication. These services are chiefly associated with telephone-based systems like call centres or hotlines for agricultural extension services. The media used are typically landlines (which, in Fiji’s case, is a direct contact line to the extension offices) and mobile phones (used to contact individual extension officers). It is noted that farmers use phones to seek information regarding supply and demand in local markets, produce prices, farm inputs, and information on pest and disease control [9,10,11,12], subsequently reducing the need for travel. Fijian farmers residing in remote areas prefer telephone services as many agricultural offices are near urban areas or otherwise distant from farming communities.

Moreover, mobile phones provide farmers with the ability to communicate using both voice and text. Mobile phone short message service (SMS) is an especially cheap, efficient, and immediate means of communicating real-time information [13]. In Fiji, SMS is used by government agencies to inform farmers about impending natural disasters and serves as a platform for surveys and online application links. Similarly, video media can be shared using mobile phones. Studies have shown that increased access to and consumption of video media leads to changes in behaviours [14], resulting in positive outcomes [15]. Correspondingly, television is a longstanding and efficient example of video media. Through television, Fijian channels present multiple programmes for local agricultural farming in all three major languages: Fijian, Fiji Hindi, and English. However, they are aired at scheduled times, which may conflict with other activities.

Furthermore, like many traditional media, radio is a prominent and influential tool for transferring information to the masses, especially for older generations [16]. In Fiji, radio programmes are broadcasted in Fijian on Radio Fiji One and in Fiji Hindi on Radio Fiji Two. It is considered advantageous for reporting news, disseminating new skills, making announcements about meetings, discussing farming practices, and giving weather forecasts [17,18]. Radio has long been crucial for Fijian farmers, especially during cyclone seasons. The content broadcasted by radio hosts is typically based on the context demand of the public. Moreover, with the support of the MoA, the Fijian people have the weekly opportunity to participate in broadcasted programmes [19]. Furthermore, the broadcasters mediate the information through in-person visits to their listeners and offer a participatory approach in which the listeners can call in and ask questions [20]. This provides an opportunity for the mass dissemination of information to farmers who may be facing similar difficulties and are seeking guidance. However, radio cannot provide immediate information unless the program is designed specifically to include call-in segments or interactive broadcasts.

Lastly, the internet has paved the way for developing long-distance networks and maintaining relationships between actors from various backgrounds [21,22,23]. It functions as a channel for many forms of communication. Typical internet usage among Fijian farmers includes email (especially for grant applications and official agricultural business), social media, and the MoA and private agriculture company websites. The potential for online networking to assist development has been evident and significant [24]. Social media platforms have been recognised as an emerging learning network for the agricultural community [25]. Facebook, WhatsApp, Instagram, Twitter, and WeChat are especially prevalent in farming communities.

Farmers can utilise these apps to message and share information through texts, videos, and voice messages [16,26]. Fijian farmers prefer obtaining information through Facebook as their primary online source of communication and news, and they generally follow the MoA and other farming communities’ pages. Viber and Messenger are two prominent apps used in the farming community to pass information on to both farmers and extension service providers [27]. Similarly, the increasing prominence of farmers’ recognition by influencers and social media celebrities has placed farming at centre stage [28]. For example, farmers with YouTube channels have become effective resources for Fijian farmers, who have gained valuable techniques through these channels. Farmers of Indian descent (Indo-Fijians) tend to follow Indian agricultural influencers to learn farming techniques due to a shared language.

2. Literature Review and Conceptual Framework

2.1. Determinant Factors in ICT Adoption in Developing Countries

Research from various developing countries has identified determinants of ICT use in farming. This section will discuss some of the more prevalent determinants in most prior studies. The first factor is age. Studying technology use among small-scale farmers in Zimbabwe [29] explained that new technology tends to be most accepted by the generation born into it. Younger farmers are more inclined to accept and be interested in ICT [16,30] and invest in it [31]. In Tanzania, it was found that the rate of mobile phone use to seek information decreased with age for rice farmers [1]. Conversely, the tendency to use traditional media like radio and television increased [16]. Older farmers also tend to have more experience that informs their decisions. Meanwhile, younger farmers may need ICT to counterbalance their lack of agricultural experience and depend on additional knowledge obtained through ICT for their decision-making [16].

Farming experience comes with a wealth of knowledge corresponding to the years spent in the business. Nevertheless, the adoption of ICT is negatively associated with years of experience. As seen in [1], the more experienced a farmer is, the less likely they are to use mobile phones for agricultural information. Farmers with higher farming experience, especially those strictly akin to cultural beliefs and dependent on indigenous knowledge, tend to refrain from utilising present ICT tools [16]. However, first-hand experience in technology dispersion helps encourage farmers to adopt that technology [15]. As noted above, younger farmers with less experience can use ICT in various ways to gain various benefits. The availability of market information can empower farmers to overcome negative trends related to input prices and capitalise on their output prices [32]. Tonny et al. [9] found that Bangladeshi ICT users were getting better quality information than their peers and thus made significantly better decisions at all stages of agricultural marketing. ICT ownership additionally influenced farmers’ buying and selling abilities, as found with smallholder farmers in Kenya [33]. Moreover, memberships in social groups for farmers create a positive environment for ICT tool adoption [33,34]. In Kenya, it was found that belonging to a social group increases the chances of using ICT tools by 23.6% [31]. The information dissemination process is thus improved for farmers belonging to social media groups.

A farmer’s level of education also influences the types of ICT they utilise. Notably, with the increase in global education, farmers have shifted from gathering information directly from other farmers to modern ICT sources [35]. The relatively better a farmer’s education is, the more likely they are to use ICT [9,31,33,34,36,37,38]. A study of Tanzanian rice farmers indicated that a one-unit increase in education produced a 22.26% increase in the likelihood of using a mobile phone [1]. This is likely due to the increased understanding of ICT. Likewise, the farmers who preferred online advisory services were also most likely to be literate [16]. Moreover, the overall number of educated members in a household increased the chances of ICT use, as evident in the study by [37] on mobile advisory services in Pakistan.

While women make use of ICTs for farming information and communication with agricultural extension services, access to those extension services does not automatically indicate gender equality. It is vital to address the underlying inequalities in gendered bargaining power [15]. The financial independence of women farmers can influence the ownership and usage of ICT for women [33]. Research in South Africa [18] and Vietnam [39] indicated that men were more likely to adopt and benefit from mobile phones in agricultural marketing, which was most likely due to their greater freedom of movement. A similar phenomenon was noted by [30] in their research on livestock farmers in Eastern Cape, South Africa. In some societies, domestic duties consume women’s free time, leaving little room for using or learning about ICT, as demonstrated by [31] through their research on ICT in Bungoma County, Kenya. Their results demonstrated that men were 0.23 times more likely to use ICT tools than women. However, ref. [40] found that when it comes to the use of smartphones among rural Chinese farmers, female household heads were more likely to use smartphones, indicating that cultural context likely has a significant impact on ICT usage patterns.

Income is a further factor in accepting ICT services [29]. Farmers with higher overall incomes are more likely to adopt ICT, as evidenced by research in Vietnam [39], Zimbabwe [29], India [41], and South Africa [18]. ICT investment is more likely if agriculture is a farmer’s primary occupation [16]. Ntiri et al. [36] noted that households that earned more than half their income through farming were heavily influenced by using mobile phones for information gathering. Increased income and ICT use create a positive feedback loop. For example, references [33,34] demonstrated the positive impact of such tools on milk production, consequently improving dairy farmers’ incomes. Increased overall income creates more disposable income, allowing for ICT investment, which can further increase income. Off-farm income earned by farmers can also impact ICT adoption. For example, farmers seeking off-season work may seek external employment, which can enable and benefit from ICT use. For instance, ref. [1] found that an increase in external earnings resulted in the increased adoption of mobile phones, leading to better decision-making regarding both primary and secondary income sources.

Interestingly, farm size also plays a significant role in the adoption of ICT tools, as was made evident in research by [1,34]. In India, farmers with more land were most likely to seek information from all available sources, possibly due to being more resourceful, aware, and connected with available sources of information [35]. Similarly, ref. [41] found that owners of larger farms in India were more likely to use various sources to access agricultural information. Accessing information from various sources is necessary for farm risk management, leading these farmers to maximise their use of available ICTs [35,41]. Farmers with larger farms were also more inclined to use TV-based farm advisory and internet-delivered information [16]. This may be because of the combined display of graphics and audio, which helps people better grasp concepts. Similarly, more land availability may influence farmers to use the recommended practices demonstrated on television programmes [16]. In Bangladesh, the analysis demonstrated that more cultivated land positively impacted the utilisation of ICT [9]. Their model predicted that for every acre of land, the use of ICT in agricultural marketing increased by 0.4 units. In addition, ref. [16] explained the significance of land size associated with the adoption of ICT. Simply put, an increase in land size increases the chances of using ICT. Hence, farmers with large farms may use ICT to gather information that could impact their operations.

Overall, the factors consistently associated with ICT adoption across the literature on various developing countries are age, experience, education, gender, income, and farm size. It is vital to better understand these determinants to increase the capacity of ICT implementation and utilisation in other developing countries. Focusing on individual factors will create better chances of adoption, efficiency, and efficacy.

2.2. Conceptual Framework

Immediate and efficient facilitation of information through ICT can build towards the possibility of quicker agricultural development in a country. The quick dissemination of information to farmers based on their needs—such as supply and demand for marketing or information on current grant schemes—presents a significant opportunity for local agricultural development. The availability of necessary, timely information may incentivise farmers to invest further in agricultural growth. The conceptual framework (Figure 1) illustrates the factors influencing ICT use by farmers seeking agricultural information for their farm productivity. The ICT tools most utilised by farmers in Ba Province were the focus of research, and they enable access to information services. The use of these tools then leads to an increase in productivity.

3. Materials and Methods

3.1. Study Area

3.1.1. Fijian Agriculture

Fijian agricultural households are typically defined as any households directly involved in the farming community that derive their primary income from farming [42]. It may also include the practice of any kind of agricultural activity, such as crops, livestock, fishing, and forestry. The 2020 agricultural census for Fiji stated that 190,460 households participated in at least one of these activities [42]. Household members aged 10 and above participated in at least one agricultural activity, such as livestock or vegetable farming. Most Fijian farmers are unpaid family workers or subsistence farmers, followed by self-employed farmers [42]. Despite Fiji being a small country with a population of less than a million, the difficulty of passing information to and between farmers on the archipelago’s many islands remains a persistent issue in agricultural development.

3.1.2. Ba Province

This study was conducted in Ba Province following preliminary consultation with the Ministry of Agriculture on farmers’ utilisation of ICTs. Fiji comprises of two major Islands, which are Viti Levu and Vanua Levu. Ba Province is Fiji’s largest province by land mass. The province has twenty-one districts, sixteen of which are on Fiji’s main island, Viti Levu. The major crops grown in Ba Province include cassava (Manihot esculenta), dalo (Colocasia esculenta), kumala (Ipomoea batatas), bananas (Musa), eggplants (Solanum melongena), pineapples (Ananas comosus), watermelons (Citrullus lanatus), and tomatoes (Solanum lycopersicum). Farmers in Ba Province also raise cattle, poultry, ducks, goats, and sheep. However, livestock husbandry is mostly performed as part of mixed farming.

3.2. Data Sampling and Collection

This study used both qualitative and quantitative methods of data collection. In total, 320 farmers from the 16 districts of Viti Levu have completed a mixed questionnaire for the collection of qualitative and quantitative data (Figure 2). Prior to administering the final survey, the questionnaire was compared with the existing literature, evaluated by experts in the field, and subjected to random farmer testing (testing results are not included). A random sampling method was applied for this research. To avoid collecting centralised data only, farmers within the districts were randomly selected regardless of village location and topography. The data collected were stored in Excel format.

Prior arrangements were made with the MoA, and the district extension officer’s liaison was notified. During the research, agricultural extension officers were present for the journeys within and between districts to facilitate village entry and inform each district’s office about the study. To enter villages in Fiji, a Sevusevu must be presented to the village chief. Sevusevu, the ritual presentation of kava (a plant with mild narcotic properties), is essential to all Vanua rituals. By accepting a Sevusevu, the village chief symbolically welcomes visitors and offers them hospitality and protection. Twenty questionnaires were randomly dispersed to farmers from various villages and settlements within each district. Translators/interviewers were present for farmers who needed assistance understanding questionnaires in Fiji Hindi or Fijian.

3.3. Variable Coding

Table 1. Variable descriptions and coding used in logistic regression.

Dependent Variable
Variable				Categories	Code
Improvement of farm productivity with ICT use				No	0
				Yes	1
Independent Categorical Variables
Variables	Categories	Code	Variable	Categories	Code
Age	≤35	0	Education Level
	≥36	1		Primary	0
				Secondary	1
				University	2
Marital Status			Business Model
	Married	0		Commercial	0
	Single	1		Semi-Commercial	1
				Subsistence	2
Type of Farming			Outside Income
	Crop	0		No	0
	Livestock	1		Yes	1
	Mixed	2	Gender	Male	0
	Vegetables	3		Female	1
Land Tenure
	Freehold	0
	Native lease	1
	State lease	2
Independent continuous variables: household size, experience, land owned (acres), and land used for farming (acres).

lists the coding for the dependent and independent categorical variables. The continuous independent variables were not coded for analysis to maintain an accurate result. The dependent variable was the use of ICT for increasing farm productivity, and it was dummy coded. At the same time, independent variables that used dummy coding were age, education, marital status, business model, farming type, outside income, land tenure, and gender. No dummy coding was done for continuous variables such as household size, experience, land ownership, and land used for farming.

3.4. Logistic Model

A logistic, binary choice model was used to estimate the probability of farm households using ICT to increase their productivity. The logistic regression or logit model is a qualitative choice model used to describe the relationship between a dependent variable and independent variables. This model is beneficial because it works well with yes/no outcome variables and does not require assumptions about the relationships between variables. The dependent or response variable is dichotomous or binary, taking a 1 or 0 value. For example, Ba Province farmers’ use of ICT either increased their farming productivity or did not. Thus, the dependent variable (increased productivity with ICT use) can take only one of two values: 1 if the farmer had an increase in productivity with ICT use and 0 if the farmer did not. The logistic regression is depicted in Equation (1):

$$log\left[{\displaystyle \frac{P_i}{1-P_i}}\right]={\beta }_0+{\beta }_1{x}_i+{\beta }_2{x}_2+\cdots +\beta$$

(1)

Let P_i be the probability of the default of an adopter i, and let β₀ be the intercept term. β_i represents the respective coefficient in the linear combination of independent variables x_i for i = 1 − n. The dependent variable is the logarithm of the odds of the ratio of the two probabilities of the outcome of interest $log\left[{\displaystyle \frac{P_i}{1-P_i}}\right]$. Given the set of independent variables, the probability of a value of one (1) for the dichotomous outcome is shown in Equation (2):

$${\displaystyle \frac{P_i}{1-P_i}}={\displaystyle \frac{1}{1+e_{-z}}}$$

(2)

here,

z = β₀ + β₁x_i + β₂x₂+ … + β_nx_n + ε

(3)

The purpose of logistic regression is to determine the conditional probability of a specific observation within a class given the values of the independent variables of the ICT used to increase productivity. The statistical software SPSS version 29.0 was used to analyse data with reference to the logit model to identify the determinant factors influencing ICT use to increase productivity.

4. Results

4.1. Demographic Data

Figure 3 presents respondents’ age (A) and marital status (B). Farmers 35 or younger (green) represented 23.78% of the participant population. The majority (76.22%) were 36 or older (red). Of the 320 farmers in the research, nearly 86% were married, while the remaining 14% were single, as presented in the pie chart (B).

Figure 4 reflects respondents’ gender and education level. Most farmers, both men and women, completed secondary education. However, it is vital to note that most of the participants were men. Education levels, therefore, cannot be assumed to be equal for both genders.

Figure 5 illustrates information regarding the business model and presence of outside income (A) along with land tenure and form of farming (B). The business models included commercial, semi-commercial, or subsistence, and farmers of any model may have additional income sources. Most respondents were involved in semi-commercial agricultural production, with subsistence farming following. A smaller segment of the farmers was engaged in commercial farming. Moreover, the results indicated that regardless of the business model, most farmers relied on farming as a sole source of income.

Land tenure is an essential point of consideration for would-be farmers. In Ba Province, a significant number of farmers held native leases. Since the land originally belongs to iTaukei natives, it is leased out to others for certain periods. Ownership of freehold land titles was less common, likely due to the rare availability of this type of land and its high price. Furthermore, Ba Province farmers tended to prefer a mixed form of farming. The high level of subsistence farming potentially influenced the preference for mixed farming, as it ensures a nutritious diet and food security for the farmer.

4.2. ICT Ownership and Usage

Overall, 302 of 320 farmers (94.37%) stated that they owned mobile phones (

Table 2. Ownership and usage of ICT.

ICTs	Ownership			Usage for Agriculture
	N	Percent	E	N	Percent	E
Mobile	302	29%	94.4%	163	49.4%	50.9%
Radio	252	24.2%	78.8%	39	11.8%	12.2%
Television	277	26.6%	86.6%	61	18.5%	19.1%
Internet	211	20.2%	65.9%	67	20.3%	20.9%
Total	1042	100%	325.6%	330	100%	103.1%

N represents the total number of times a farmer says yes to either ownership or use of the stated ICT. The percentage is calculated for all ICTs together to be 100%. E represents the total % of each ICT analysed on a singular basis. For example, mobile = (302/320 farmers) × 100 = 94.4%.

). However, of these 302, only 163 farmers used their mobile phones to gain agricultural information. Despite high rates of mobile phone ownership, only 50.9% of the farmers surveyed used them for gathering agricultural information. Similar usage rates can be seen for the rest of the ICTs owned by farmers.

In other words, owning ICTs does not necessarily indicate that they are used for farming. This demonstrates that technology availability does not necessarily lead to acceptance of its usage in a traditional agricultural business. For instance, most farmers believe that internet access’s ultimate purpose is to use social networking sites (SNSs) to contact family members. Meanwhile, their main reason for owning a radio was to receive hourly news or entertainment. Similar statements were given for television and mobile phones.

4.3. Socioeconomic Analysis

In terms of model fit, the Hosmer–Lemeshow goodness-of-fit test yielded a statistic of 0.726, indicating that the model’s estimates align with the data at an acceptable level. Given that R-squared cannot be precisely calculated for logistic regression [43], a pseudo-R-squared was computed. In this study, the Nagelkerke R-squared was utilised as a proxy for R-squared in ordinary least squares (OLS) regression, which, as noted by [43], assesses the proportion of variation in the response variable explained by the model. The Nagelkerke R-squared value of 0.233 suggests that approximately 23.3% of the variance in the outcome can be accounted for by the predictors included in the model.

Overall, socioeconomic factors heavily influenced ICT use and its subsequent impact on farm productivity (

Table 3. Logit analysis of socioeconomic factors.

Estimation Results of the Logistic Model (n = 320)
Variables	B	S.E.	Wald	df	Sig.	Exp (B)
Gender	0.087	0.324	0.072	1	0.789	1.091
Marital status	0.987	0.489	4.071	1	0.044 *	2.683
Primary 1			7.704	2	0.021 *
Secondary	0.359	0.328	1.194	1	0.274	1.432
Tertiary	1.782	0.642	7.703	1	0.006 **	5.94
Age	−0.081	0.423	0.037	1	0.848	0.922
Freehold 2			4.045	2	0.132
Native lease	−19.815	14,941.994	0	1	0.999	0
State-owned	−20.618	14,941.995	0	1	0.999	0
Land size owned	−0.028	0.013	4.394	1	0.036 *	0.972
Land sized used	0.049	0.023	4.403	1	0.036 *	1.05
Subsistence 3			8.804	2	0.012 *
Semi-commercial	1.031	0.449	5.279	1	0.022 *	2.805
Commercial	0.842	0.321	6.867	1	0.009 **	2.32
Additional income	−0.091	0.322	0.081	1	0.777	0.913
Household size	0.11	0.07	2.451	1	0.117	1.116
Experience	−0.026	0.012	5.022	1	0.025 *	0.974
Crop 4			8.184	3	0.042 *
Livestock	0.611	0.659	0.86	1	0.354	1.843
Mixed	0.86	0.399	4.654	1	0.031 *	2.362
Vegetable	1.352	0.483	7.828	1	0.005 **	3.864
Constant	19.096	14,941.995	0	1	0.999	196,563,251.994

¹ Primary education is the base category for education. ² Freehold is the base category for tenure. ³ Subsistence farming is the base category for the business model. ⁴ Crop farming is the base category for the type of farming. Pseudo R-square = 0.233. B (Coefficient)—explains the relationship between an independent variable and a dependent variable. SE—measures the variability or precision of the estimated B value. DF (degrees of freedom)—is the number of predictors in the model. Sig (significance)—refers to the p value. p < 0.05 is indicated by *, p < 0.01 is indicated by **.

and Figure 6). Through the analysis, it is evident that socioeconomic factors influence ICT use to increase productivity. Marital status (<0.05) implies that married farmers have a positive relationship with ICT use. While for education, primary and tertiary education (p < 0.05 and p < 0.01, respectively) have an influence on ICT use. However, interestingly enough, secondary education has no significant influence on ICT use. Experience, on the other hand, had a positive (p < 0.05) relationship. When the years of experience increase, so do the chances of the use of ICTs. Land size owned and land size used for farming both have a positive (p < 0.05) impact on ICT use. These results indicate that the bigger or the more a piece of land is used for farming, the better the chance of an increase in ICT use for productivity.

The business model variable also shows a positive indication of farmers’ use of ICT. Subsistence and semi-commercial (p < 0.05) farmers understand the importance of ICT in farming; however, commercial farmers (p < 0.01) seem to be more understanding of its usefulness and application for productivity. Type of farming, according to the analysis, shows a positive influence on crop, mixed (p < 0.05), and vegetable farming (p < 0.01) for influencing ICT use. However, livestock farming has no significant impact; this may be due to a lack of farmers investing or specialising in livestock farming only. All in all, each of these tested variables with an indication of significance has, at one point in time in the farmer’s life, influenced them in the utilisation of ICT to increase their farming productivity. It is additionally important to note that gender, age, household size, land tenure, and other income sources (external) have no influence on the use of ICT to increase farm productivity.

5. Discussion

The use of modern ICT applications in Fiji has gained popularity; however, its use in agriculture among farmers remains limited. Nevertheless, there is recognition of ICT integration as a vital component for enhancing agricultural productivity. This discussion examines the multifaceted influence on tested variables, specifically highlighting the determinants of ICT use, the rationale behind the tested value, and their significance in relation to previous research, both in support of and in contrast to our findings.

According to the analysis, marital status is shown to impact ICT use for farm productivity significantly. This finding aligns with [44] as the study explained that the movement from single to being married increased the level of ICT adoption. This is further supported by the results of [45] that claimed that marital status can influence a farmer’s experience and ICT use due to family responsibilities. Similarly, the division of labour can also influence the significance of the role of married farmers. The division of labour between two members may give a farmer a better opportunity to use ICT on their farm as the time is efficiently divided between two individuals for work required on the farm. A married farmer may also have more opportunity to participate in training as their partner can care for their farm in their stead during their absence.

Education influenced ICT use by increasing productivity for farmers who only went to primary school and farmers with higher education, but not farmers with only secondary level education. This may be because many who finished secondary school received appropriate information regarding farming and thus did not see the need to use ICT to gather further information. Meanwhile, farmers with only a primary school education may have seen the need for ICT because their initial level of information was not enough to increase productivity. Additionally, primary-level education might reflect the influence of older, more traditional farmers who are supported through family or community networks, where ICT diffusion happens informally. Hence, this further confirms the need to provide ICT-based training for farmers to improve their digital literacy. In the case of farmers with higher education, the use of ICT presented an easier means of gathering and applying information in their field as they were likely already familiar with ICT’s use and importance. Hence, education changes how people approach certain situations [46,47]. Interestingly, some studies [48,49] showed a counterintuitive result where the higher an education was, the less likely ICT was used. A study [48] elucidated that individuals with more education had resources available to them beyond ICTs’ need for use. In addition, ref. [49] explained that the more educated a farmer is, the likelihood of them having an existing ICT platform is higher. Hence, the farmer sees less need to adopt a new ICT. On the other hand, other studies [37,50] found that education level had no significant impact on ICT use [46,47].

Land size owned and land size used for farming are also significantly portrayed in the analysis. This is in agreement with findings by [51,52] in Nigeria and Nepal. The size of the land owned may contribute to the business model a farmer uses while farming. In Fiji, farmers with large lands are more likely to invest in commercial farming and, similarly, more likely to apply for online farming grants. Likewise, as the size of land used for farming increases, the likelihood of the farmer to use ICTs to further his or her productivity also increases, matching findings of research in Pakistan [16]. The findings by [53,54] explain that farmers are more inclined to use social media for farming activities to support their expanding farms. However, contrasting results were found in Africa, as a negative relationship was seen between the increase in farm size and digital media use [55,56]. This difference may be attributed to the research region and cultural differences.

A farm’s business model also played a significant role, but only for farmers involved in commercial and subsistence farming. There was no apparent significance for semi-commercial farmers. Subsistence farmers were more inclined to use ICT based on their need for food security and sustainable agriculture. Additionally, the capacity of ICTs to reach remote areas [57], where livelihoods are predominantly reliant on subsistence farming, plays a significant role in inclination. For commercial farmers, the use of ICT is highly driven by marketing concerns [58]. The study conducted by [59] corroborated these findings, indicating that farm households that adopted ICTs experienced an increase in the percentage of their marketed output. Likewise, ref. [60] further stated that farmers that perceive farming as a business are more likely to adopt ICTs for decision-making. It has been noted [61,62] that competitive pressure serves as an additional factor driving the adoption of ICT among small and medium enterprises, a phenomenon that may also be relevant to the agricultural sector. The study conducted by [32] in Kenya provided strong evidence for the positive impact of ICT on agricultural commercialisation, while also indicating its beneficial effects on household per capita income.

Experience was a crucial variable in many cases for farmers’ adoption of any form of ICT. Farmers with more experience tended to better understand their productivity needs and the factors associated with ICT’s potential impact on their farms. This result is consistent with finding in Nigeria [45]. This may become a factor pushing them towards the use of technology to improve their farming as farmers are more likely to have understood the deeper need for information [51]. Moreover, experienced farmers likely attended training or belonged to organisations that utilise ICT. They, therefore, had prior knowledge on the effectiveness of using a particular ICT. In contrast, a study conducted by [63] indicated that experience had a positive influence on the adoption of ICT, although the effect was not substantial enough to be deemed significant. However, the findings of [64] showed contradictory results and indicated that social media as an ICT platform for agricultural services has a negative relationship with years of experience. This was because youths did not need to spend several years in farming before using social media. Interestingly enough, a study by [65] revealed a negative and significant relationship between experience and adoption of ICT. The findings indicated that cocoa farmers with less experience demonstrated a greater inclination to utilise ICT, potentially reflecting their increased openness to learning and adopting new technologies.

The type of farming conducted by a farmer was a significant factor associated with using ICT to increase productivity. Farmers engaged in crop, mixed, and vegetable farming were more likely to use ICT to increase their productivity, while livestock farmers were not. This phenomenon can be attributed to the high marketability of crops, akin to findings observed in a Kenyan study [66]. This trend may particularly be evident in western Fiji, as the west is a major tourist destination and hosts a significant tourism infrastructure. A study conducted in Africa [67] reported results consistent with our findings, indicating that internet usage for crop production was statistically significant, whereas its impact on livestock production was deemed insignificant. One study [68] described significant difference with digitisation between crop farms and livestock farms. It was clarified by [69] that the rationale behind crop farmers’ utilisation of ICT may stem from the greater emphasis placed on developing ICT resources for crop production compared to those for livestock management. The findings by [70] indicate that farmers engaged in crop production are significantly influenced by digital platforms, further corroborating our results. However, it was observed that as the number of platforms increased, the utilisation of media for information dissemination decreased. Conversely, the relevance of mixed farming appears to be associated with the imperative to maintain food security. This was further confirmed by [71] as it was revealed that integrated crop–livestock (mixed) farming positively and significantly affects food security. For vegetable farming, ref. [72] highlighted the significance of utilising ICT in vegetable farming, particularly regarding its impact on primary marketable produce that influences profit margins. Nonetheless, one study [73] stated that the type of farming had no influence on ICT use.

Overall, by addressing these various factors, the discussion underscores the importance of targeted ICT training and support services that cater to the diverse needs of different agricultural stakeholders, ultimately aiming to improve digital literacy and enhance the overall productivity of the agricultural sector [56].

6. Conclusions

ICTs are becoming increasingly prevalent as a tool for delivering agricultural services. In Fiji, the MoA uses ICTs to both broadcast and gather agricultural information. In the Fijian agricultural context, the employment of social media, mobile phones, and ministerial and private sector websites is slowly overtaking traditional technological communication methods such as radio and television. However, the complete replacement of human interactions and traditional ICTs as advisory services remains a distant prospect for most Fijian farmers. This study highlights that socioeconomic factors significantly influence farmers’ acceptance and usage of ICTs. Hence, these factors must be considered before major implementations lead to heavy reliance on digital solutions for agricultural extension. Understanding these dynamics is essential for developing inclusive and effective ICT policies across Fiji’s provinces.

This research, therefore, presents significant implications for policymakers and government institutions. Decision-makers involved in ICT implementation need to recognise education level as a critical factor influencing successful integration. To support effective ICT adoption, policymakers and agricultural extension service providers should prioritise increasing digital literacy among farmers, particularly older and less-educated individuals, to bridge the ICT skill gap. In addition, ICT platforms already widely used by Fijians, such as Viber, Messenger, and WhatsApp, can be leveraged for advisory services and farmer-to-farmer communication. Integrating these platforms into routine extension activities and training extension officers to use tools like SMS, mobile apps, and radio broadcasts effectively will improve outreach and engagement.

Moreover, addressing financial barriers is equally important. Subsidising mobile data plans, ICT tools, and electricity can significantly improve affordability and encourage adoption, especially in remote or disadvantaged areas. Alongside this, investing in infrastructure such as mobile towers and consistent electricity access is vital for regions like Ba Province, where climate and geography often hinder connectivity.

Furthermore, the development of localised ICT content, such as crop advice, weather alerts, or market prices in native languages, will enhance the relevance and accessibility of these services. Community consultation should be embraced before any new ICT initiative to ensure that solutions reflect farmers’ actual needs, preferences, and cultural contexts. Inclusive participation will foster trust, ownership, and ultimately more sustainable outcomes. These policy considerations together support a more inclusive, accessible, and responsive ICT-based agricultural extension system that aligns with the realities of Fiji’s diverse farming communities.

Nonetheless, this study has a few limitations that should be considered in future research. While the study focused primarily on ICT adoption, it did not incorporate cultural perspectives, which could influence how different ethnic groups in Fiji approach technology use. In particular, understanding the intersection of traditional farming practices and ICT adoption would provide a more comprehensive picture of the barriers and opportunities for ICT integration in Fiji’s agricultural sector. Moreover, the gender imbalance in this study, due to the land title-based registration system, may have excluded female farmers, who are key contributors to agricultural productivity. Future research should aim to include more inclusive sampling, particularly of women and other underrepresented groups, to ensure that their unique challenges and contributions to ICT adoption are better understood.

Additionally, the scarcity of reliable, localised data on ICT adoption across the Pacific limits the ability to make broader comparisons or generalisations. Future studies should prioritise data collection, particularly through longitudinal research that tracks ICT usage over time and across different contexts. Comparative studies across Pacific Island nations could provide valuable insights into region-specific challenges, helping to refine policies and practices that address local needs. This will contribute to a more robust understanding of ICT adoption’s impact on farm productivity and its role in improving agricultural outcomes in the region.