Agricultural Information Needs and Resources of Smallholder Sugarcane Farmers in Swaziland

1.1 Problem Statement

Information is critical at every level of sugarcane production, particularly in the growing phases where smallholder farmers are involved. The Swaziland^[1] extension service initially provided agricultural information services to farmers, however, over the years, there has been a decline in the effectiveness of this service (Dlamini and Worth 2016). This resulted in the extension service losing enormous credibility with farmers (Food and Agriculture Organization 2011), yet access to adequate information is very important for increased agricultural productivity (Mgbada 2006). However, smallholder sugarcane farmers in Swaziland are inundated by a myriad of information deficiencies which results in the low adoption of new technologies, culminating in low yields (Dlamini and Dlamini 2012; Malaza and Myeni 2009; Samuel 2008; Sifundza and Ntuli 2001). To ensure efficient and effective information services, a better understanding of smallholder sugarcane farmers’ information needs and the kind of communication resources they use to access agricultural information was required. Hence, the overarching purpose of this study was to examine the provision of agricultural information services to smallholder farmers in the sugar industry, as this has the potential to increase production and reduce poverty in line with the United Nations Sustainable Development Goals.

1.2 Overview of Swaziland’s Agriculture

Swaziland is a small country landlocked by South Africa and Mozambique in the east with an area of 17,364 square kilometres. The country has a population of 1.1 million people with around 70 % of the population living in rural areas and relying on agriculture for their livelihoods. The Swaziland economy is characterized by high dependence on its membership to the South African Customs Union (SACU) and the Common Monetary Area (CMA), with the country’s agricultural policies harmonized with that of the block (African Development Bank 2019). Swaziland’s land tenure is dualistic in nature, comprising two major agricultural production systems, namely the Traditional Customary Tenure (TCT) and Title Deed Tenure (TDT), which respectively cover 46 % and 54 % of the country (Dlamini and Masuku 2011).

The World Bank and the Government of Swaziland (GOS) have been continuously engaged in reviewing means of improving the productivity of small-scale farmers and how best to integrate smallholder farmers to markets (World Bank 2011). The effort by the GOS to improve the productivity of smallholder farmers is illustrated in the Swaziland National Development Strategy (NDS) for the period 1997–2022 where agriculture was put as a key focus. One of the goals of the National Development Strategy (NDS) for agriculture is to enhance private sector involvement in the uplifting of Swazi Nation Land (SNL) farmers from subsistence to commercial farming, while maintaining economic efficiency in the production and promotion of sugarcane production by smallholder farmers (Government of Swaziland 1999).

1.3 Sugarcane Farming in Swaziland

The sugar industry is an important player in terms of its contribution to the economy of Swaziland, growing sugarcane in an area spanning approximately 59,000 ha of land (Eswatini Sugar Association (ESA) 2021). The country is amongst the largest sugarcane producers and was ranked twenty-seventh globally in 2007. Sugar dominates the country’s agricultural production by value and is a crucial export to the European Union (EU) under special preferences (Sandrey and Vink 2009). In 2014–15, the Swaziland sugar industry was ranked fourth largest in Africa, although the country is relatively small (Terry and Ogg 2017, 585). The sugar industry consists of three mills, namely Mhlume, Simunye and Big Bend Mill, all located in the eastern lowveld of the country. The Sugar Act of 1967, together with the Sugar Industry Agreement and the constitution of the Swaziland Sugar Association (SSA), regulates all aspects of the industry, from the right to grow sugarcane to how sugar is marketed. The Swaziland Sugar Association is an umbrella body that comprises the sugarcane growers and millers (United Nations Conference on Trade and Development 2000).

The Swaziland sugar industry, which had largely been dominated by large-scale estates since its inception in 1957, has seen a rapid increase in the involvement of smallholder farmers, with efforts made to extend participation to small-scale growers, with the aim of reducing poverty in rural areas (Terry 1997). Presently, there is a substantial number of smallholder sugarcane farmers along the Komati and Great Usutu River in the Lowveld as well as some parts of the Middleveld, distinguishable by cane stands. These farmers operate as individuals and some in groups under the communal land arrangement administered by the chief, funded at varied levels by the European Union (EU) and state financial institutions (Samuel 2008).

1.4 Research Objectives

The objectives of the study were to:

1. Determine the information needs of smallholder sugarcane farmers in Swaziland.

2. Determine information needs differences between the milling groups of smallholder sugarcane farmers.

3. Determine the preferred communication resources/channels of agricultural information for the smallholder sugarcane farmers.

In addressing the above objectives, the following research questions were asked:

1. What are the types of information needs required by smallholder sugarcane farmers?

2. What are the information needs differences between the milling groups of smallholder sugarcane farmers?

3. What are the preferred communication resources/channels of agricultural information for the smallholder sugarcane farmers?

2.1 Theoretical Framework

User studies interrogating information needs abound in literature, culminating into models of information behaviour (Ellis 1989; Kuhlthau 1991; Wilson 1981, 1997). According to Dervin (1998) information needs is a “gap” that which hinders movement, whether physically or cognitively. Later, Case (2012, 5) acknowledged the cognitive aspects in his overlapping definition of information needs, or simply put, as the “recognition that your knowledge is inadequate.” In other words, an information need could just mean a lack of requisite information or knowledge to resolve a certain issue. Elsewhere, Wilson (1981) argued that information needs is interwoven in the concept of information behaviour and, therefore, the study adopted Wilson’s 1996 model of general information behaviour (Figure 1) to investigate the information needs and resources of smallholder sugarcane farmers in Swaziland. The model is broad, comprehensive, and constitutes many important elements of information behavior concerned with how people seek and make use of information as well as the channels they use to get information (Wilson 1997). It qualifies to study different kinds of information users because it is interdisciplinary and general (Robson and Robinson 2013; Wilson 1999). The model pictures the cycle of information activities from the inception of an information need to the final phase where information is processed and used (Niedzwiedzka 2003). In the model, the person-in-context, in this case the smallholder sugarcane farmers, constitutes the focus of the information need. The model includes intervening variables, namely demographic variables, role related or interpersonal factors, environmental factors, and source characteristics that influence motivation to seek information. Wilson further integrated theories that explain why people pursue a goal more successfully than others, and the model further depicts information sources as communication channels in which information is sought or received through formal and informal sources/services. In the event information is not received, the search process is repeated (Wilson 1997). Equally, Kuhlthau (1991) averred that knowledge retained in the process of satisfying the information needs by the information seeker may invoke renewed information needs. Information needs are unstable and change over time (Bothma and Bergenholtz 2013); in the same vein, information needs identified a long time ago may not be applicable in the present period and may invariably become the present period’s gap (Dervin 1998). In the sugar industry in Swaziland there have been changes in policies resulting in the involvement of subsistence farmers in the commercial growing of sugarcane. One can expect that changes in the farmers’ environment relatively influence their information needs which are thus construed to be unstable and changing over time.

Figure 1:

Wilson’s 1996 model of information behaviour (Wilson 1999).

2.2 Information Needs

Information needs studies are long-standing in the information science field, even though the concept is misunderstood (Cole 2011), with the confusion regarding the definition of information needs emerging from the confusing concepts of “information” and “need” (Wilson 1981). Sometimes information can mean a thing, a piece of knowledge, or a process at once (Byström and Järvelin 1995). The onus is on the researcher to contextualize the meaning of “information” in relation to his or her study (Bawden 2006, 673).

Shedding some light on the concept of information needs, Belkin, Oddy, and Brooks (1982) viewed it as an anomaly in the user’s state of knowledge, dubbed as “Anomalous State of Knowledge” (ASK). This approach suggests that the user is in a dilemma regarding the specificity of the need, and how to resolve the anomaly, with the authors placing more emphasis on the role of the users’ state of knowledge and their desires to recognize the problematic aspects, as well as to articulate specific information needs without the involvement of third parties, such as scholars’ points of view. Unfortunately, in most cases, users cannot precisely specify the kind of information required to resolve their information needs (Belkin, Oddy, and Brooks 1982, 61). Therefore, the role of scholars becomes indispensable in the process, which perhaps suggests the reason why Wilson (1981) and Chowdhury and Chowdhury (1999) argue that examining information needs can be a challenge, since this involves cognitive phenomena typically performed beyond the consciousness of the individuals, which makes it difficult to observe let alone define their information needs. The difficulty in defining information needs could negatively impact on the output of those responsible for developing information services that would be helpful in meeting users’ needs.

Krikelas (1983, 6) defined information needs as the recognition of the existence of an uncertainty in the personal or work-related life of an individual (person or an organization) that can be solved using information. In the same vein, Wilson (1981, 2000, Godbold (2006), and Järvelin and Wilson (2003) describe information needs as a gap that requires bridging using information.

The literature reveals that information needs of individuals are triggered by an interaction between personal and environmental factors (Du Preez 2015; Robson and Robinson 2013). The above assertions corroborate Wilson’s 1996 model of general information behaviour which focuses on the information user depicting the events that take place before an information need arises until the time it is processed and used (Wilson 1999). Some of the antecedents in the case of the Swaziland sugarcane farmers that give rise to their information needs are environmental (context) related – events such as policy changes by the government that resulted in the transition of small farmers from maize growers to sugarcane growers of which they might have very little understanding, as well as the failure of the agriculture extension services in providing adequate information services to rural communities.

2.3 Importance of Studying Information Needs of Farmers

Regularly studying the information needs of farmers is vitally important because farmers’ information needs change in response to changes in the agriculture industry and policy. As a result, farmers need to address issues related to strategic positioning, capacity building, performance monitoring, evaluating new technology, managing information, and accountability (Klair, Boggia, and Richardson 1998). Therefore, it is important for scholars and system developers to understand how people interact with information in their environments or in a sociological group (Choo and Auster 1993; Hepworth 2007). According to Babu et al. (2012), information needs of farmers are not static, an assertion echoed by Bothma and Bergenholtz (2013) who also stated that information needs change over time, implying that information needs are unstable in nature and may change from time to time in relation to a particular context, which can be personal, role-related, or environmental. For instance, the involvement of smallholder sugarcane farmers in the sugarcane growing sector in Swaziland exposed farmers to a new competitive agribusiness terrain which requires a high level of decision-making (Terry 1997), illustrating why studying their information needs cannot be overemphasized.

2.4 Information Resources Used by Farmers

Farmers use a variety of resources to obtain information. These include browsing the internet (e-learning, social media), excursions (field days), colleagues (other farmers), exhibitions (fairs), workshops, agricultural courses or trainings, seminars or conferences, and consulting farmers’ journals (Zondag et al. 2015). According to Chen and Lu (2020), information access channels for farmers can be categorized into traditional media channels, modern media channels, interpersonal communication, and organizational communication. In their study of farmers from rural Tanzania, Msoffe and Ngulube (2016) reported that farmers mostly preferred sourcing of information from family, friends, neighbours, extension officers, researchers, and the radio. Similarly, Al Musawi (2013) cited that farmers have a propensity to use sources of information they are familiar with or rely on sources used by their neighbours. Similarly, in Pakistan, most small-scale sugarcane farmers obtain information about sugarcane production technologies from fellow farmers and progressive farmers (Abbas et al. 2003). Elsewhere, Sifundza and Ntuli (2001) detailed how smallholder sugarcane farmers in Swaziland obtained the latest sugarcane production information from newsletters and bulletins published by the Swaziland Sugar Association (SSA), millers, SSA extension officers, smallholder administrators, and the Ministry of Agriculture. The literature above reveal that farmers have a variety of information resources/channels that farmers use to obtain information, hence it is important for information service providers to determine the most effective ones to ensure effective delivery of information packages.

4.1 Respondents’ Characteristics

Findings revealed that most of the smallholder sugarcane farmers surveyed were males (63 %), with 37 % being females. The disparities between male and female respondents confirm the male dominance in the smallholder cane growing sector as reported by Dlamini and Worth (2018) and Ndlovu, Dlamini, and Nkambule (2014), despite international organizations such as the Food and Agriculture Organization (2012) and the World Bank (2007) encouraging women’s involvement in rural development schemes. The majority of the respondents were aged between 20 and 60 years (85 %) and most had formal education (97.6 %), with 48 % having post high school qualifications. A majority of the farmers (66, 39 %) surveyed were between the ages of 41 and 50 years of age, followed by those between 31 and 40 years (24 %) and 51 and 60 years (21 %), with the least represented being farmers between 20 and 30 years and those above 60 years (15 %). The findings show that the farmers were predominantly within the average active working age, with more farmers between the ages of 20 and 50 years. The study further indicates that a majority (65.5 %) of farmers were supervisors, followed by 23.8 % of managers and 5.4 % executive committee members. Results also revealed that the mean average farm size for smallholder farmers was 47.7 ha, and that a third of the farms have above 50 ha of farmland, which suggests that, technically, they are medium scale farmers as per the SSA farmer classification scheme. The current study indicated that a large proportion of farmers (91 %) had more than five years farming experience, followed by those with two to five years (13, 8 %) and below two years (1 %). The study further revealed that a majority of farmers (67 %) have more than ten full-time employees.

4.2 Information Needed by the Farmers: Factor Analysis

Exploratory factor analysis (herein referred to as factor analysis) was used to determine the type of information needs required by smallholder farmers. As a prerequisite to factor analysis, the dataset was assessed to ensure that it met the preliminary assumptions for further factor analysis. Several considerations were taken into account to ensure the factorability of the dataset.

The first consideration was validation of the sample (sample size). To assess the adequacy of the sample size of the dataset, the Kaiser–Meyer–Olkin measure of sampling adequacy (KMO) and Bartlett’s test of sphericity were used. The KMO test was used to verify whether the data collected qualified for factor analysis whereas the Bartlett’s Test of Sphericity test was conducted to assess whether the dataset had patterned relationships (Field 2005, 652). As shown in Table 1, the KMO values are 0.736 above the recommended threshold of 0.6 and Bartlett’s test of sphericity is significant (p < 0.05), and the acceptable sample sizes to yield acceptable factors were those with a KMO above the threshold of 0.6 (Kaiser 1974). The Bartlett’s test should be significant (p < 0.05) for the data to be factorable (Field 2005, 652).

Another important test was to determine if there were no issues with the dataset by ensuring that there were patterned relationships between the variables in the correlation matrix. Items that lack patterned relationships have correlation coefficients that are less than r = 0.3 (Yong and Pearce 2013). Table 2 shows the correlation matrix for the 11 types of information needs with correlation matrix values of less than one, which implies that correlation values (0.901 … 0.301) are acceptable. In addition, there is “quite a number” (more than half) of correlation values that are greater than 0.3 in magnitude, indicating that factor analysis is appropriate (Hooper 2012). Furthermore, Table 3 depicts the anti-image correlation matrix to measure sampling adequacy for all 11 types of information needs/factors with values between 0.791 and 0.566 above the threshold of 0.5, suggesting that the dataset is suitable for factor analysis (Yong and Pearce 2013).

The third determinant of dataset eligibility for a good factor analysis is the adequacy of the sample size, that is, the number of valid and usable collected questionnaires or cases. Many researchers have given wide varying “rules of thumb” relating to the determination of a suitable sample size. Comrey and Lee (1992) classified adequate sample sizes for factor analysis as follows: 50 = very poor, 100 = poor, 200 = fair, 300 = good, 500 = very good, 1000 or more = excellent. However, some authors are not content with these guidelines (MacCallum et al. 1999; Worthington and Whittaker 2006); Worthington and Whittaker (2006) provided alternative guidelines for determining the minimum necessary sample size, N, in factor analysis, with one of the overarching guidelines recommending that “sample sizes of 150–200 are likely to be adequate with datasets containing communalities higher than 0.50” (Worthington and Whittaker 2006, 817). This view is shared by Pallant (2010), who suggests that, ideally, the data should have at least 150 cases. In this study, the number of respondents or cases used were N = 168, with item commonalities ranging between 0.590 and 0.918 (Table 4), representing a good fit within the low-to-moderate range characterizing social sciences research (Costello and Osborne 2005, 7). Table 4 depicts the commonalities for the variables, with the results showing that after all the factors were extracted through the use of the Principal Component Analysis (PCA) method, all the values are larger than 0.5 which indicates approval of values (Worthington and Whittaker 2006).

Factor extraction was conducted using the principal component analysis (PCA), which Costello and Osborne (2005) argued is the most preferable factor extraction method in factor analysis. As a result, four factors were extracted, which explained about 71 % of the total variance (Table 5). In social science, a solution that represents above 60 % of the variance is considered appropriate (Maskey, Fei, and Nguyen 2018); the higher the value of the factor loading and the total accumulative variance of factor, the better the explanatory power of the data (Asafo-Adjei and Buabeng 2016). Table 5 shows the total variance explained for the types of information needs factors, which shows that four components were extracted, which were above the recommended eigenvalue of 1 (Hooper 2012; Yong and Pearce 2013). The totals for the extracted components were 4.128, 1.408, 1.215 and 1.034 respectively, with variance percentages of 37.525, 12.801, 11.043 and 9.397, with the components having a cumulative percentage of 70.766 before and after rotation using the varimax method. Elsewhere, findings from the factor extraction (Table 6) reveal that the information needs of smallholder sugarcane farmers can be grouped into four factors, which are categorized and labelled as follows: legislative compliance, crop husbandry, markets and transport, and financial advice information. The cumulative variance explained by these four factors was 71 %.

The first factor, “legislative compliance,” had higher loading, accounting for 23 % of the variance, indicating that farmers required information related to policy, government regulation, and training. Farmers have to comply with everchanging rules and regulations on environmental, financial, safety, and employment legislation. The factor suggests that farmers require training information on laws and regulations set out by the government regarding their agricultural operation; it is important that farmers understand all the relevant laws, regulations, and policies they must follow such as the storage and disposal of pesticides and pesticides containers. The consequences of not complying with industry’s legislative requirements can be very costly in terms of penalties and fines. In South Africa, a study by Netnou-Nkoana et al. (2015) found that smallholder farmers were not aware of the existence of the legislation on plant breeders’ rights and its implications. By the same token, Mbagwu, Benson, and Onuoha (2018) observed that in developing countries, there was a lack of a designated information providing agencies, whether from government or the private sector, which is a bottleneck to access to government-related information.

The second factor is labelled “crop husbandry” and accounted for 21 % of the 71 % variance, indicating that farmers are concerned about information on sources of agricultural inputs, diseases and pest control, agronomic information, and weather information. This factor loaded highly, almost at the same level of the first factor. It is widely acknowledged in the farming industry that access to and understanding of the agronomy (e.g. variety selection, fertilizer application) of crops is crucial in producing the highest possible yield of high-quality produce that can generate good profits. The findings are supported by data from interviewees from support agricultural organizations (Table 4) that indicate that farmers were more concerned about information related to crop husbandry such as the agronomy, weather, pests, and disease. One informant mentioned that: “They also need to know how to plant their crops, what kind of planting material to use, where to source such material, how to manage crops in terms of diseases, how to prevent crop diseases and when to use specific chemicals” (Male, Agriculture Economist, in 40s).

These findings are also in line with those of Mtega, Ngoepe, and Luyanda Dube (2016), who found that rice farmers in rural Tanzania required agricultural knowledge on crop husbandry techniques, such as land preparation, seed selection, and planting. Similar findings were revealed in a study by Phiri, Chipeta, and Chawinga (2019), who investigated the information needs of rural smallholder farmers in Malawi and found that a majority of farmers needed crop husbandry information. Another study by Elly and Silayo (2013) assessed the agricultural information needs of rural farmers in Tanzania and discovered that close to one third of farmers’ information needs bordered on crop husbandry, while that of Naveed and Anwar (2013) to identify agricultural information needs of Pakistani farmers found that the information needs of the farmers centered around crop husbandry. The results of these studies perhaps demonstrate the increasing need for dissemination of information that provides more efficient techniques of growing field crops such as sugarcane.

While the study suggests that a lack of access to agronomic information might be an issue on the one hand, the overarching issues seem to relate to the level of information required which could be linked to the education level of the farmer. However, the level of information required for a specific task depends on the information requirements of said task, which has implications for the design of an information provision strategy. Some key informants are of the view that farmers may not comprehend the information, especially from third parties, who are not experienced in dealing with farmers. One key informant said: “Some of them are basic farmers who are not highly educated to be able to understand some of the information they receive, especially from government and other stakeholders who are not necessarily trained to communicate with farmers” (Male, CEO, in 50s).

This assertion is shared by Ndlovu, Dlamini, and Nkambule (2014) and Dlamini and Worth (2016), who were quite concerned about the education level of some farmers which was rather low, which may negatively affect their understanding of technical sugarcane management practices. On the other hand, most technical sugarcane practices could be learned provided that the receiver’s sense making and world views are considered in the transfer of information by applying the relevant information transfer mechanisms with which receivers can relate. According to Meyer (2009), people in rural communities are, to a larger extent, heterogeneous in terms of their information practices, with some information users semi-literate and others illiterate, with smallholder sugarcane farmers falling into the latter category. Meyer argued that indigenous people over the years have developed specific communication mechanisms such as storytelling, memorising, repetition, dancing, acting, observing, and demonstrations to collect, store, and share information. It can therefore be assumed that farmers with a low level of education have the potential to understand sugarcane management practices as long as their oral context is taken into consideration in the formulation of information provision strategies. Findings, on the characteristics of respondents, show that less than one fifth of the farmers had formal tertiary training, in spite of the fact that the agriculture sector is knowledge intensive and complex (Babu et al. 2012). The complexity perhaps comes from the fact that the farmers were initially growing subsistence crops on a small scale, yet now, under a new regime, are obliged to grow sugarcane on commercial basis, and to follow business principles that might be a challenge if they are not well versed on the proper management of the crop.

In the farming industry, farmers must make several decisions that include adoption of more productive technologies, management of these technologies, human resource management, as well as skillful manipulation of the limited resources to improve output (Van den Ban 1998). On the contrary, when information is transferred from sender to receiver, this only applies where both have the same background knowledge of a practice. Where the transfer process spans across cultural boundaries (implying traditional practices versus modern industrial practices), other strategies need to be harnessed to accommodate the receiver (smallholder sugarcane farmer) (Meyer 2009).

The third factor, which explains 13 % of the variance, is labelled “market and transport” with two variables loaded, namely marketing and transport information. Notwithstanding that the marketing of sugar to international markets is the prerogative of the Swaziland Sugarcane Association (Masuku 2003), farmers are still concerned about the uncertainty of the markets; this was concurred by Terry and Ogg (2017) who identified market volatility as a major threat to the profitability of the sugar industry in Swaziland. This factor also reveals that farmers are concerned about transportation of their produce to the (market) mills. Despite the government of Swaziland intensively investing in road infrastructure in the sugarcane belt at the time when there was increased introduction of smallholder sugarcane development projects in the Lowveld, unsubsidized transport cost, distance from the mill, and regulatory regimes on transport affect the profitability and performance of the sugarcane farmers in the Swaziland sugarcane supply chain (Masuku 2011; Mhlanga-Ndlovu and Nhamo 2017). Therefore, access to an efficient transport system is essential for efficient haulage of sugarcane to the mills.

The fourth factor is labelled “financial advice,” which has a positive loading (financial advice) and a negative loading for “other information,” indicating that there was negative correlation between the variables and the component/fourth factor. Simply, there was no other information required by the farmers. Many studies suggest that access to finance is the major obstacle among smallholder farmers (Elly and Silayo 2013; Galadima 2014; Samuel 2008); farmers use financial resources to pay for the operational costs of growing sugarcane such as irrigation equipment, machinery and fertilisers, and harvesting. For instance, Mhlanga-Ndlovu and Nhamo (2017) reported that smallholder farmers spend 38 % of their finances on harvesting. The EU’s financial support, combined with the Swaziland government’s support towards irrigation infrastructure, makes it attractive for a financial institution to finance smallholder sugarcane farmers given their potential to repay loans (Malaza and Myeni 2009; Samuel 2008; Shongwe 2009; Terry and Ogg 2017). According to Shongwe (2009), smallholder sugarcane farmer received grants from the EU accounted for 70 % of the costs involved in the production of sugarcane by the respective farming companies. As a result, financial risk becomes minimal to private financial institutions interested in the financing of farmers because the larger burden is borne by the EU.

Unbundled, the survey shows that the information needs of the smallholder sugarcane farmers are: government policies on farmer development, regulations, training, sources of agricultural inputs, agronomy, disease and pest control, weather, the markets, transport, and finance information. The findings of the study showed that the information needs of farmers are not unique to that of other smallholder farmers in other developing countries, apart from the way they are categorized and the type of enterprise undertaken (Elly and Silayo 2013; Lwoga, Ngulube, and Stilwell 2010; Msoffe and Ngulube 2016; Tena et al. 2016).

The results presented in Table 6 are supported by responses to the interview schedule, where the officials from agricultural support organisations were asked to indicate the kind of requests they receive from smallholder sugarcane farmers. In the same vein, most agricultural support organizations indicated that most farmers required information related to the management of the sugarcane crop (crop husbandry). Table 7 below shows the results of the thematic analysis on the type of information needs of farmers from the perspective of agricultural support organisations.

It became evident from the data that all the officials who were interviewed confirmed that they receive various information requests from SSFs such as agronomy, weather, finance, pests and disease, markets, irrigation, and policy related information. Among these information requests, agronomic information (50 %) emerged as the most important information need category during content analysis. This was followed by finance (33.3 %), pests and disease (33.3 %), and irrigation (33.3 %), while the least important requests were weather (16.6 %) and policy information (16.6 %). It is clear from the results that a majority (four out of seven) of information need categories fall within the theme of crop husbandry; these include agronomy, weather, pests and disease, and irrigation information. Other themes included finance, markets, and policy information, however, these emerged as least important.

4.3 Information Needs Differences Between the Milling Groups

An analysis of variance (ANOVA) was computed to determine if there was a statistically significant difference in the information needs of farmers between the mill groups (Mhlume, Simunye and Big Bend) where smallholder sugarcane farmers belong. The mill groups were used as independent variables. As presented in Table 8, the results show the significance value (F = 0.418, p = 0.659) which is above 0.05. Therefore, there was no statistically significant difference between the information needs of farmers between the three mill groups.

Scholars have asserted that the need for information is heavily influenced by an individual’s association to a social or professional group (Choo and Auster 1993, 284; Du Preez 2015, 242) or a task or role in the work-based environment (context) (Leckie, Pettigrew, and Sylvain 1996). In the same vein, Lwoga, Ngulube, and Stilwell (2010, 97) aver that the information needs of rural farmers are location-specific, implying that individuals may experience different information gaps, defined along their social, professional or geographic affiliations. On that basis, one-way ANOVA was performed on the dataset to ascertain whether there were differences in information needs between the Mhlume, Simunye, and Big Bend Mill groups (Table 8). The findings of the study indicated that there was no statistically significant difference between the information needs of farmers between the mill groups, suggesting that the geographic context had no effect on the information needs of farmers. This means that the information needs of smallholder sugarcane farmers are the same across the mill groups, which could be explained by the fact that most of the farmers under the auspices of these mill groups are mainly located in the semi-arid Lowveld of Swaziland, where production challenges for sugarcane farmers are relatively similar.

Contrary to these findings, Lwoga, Ngulube, and Stilwell (2010), in a study conducted from 12 districts in Tanzania, assessed the information needs of small-scale farmers and discovered that the information needs of farmers were location specific to a certain extent. Similarly, Elly and Silayo (2013) found that there was a significant difference between two wards in farmers’ information needs in Tanzania. The disparities between the results of the present study and that of Lwoga, Ngulube, and Stilwell (2010) and Elly and Silayo (2013) could be attributed to the fact that these studies were conducted in diverse agricultural contexts (crops and livestock enterprises) and the populations of their studies were largely based in diverse agro-ecological zones.

4.4 Information Resources/Channels Used by Farmers

The study analyzed resources of information used by smallholder sugarcane farmers to access agricultural information. In the questionnaire, farmers were asked to indicate the information resources they currently use to access agricultural information, with a list of possible resources provided, with the option to choose more than one. The findings (Table 9) show that smallholder sugarcane farmers prefer to obtain agricultural information through extension officers, farmer groups, mobile phones, input suppliers, personal knowledge, the internet, other farmers, newspapers, and the radio.

A substantial number of farmers expressed high preference for extension officers, indicating that farmers prefer traditional mechanism for accessing information. This is consistent with the work of Dlamini and Worth (2016) and Jibowo and Dube (2008), who reported that extension work still emerges as the main pathway for communicating agricultural information to rural farmers in Swaziland. Literature reveals that the sugar industry in Swaziland heavily relies on the use of extension officers to communicate information to smallholder farmers; SSA, SWADE, millers, and GOS provide extension services to farmers to equip them with skills necessary to be successful in the industry (Dlamini and Dlamini 2012; Malaza and Myeni 2009; Terry and Ogg 2017). Similar findings were observed in a study by Galadima (2014) which investigated the constraints on farmers’ access to agricultural information in Nigeria and reported that 62 % of the farmers consulted extension officers for agricultural related information. The results of this study together with the findings of Galadima’s study clearly showed that farmers prefer to get their information through extension officers. The findings suggest that information providers have to first determine the appropriate means of disseminating information to farmers to maximize the chances of having more farmers receiving information.

When agricultural support organizations were asked to state the mechanism they use to disseminate information, a majority of key informants (83 %) of the study indicated that they do so through face-to-face communication. One key informant claimed that it is very useful to meet the farmer in person so that they can explain the concepts and in turn give the farmer a chance to ask questions. An official from FINCORP stated that: “We interact with our farmers through meetings, one on one. Most of the sugarcane farmers have companies so we meet with their boards of directors to share any information updates on loans” (Male, Branch Manager, in 40s). In the same vein, an official from SWADE stated that: “When we want to update them on certain things which we feel they should know, we organise meetings with them. We go to them and we talk to them in those meetings face to face” (Male, Extension Officer, in 30s).

In Swaziland, extension officers, primarily from the large-scale sugarcane producers and including those from government, aid farmers by ensuring that they adhere to basic good management practices (Dlamini and Worth 2016). The results of the present study are corroborated in a study by Mtega, Ngoepe, and Dube (2016), which investigated the accessibility of agricultural knowledge among smallholder farmers in Tanzania and found that smallholder farmers primarily used agricultural extension officers to access agriculture related information. In agreement, Lwoga, Ngulube, and Stilwell (2010) reported that physical communication is important in sharing agricultural knowledge because farmers are able to internalize knowledge gained from tacit sources compared to explicit sources of knowledge such as ICTs. Thus, information intermediaries must consider the information behavior of farmers in terms of how they handle information as they prepare information products to help farmers improve their farming activities.

Moreover, the findings revealed that the second most preferred information resource is farmer groups, presenting an alternative to only relying on extension officers. The findings of this study suggest that farmers prefer to access information from familiar communication channels, with which they are able to establish relationships. One key informant from SWADE indicated that farmers are continually forming groups to collectively bargain when making purchases for farm inputs like fertilizer. Elsewhere, Dlamini and Worth (2016) averred that extension work plays an important role in the promotion of farmer group formation, while according to Meyer (2009) group action to exchange/share information is a typical information behaviour pattern within rural communities where people are hesitant to act alone, which stems from the oral tradition. Meyer further pointed out that group dynamics play an important role in the acceptance of information packages, and hence an innovation would be more easily adopted in a group setup than it would by an individual acting in isolation.

The study also found that the third most used information resource is mobile phones, which is consistent with the finding of the study that most participant farmers owned mobile phones. The findings provide evidence for the increasing influence of ICTs in rural communities and further support the assertion by the Food and Agriculture Organization (2012) and Dlamini and Worth (2018) that mobile phones have now penetrated even rural communities. The Food and Agriculture Organization (2012) reported that mobile technology is ubiquitous and holds great promise in rural development, and Dlamini and Worth (2018, 25) detail that more than 60 % of farmers in the sugar industry have mobile phones. Moreover, mobile phones can easily deliver undiluted information packages, and in a cost-effective manner (Anunobi and Anunobi 2018). The study further confirmed that while ICTs may be widely used, however, face-to-face interaction is still a predominant mode of communicating agricultural information to smallholder sugarcane farmers. In contrast, results also revealed that the library is the least used information communication channel in the sugar industry. The key informant from the Swaziland National Library Services (SNLS) disclosed that the library receives publications from the Swaziland Sugar Association such as annual reports and periodic agricultural research newsletters. In addition, the library has a collection of government gazettes and has access to a digital library in the field of agriculture called Access to Global Online Research in Agriculture (AGORA). However, the minimal use of the library was caused by the lack of dedicated agricultural information resource sections in the Swaziland National Library Service (SNLS), which is compounded by the fact that the libraries are under-resourced. This was confirmed by the interviewed official from SNLS who said: “The budget for procuring agricultural information is not enough, so we can only buy resources once in a while” (Female, Branch Librarian, in 40s).