1. Introduction
Over the past decades, adoption of novel technologies has fueled rapid growth in global agricultural production and supported development of more efficient food production. Use of advanced information and computer technologies has contributed to enhanced performance in the economy and in the agricultural sector [1]. With the digitization of technology, the world now stands on the cusp of the next agricultural revolution [2].
Technological innovation and applications, such as digitalization and agroecological farming, could play a major role in catalyzing the transformation of agri-food systems towards sustainability and resilience [3]. Indeed, the availability of sensors, mapping technology, and tracking technologies has changed many farming systems and the management of the food system as it flows from producers to consumers [4].
Digital agriculture has advanced rapidly over the last two decades, particularly in the United States and Brazil, which are world leaders in soybean production [5]. And the use of digital agriculture will likely continue or accelerate given the increased venture capital devoted to developing these technologies [6]. The advancement of digital technologies is stimulated by large agricultural companies and start-ups, who collectively invest millions of dollars in technologies that use data on soil types, seed variety, and climate to help farmers increase productivity [7].
However, studies have shown that the need for more understanding of appropriate technologies and skill in using these tools contributes to some farmer apprehension about digital technology [8,9]. The gradual development of the new processing tools has led to slower adoption patterns, reduced increases in information flows and more skeptical perceptions of new technology among some farmers. Adoption of some technologies requires more knowledge about how technology works, and other technologies demand more scale [10].
The factors influencing the adoption rate of agricultural technologies, such as farmer age, farm size, technology cost and complexity, and level of farmer education, have been a focus of several research efforts [11,12]. Barriers to the adoption of Internet of Things-based precision agriculture technologies found in the literature include also factors such as cost effectiveness, power requirements, wireless communication range, data storage and data processing [13].
However, the role of communication, as it impacts broader adoption decisions, remains of interest but is somewhat understudied. While social media channels have emerged as valuable ways to connect farmers and promote discussion, they remain underused in the dissemination of knowledge on smart farming technology [14]. In response, this study investigates how soybean farmers’ use of communication channels affects digital agriculture adoption decisions in the United States and Brazil.
Based on the crucial role played by information in agriculture’s technology adoption and the premise that diffusion is the process by which an innovation is communicated over time among social system participants [15], this research focuses on an important factor influencing the spread of an innovation: communication channels. The theoretical foundation for this study includes aspects from the diffusion of innovations [15]. While this theory speaks to the adoption of technologies in general, this study focused specifically on the role of communication channels in digital technology adoption in agriculture. As such, the diffusion of innovation theory discusses this specifically, yet with little empirical support. Here, we show that the relative influence (we used the word “influence” as a synonym for “relationship”, “association”, or “correlation”. In the context of this study, our intention is to show an indirect impact rather than causality (direct cause-and-effect relationship)) of a specific channel impacts decision-making differently in two important agricultural settings.
The types of communication channels available can affect the success of any innovation diffusion. There are three types of channels discussed in this study—interpersonal meetings (field days, conferences, retailers, extension agents, peer groups, conversations with neighbors, etc.), mass media (newspaper, magazine, radio, television, website/blog, etc.), and social media channels (YouTube, WhatsApp, X/Twitter, Facebook, Instagram, LinkedIn, etc.). Choosing among different types of channels is necessary because we are living in a world experiencing a wider choice of channels and more information overload [16].
This study builds on previous research in which we interviewed soybean farmers in Brazil about the use of technologies and the influence of communication channels on the decision to adopt digital technologies [17]. The relevance of the results in Brazil encouraged the authors to replicate the survey and its research in the United States. Therefore, this study both documents and compares the influence of various communication channels on Brazilian and American farmers’ decisions to adopt digital technologies in soybean production. While U.S. results are of interest by themselves, the comparison across Brazil and the United States is particularly valuable because of the combined share of these two countries in world soybean production, around 70% [18].
Therefore, this study investigated the influence of communication channels—mass media, social media, and interpersonal meetings—on farmers’ adoption, decision-making, and benefits obtained through the use of digital technologies in the United States and Brazil. For the purposes of this article, the term digital technology is used to refer both to precision and digital technologies applied on-farm.
2. Materials and Methods
2.1. Study Region
This comparative study was conducted in Brazil’s top five soybean-producing states (Mato Grosso, Paraná, Rio Grande do Sul, Goiás and Mato Grosso do Sul) and in the United States’ top nine soybean-producing states (Illinois, Iowa, Indiana, Minnesota, Nebraska, Ohio, Missouri, North Dakota, and South Dakota) (Figure 1). These states account for about 75% of soybean production in each country. Completed surveys were obtained from 461 farmers in Brazil and 340 farmers in the United States.
Soybean production in Brazil originated in the states of Rio Grande do Sul and Paraná, southern states with a sub-tropical climate. Following the introduction of the crop, the Brazilian Agricultural Research Corporation, Brasília, Brazil (EMBRAPA, the acronym in Portuguese) started to develop seeds suited to the tropical climate of the savanna. Consequently, in the 1970s and 1980s, soybean production migrated to the Center West states of Mato Grosso do Sul, Mato Grosso, and Goiás. Development of cultivars well adapted to the low latitude of the region, being less sensitive to the length of daylight, enabled this shift [19].
The study area in the United States comprises a portion of its Midwestern states. Known for its fertile land and agricultural prominence, 75% of the area’s cropland is planted in corn and soybeans [20]. The U.S. Midwest is one of the world’s most intensive agricultural production areas, consistently impacting the global economy [20]. The adoption and implementation of digital agriculture technology have the potential to contribute to on-farm economic and environmental benefits in this region as well as in Brazil [21].
2.2. Survey Instrument
The questionnaire survey employed in this study was based on a literature review and previous research focused on digital agriculture adoption. The online survey (the surveys, in English and Portuguese versions, are available from the corresponding author on request) was prepared in English and Portuguese, the official languages in the United States and Brazil, respectively. The survey instrument was hosted on the Qualtrics virtual platform.
The first section asked about the digital technology tools used on the farm, the decisions influenced by use of these tools, and the perceived benefits of use of the tools. The variables in each question were adapted from the Precision Agriculture Dealership Survey conducted by CropLife magazine and Purdue University [22]. The CropLife survey is extensive, so we adapted it to cover key precision agriculture technologies. We employed the technologies with the highest percentages of use in the results of the 2017 and 2019 Purdue/CropLife Precision Agriculture Dealer Surveys [22,23].
The second section explored the influence of mass media, social media, and interpersonal meetings on the decision to adopt a new digital technology on-farm. In the questionnaire applied in Brazil, the variables regarding social media were based on Brazilian Association of Rural Marketing and Agribusiness (ABMRA, São Paulo, Brazil) reports, a study of farmers’ media. The ABMRA study considered nine social media platforms. The current study chose the top six platforms in terms of use according to their results in 2017 (data were obtained from ABMRA and are available from the corresponding author with the permission of the ABMRA board).
In the questionnaire employed in the U.S., the variables regarding social media were based on the Social Media Use in 2021 report from the Pew Research Center [24]. The report considered 11 social media platforms. The current study chose the top seven to investigate.
The last section focused on the demographic profile of the respondents. In both countries, we collected data such as the number of hectares planted, age group, and level of education. Soybean farmers in both countries were asked to weigh the use of eight digital technologies on interval scales in numeric format, from 1 being “never use” to 5 being “always use”, and to weigh the influence of different types of communication channels on interval scales in numeric format, from 1 being “not at all influential” to 5 being “extremely influential”. The scale’s reliability was validated with the help of a pilot survey of 10 respondents in Brazil and 8 in the United States, leading to minor alterations in the final questionnaire.
2.3. Data Collection
Online surveys were distributed to soybean farmers in both countries. In Brazil, the data were collected through an online questionnaire available to the farmers from March to June 2021. In the United States, data were collected through an online questionnaire open to the farmers from August 2021 to April 2022.
A convenience sampling method, facilitated by cooperating agricultural associations and universities, was the method used to obtain the data used in this study. A main advantage of this method is that it can be used in a cost-effective manner when a list of all population estimates is not available. However, convenience sampling has limitations, the foremost being that variability and bias cannot be measured or controlled. Secondly, results from the data cannot be generalized beyond the sample [25]. Members of the target population met specific criteria, such as easy accessibility, availability at a given time, or the willingness to participate [26].
In Brazil, the online survey was distributed with help from the Brazilian Association of Soybean Growers (Aprosoja Brasil), National Supply Company (Conab, Brazil), state associations of rural producers, and agricultural cooperatives. In the United States, the online survey was distributed primarily among Farmdoc subscribers. Farmdoc is an outreach program of the Extension Service of the University of Illinois. In addition, state commodity associations, including state soybean associations in Illinois, Iowa, Ohio, and North Dakota, distributed survey links.
In both countries, these organizations sent information about the study to potential farmer participants. An invitation message, including a link to the survey, explained the survey’s objectives. In some cases, farmer participants distributed the study further to other farmers as potential research participants. Therefore, membership from those organizations was not a criterion for participation in the survey.
2.4. Data Analysis
All survey data were consolidated in a Qualtrics platform report, exported in CSV, and imported into a spreadsheet. The data were analyzed using descriptive statistics, Spearman rank correlation (ρS), and one-way analysis of variance (ANOVA). We used the Statistical Package for Social Science (SPSS), Version 18.0 for Windows, to analyze data with a minimum level of statistical significance of p < 0.05.
The Spearman rank correlation was used to determine the significance of correlations of individual variables. Spearman correlation is recommended for ordinal variables, and when data follow a non-normal distribution. The first construct, “use of digital technologies”, and its three variables were correlated with the second construct, “influence of communication channels”, and its three variables. However, it is important to point out that these relationships between different variables do not imply causality.
The underlying rationale for this research was the recognition that without a clearer understanding of the role of communication in farmers’ technology adoption, it is difficult to address the persistent lack of understanding surrounding smart farming technologies in agriculture and consequent low adoption rates [8,9]. This investigation correlating use of digital technologies and communication channels is crucial because any innovation’s successful diffusion depends upon the communication channels available [27]. Moreover, information and communications technologies are playing an increasingly important role in keeping farmers and rural entrepreneurs informed about agricultural innovations [28].
The ANOVA was employed to identify statistically significant differences between means in Brazil and the United States about the level of use of digital technologies, the decisions influenced by using these technologies, and their perceived benefits. Also, the ANOVA was employed to identify statistically significant differences in demographic profiles (age, education, and farm size) about the use of technologies and influence from mass media, social media, and interpersonal meetings on farmer decisions. We used ANOVA, a parametric test, because our number of observations is large for the comparative analyses between the groups. Moreover, these observations are symmetrical, i.e., they are not skewed to one side or the other of the distribution. Both of these factors reduce the potential for bias
Before the analysis, the reliability of the scales used to measure the variables was investigated using Cronbach’s α coefficient. A Cronbach’s α coefficient higher than 0.7 indicates that the different items can be summed and that the median can be used to represent these constructs. In Brazil, the Cronbach’s α indicated that all the scales used for digital technologies (α = 0.83), making decisions (α = 0.87), perceived benefits (α = 0.88), mass media (α = 0.78), social media (α = 0.76), and interpersonal meetings (α = 0.77) were within the acceptable limit. In the United States, the Cronbach’s α also indicated that all the scales used were within the acceptable limit, as follows: digital technologies (α = 0.80), making decisions (α = 0.85), perceived benefits (α = 0.86), mass media (α = 0.78), social media (α = 0.75), and interpersonal meetings (α = 0.78).
The following formula for sampling from an infinite population was used to calculate the number of elements of the sample according to the level of confidence and margin of error:
n = 4·p·q/E2
n = number of elements in the sample;
p = probability of finding the phenomenon studied in the population;
q = probability of not finding the phenomenon studied in the population; and
E = margin of error.
The “p” and “q” in the formula were determined by the probability of finding or not finding potential soybean farmers in the population investigated. Results gathered with 461 farmers in five Brazilian states reached a 95.3% confidence level and a margin of error of 4.7%. In the United States, 340 farmers were interviewed in nine American states, obtaining a confidence level of 94.5% and a margin of error of 5.5%.
2.5. Sample Characteristics
Demographic characteristics assessed in the study were age, educational level, and farm size. Differences exist in the farmer profile between Brazil and the United States. Farmers in Brazil tend to be younger than in the United States. Among the farmers who participated in the survey in Brazil, 43.2% are under 41 years old. Conversely, in the United States, only 17.1% are under 41 years old. The percentage of farmers over 56 is higher in the United States: in Brazil, 21.4% are more than 56 years old, while in the United States, 61% are more than 56 years old (Figure 2). The age difference is consistent with the average age of producers in both countries. In the United States, 62% of farmers are older than 55 years old, according to the US Census of Agriculture [29]. In Brazil, the percentage of farmers older than 55 years old is 46%, according to data from the Census of Agriculture in Brazil [30].
Figure 3 shows the respondents’ level of education in each country. The proportion of the respondents with a high school degree or less is similar in both Brazil and the United States: 25.2% in Brazil compared to 29.4% in the United State. However, the mix of education levels beyond high school differed substantially. Regarding the bachelor’s degree, 39.7% of the respondents in Brazil have obtained that degree, while 53.2% of the respondents have done so in the United States. In Brazil, 35.1% of respondents have a postgraduate degree (MBA, master’s, or doctorate). Conversely in the United States, the percentage of postgraduate degree holders is only 17.1%.
Figure 4 shows the respondents’ farm sizes in both countries. Note that the farm size categories are not identical because of the difference between the metric and imperial systems. Most countries, including Brazil, use the metric system, where the metric for area is hectares. In the United States, the imperial system is used, and area is measured in acres. To minimize confusion among the respondents, we used hectares in Brazil and acres in the United States. One hectare is equal to 10,000 square meters or 2.471 acres. Fortunately, it was possible to create three categories that are roughly similar to compare responses between the countries.
In Brazil, 50.9% of the soybean producer respondents farm less than 500 hectares. Meanwhile, in the United States, 38.6% farm less than 405 hectares. The percentage of respondents who farm more than 2000 hectares in Brazil is almost double that of the United States (Figure 4). The sample of respondents in Brazil is 62% from the South, characterized by small and medium properties, and 38% from the Center West, characterized by large farms. The distribution of the sample is relatively consistent with the distribution of soybean farms in Brazil [30].
3. Results and Discussion
This section presents and discusses the findings regarding the level of the use of digital technology tools by the respondents, the decisions influenced by the use of these tools, and the perceived benefits of use of the tools. We also describe and analyze the influence of mass media, social media, and interpersonal meetings on farmer adoption of new technology. Moreover, the section reports on the relationships between the demographic profile and adoption of the technologies of interest in this study. Lastly, the section presents the association between communication channels and levels of technology adoption by soybean producers in Brazil and the United States.
3.1. Technology Adoption, Decisions, and Benefits
The survey in Brazil and the United States asked about the level of the use of digital technology tools on-farm on an interval 5-point scale, from 1 = Never use to 5 = Always use. Table 1 shows the mean responses for each technology in both countries.
Respondents reported a higher adoption rate for digital technology tools in the United States for seven of the eight digital technologies analyzed in the study: Guidance/Autosteer, Yield monitors, Soil electrical conductivity mapping, Wired or wireless sensor networks, Electronic records/mapping for traceability, Sprayer control systems, and Automatic rate control telematics (Table 1). The results of the one-way ANOVA (All numeric results of the one-way ANOVA from this section are available from the corresponding author on request) did show a significant difference between the country means relative to the seven technologies just listed. The exception was the Satellite/drone imagery, the only technology with higher adoption in Brazil (M: 2.99) than in the United States (M: 2.94).
The Global Navigation Satellite Systems (GNSS) technology enabled the beginning of precision agriculture worldwide. The combination of GNSS-enabled soil sampling, variable rate fertilizer applications, and yield monitoring was the “classic precision agriculture” package in the 1990s, and some studies focus on whether that classic package has been adopted [31]. Not by coincidence, Yield monitors had the second-highest mean in the survey in the United States, and the third-highest mean in Brazil (Table 1).
In the United States, two technologies reached an average of over 4 points: Guidance/Autosteer (M: 4.23) and Yield monitors (M: 4.31). In Brazil, no technology had a mean above 4 points, and only one technology had a mean exceeding 3 points: Guidance/Autosteer (M: 3.56). The introduction of Guidance/Autosteer systems trailed the adoption of the classic precision agriculture package in the early 1990s. Autosteer has numerous advantages, including less operator fatigue, more time focused on the operating equipment, and less waste of applied inputs [22].
The same technology has the lowest mean respondents in both the United States (M:1.81) and Brazil (M:1.50): Soil electrical conductivity mapping. This technology is newer, and its capabilities are less well known in comparison to the more established ones.
In summary, the study shows a higher use level in the United States than in Brazil in seven of eight digital technologies analyzed. This result in use level is consistent with the time that these agricultural technologies have been available in each country. Technology adoption is a process that occurs over time among the participants in a social system [15]. Introduction of precision agriculture in the United States tended to precede the corresponding introduction in Brazil. Throughout the 1990s, advances in information technology and new uses of that technology were a constant feature of American society. Within production agriculture, the concept of precision agriculture also has emerged over this time period [32].
Many studies have been conducted on the adoption rate of precision agriculture technologies in the United States and Brazil. Although results among these studies vary, adoption rates have generally increased over the last two decades, despite being behind what many researchers expected. In the United States, the overall adoption rates rarely exceed 50% of farms or even 50% of planted areas [33,34]. Although precision agriculture is already a reality for professional and rural producers in Brazil, there are still gaps in the adoption process [8,9]. A study conducted in 2021 by IHS Markit—Business Intelligence showed an adoption rate of 34% among soybean farmers in Brazil [35].
Respondents in Brazil and the United States also indicated the influence of the use of digital technology on making decisions, on an interval 5-point scale, from 1 = Not at all influential to 5 = Extremely influential. Table 2 shows the mean responses for each technology in both countries.
The results of the one-way ANOVA (All numeric results of the one-way ANOVA from this section are available from the corresponding author on request) showed a significant difference between the means in both countries for five of the seven decision types analyzed in the study. The exceptions were Overall hybrid/variety selection and Overall crop planting rates, which have similar means among U.S. and Brazilian farmers (see Table 2). These decisions involve crop management according to field variability and site-specific conditions. Improved decision-making can boost the efficient use of resources, reduce input costs, and improve yields [36].
Despite having reached a higher mean in the United States (M:3.93) than in Brazil (M: 3.64), the Nitrogen, phosphorus, potassium (NPK) fertilization and liming applications had the highest means among the respondents in two countries. Nutrient application can be variably applied based on soil test results and analysis of prior yield data. The willingness and ability of agricultural retailers to provide variable rate applications of various fertilizers has grown over time [5].
The means in Brazil were higher than in the United States for Pesticide selection (herbicides, insecticides or fungicides), Cropping sequence/rotations, and Irrigation (Table 2). These results can be understood by the more intensive use of agricultural land in some regions of Brazil, where two crops can be grown in one year. The tropical environment in Brazil allows pest populations to go through multiple generations per year, consequently increasing selection pressure [37].
Although with a higher mean in Brazil (M: 2.02) than in the United States (M: 1.41), Irrigation was the decision with the lowest means in both countries. This likely can be associated with the smaller number of farmers using these technologies in relation to the other groups of technologies evaluated. For example, in Brazil, less than 5% of the total soybean area harvested was irrigated in 2020, according to data from Embrapa and Conab. In the United States, this proportion was less than 10%, higher than in Brazil and still low relative to the total production area [20].
Another aspect of the survey investigated the benefits obtained through the adoption and use of precision and digital technologies as perceived by the farmer respondents. Table 3 shows the mean responses for each technology in both countries, from 1 = Not at all influential to 5 = Extremely influential.
The results of the one-way ANOVA (all numeric results of the one-way ANOVA from this section are available from the corresponding author on request) showed a significant difference between the means in Brazil and the United States for six among eight benefits analyzed in the study. The exceptions were Cost reductions and Purchase of inputs, which had means similar among U.S. and Brazilian farmers (see Table 3). Both decisions are a constant concern in a commodity market with thin profit margins.
Looking at the differences between both countries regarding benefits obtained, Autosteer (less fatigue/stress) had the highest mean among the respondents in the United States (M:4.18). Meanwhile, in Brazil, the highest mean was reported for Increased crop productivity/yields (M: 3.70).
Another significant difference between the countries was noted for the Lower environmental impact category, with a higher mean in Brazil (M: 3.34) than in the United States (M: 2.99). This result may be because of the considerable concern focused on the potential environmental impacts of soybean production in Brazil’s Center West and North. Soybean production and its supply chain depend highly on land, fertilizer, fuel, machines, pesticides, and electricity [38]. The pressure from end-use consumers to adopt digital technology on the farm is mounting. One motivation for that pressure is the desire for more sustainable cropping systems.
Of course, the use of technology varies from farmer to farmer. Producers are heterogeneous in their perceptions of digital farming technologies, and their perceptions are also influenced by the technologies they use [5]. Despite the differences, the results suggest that farmers in Brazil and the United States perceive substantial benefits from using technologies in soybean production.
3.2. Level of Influence from Mass Media, Social Media, and Interpersonal Meetings
Soybean farmers in the United States and Brazil also were asked to report on the level of influence of mass media, social media, and interpersonal meetings on their decisions to adopt digital technology on an interval 5-point scale, from 1 = Not at all influential to 5 = Extremely influential. Table 4 shows the means regarding each question in both countries:
In relation to the mass media group, the results of the one-way ANOVA showed a significant difference between the means in Brazil and the United States for four among six channels analyzed in the study. The exceptions were Television and Website and blog, which reached similar means in both countries (Table 4). Website and blog, for example, had the highest average in Brazil (M: 3.38) and the United States (M: 3.41). The result is consistent with the rapid growth of the internet, the ease of global communication, and the ability of information to spread with surprising speed and intensity [39]. In addition, as more agribusinesses become digital, the power of the internet is likely to continue to grow [40].
Looking at the differences between countries regarding the influence of mass media on farmers’ adoption decisions, the Radio remains more relevant to farmers in the United States (M: 2.40) than in Brazil (M: 2.17). This channel can be accessed in tractors, cars and trucks while operating agricultural equipment. The Radio also usually brings local news that interests the grain producer, such as weather forecasts. Another difference between both countries was in relation to the Newspaper and Magazine channels, which reached higher levels of influence among the respondents in the United States than in Brazil, but still below 3 points. In addition, Cable television had the second-highest mean among Brazilian respondents (M: 2.41), whereas in the U.S., it had the lowest mean among the mass media group (M: 1.55).
In relation to the social media group, interestingly, there was a noticeable difference in the overall influence attributed between the Brazilian and United States respondents. For each channel for which respondents in both countries could respond, the level of influence reported in Brazil exceeded that reported among U.S. respondents (Table 4). Age differences—with Brazil generally having younger producers than the United States—likely influence differences between countries. The results of the one-way ANOVA (all numeric results of the one-way ANOVA from this section are available from the corresponding author on request) showed a significant difference between the means in Brazil and the United States for all four social media investigated in both countries (YouTube, Facebook, LinkedIn, and Instagram). If social media categories were not tested in both countries, comparisons were not possible.
WhatsApp stands out in Brazil with the highest mean in the social media group, with a 3.65 average. WhatsApp is an online instant messaging service for mobile devices. With approximately 2 billion monthly active users, WhatsApp is the most popular mobile messenger app worldwide, surpassing WeChat at 1.2 billion and Facebook Messenger at 988 million global users [41]. In the United States, YouTube was the highest-ranked social media channel, with a 2.52 average. The number of YouTube users worldwide equaled 2.1 billion in 2022, up from 1.47 billion in 2017 [41].
For the interpersonal meetings group, the results of the one-way ANOVA (All numeric results of the one-way ANOVA from this section are available from the corresponding author on request) showed a significant difference between the means in Brazil and the United States for four among six benefits analyzed in the study. The exceptions were Extension agents and Peer groups, which had similar results among U.S. and Brazilian farmers (Table 4).
Specialist connections and visits to farms, for example, help the dissemination of innovative technologies. Agricultural extension services, which provide consultancy and education, can have an impact on technology acceptance [42,43]. Peer groups facilitate sharing specific context-sensitive knowledge that makes intuitive, practical sense [44]. Also, as more farmers get online, they are building digital relationships with other farmers to form virtual communities of practice [40].
Interpersonal meetings were more highly ranked in Brazil than in the United States—Field days, Conferences, forums and seminars and Conversations with neighbors, although each category exceeded 3 on a 5-point scale in both countries. These events typically offer educational opportunities for producers interested in crop production, farm management, land use, and other topics. Furthermore, these are opportunities for farmers and other agroindustry chain agents to share information and experiences. Many assume that farmers learn by observing experimentation of their neighbors [44].
3.3. Relationship between the Adoption of Technologies and Communication Channels
In this last section, the results are discussed using Spearman’s correlation to measure the strength of the association between the communication channels and the level of technology adoption. In both countries, positive correlations exist between all digital technologies and several mass media, social media, and interpersonal meetings; however, the communication channels most associated with using technologies are quite different in the United States than in Brazil. Table 5 shows the three communication channels with the highest correlation coefficients for each of the eight technologies.
Although the correlation coefficients were relatively low, below ρS 0.300, this was expected given the nature of the analysis involving variable and ranking categories [45] 2020. Our goal was not to find high correlations but to identify the most significant correlations. First, these kinds of correlations (between communication channels and the use of technologies) are not widely available. Second, this information can be valuable for agribusiness decision-makers to address the persistent lack of understanding and low adoption rates of smart farming technologies in agriculture.
To summarize the results from the Spearman’s correlation between use of digital technologies and communication channels, we show in Table 5b only the communication channels listed in Table 5a and the number of times listed with significant impact determined by the Spearman’s rank correlation coefficient (ρS). Channels are listed from most to least cited within their category (mass media, social media, and interpersonal meetings).
Differences in influence exist between Brazil and the United States, as shown in Table 5b. Mass media channels have more influence on technology adoption in the United States, appearing as significant seven times, while those channels appeared only twice in Brazil. Meanwhile, interpersonal meetings channels are more significant in Brazil, showing up eight times, while interpersonal meeting only appear three times in the United States. Furthermore, social media are significant 14 times in both countries, but channels vary across countries. LinkedIn is significant in Brazil, while YouTube is significant in the United States (Table 5b).
LinkedIn had the highest positive correlation in Brazil for seven of the digital technologies (Table 5a). This result may be somewhat surprising because the social media for business professionals had a low mean (M: 2.03 points) in the question about the level of influence of social media on farmer adoption decisions (see Table 4). Although the low mean indicated that fewer farmers use this channel, LinkedIn may have the strongest association with producers who use these technologies. In other words, farmers who said LinkedIn influences their decision-making have the highest rates of on-farm technology adoption.
Moreover, respondents’ education level may have influenced the results around LinkedIn. Among the farmers surveyed in Brazil, 39.7% have a bachelor’s degree, and 35.1% have a graduate degree. According to the Social Media Use in 2021 report, people with higher levels of education are more likely to use LinkedIn than those with lower levels of education [24].
Instagram also showed a positive association with the use of digital technologies in Brazil. Instagram began as a photo-sharing platform in 2010, quickly gaining popularity and attracting many followers, leading to its creative application by bloggers and marketers. Instagram has expanded from photo sharing to video and live streaming. Also, it is the main platform used by digital influencers and digital media content creators to influence audience behavior.
Meanwhile in the United States, YouTube had the highest positive correlation with four of eight digital technologies analyzed among American farmers (Table 5a). Note that the four technologies with the highest positive correlation with YouTube—Automatic rate control telematics, Guidance/Autosteer, Yield monitors, and Spray control systems—are the same ones that reached the highest means in relation to the use of on-farm digital technologies (Table 1). The results suggest an association among adopters of these long-used technologies, present since the beginning of the implementation of precision agriculture, with YouTube.
During the pandemic, YouTube grew the fastest of any social media app among American users [24]. Farmers frequently use YouTube to learn about agricultural innovations, emerging technologies, and specialized skills. In addition, the live streaming service is popular among producers, especially younger internet users.
Website and blog also showed a positive association with six of eight digital technologies analyzed in the United States (Table 5a). The result is in line with the answers from respondents that indicated Website and blog as the most influential mass media channel in their decisions to adopt digital technology on-farm. In addition, this channel was the only one that reached a mean above 3 points among the respondents within the mass and social media groups, at the same level as the interpersonal meetings (Table 4).
Another difference between Brazil and the United States pointed out in the findings is the interpersonal meetings most associated with the use of technologies. In Brazil, Conversation with neighbors and Conferences, forums, and seminars had the highest correlations with the use of technologies. In contrast, Peer groups and Retailers and extension agents appear as the most influential interpersonal meetings in the United States. Regardless of the nature of the interpersonal meetings, the results reinforce the role of in-person social networks in influencing the respondents’ propensity for innovation adoption.
The communication channels most associated with perceived benefits using technologies on-farm are quite different in the United States compared to Brazil. The Spearman’s correlation (the Spearman’s correlations between perceived benefits of using technologies on-farm and communication channels are available from the corresponding author on request) showed a higher association with mass media channels in the United States than in Brazil. Website and blog had the highest positive correlation with seven among eight perceived benefits of using technologies analyzed in the United States. In Brazil, Conferences had the highest positive correlation with six perceived benefits of using technologies on-farm among the eight analyzed in the study.
Lastly, one important finding concerns the link between decision-making and communication channels. Another Spearman’s correlation (the Spearman’s correlations between making decisions and communication channels are available from the corresponding author on request) showed a higher relevance for interpersonal meetings in Brazil than in the United States, yet interpersonal meetings are still important in a U.S. context. The findings may imply that adopters of more established digital agriculture technologies prioritize in-person connections. On the other hand, adopters of emerging technologies tend to prefer social media. Website and Cable television stood out with the highest correlation coefficients in the United States.
4. Conclusions
This study analyzed relationships among communication channels and digital technologies most used by soybean farmers in Brazil and the United States. Based on the survey data of 461 Brazilian soybean farmers and 340 American soybean farmers, we found differences and similarities in Brazilian and American producer behavior regarding the adoption of technologies. The descriptive results showed a higher adoption level in the United States than in Brazil in seven of eight digital technologies analyzed. This result is consistent with the length of time that these digital agriculture technologies have been available in each country.
The results in the United States and Brazil are relatively similar regarding the influence of digital technologies in making decisions and perceiving benefits. The results also suggest that farmers perceive substantial benefits from using technologies in soybean production in Brazil and the United States, especially regarding the potential for increases in efficiency and profitability.
Conversely, there is a noticeable difference in the overall influence attributed to social media between the Brazilian and United States respondents. For each channel for which respondents in both countries could respond, the level of influence reported in Brazil exceeded that reported among U.S. respondents. This result could be correlated to the respondent age groups in both countries, with younger respondents in Brazil than in the United States.
The results overall offer essential insights into current farmer behavior regarding adopting new technologies, helping analyze strategies for generating and disseminating information about digital technologies for agriculture in Brazil and the United States. A better understanding of the role of communication around technologies continues to be necessary, especially because new agriculture technologies are becoming available, such as blockchain, traceability, robotics, artificial intelligence, etc. Hopefully, findings in both countries will enable farmers, agribusiness managers, academics and government decision-makers to use communication channels more effectively in evaluating and adopting digital technologies.
Several limitations of the study should be noted. The results are contingent upon the respondent sample. Therefore, caution should be exercised when extrapolating or generalizing the results to all soybean farmers in these countries. In addition, the convenience sampling method implies self-selection, which means that it is completely left to individuals to select themselves for the survey. Therefore, this process could lead to self-selection bias. Those who are more interested or have stronger opinions about digital agriculture might be more likely to respond.
Another limitation is data collection, which was conducted online due to the pandemic and in different times in Brazil and the United States. Consequently, the findings may have a bias influenced by the profile of online respondents, who are typically more adept at technologies, and by the different timings of the survey in the two countries.
The study presented here did not analyze the difference between heavy and light users of technologies. Future research could fill this gap by providing insights on how their communication channels differ from those with the same demographic profiles but differences in the extent of use. Moreover, the study did not evaluate the causal relationship between the use of technologies and the influence of communication channels. Further research could expand to a deeper analysis of soybean farmers’ behavior regarding technology adoption and the influence of communication channels.
Finally, the study should be repeated in agricultural developing countries, such as South Africa, Nigeria, and Zambia—the top three soybean producers on the African continent. As the behavior of the individual actors may differ, the study will be essential to enable agribusiness managers to be more effective in the evaluation and potential adoption of digital technologies in these regions—where an increase in technology adoption likely could materially advance agricultural development.
Author Contributions
J.C., S.S. and G.D.S. conceived and designed the study, and E.L.M. completed the paper in English and revised it critically for intellectual content; A.D.P. provided research advice and revised the manuscript. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
Not applicable for studies not involving humans or animals.
Data Availability Statement
The data used to support the findings of this study are available from the authors upon request.
Acknowledgments
We kindly acknowledge the statistical support of Lourdes Odete dos Santos and Paulo Waquil, both from the Federal University of Rio Grande do Sul (UFRGS). We also gratefully acknowledge the National Council for Scientific and Technological Development (CNPq) from Brazil and the Department of Agricultural and Consumer Economics at the University of Illinois Urbana-Champaign for their research support. Data used in this analysis were collected in collaboration with the Brazilian Association of Soybean Growers (Aprosoja); Brazil’s National Supply Company (Conab); Union and Organization of the Cooperatives of the State of Paraná (Ocepar); Federation of Agriculture and Livestock in Mato Grosso (Famato); Federation of Agriculture in Rio Grande do Sul (Farsul); Federation of Agriculture in Paraná (Faep); Federation of Agriculture in Goiás (Faeg); Brazilian Soybean Strategic Committee (Cesb); Sustainable Agriculture Associated Group; Aquarius Project from the Federal University of Santa Maria (UFSM); Farmdoc/University of Illinois; Illinois Farm Business Farm Management Association (FBFM); Illinois Soybean Association; Illinois Farm Bureau; Iowa Soybean Association; Nebraska Soybean Board; North Dakota Soybean Growers Association; Ohio Soybean Association; Ohio State University; Missouri Soybeans; and Minnesota Soybean Growers Association.
Conflicts of Interest
The authors declare no conflicts of interest.
References
- Sonka, S. Big data and the ag sector: More than lots of numbers. Int. Food Agribus. Manag. Rev. 2014, 17, 1–20. [Google Scholar] [CrossRef]
- Sridhar, A.; Balakrishnan, A.; Jacob, M.M.; Sillanpää, M.; Dayanandan, N. Global impact of COVID-19 on agriculture: Role of sustainable agriculture and digital farming. Environ. Sci. Pollut. Res. 2023, 30, 42509–42525. [Google Scholar] [CrossRef] [PubMed]
- Mao, R.; Shanoyan, A.; Brooks, K. Special issue: Agri-food systems transformation: Sustainability, resilience, and the role of technology. Int. Food Agribus. Manag. Rev. 2023, 26, 373–377. [Google Scholar] [CrossRef]
- Coble, K.H.; Mishra, A.K.; Ferrell, S.; Griffin, T. Big data in agriculture: A challenge for the future. Appl. Econ. Perspect. Policy 2018, 40, 79–96. [Google Scholar] [CrossRef]
- Thompson, N.M.; Bir, C.; Widmar, D.A.; Mintert, J.R. Farmer perceptions of precision agriculture technology benefits. J. Agric. Appl. Econ. 2019, 51, 142–163. [Google Scholar] [CrossRef]
- Boehlje, M.; Langemeier, M. The Role of Information in Today’s Uncertain Business Climate. Farmdoc Dly. 2021, 11, 41. Available online: https://farmdocdaily.illinois.edu/2021/03/the-role-of-information-in-todays-uncertain-business-climate.html (accessed on 18 December 2022).
- Faulkner, A.; Cebul, K.; McHenry, G. Agriculture Gets Smart: The Rise of Data and Robotics; Cleantech Agriculture Report; Cleantech Group: San Francisco, CA, USA, 2014. [Google Scholar]
- Bolfe, É.L.; Jorge, L.A.D.C.; Sanches, I.D.A.; Luchiari Júnior, A.; da Costa, C.C.; Victoria, D.D.C.; Inamasu, R.Y.; Grego, C.R.; Ferreira, V.R.; Ramirez, A.R. Precision and digital agriculture: Adoption of technologies and perception of Brazilian farmers. Agriculture 2020, 10, 653. [Google Scholar] [CrossRef]
- Gelb, E.; Voet, H. ICT Adoption Trends in Agriculture: A summary of the EFITA ICT Adoption Questionnaires (1999–2009). Letzter Zugriff 2010, 20, 1–11. Available online: http://departments.agri.huji.ac.il/economics/voet-gelb.pdf (accessed on 10 May 2023).
- Pivoto, D.; Barham, B.; Waquil, P.D.; Foguesatto, C.R.; Corte VF, D.; Zhang, D.; Talamini, E. Factors influencing the adoption of smart farming by Brazilian grain farmers. Int. Food Agribus. Manag. Rev. 2019, 22, 571–588. [Google Scholar] [CrossRef]
- Lambert, D.M.; English, B.C.; Harper, D.C.; Larkin, S.L.; Larson, J.A.; Mooney, D.F.; Roberts, R.K.; Velandia, M.; Reeves, J.M. Adoption and frequency of precision soil testing in cotton production. J. Agric. Resour. Econ. 2014, 39, 106–123. [Google Scholar]
- Roberts, R.K.; English, B.C.; Larson, J.A.; Cochran, R.L.; Goodman, W.R.; Larkin, S.L.; Marra, M.C.; Martin, S.W.; Shurley, W.D.; Reeves, J.M. Adoption of site-specific information and variable-rate technologies in cotton precision farming. J. Agric. Appl. Econ. 2004, 36, 143–158. [Google Scholar] [CrossRef]
- Hundal, G.S.; Laux, C.M.; Buckmaster, D.; Sutton, M.J.; Langemeier, M. Exploring barriers to the adoption of internet of things-based precision agriculture practices. Agriculture 2023, 13, 163. [Google Scholar] [CrossRef]
- Dilleen, G.; Claffey, E.; Foley, A.; Doolin, K. Investigating knowledge dissemination and social media use in the farming network to build trust in smart farming technology adoption. J. Bus. Ind. Mark. 2023, 38, 1754–1765. [Google Scholar] [CrossRef]
- Rogers, E.M.; Singhal, A.; Quinlan, M.M. Diffusion of innovations. In An integrated Approach to Communication Theory and Research; Routledge: London, UK, 2014; pp. 432–448. [Google Scholar]
- Bawden, D.; Robinson, L. Information overload: An overview. In Oxford Encyclopedia of Political Decision Making; Oxford University Press: Oxford, UK, 2020. [Google Scholar] [CrossRef]
- Colussi, J.; Morgan, E.L.; Schnitkey, G.D.; Padula, A.D. How Communication Affects the Adoption of Digital Technologies in Soybean Production: A Survey in Brazil. Agriculture 2022, 12, 611. [Google Scholar] [CrossRef]
- US Department of Agriculture, Foreign Agricultural Service. U.S. Soybean Meal Exports Forecast at Consecutive Records in 2022/23 and 2023/24. Oilseeds: World Mark. Trade 2024. Available online: https://fas.usda.gov/data/oilseeds-world-markets-and-trade (accessed on 15 March 2024).
- Cattelan, A.J.; Dall’Agnol, A. The rapid soybean growth in Brazil. EDP Sci. 2018, 25, D102. Available online: https://www.alice.cnptia.embrapa.br/bitstream/doc/1091243/1/2018Therapidocl170039.pdf (accessed on 12 December 2023). [CrossRef]
- US Department of Agriculture, USDA; Midwest Climate Hubs. Agriculture in the Midwest. 2020. Available online: https://www.climatehubs.usda.gov/hubs/midwest/topic/agriculture-midwest (accessed on 18 May 2022).
- Kolady, D.E.; Van der Sluis, E.; Uddin, M.M.; Deutz, A.P. Determinants of adoption and adoption intensity of precision agriculture technologies: Evidence from South Dakota. Precis. Agric. 2021, 22, 689–710. [Google Scholar] [CrossRef]
- Erickson, B.; Lowenberg-DeBoer, J.; Bradford, J. Precision Agriculture Dealership Survey 2017; Purdue University: West Lafayette, IN, USA, 2017; Available online: https://ag.purdue.edu/digitalag/_media/croplife-precision-agriculture-report-2017.pdf (accessed on 20 October 2020).
- Erickson, B.; Lowenberg-DeBoer, J.; Bradford, J. Precision Agriculture Dealership Survey 2019; Purdue University: West Lafayette, IN, USA, 2019; Available online: https://ag.purdue.edu/digitalag/_media/croplife-precision-agriculture-report-2019.pdf (accessed on 20 October 2020).
- Pew Research Center. Some Digital Divides Persist between Rural, Urban and Suburban America. Survey of U.S. Adults Conducted Jan. 25–Feb. 8, 2021. Available online: https://www.pewresearch.org/fact-tank/2021/08/19/some-digital-divides-persist-between-rural-urban-and-suburban-america/ (accessed on 8 May 2022).
- Acharya, A.S.; Prakash, A.; Saxena, P.; Nigam, A. Sampling: Why and how of it. Indian J. Med. Spec. 2013, 4, 330–333. [Google Scholar] [CrossRef]
- Etikan, I.; Musa, S.A.; Alkassim, R.S. Comparison of convenience sampling and purposive sampling. Am. J. Theor. Appl. Stat. 2016, 5, 1–4. [Google Scholar] [CrossRef]
- Littlejohn, S.W.; Foss, K.A.; Oetzel, J.G. Theories of Human Communication, 12th ed.; Waveland Press: Long Grove, IL, USA, 2021. [Google Scholar]
- Food and Agriculture Organization of the United Nations (FAO). The Future of Food and Agriculture—Trends and Challenges; FAO: Rome, Italy, 2017; Available online: https://www.fao.org/3/i6583e/i6583e.pdf (accessed on 11 October 2023).
- US Department of Agriculture. United States Census of Agriculture; USDA: Washington, DC, USA, 2017. [Google Scholar]
- IBGE, Instituto Brasileiro de Geografia e Estatística. Censo Agropecuário; IBGE: Brasília, Brazil, 2017. [Google Scholar]
- Lowenberg-DeBoer, J.; Erickson, B. Setting the record straight on precision agriculture adoption. Agron. J. 2019, 111, 1552–1569. [Google Scholar] [CrossRef]
- Sonka, S.T.; Lins, D.A.; Schroeder, R.C.; Hofing, S.L. Production agriculture as a knowledge creating system. Int. Food Agribus. Manag. Rev. 1999, 2, 165–178. [Google Scholar] [CrossRef]
- Pathak, H.S.; Brown, P.; Best, T. A systematic literature review of the factors affecting the precision agriculture adoption process. Precis. Agric. 2019, 20, 1292–1316. [Google Scholar] [CrossRef]
- Schimmelpfennig, D.; Ebel, R. Sequential adoption and cost savings from precision agriculture. J. Agric. Resour. Econ. 2016, 41, 97–115. [Google Scholar]
- IHS Markit. Um Estudo Sobre a Percepção e uso Destas Ferramentas Durante a Pandemia. Mercado Brasileiro Agricultura de Precisão/Digital, São Paulo, SP. 2021. Available online: https://cdn.ihsmarkit.com/www/pdf/0122/ihs_markit_agric_precisao_2021_asbraap_teaser_v2.pdf (accessed on 7 July 2022).
- Tey, Y.S.; Brindal, M. Factors influencing the adoption of precision agricultural technologies: A review for policy implications. Precis. Agric. 2012, 13, 713–730. [Google Scholar] [CrossRef]
- Fatoretto, J.C.; Michel, A.P.; Silva Filho, M.C.; Silva, N. Adaptive potential of fall armyworm (Lepidoptera: Noctuidae) limits Bt trait durability in Brazil. J. Integr. Pest Manag. 2017, 8, 17. [Google Scholar] [CrossRef]
- Da Silva, V.P.; van der Werf, H.M.; Spies, A.; Soares, S.R. Variability in environmental impacts of Brazilian soybean according to crop production and transport scenarios. J. Environ. Manag. 2010, 91, 1831–1839. [Google Scholar] [CrossRef] [PubMed]
- Easley, D.; Kleinberg, J. Networks, Crowds, and Markets: Reasoning about a Highly Connected World; Cambridge University Press: Cambridge, UK, 2010; Volume 1. [Google Scholar]
- Rust, N.A.; Stankovics, P.; Jarvis, R.M.; Morris-Trainor, Z.; de Vries, J.R.; Ingram, J.; Mills, J.; Glikman, J.A.; Parkinson, J.; Toth, Z.; et al. Have farmers had enough of experts? Environ. Manag. 2022, 69, 31–44. [Google Scholar] [CrossRef] [PubMed]
- Statista. WhatsApp—Statistics Facts. Statista Research Department. 8 February 2022. Available online: https://www.statista.com/study/20494/whatsapp-statista-dossier (accessed on 13 April 2022).
- O’Donoghue, C.; Heanue, K. The impact of formal agricultural education on farm level innovation and management practices. J. Technol. Transf. 2018, 43, 844–863. [Google Scholar] [CrossRef]
- Deichmann, U.; Goyal, A.; Mishra, D. Will digital technologies transform agriculture in developing countries? Agric. Econ. 2016, 47, 21–33. [Google Scholar] [CrossRef]
- Maertens, A.; Barrett, C.B. Measuring social networks’ effects on agricultural technology adoption. Am. J. Agric. Econ. 2013, 95, 353–359. [Google Scholar] [CrossRef]
- Béné, C.; Fanzo, J.; Prager, S.D.; Achicanoy, H.A.; Mapes, B.R.; Alvarez Toro, P.; Bonilla Cedrez, C. Global drivers of food system (un) sustainability: A multi-country correlation analysis. PLoS ONE 2020, 15, e0231071. [Google Scholar] [CrossRef]
Figure 1. Maps of the study area in Brazil and the United States. Source: IBGE, 2021 and National Weather Service, 2022.
Figure 1. Maps of the study area in Brazil and the United States. Source: IBGE, 2021 and National Weather Service, 2022.
Figure 2. Age distribution of survey respondents in Brazil and the United States.
Figure 2. Age distribution of survey respondents in Brazil and the United States.
Figure 3. Level of education of survey’s respondents in Brazil and the United States.
Figure 3. Level of education of survey’s respondents in Brazil and the United States.
Figure 4. Farm size of survey’s respondents in Brazil and the United States.
Figure 4. Farm size of survey’s respondents in Brazil and the United States.
Table 1. Level of use of digital technologies on-farm.
Table 1. Level of use of digital technologies on-farm.
| Brazil | United States | |
|---|---|---|
| Use of Digital Technologies | Means | Means |
| Guidance/Autosteer | 3.56 *** | 4.23 *** |
| Yield monitors | 2.92 *** | 4.31 *** |
| Satellite/drone imagery | 2.99 ns | 2.94 ns |
| Soil electrical conductivity mapping | 1.50 *** | 1.81 *** |
| Wired or wireless sensor networks | 2.10 ** | 2.36 ** |
| Electronic records/mapping for traceability | 2.09 *** | 3.26 *** |
| Sprayer control systems | 1.98 *** | 3.93 *** |
| Automatic rate control telematics | 2.11 *** | 3.36 *** |
Note: Statistical significance levels: *** p ≤ 0.001, ** p ≤ 0.01, and non-significant (ns) p > 0.05.
Table 2. Level of influence of the use of digital technology in making decisions on-farm.
Table 2. Level of influence of the use of digital technology in making decisions on-farm.
| Brazil | United States | |
|---|---|---|
| Making Decisions | Means | U.S. Means |
| NPK fertilization and liming application | 3.64 *** | 3.93 *** |
| Overall hybrid/variety selection | 3.49 ns | 3.53 ns |
| Overall crop planting rates | 3.44 ns | 3.45 ns |
| Variable seeding rate prescriptions | 2.38 *** | 2.72 *** |
| Pesticide selection (herbicides, insecticides or fungicides) | 3.26 *** | 2.91 *** |
| Cropping sequence/rotation | 3.12 *** | 2.69 *** |
| Irrigation | 2.02 *** | 1.41 *** |
Note: Statistical significance levels: *** p ≤ 0.001, and non-significant (ns) p > 0.05.
Table 3. Level of influence of using digital technology on-farm on the benefits obtained.
Table 3. Level of influence of using digital technology on-farm on the benefits obtained.
| Brazil | United States | |
|---|---|---|
| Benefits | Means | Means |
| Increased crop productivity/yields | 3.70 ** | 3.92 ** |
| Cost reductions | 3.63 ns | 3.78 ns |
| Purchase of inputs | 3.38 ns | 3.40 ns |
| Marketing choices | 3.31 *** | 2.96 *** |
| Time savings (paper filing to digital) | 3.51 *** | 3.17 *** |
| Labor efficiencies | 3.57 *** | 3.30 *** |
| Lower environmental impact | 3.34 *** | 2.99 *** |
| Autosteer (less fatigue/stress) | 3.54 *** | 4.18 *** |
Note: Statistical significance levels: *** p ≤ 0.001, ** p ≤ 0.01, and non-significant (ns) p > 0.05.
Table 4. Level of influence of mass media, social media and interpersonal meetings on farmers’ adoption decisions.
Table 4. Level of influence of mass media, social media and interpersonal meetings on farmers’ adoption decisions.
| Brazil | United States | |
|---|---|---|
| Mass Media | Means | Means |
| Newspaper | 1.75 *** | 2.11 *** |
| Magazine | 2.11 *** | 2.78 *** |
| Radio | 2.17 ** | 2.40 ** |
| Television | 2.15 ns | 2.10 ns |
| Website and blog | 3.38 ns | 3.41 ns |
| Cable television | 2.41 *** | 1.55 *** |
| Social Media | Means | Means |
| YouTube | 3.17 *** | 2.52 *** |
| 3.65 | - | |
| 2.40 *** | 1.74 *** | |
| - | 1.89 | |
| 2.03 *** | 1.47 *** | |
| 2.61 *** | 1.26 *** | |
| Snapchat | - | 1.26 |
| Messenger | 1.71 | - |
| Interpersonal Meetings | Means | Means |
| Field days | 3.87 *** | 3.51 *** |
| Conferences, forums, seminars | 3.86 *** | 3.53 *** |
| Extension agents | 3.63 ns | 3.50 ns |
| Retailers | 3.20 *** | 3.50 *** |
| Peer groups | 3.42 ns | 3.41 ns |
| Conversations with neighbors | 3.62 ** | 3.40 ** |
The blank (-) means that this option was not included in one of the countries following the criteria described in the methodology. Note: Statistical significance levels: *** p ≤ 0.001, ** p ≤ 0.01, and non-significant (ns) p > 0.05.
Table 5. (a) Relationship between use of digital technologies and communication channels. (b) Number of times a communication channel had a statistical impact on the use of the eight digital technologies.
Table 5. (a) Relationship between use of digital technologies and communication channels. (b) Number of times a communication channel had a statistical impact on the use of the eight digital technologies.
| (a) | ||
| Brazil | United States | |
| Digital Technologies | Communication Channels (Spearman’s Rank Correlation CoefficientρS) | Communication Channels (Spearman’s Rank Correlation CoefficientρS) |
| Guidance/Autosteer | 1st Conversation with neighbors (ρS 0.209) | 1st YouTube (ρS 0.208) |
| 2nd Conferences, forums, seminars (ρS 0.120) | 2nd Twitter (ρS 0.159) | |
| 3rd Field days (ρS 0.096) | 3rd Website and blog (ρS 0.154) | |
| Yield monitors | 1st LinkedIn (ρS 0.178) | 1st YouTube (ρS 0.181) |
| 2nd Conversation with neighbors (ρS 0.170) | 2nd Peer groups (ρS 0.163) | |
| 3rd Cable television (ρS 0.145) | 3rd Website and blog (ρS 0.145) | |
| Satellite/drone imagery | 1st LinkedIn (ρS 0.253) | 1st Website and blog (ρS 0.225) |
| 2nd Conferences, forums, seminars (ρS 0.246) | 2nd Twitter (ρS 0.180) | |
| 3rd Instagram (ρS 0.226) | 3rd YouTube (ρS 0.165) | |
| Soil electrical conductivity map | 1st LinkedIn (ρS 0.228) | 1st Cable Television (ρS 0.199) |
| 2nd Instagram (ρS 0.183) | 2nd YouTube (ρS 0.163) | |
| 3rd Messenger (ρS 0.182) | 3rd Peer groups (ρS 0.141) | |
| Wired or wireless sensor networks | 1st LinkedIn (ρS 0.261) | 1st Instagram (ρS 0.271) |
| 2nd Instagram (ρS 0.208) | 2nd YouTube (ρS 0.231) | |
| 3rd Conferences, forums, seminars (ρS 0.183) | 3rd Twitter (ρS 0.209) | |
| Electronic records/mapping for traceability | 1st LinkedIn (ρS 0.224) | 1st Website and blog (ρS 0.252) |
| 2nd Instagram (ρS 0.180) | 2nd YouTube (ρS 0.190) | |
| 3rd Conferences, forums, seminars (ρS 0.148) | 3rd Facebook (ρS 0.158) | |
| Sprayer control systems | 1st LinkedIn (ρS 0.221) | 1st YouTube (ρS 0.165) |
| 2nd Cable television (ρS 0.189) | 2nd Website and blog (ρS 0.164) | |
| 3rd WhatsApp (ρS 0.151) | 3rd Retailers and extension agents (ρS 0.133) | |
| Automatic rate control telematics | 1st LinkedIn (ρS 0.246) | 1st YouTube (ρS 0.238) |
| 2nd Instagram (ρS 0.186) | 2nd Website and blog (ρS 0.204) | |
| 3rd Peer groups (ρS 0.135) | 3rd Facebook (ρS 0.145) | |
| (b) | ||
| Brazil | United States | |
| Mass Media | Number of times listed | Number of times listed |
| Website and blog | 0 | 6 |
| Cable television | 2 | 1 |
| Total | 2 | 7 |
| Social Media | Number of times listed | Number of times listed |
| YouTube | 0 | 8 |
| 7 | 0 | |
| 5 | 1 | |
| 0 | 3 | |
| 0 | 2 | |
| 1 | 0 | |
| Messenger | 1 | 0 |
| Total | 14 | 14 |
| Interpersonal Meetings | Number of times listed | Number of times listed |
| Conferences, forums, seminars | 4 | 0 |
| Conversation with neighbors | 2 | 0 |
| Peer groups | 1 | 2 |
| Field days | 1 | 0 |
| Retailers and extension agents | 0 | 1 |
| Total | 8 | 3 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
