Ontario - Canada Logos
SWEEP Banner

SWEEP Report #8

Social Structure and the Choice of Cropping Technology:
Influence of Personal Networks on the Decision
to Adopt Conservation Tillage

G.K. Warriner and G.M. Moul, Department of Sociology, University of Waterloo, Waterloo, Ont.

Executive Summary

Evaluation Summary (Tech. Transfer Report Summaries)

Associated SWEEP/LSP Research

View/Download Report  [295 KB pdf]



List By Number | List By Sub-Program

Completed: June, 1989

Key Words:

adoption, conservation technology, social structure, survey, adoption diffusion models, social networks, farmer attitudes

Executive Summary


This report employs a multifaceted approach to examining the decision framework of farmers in relation to the decision to adopt or not to adopt conservation tillage. Responses from a representative cross-section of 259 farms surveyed in 1988 in the SWEEP project area are compared to those from a second, nonprobability sample of 55 known conservation tillage adopters for the purpose of examining the relative importance of various on-site, economic, sociodemographic, attitudinal and social network influences leading to the adoption of soil conservation practices.

A distinguishing aspect of this research is the focus of the examination on the farmer's social frame of reference constituted of family, friends and acquaintances to whom he may turn for information or guidance concerning the choice of cropping technology. In this regard we examine both the membership of this network and its structure. The membership of the network, ranging from spouse, through other family, friends, neighbors, and farm advisors constitute important sources of information on farming to which the operator may turn. In addition, the structure, or form, of this network facilitates or impedes the receipt of information concerning innovative farming practice.

Thus the research goals are: (1) to provide evidence of the influence of the structural characteristics of the farmer's communication network in enabling the diffusion of innovation of conservation tillage; (2) to assess information concerning what actors typically constitute the reference group of farmers with regard to decisions on tillage and other farm practices; (3) to assess the relative influence of social, economic, site and cognitive factors in leading to the adoption of new cropping technologies; (4) to compare known adopters with the general cash crop farming public of south-western Ontario on farm and operator characteristics; and (5) to estimate the relative rates of adoption of conservation tillage and other conservation practices among rowcrop farmers during the 1987 growing season.


Funding for this research was provided by a grant from the Socio-Economic Evaluation subprogram of the Soil and Water Environmental Enhancement Program (SWEEP). SWEEP is a $30 million jointly funded federal-provincial program intended to reduce phosphorous loadings from non-point cropland soil erosion to Lake Erie by 200 tonnes annually by 1990. The Socio-Economic Analysis subprogram funds projects to evaluate the impact of SWEEP programs and to investigate the conditions under which farmers may choose to adopt conservation forms of farming.


A two stage sampling plan was employed to select farms from the SWEEP project area for participation in a comprehensive survey of rowcrop farming practices. At the first stage a probability sample of cash crop farm operations was selected using conventional proportionate stratified, random sampling procedures. A random sample of 497 farms stratified by farm size and county location was drawn from the population of grain corn operations in the twelve counties of the SWEEP project area. Then, on the assumption that such a sample would provide too few farms on which conservation tillage was actually practiced, a second, smaller sample was selected made up of 85 farmers previously identified as using conservation tillage.

The 582 sampled farms were surveyed by mail using a survey instrument comprising more than 200 questions for which information on the following topics was requested: (1) farm characteristics, (2) household composition, (3) soil and erodibility conditions, (4) farming practices, (5) attitudes, values and beliefs, and (6) personal network. With three follow-up mailings to encourage participation, 350 questionnaire from the 582 originally mailed were returned, for an overall response rate of 60.2 percent. Following exclusions for ineligible and uncodeable responses, the analysis group constituted 314 farms -- 259 farms from the probability sample and 55 from the conserver sample.


  1. Farm Characteristics

    A sampling objective was to select farms which were mainly row crop operations. Nearly 93 percent of the farms in the probability sample report producing corn, and 74.4 percent produce soybeans. There were no statistically significant differences between farms in the probability sample and the conserver group with respect to the type of products being farmed. However, the farms of the conserver sample are substantially larger than those in the probability sample, averaging 609 acres versus 339 in the probability sample.

    The conserver sample farms were also somewhat more likely to be operated with another family member, not their spouse, or for the legal status of the farm to be a corporation, while self-owned, or ownership with spouse was more common in the probability sample. Most farmers, however, share ownership with at least one other. Few respondents to this survey, only 3.8 percent, reported renting or leasing their farm.

    Once the difference in size of farms between conservers and the probability sample is taken into account there is no difference between farms in the two samples in terms of farm assets. However, the conserver farms had significantly more debts, even in consideration of the larger average sizes of these farms. The average debt to asset ratio across all farms in this survey is 34 percent, somewhat in excess of the thirty percent often seen as an acceptable debt to asset balance. Further, the conserver group had significantly lower returns on investment from their farming operations in 1987 in comparison to the farms of the probability sample. On balance the data on the financial aspects of the farm provide little support for a hypothesis that some farmers are conserving because they are financially better off.


  2. Operator/Household Characteristics

    There were no differences between the farms of the conserver group and the probability sample in terms of number of people in the household, or the age the operator began farming, years of experience, how long the farm has been in the family, or the importance of off-farm income to the operation of the farm. Operators in the conserver group had more years of schooling, 13.3 years on average, versus 11.7 years among operators in the probability sample. Those in the conserver sample were younger, although the difference is not statistically significant.

    There are some differences between the probability and conserver group with respect to what are the future plans for the farm. Over seventy-two percent of the operators in the conserver group predict that the farm will remain in the family and be farmed, while 56.6 percent of the probability sample claim this, with the remainder believing the farm will be leased or sold.


  3. Soil Characteristics

    The survey asked respondents about soil conditions on their farms and the existence of various types of erosion. Problems of water erosion and other soil problems are far more likely to be reported on the farms of the conserver group than the probability sample. Compared to farmers in the probability sample, operators in the conserver sample more often report their farms to have finer soil types, steeper and longer slopes, more soil erosion potential, more severe damage from erosion, and greater likelihood of both poor soil structure and soil compaction.

    The operators in the conserver group were also more likely to admit their own farming practices to be contributing to soil erosion. In general, the conserver group had a greater tendency to recognize the potential for erosion, while ascribing personal responsibility for its existence. Only on-site inspection (not conducted) can ascertain whether the differences in the erosion levels of the conserver and the probability samples are actual or are simply perceived. However, it is likely that the farmers in the probability sample do not practice conservation as frequently as those in the conserver group because they are less able to recognize when it is needed.


  4. Conservation Practices

    The survey also examined the tillage and cropping practices used by operators in 1987 for evidence of conservation forms of farming. On this evidence there is some basis for optimism concerning the potential for soil conservation in Ontario. Although 86.5 percent of the farmers in the probability sample reported using the conventional moldboard plough in the fall, a surprising proportion, 47.6 percent, also reported using reduced tillage on some or all of their fields. Almost one-third of cropland for harvest for this sample, 31.3 percent, was affected by these methods. For the conserver sample, more than one-half, 51.9 percent, used conventional tillage on part of their farm in 1987, while 59.3 percent used reduced till, 22.2 percent used ridge till, 27.8 percent practiced no till, and 31.5 percent used modified no till.

    With the two samples combined, 44.6 percent of respondents reported using only conventional tillage in 1987. A further 36.3 percent used some combination of conservation tillage and conventional tillage, and 19.1 percent used only conservation tillage during 1987. Among farmers in the probability sample, 51.4 percent used only conventional tillage, while 35.7 percent combined conventional tillage with conservation tillage, and 12.9 percent used only conservation tillage.

    Overall, nearly one-half of the farms in the probability sample, 48.6 percent, reported using some type of conservation tillage on their farms in 1987, and about one in eight farmers reported using only conservation tillage. In addition, 52.2 percent of these farmers reported leaving significant crop residue following planting in 1987 (versus 83.6 percent in the conserver group), with 37.8 percent of the cropland in harvest for that year being affected.


  5. Attitudes

    Environmental attitudes can be important pre-determinants to conservation behaviour. While in the soil conservation field most interest has been paid to factors involving the financial incentives for conservation, there is significant doubt that conservation will ever be seen by the majority of farmers to be financially attractive. Hence attempts to make conservation financially viable through government programs involving subsidies may be misdirected. While such programs may encourage farmers to adopt conservation, they also imply that the sole basis for deciding to practice conservation rests on its financial merits, while at the same time suggesting the government will accept responsibility for ensuring conservation is financially worthwhile. What this ignores is an approach to conservation which involves an individual's long-term willingness to prevent soil loss through conservation which is based on an ethical and moral commitment to the land, and which is practiced in spite of the extra effort and financial sacrifices which may be involved.

    The survey found evidence that factors involving beliefs about soil loss, and attitudes about conservation and land stewardship were important to the farmer's decision to practice or not to practice conservation. There were statistically significant differences between the conserver and probability samples on a number of such beliefs, as well as the intention to practice conservation. Thus, on average, the conserver group was more likely to believe in the existence of soil loss problems in Ontario, while denying that farmland is mainly an economic commodity, or that the government, or other farmers, should be mainly responsible for solutions to soil loss problems. Farmers in the conserver group also saw themselves as being innovative with regard to new farming practices, in addition to believing that the overall benefits of conservation would, in the long run, outweigh its shortcomings, while stating they held a moral obligation to protect the land, and being committed to soil conservation through the intention to practice it.

    The analysis also showed a relationship between these attitudinal and belief dimensions and the farming practices of operators. The majority of the attitudinal and belief factors were related to the intention to practice conservation, in addition to predicting the likelihood of having used conservation tillage in 1987. Further, farmers who felt that conservation tillage was a bad idea from an economic perspective, were those who were least likely to try conservation tillage, while those who believe in its long-term benefits and need were those who were most likely to practice it, or want to. It is also the case that the geographical conditions of the farm with respect to soil and slope conditions, and the perceived evidence of soil loss on the farm, were positively related to the practice of conservation. This points to a need for the development of programs to help farmers recognize the existence of soil erosion on their own farms. The survey findings suggest that farmers are often committed to conservation, but are inhibited from practicing it by being unable to recognize when it occurs on their own farms.


  6. The Social Network

    The notion of the social network in the context of adoption of conservation tillage adds a further dimension of understanding to the problem of how the decision to adopt conservation tillage is made. The idea that the social network may be important with respect to the farmer's decision to adopt or to not adopt conservation tillage is straightforward. Since before any change in farming practice can be expected the farmer must first receive information about the practice, it is through the social network that new information will be channelled.

    Certain aspects of the network may facilitate more effective communication. These include the farmer's location with a sphere of acquaintanceship, as well as the size of the network and its density (proportion of reciprocal ties). Dense networks provide greater opportunities for communication transfer since there are more communication paths. On the other hand, dense networks may also be more "closed", to the extent that their tightly interlocking natures can prevent entry of innovative ideas. Finally, when dealing with innovative, or risky, new farming technology, there is likelihood that individuals with whom the farmer has a "weak" link will serve as a better source of information than with those with whom the link is "strong".

    This latter idea centers on the understanding that each farmer's network may be differentiated according to both (1) degree of homophily (similarity) between members of the network, and (2) level of attraction (strength of tie) between members. Thus homophily and close attraction facilitate effective communication, while also acting to prevent new ideas from entering the system. Alternatively, heterogeneity within networks, and weak ties, give the system openness, enabling communications on farming innovations to flow freely.

    Responses to the survey suggest the majority of farmers feel their friends and neighbors want them to practice conservation, and there are apparently few internal divisions within households to prevent this. In these respects both the conserver and probability samples are very similar. These factors, in addition, are related, weakly, to the farmer's intention to practice or not to practice conservation.

    On the factors relating to the sources of information seen as important with respect to deciding on farming practices, there are no differences between the responses of the conserver and probability samples. The most important source of information is the farmer's own personal experience, followed by neighbors, the farm media, government and educational agricultural specialists, representatives of commercial firms, farm organizations, family members and one's spouse.

    With respect to the personal network, the average size of the networks of the conserver sample is significantly larger than those of the probability sample. This is consistent with network theory generally which argues that larger networks provide more sources of information, and more channels along which it can be communicated, therefore facilitating the diffusion of innovation. Respondents in the conserver sample also had known members in the network for a shorter period of time, but there were no significant differences between the two samples concerning the frequency of interaction between network members, or their occupations. The majority, 64 percent, of network members are active farmers, followed by representatives of commercial or financial agencies, and government agricultural specialists. A relatively high percentage of the network members in both samples, over 40 percent, are people who either practice conservation or who belong to a farm organization which supports it.

    The size of the network is also positively associated with the likelihood of having used conservation tillage in 1987, as well as with the intention to practice conservation and with the perception that soil erosion is a problem on the farm. Larger networks are also positively associated with younger and better educated farm operators, who have more debts, less returns on investment, and smaller farms. The density of the network is inversely related to membership in the conserver sample (nonsignificant), as well as with conservation tillage in 1987, and with intention to practice conservation, and the perception that soil erosion exists on the farm. Thus it seems that larger networks encourage conservation, while those which are dense inhibit it, either by being more "closed" and preventing the entry of information, or by being comprised mainly of homophilous relations acting to resist ideas on novel farming techniques.

    Closer inspection of the individual networks of farmers reveals that in terms of whom the farmer turns to most often, and mentions as most important, when seeking information on farming matters, it is likely to be another farmer. However, if the network contains a known conserver, or someone who is a member of an association which promotes conservation, then this person is likely to be referred to. On the other hand, when it comes to deciding something on the farm, the network, or reference group to whom the operator turns, is very likely to be comprised mainly of family, kin and other farmers, all individuals with whom the farmer, on grounds of obvious similarity, is strongly connected. Thus if the issue is whether to adopt a new and innovative farming practice, such as conservation tillage, networks comprised of member similarity and strong ties tend, on the whole, to counteract the chances of adoption.

Conclusions and Recommendations

On the basis of these findings, the following conclusions are warranted:

  1. The decision to adopt conservation tillage is multi-faceted. Across the range of dimensions investigated it is almost always the case that one or several factors stand out to distinguish members of the probability sample from the conserver group, or to identify conservers. Some of these --education, age, farm size -- are already widely known from the forty year research tradition on diffusion of innovation. Other factors, particular those relating to attitudes, social normative beliefs, and network factors, are much less often recognized. However their relationships with adoption were, in several cases, larger than the more commonly cited predictors of conservation practices. Thus, farmers make up their minds to practice conservation, or not to practice it, based on many things, suggesting the need for the complexity of this decision framework to be taken into account in the development of agricultural programs.


  2. Conservation adoption is unpredictable. In spite of a number of significant associations across several dimensions, the correlations are, for the most part, not sizeable. Correlations in the range of 0.12 to 0.20 are common and, while statistically significant, can hardly be regarded as large. On these grounds it should be noted that identifying conservers on the basis of farms, or socio-demographics, or financial situation, or beliefs and attitudes, or social networks, or anything else, is never going to be easy.


  3. There was little to suggest the financial conditions of farms are related to conservation. Large farms are more likely to practice conservation, true, but several financial indicators were unrelated to conservation adoption, while adopters also had significantly higher debts and lower returns on investment. However, the one factor which strongly predicts the intention not to adopt conservation tillage was the belief that such a practice would be uneconomical.

    On grounds of these findings it may be concluded that programs to encourage conservation based on its questionable financial benefits are misguided. It often appears that in developing programs to promote soil conservation planners have assumed the need to convince farmers of its financial benefits. However, an important factor predicting adoption of conservation practices in this survey was the farmer's moral commitment to conserve, and it is notable that such a factor is additionally inversely related to the belief that conservation is uneconomical. Soil conservation in Ontario may in the long run be better served by replacing an emphasis on economic returns with one which stresses land stewardship.


  4. There may be more farmers in south-western Ontario practicing conservation than was previously believed. Among the probability sample 12.9 percent of farmers reported using only conservation tillage, and a further 35.7 percent combined conventional tillage with conservation tillage. Further, 44.1 percent of the farms in this survey reported either now practicing conservation tillage, or intending to practice it in the near future.

    Still, it is also the case that the great majority of respondents, 86.5 percent, did use the moldboard plough in the fall on at least some portion of their farm in 1987. On balance it can be concluded there is some evidence of increased conservation occurring in south-western Ontario farming, and a willingness by farmers to do more.


  5. A factor which appears consistently related to the likelihood of adoption of conservation forms of farming is the farmer's perception that soil loss is occurring on the farm. The belief that soil loss is occurring on one's own farm is a necessary condition for taking steps to prevent it. However, farmers cannot believe in soil loss until they experience it, and cannot experience it until the phenomenon is understood and recognised. If more farmers could be shown how to recognize soil erosion and to estimate its cumulative effects, they would believe in soil erosion and practice conservation.


  6. An important goal of this research was to investigate the influence of the farmer's social network on the decision to adopt or to not adopt conservation tillage. In this regard the findings are suggestive, but not conclusive. The structural properties relating to the size and density of the network appear to be related to the dissemination of new information, as well as influencing the network's receptiveness toward innovative farming technology. The networks in the sample are characterized by highly homophilous and strong ties, properties which may impede entry of novel information. However, comparisons between the networks of adopters and nonadopters revealed little which distinguishes their behaviours, excepting the networks of adopters were larger. While the findings from this approach are not conclusive, it should be understood that these network variables introduce a new level of analytical complexity, and with continued investigation are still potentially rich in their ability to explain conservation adoption.


Evaluation Summary

(From Technology Transfer Report Summaries - A. Hayes, L. Cruickshank, Co-Chairs)

The intent of this study was to explore the factors affecting the decisions of producers regarding conservation technology. The goals of the study were: (1) to determine the influence of social structure on producers' decisions; (2) to define composition of reference groups; (3) to assess the relative influence of environmental, social, personal and business factors on decisions for changing cropping technologies; (4) to contrast the above with traits of known adopters; and (5) to establish the relative rates of adoption of conservation practices. A multi-faceted approach was chosen to address the diversity of factors identified.

A mailed questionnaire containing over 200 questions was mailed to two groups of cash-crop farmers in the SWEEP area: a randomly selected group of 497 grain producers (probability sample) and a purposive sample of 85 producers known to have practised conservation tillage. Following a cull process to eliminate unusable returns, 259 probability and 55 conserver, for a total of 314 of the 350 returned questionnaires (return rate of 60.2%) were found to be acceptable to be used in the database. The questionnaires and findings are categorized into the following areas: farm characteristics, household composition, soil and erodibility conditions, farming practices, attitudes and personal networks.

  1. Farm Characteristics - most features were similar between the two groups except that the conserver group had on average larger farms and lower return on their investments in 1987.

  2. Operator/ Household - most features were similar except that conservers were younger and better educated and predicted more often that the farm would stay in the family.

  3. Soil Characteristics - conserver farmers had soil features with a high potential for erosion and had changed practices because they had observed erosion and had decided to take responsibility. It is postulated that the probability group may not be able to recognize erosion problems and there exists a causal relationship between this and motivation for change.

  4. Conservation Practices - although more conserver farmers use conservation or no tillage more frequently than the probability sample, most farmers use a variety of tillage methods and planting methods (nearly one-half of the probability sample reported using some sort of conservation tillage in 1987). At this time people were experimenting or had selected tillage/cropping practices to reflect the range of conditions they farmed.

  5. Attitudes - the conserver group had a stronger stewardship ethic and commitment to innovation in conservation than the probability group. Recognition or acknowledgement of soil problems was once again correlated with belief and commitment for change.

  6. The social network - peer pressure was found to have a positive influence on adoption rates. Both groups rank the relative importance of information sources in a descending order of importance as follows: personal experience, their neighbours, the farm media, specialists, agribusiness representatives, farm organizations, family and spouse. Larger networks were positively correlated with adoption of conservation tillage. Smaller groups had a tendency to be closed. Thus, if another farmer is the best source of information for other farmers, those with conservers in their network would have a higher probability of adopting.


The strength of this study was that the approach reflected the very nature of the phenomena itself; multifaceted. Financial or personal characteristics alone do not explain an individual producers reason to change. Of note they recommend:

  1. Farmers need to understand, be able to recognize and feel comfortable admitting to problems on their farm before considering changes to their operations. This should be central to conservation/environmental farm planning exercises.

  2. The stewardship ethic is as important as financial considerations when fostering change - it should be fostered in conservation programming.

  3. The role of the extension agent in conservation has been verified as to support innovators and leaders in their trials and to promote their leadership in their community, to provide the farm press with accurate information and case studies and to work with agri-business and farm organizations to promote conservation practices and beliefs.

Associated SWEEP/LSP Research:

  • SWEEP Report #0 - Cropping, Tillage and Land Management Practices in Southwestern Ontario 1986

  • SWEEP Report #3 - An Economic Assessment of the Distribution of Benefits Arising from Adoption of Conservation Practices in Crop Production in Southwestern Ontario

  • SWEEP Report #6 - A Survey of Crop Residue in Southwestern Ontario 1987

  • SWEEP Report #7 - Sources of Motivation in the Adoption of Conservation Tillage

  • SWEEP Report #9 - Conservation Practices in Southwestern Ontario: Barriers to Adoption

  • LSP7008 - Differences in Soil Conservation between Operator-Owned and Rented Land

Other Publications and Theses from this Research:

Moul, G. M. 1991. Power relations of spouses farming in southwestern Ontario. Master's Thesis, University of Waterloo, Waterloo, Ont.

Warriner, G. K., and G. M. Moul. 1992. Kinship and personal communication network influences on the adoption of agriculture conservation technology. Journal of Rural Studies 8: 279-291.

Future Research: ( ) indicates reviewers suggestion for priority, A - high, C - low.

(C) Similar work should be initiated to determine if other areas of conservation (pollution control, nutrient management, etc.) follow similar trends of adoption and diffusion.





List By Number | List By Sub-Program | LSP Report List


Created: 05-28-1996
Last Revised: Thursday, May 19, 2011 01:11:35 PM