Front Cover
        Front Cover
    Main
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
    References
        Page 21
        Page 22

~~i~ii~l~t~L~


~~ 906


COoperative Extension Service
Institute of Agriculture and Natural ResourceS


Department of Agronomy
Plant Science Building
Lincoln, NE 68583-0910


University of Nebraska-Lincoln




CLOSING THE INFORMATION CYCLE:
PARTICIPATORY METHODS FOR ON-FARM RESEARCH




C. A. Francis, P. E. Rzewnicki, A. Franzluebbers,
A. J. Jones, E. C. Dickey, and J. W. King



Departments of
Agronomy
Agricultural Engineering
Agricultural Communications



Presented in Workshop
"Farmer Participation in Research for
Sustainable Agriculture"

Fayetteville, Arkansas
October 8, 1989


UNIVERSITY OF NEBRASKA-LINCOLN, COOPERATING WITH THE COUNTIES AND THE U.S. DEPARTMENT OF AGRICULTURE


The University of Nebraska-Lincoln


The University of Nebraska Medical Center


The University of Nebraska at Omaha










CLOSING THE INFORMATION CYCLE:
PARTICIPATORY METHODS FOR ON-FARM RESEARCH

C. A. Francis, P. E. Rzewnicki, A. Franzluebbers,
A. J. Jones, E. C. Dickey, and J. W. King 2/


ABSTRACT

Information resources are among the most valuable inputs used by today's
successful manager in agriculture, yet they can be among the most confusing resour-
ces. A classical process has scientists setting research priorities, conducting field
studies, interpreting results, developing recommendations, and moving these through
extension specialists to the farm. The farm manager has been seen as a passive
receiver of information who occasionally provides a credibility check through
questions at meetings or participation on a grower's advisory committee. In our
quest for research relevance, for efficiency in extension, for facilities and inputs to
conduct research, and for a broader spectrum of ideas scientists are inviting greater
participation by farmers in the research and extension process. This is consistent
with farming systems research/extension methodology, and a logical implementation
of a sustainable agriculture philosophy.

It is important to consider decisions about the research agenda, responsibilities
of various participants, designs for relevant field activities, and extension methods
for sharing information. In this process, we need to design and carry out research
on the most important new components of production technology without ignoring
the complex interactions of these components as a part of total systems. Models
currently being used in the Midwest, especially in Nebraska and Iowa are presented
to illustrate the process: conservation tillage tests and demonstrations, nitrogen rates
and cropping systems alternatives, neighbor to neighbor information exchange, and
on-farm strip tests. Some of the immediate challenges include developing research
priorities for this work, seeking financing and other support for field work, and
disseminating information. New communications theories and ideas about practical
application need to be developed as a part of this process. One on-going challenge
is how to evaluate the credibility of results before they are included in an
information base for others to use. In the near future, the farmer/manager will
become an increasingly active and valued participant in the research/extension
process as more methods and confidence are developed in this dimension of the
sustainable agriculture philosophy.

1] Keynote Address presented in workshop, "Farmer Participation in Research for Sustainable Agricut
turc",Fayetteville.Arkansas, October 8, 1989
2) Contributions from Departments of Agronomy, Agricultural Engineering, and Agricultural Communications, Cooperative
Extension Service and Agricultural Research Division, Institute of Agriculture and Natural Resources, University of Nebraska,
uincoln, Nebraska.









INTRODUCTION


Discussions with farmer groups reveal an increasing concern about relying
completely on controlled condition, laboratory research results to solve real-world.
on farm problems. Likewise, there is a ground swell of farmer demonstration plots
and "research" activities on farm that often do not meet the accepted criteria for
repeatable and confident results on which management decisions should be made.
One of the biggest challenges to farmers today is sorting through the multiplicity
of recommendations that are available from industry, farm press, and public
agencies -- often these contain conflicting advice. A4n emerging notion is that of
participatory methods in the research process, a topic discussed many times in
previous farming systems research/extension symposia.

For review, it is useful to summarize some of the purposes of on-farm research
as described by Hildebrand and Poey (1985). On-farm research can pro vide:

-- linkage between on-going research and extension
-- more purposeful research by integrating components into systems
-- orientation of research and selection of priorities
-- more understandable research that is accessible to decision makers
-- checks and balances in the real world to improve research management
-- hands-on experience to improve the image of researchers
-- added dimensions to current component research
-- a learning experience for researchers, extensionists, and farmers

To imply that all of the above are not addressed in the current experiment station
research and extension paradigm would be an error. Yet on-farm, participatory
research is useful in bringing people together to seek solutions to common
problems, and in stimulating greater communication among those who develop and
use new information in farming. On-farm research also has the potential to bring
more resources and people together to focus on practical constraints in farming.
There also are comparative cost efficiencies in on-farm research to meet some types
of objectives.

This paper outlines some of the current activities in the Midwest of the U.S.
with farmers as full participants in the research process. Traditionally, the
agricultural information dissemination process moves from researcher through
extension to the farmer. There are mechanisms for feedback to researchers through
advisory boards and meetings with extension. Yet to truly have an effective
communication system reflecting the needs of farmers, these same people must be
involved to some degree in setting research agendas, doing on-farm studies, and in
demonstrating and extending results.










On-farm research is one multi-faceted technique to bring together all actors with
a role in the drama of farming systems. With direct farmer participation in research
and extension, information dissemination can be improved and our institutional
image enhanced. The information cycle can be closed between the researcher and
farmer. Closing the information cycle actually opens and accelerates communica-
tions and promotes greater participation. The examples that follow provide models
of how this philosophy can be applied.

Research and demonstration activities in conservation tillage have helped
accelerate a change toward reduced primary land preparation and increased residue
on the soil surface to help solve erosion problems. A farmer-oriented demonstra-
tion/research project on nitrogen levels and alternative cropping systems has been
implemented over the past two years. A new neighbor-to-neighbor demonstration
activity was initiated in 1989 to make conservation oriented practices on farms more
accessible to other interested farmers. Data from large, replicated plot trials in
Nebraska and Iowa have been analyzed statistically and compared favorably with
small plot research. We have also surveyed farmers' attitudes toward on-farm
research and whether they would be willing to participate in future activities of this
type. Finally, we have looked at the broader information environment and how this
is changing as a result of new communication technologies.

WHAT DO FARMERS AND RESEARCHERS BELIEVE?

Farmers like to see demonstrations in the field. They feel that the closer to
home these plots are found, the greater relevance they have for an individual's
operation (Rzewnicki and Rockwell, 1989). What do farmers believe, and what
specifically do they want to see in the field? Some of the characteristics of research
trials or demonstrations that make them meaningful or that make them easier for
farmers to implement under their conditions include (Francis, 1986):

-- plots or strips large enough to have a strong visual impact, often across the
entire field
-- plots or strips that are one width or multiple widths of standard field
equipment to minimize extra time needed to set up trial
-- new or modified practices that require minimal changes from current equip
ment and farming systems
-- changes that focus on cutting production costs without reducing yields per unit
area
-- focus on crop yields, crop quality, and economics of production to assure
profitability and reduce risk
-- participation of the farmer in all steps of the process, from problem
identification to design of alternative practices to field implementation









Demonstration plots or comparisons with these characteristics tend to generate
a high level of farmer interest and participation, even if they are not replicated or
randomized in the field nor subjected to statistical analysis (Francis, 1986).

Researchers often view the world from a different perspective, although they
may focus on some of the same basic questions. Through academic training and
research experience, the investigator learns to define "experiments" in a very
different way. Some of the criteria that are important to the researcher who is
concerned with reliability of the data and an opportunity for technical publication
of the results include (Francis, 1986):

-- replication of treatments at least three or four times in the field
-- randomized placement of treatments within each replication
-- relative uniformity of fields within blocks or replications
-- uniformity of treatment variances
-- use of an accepted treatment and experimental design
-- accuracy and uniformity in planting, imposing treatments and data collection
-- accessibility of field or site to researcher
-- capability and funding to run trials over multiple locations and/or several
years
-- conditions of research site representative of region so that results can be
extrapolated to a wide area

These conditions are most likely met on an experiment station, with as many
environmental controls employed as practically are possible. Research conducted
on farms by technical university specialists often meet the same criteria, while most
on-farm demonstrations do not.

In comparing these two lists, we find a difference -- that criteria important to
researchers and those identified by farmers are not the same. This is one reason
why communication often breaks down between groups. Often there are far
different perceptions between researchers and farmers about what constitutes an
experiment. There may be barriers here that are difficult for an extension person
to bridge! Is there a middle ground where criteria of both groups can be met in a
meaningful and cost effective way?

SETTING RESEARCH AND DEMONSTRATION OBJECTIVES

The first step is to carefully evaluate the objectives of a given field demonstra-
tion or research activity. Rather than argue about what does or does not constitute
"research", it is far better to define what information is needed and what conclusions
are to be reached on the basis of a given field planting.










Often the objective is to take prior information on a proven practice and to
apply this in a new area, and there is little doubt that the innovation will improve
productivity, resource use, or profit. In this case there may be no reason to set up
a replicated trial or even a side-by-side comparison with a previous practice. This
is demonstration of proven technology. Farmers accept this approach every year
with a new corn hybrid or soybean variety, a new generation herbicide, or a planter
with better features for their conditions. The success of a new practice or system
is compared with conventional wisdom about the old system, last year's results or
those from another field, or how the neighbor is doing without the innovation. This
approach is valid for looking at qualitative changes in the system, for differences
that are easily observed or not influenced much by specific climate or soil type.
The quantitative comparisons with another year, a different field, or the neighbor's
farm, of course, are not valid since there is no means to statistically evaluate the
results of such an experience. We need educational or extension programs that
describe the importance of replication and randomization in terms that are easily
understood within the context of the farmer's experience.

In contrast to qualitative observations on the farm, research plots are set up to
make valid comparisons between or among alternative components of technology
-- hybrids, varieties, fertilizer levels, herbicides, tillage methods. Treatments are
replicated and assigned to plots within each replication at random to assure that
any variation in the field will be a random occurrence and not be confounded with
any specific treatment -- thus introducing a bias for or against any specific
alternative. Careful field implementation of the experiment provides credible data
and a valid comparison among the treatments -- at least under the conditions of
that year and that location. Researchers are hesitant to extrapolate from these
results to a broader zone of inference, having experienced the types of variation that
occur from year to year and across different soil types and management systems.
For this reason, trials are run for several years and over multiple locations.
Optimum number of years and locations have been studied for a number of crops,
with results dependent on the nature of the treatments under study and the magni-
tude of differences that the researcher wants to measure (Saeed et al., 1984). Thus
the decision on what type of demonstration or research trial is optimum depends
on the objectives of the exercise, the confidence that is needed in the results, and
the range of conditions across which we want the results to apply.


EXAMPLE ONE: CONSERVATION TILLAGE EXTENSION:
A NEBRASKA MODEL

Conservation tillage has rapidly been adopted in much of the Midwest, due in
part to an effective extension program organized by specialists in the region.
Farmers in Nebraska and elsewhere wanted to learn more about these practices and
how management could be fine-tuned to their specific operations. They wanted










specific information on available equipment, chemicals for weed and insect control,
potential yields, and impacts of the new systems on erosion.

To meet these needs, Area Conservation Tillage Meetings were initiated in
1982 to provide information in a lecture/discussion format -- the programs included
extension specialists from several key disciplines. County and district extension
personnel as well as local SCS staff joined farmers in setting the program for local
meetings. Farmers attending the meetings were surveyed to estimate how many of
them were planning to adopt new practices as a result of the meeting. A summary
since 1984 of number of meetings, people attending, and evaluations received is
shown in Table 1 (Dickey et al., 1989).

Table 1. Survey of attendees at planning to adopt conservation tillage
practices.

Year 1984 1985 1986 1987 1988 1989


Number of Area Meetings 10 8 10 14 14 17
Attendees 925 875 900 825 1150 990
Evaluations Received 425 403 432 421 565 495
Plan to adopt 75% 80% 84% 80% 80% 76%


Over the six years that these meetings have been held in Nebraska, more than
5,500 farmers have attended and close to half of those filled out evaluation forms.
Of those who completed the evaluations, approximately 80% indicated that they
would be changing their tillage program as a consequence of attending the meeting.
Most frequent changes mentioned were reducing the number of tillage operations
and adoption of no-till planting. In 1989, about half of the attendees had not
attended a previous meeting, and 97% said they would attend a similar meeting in
the future.

In conjunction with these meetings, a number of field activities were conducted
through the Agricultural Energy Conservation Project, a grant-funded program
(Dickey et al., 1989). The types of activities and demonstrations included in this
project are summarized in Table 2. These activities were attended by more than
11,000 people in the period between December 1983 and March 1989. Unlike
many educational programs, this conservation tillage program was concentrated in
three relatively homogeneous regions having substantial erosion problems. As the
program evolved, farmers adjacent to the project area developed an interest in the
program and supporting educational activities were conducted.












Inside the Adjacent to
Target Areas Target Areas Total
Activity No. Attend. No. Attend. Noe. Attend.


Area Conservation Tillage Meetings 15 1480 17 1410 32 2890
Coffee Shop Meetings 33 407 16 251 49 658
Other Meetings 13 376 10 500 23 877
Planter/Equipment Demonstrations 18 793 5 635 23 1428
Rainfall Simulator Demonstrations 19 1003 22 2040 41 3043
Conservation Tours 48 1799 13 652 61 2451



Each year of the project, demonstration plots were organized with 50 to 75
farmers. These demonstrations included side-by-side comparisons of no-till planting
versus conventional planting methods, fertilizer application methods, and different
herbicide programs. Plots were planted and tilled by farmers using their own
equipment, with some support from Extension assistants where needed. Yield and
cost data were collected and this information used in field tours and subsequent
meetings as part of the total educational program. With this approach, farmers
could see planting equipment in action, follow the crop through the season, and
learn from the final data that was collected in a site near them. Details of the
results are shown in Table 3 (Dickey et al., 1989).


Table 2. Education Activities used in the AECP.










Table 3. Yield and cost summary for side-by-side comparisons of no-till
and tilled demonstration plots.

Crop Number of Fields
(No-till/Tilled Comparison) Yield Cost
Comparison Comoarison

Corn (35 Fields)
No-till Greater than Tilled 14 6
Same (within 5 bu/ac or $5/ac) 14 4
Tilled Greater than No-till 7 25

Soybeans (18 Fields)
No-till Greater than Tilled 2 3
Same (within 5 bu/ac or $5/ac) 15 8
Tilled Greater than No-till 1 7

Grain Sorghum (7 Fields)
No-till Greater than Tilled 2 0
Same (within 5 bu/ac or $5/ac) 3 2
Tilled Greater than No-till 2 5



Yield and production costs data shown in this table helped to enlighten farmers
about the perception that no-till planting would reduce yields and increase costs. In
fact, corn produced equal or higher yields in no-till in 28 of 35 comparison sites,
while costs were at least $5/acre lower in no-til in 25 of the 35 sites. No-till
soybeans produced the same or higher yields than conventional soybeans in 17 of
18 sites, and costs per acre were equal or lower in 15 of the 18 sites. There were
too few sites with sorghum to gain the same level of confidence in the results.
Although these data do not come from replicated plots in any specific location, the
number of locations and consistency of the data makes this a valid approach for
meeting the extension objectives of the project.

EXAMPLE TWO: ENERGY CONSERVATION IN CROP PRODUCTION

A research/demonstration project in Nebraska is focused on helping farmers
to design and test different cropping patterns as well as test reduced nitrogen
application levels to save energy and production costs. Following meetings with
groups of farmers as well as with individual collaborators, farmers decide what
practices they would like to test and a coordinator from the project helps in the
design of a field comparison. In some cases these are replicated trials, while in
others they are non-replicated observation plots. During the first two years of the
project, 1988 and 1989, nineteen different types of comparisons were made on one
to 36 different farms. These are summarized in Table 4 (Franzluebbers, 1989).











Table 4. Replicated trials and unreplicated observations of
alternative cropping practices compared with conventional
systems of farmers.


Replicated trials

Zero fertilizer use in crop rotations
Broadcast soybeans with rye as weed control
Vanrying fertilizer rates in crop rotation
Starter versus no starter fertilizer
Cover cropping at cultivation time
Use of animal manure
Strip cropping corn and soybeans
Relay cropping soybeans into small grain
Drili~ng versus row planting
No pesticide use

Non-replicated demonstrations

Contour strip cropping with terraces
Contour strip cropping without terraces
Banded herbicide use
No pesticide use
Tree windbreaks
Cover crop planting after harvest
Planting grain following cover crop
Ridge planting without chemicals
Ridge planting with banded chemicals
No-till drilling small grains
Use of animal manure



In these trials, the project technician assisted farmers with soil sampling using
a deep probe unit mounted on a pickup. The project paid for the soil sample
analysis. Farmer's observations were supplemented by those of the technician
through the course of the cropping season. Often the farmer received assistance at
harvest with a weigh wagon and some hours of labor to collect the data. Although
analyses of data from the replicated trials were conducted at the university, the
interpretation was left to the farmer collaborator. In the first year, there was no
response to applied fertilizer on all but one of the sites. Even with the conservative










applications used by farmers who are in the cooperator group, the fertilizer levels
they applied were higher than what is economically sound, based on data from deep
soil samples and soil tests. Most farmers were surprised by the lack of response,
even to modest applications of nitrogen. Comparisons of the alternative cropping
systems were more varied, and conclusions must await the three year's data from
the project.

EXAMPLE THREE: NEIGHBOR TO NEIGHBOR CONSERVATION FARMING
DEMONSTRATIONS

Soil conservation practices have long been components of Nebraska farming and
ranching systems, yet they have not been adopted by all farmers. With the
eligibility for farm programs tied to conservation compliance, many more farmers
need first hand experience with these practices in order to continue to qualify for
assistance. Project objectives are to establish a neighbor-to-neighbor guided tour
throughout the state; to encourage farmers to take tours of nearby farms; and to
help farmers become more familiar with proven conservation practices in their area.
In the longer term, this activity will serve to encourage more farmers to become
involved in more rigorous on- farm testing of new technologies.

In each county in Nebraska, the Cooperative Extension System has worked with
SCS, ASCS, FmHa, Natural Resource Districts, the Natural Resources Commission,
and industry to identify collaborators willing to share their conservation practice
experience and their farms. In most cases, the farm/ ranch collaborator has been
using the practice for at least five years and is willing to communicate about
advantages, disadvantages, costs, and profitability of the practice. A large sign
identifying the practice and collaborator is located along the road, with a set of fact
sheets in a waterproof mailbox at the site. This sheet includes details about the
specific practice, including costs and comments from the cooperator, and the name,
address, and phone number of the collaborator. Table 5 lists the practices that are
on demonstration during 1989, the first year of the project (Jones, 1989).













Table 5. Conservation practices in neighbor-to-neighbor demonstration
project in Nebraska, 1989. 1


Tillage Systems
No-till
Conservation tillage
Ecofallow
Contour farming

Pest Management
Leafy spurge control
Prairie dog control
Chemigation

Alternative Cropping or
Enterprise Patterns
Windbreaks
Strip cropping
Tree plantings
Fish farming
Cover crops
Alternate cropping
Grass seed production

Reduced Inputs
Low-input agriculture
Fertilizer management

Irrigation Management
SSurge valves
Irrigation water
Groundwater recharge


Total Demonstrations in 1989 359


& This program has received strong recognition by the Nebraska Farmer magazine, as well as from many local newspapers. A
large number of the single-sheet handout materials have disappeared from the mailboxes located at each demonstration site. We
hear discussions of these demonstrations curing our tours and field days. From the response and interest in this program in the
first year, we conclude that farmers are anxious to gain more information from their neighbors, and that most are willing to share
experience when there is something to show in the field. These are not replicated trials, nor will data be collected from the fields.
yet they meet the objectives of demonstrating known technology.


General Conservation Practices
Conservation farm
Conservation reserve
Terraces
Dam
Diversion
Waterways
Wildlife habitat
Stream stabilization
Water treatment plant
Grass buffer strip
Grass planting
Contour bench leveling
Farm pond
Waste storage structure
Grade stabilization

Grazing Systems
Planned grazing systems
Native grasses
Cross fencing
Range seeding
Pasture management
Prescribed burning
Solar livestock pipeline
Hay management
Brush control










ON-FARM EXPERIMENT DESIGNS


An important issue in closing the information cycle is the quality of data and
interpretations. There has long been concern about the use of demonstration plots
in extension, and the credibility of the results that farmers perceive from these plots
(Nafziger, 1984). Recent analyses suggest that it is possible to combine objectives
of demonstration with credible statistical comparison. If the objectives suggest that
full-scale experiments are needed on the farm, including randomized treatments in
each of several replications, then there is need for use of credible designs and
analyses. The obvious advantages of large plots located on a farmer's field or on
the experiment station are their visual desirability and demonstration value for
farmers, and other characteristics of large plots listed under the "farmer credibility
section" given earlier. There has always been some doubt, however, about the
statistical credibility of plots that stretch 1,200 feet across a field. A number of
analyses of data from Nebraska and Iowa, from both farmer's fields and experiment
stations, were performed and reported by Rzewnicki et al. (1988). A summary of
these results from variety trials, fertilizer level comparisons, and weed control
alternatives are shown in Table 6.



Table 6. Coefficients of variation from randomized complete block
experiments using larger than average plot size in Nebraska and
lova (from Rzeynicki et al. 1988) .


Plot Number of Coefficient
length width Treatments Replications Trials of Variation


1200' 12 to 36' 2 to 3 5 to 6 23 0.7 to 5.9

125 to 800' 20 to 30' 3 4 4 4.5 to 15.2


250 to 400' 25 to 30' 2 4 farms 1 7.8

1200' 15 to 20 13 to 22 4 farms 4 3.4 to 4.0



Results from these analyses showed that large sized experimental plots from 125
to 1200 feet long, whether on farms or on stations, can provide reliable data for
research purposes. Such plots even with widths up to 40 feet per treatment level
can still provide enough control of experimental error to meet conventional










agronomic research criteria. A successful model used by the Practical Farmers of
lowa includes long, narrow strips in a randomized complete block design (first line
in the table). With limited number of treatment levels (2 or 3) and 4 to 6
replications, the range of coefficients of variation were 0.7 to 5.9 percent for 23
farmer-conducted trials. Fifteen of these 23 trials had CV's under 3.0 percent.
Advantages for the farmers were use of full scale farm machinery and potential for
demonstration with these large plots.

Experiment station trials with long-term rotations, relay cropping, and
variety/planting date combinations had treatment level plots from 20 to 40 feet
wide and 125 to 800 feet long (second line in table). Coefficients of variation from
4.5 to 15.2 percent were well within acceptable research limits.

Another option for on-farm testing design is the use of several farms as
replications for the treatment levels of interest. One experiment on soybean row
spacings used 4 farms in three adjacent Nebraska counties (third line in table).
Analysis of on-farm plots 25 to 30 feet wide and 250 to 400 feet long resulted in a
CV of 7.8 percent. Corn variety trials on four farms in Clay County, Nebraska were
conducted over four years, with a single replication on each farm and 13 to 22
hybrids in the test. Plot sizes were similar to those used by the Iowa farmer's
group. Using farms as replicates, the uniform variety trials gave analyses with
coefficients of variation of 3.4 to 4.0 percent over the four years (last line in table).

Although initially somewhat surprised by these results, we speculate that the
long narrow plots allow the farmer or researcher to measure each treatment across
a wide range of variation when moving down the long plot. In contrast, the relative
amount of variation across treatment plots in any given replication (or farm) at one
point is small. Thus the results conform well to what is found in well-conducted
trials on the experiment station in small plots. We are currently testing the
correlations between results from large and small plots taken in the same field.

RELATIVE COSTS OF ON-FARM AND ON-STATION RESEARCH

Just as there are different objectives and simpler designs used in the on-farm
research trials, there are widely divergent costs for conducting research in different
locations. In a comparison of costs in experiments conducted in Nebraska and
Iowa, we calculated salaries, operating costs, supplies, and specific equipment
needed for the trials. Land costs were not included. Three types of trial manage-
ment and implementation were used for the analysis: on-station research trials,
on-farm research trials conducted by university researchers, and on-farm validation
trials with replications conducted by farmers (Franzluebbers et al., 1988). Station
trials had costs from $5700 to $16000, with an average cost of $11263; these trials
often include large numbers of treatments and/or detailed measurement of many
crop responses. Trials conducted on farms by researchers had a range from $1900









to $3900 and an average of $2950; these most often are uniform variety trials with
large numbers of entries or fertility tests with several levels. Replicated validation
experiments planned and conducted by farmers ranged in cost from $500 to $900
per trial, and the average cost was $800. These demonstration/ research tests most
often include large plots, side-by-side comparisons, and a small number of
treatments. Needless to say, there are different amounts of data collected from
these trials, different numbers of treatments, and different levels of control of the
non-treatment management variables. Choice of location depends on objectives of
the activity, and the level of detail needed in data collection. For some types of
information, the on-farm approach, with trials planned and run by farmers, could
be a cost-effective and efficient way to collect data and demonstrate or adapt new
technology.



FARMER PERCEPTIONS OF ON-FARM RESEARCH RESULTS

A recent opinion survey of Nebraska agricultural producers covered topics
related to the conducting of research on experiment stations and on farms.
Approximately 225 responses have been received so far to each question in the
survey, and results will be reported in more detail at the annual meetings of the
American Society of Agronomy (Rzewnicki and Rockwell, 1989).

About 70 percent of the producers surveyed conduct some comparisons of new
varieties or cropping practices with their current system within fields each year.
However, more than half of these comparisons are made by testing the innovation
on more than 10 acres in a particular field. Nearly all farmers (approximately 90%
in the survey) find university research and resulting recommendations to be useful
to their farms. Nevertheless, a large proportion (85 percent) prefer to see
experiment station results tested on working farms before recommendations are
made.

Ninety-five percent of the Nebraska producers consider university demonstration
plots as useful. Of these, two-thirds insist that these plots be located within 30
miles of their farms to be applicable to their farm conditions. In terms of
convenience of travel, researchers and extension specialists should consider that 89
percent of producers would travel 11 to 20 miles to a demonstration plot; beyond
30 miles, only 28 percent would consider traveling to see the demonstration.

For usefulness of information, on-farm strip plots as used by the Iowa farmers
are favored by 63 percent of the Nebraska producers over non-replicated
demonstration plots. Eighty seven percent of lowa farmers preferred these
replicated strip plots. Over half (54%) of the Nebraska producers surveyed
considered on-farm strip plots to be more useful than experiment station trials in










providing information applicable to their farms. About one-third rate on-farm strip
plots as equally useful compared to experiment station trials.

When asked to indicate their willingness to cooperate with university staff in
conducting on-farm research using strip plots with 5 to 6 replications and two
treatments, only 25% of Nebraska producers are willing to cooperate. Forty-three
percent said they would not be willing to participate, usually commenting that it
would be a hassle for them to find time for the special work. Some of this
resistance may be due to unfamiliarity with this on-farm research method. In
contrast, 66 percent of lowa respondents who belong to the PFI expressed
willingness to cooperate, and only 14 percent were unwilling. The Practical Farmers
of lowa have been using this technique for about three years, while it is virtually
unknown in Nebraska.

When asked if they would use the replicated strip plot method on their own if
given instructions, 37 percent of the Nebraska producers said they would while 34
percent were undecided. It may be worthwhile to initiate extension efforts in
disseminating this information and methodology to farmers to encourage adding
statistical reliability to their current initiatives in looking at new components of
technology on farms.

When asked their opinion about the importance of incorporating farmer input
into the planning of university based agricultural research, 70 percent of the
Nebraska producers consider it very important to do so. When asked how important
it would be to have on-farm testing as part of the university research process, 71
percent consider it to be very important. Although the survey did not ask the
farmers to rate current university implementation of these two ideas -- farmer input
to priorities and on-farm testing -- it is apparent from the survey results that these
should be considered in our long- term planning in research and extension.

To better understand how farmers view the research process and specific trials,
Rhoades (1984) suggests that we ask the following questions:

-- is the problem addressed important to the farmer?
-- do farmers understand the trials?
-- do farmers have the time, inputs and labor to adopt the new technology?
-- does the proposed change make sense within the current system?
-- is the mood favorable for investing in new crops or technologies?
-- is the proposed change compatible with local preferences, beliefs, or
community sanctions?
-- do farmers believe the technology will hold up over the long term, or make
their systems more sustainable?










Although written in a developing world context, these are questions that are equally
relevant for farmers, researchers, and extension specialists in an industrialized
agriculture.

THE NEW INFORMATION ENVIRONMENT

Given this enthusiasm among a large number of producers to become par-
ticipants in the research process, and the emerging methodology for conducting
credible research on farms, how do we collect this information and incorporate it
into the system? Figure 1 represents the traditional information dissemination
approach we have used in agriculture. While seemingly a one-way information flow
from researchers to farmer, in reality there is movement of information back and
forth across this continuum. Even though communication feedback is not a high
priority in this model, many mechanisms have been used to foster reactions and
responses from producers. These include communication of research and extension
specialists with commodity groups and natural resource districts, involvement in the
political arena and in funding activities. This model functioned well, especially
when research questions were straight forward, information was scarce, communica-
tion channels were technologically limited, and audiences looked homogeneous.


Primary Information Flow


Source: Lionberger and Gwin (1982)


Figure 1. Conventional information dissemination approach in extension.









An improvement on this model has been the development of extension positions
in universities that also have a formal research component. Farmer concerns
expressed at an extension function can be more easily and quickly incorporated into
a research program being conducted by the extension specialists. These types of
extension/research appointments are found in a number of land-grant institutions.

Today another information dissemination model is emerging (King et al., 1989).
This pattern reflects the growing convergence among research and extension
objectives, the blending of information and communication, the increased
participation in the system by farmers/ranchers and industry managers/decision
makers, and emergence of new information dissemination technologies. This
information convergence model shown in Figure 2 builds on the communication
options available to each of the groups listed. Research and extension personnel
now have available extensive communication possibilities. Numerous agricultural
clients are collaborating in both the research and extension activities, as described
above. Farmers and ranchers presently have countless message sources for their
agricultural intelligence. The power of the communication overlap or convergence,
however, derives from all individuals in the communication process who are at once
both senders and receivers of information.


Figure 2: The Information Convergence Model










As described in earlier sections, an increasing number of research and extension
plans are including farmers and ranchers as part of the process. At the same time.
more farmers are beginning on-farm testing and experimentation. Mass rural
audiences of the past are becoming recognized as discrete groups of individuals.
many of whom are articulate and capable of communicating with a large number
of other senders and receivers of information. Two-way communi cation is the norm
in this new model, underpinned by communication technologies; this includes the
hardware of computer networks and satellites, software for a near-infinite range of
applications, and new participatory communication theories (King and Francis, 1988;
Dervin, 1989).

Traditionally, information dissemination is a combination of the total mix of
sources, messages, channels, receivers, feedbacks, and outcomes that result from the
interrelationship of components in the system. Application of this process in each
environment also influences both the mix of elements and the form of the outcome.
The following figure visually describes a number of these functions, how they have
operated in the past in contrast to how they operate in the current context of a~gri-
culture. A description of each element is given.









Figure 3: A Description of the Changing Communication Functions


Function



Source:


Past


Today


Description

A move from single or few
senders of messages to many
senders, often in dispute.


A move from audience
homogeneity to audience
segmentation.

A move from few messages
to massive volumes of
message es.


A move from few
communication channels to
multi-media and many
access points.

A move from passive receivers
to active participants.




A move from minimal feedback
to multi-feedback loops.


A move from unspecified
outcomes to behavioral, but
uncertain outcomes.


A move from known
communication environments
to unknown and changing
communication environments.


Audience:


* *


4 ---t




*---ge


Message:





Channel:


Receiver:


c-c


e-c~ e


Feedback:


*-*


*-*


?' 7


Outcome:


Environment: I -e


Sourc: J. W. King. 1989. Unpalblished Paper. Depmrman of
Agicrimnirl Communications Intimtut of Agriculale cand
NamralResorces. University of Nebrasks- Lincola Lincoln. Ne


* ~p
*-' "
.##"

**









The increasing number of information sources and complexity of channels are
obvious. Audience segmentation is a concept used by advertisers and educators to
try to fine-tune packages of information for those most interested in getting that
specific message. What is striking about the several diagrams in Figure 3 is the
increased number of interactions that are taking place and the new routes that
information may take in this communication environment. People are now both
senders and receivers of information, playing several roles in the matrix, often at
the same time. While this figure describes the complexity of the current
information age, it is intriguing to speculate about how this will evolve in the future.
The other feature of several of the small diagrams is the cyclical nature of the
communication patterns. We are truly going full circle, and in a way closing the
information cycle in agriculture as well as in other sectors of our economy and
society.

CONCLUSIONS

This discussion describes how we are closing the information cycle. There are
opportunities for researchers, extension specialists, farmers and ranchers, and
agribusiness to collaborate in future information networks. Participants in the
information cycle can come together in the planning of research and implementa-
tion of field information discovery activities. Farmers are becoming respected
participants in the generation and dissemination of information. On-farm, internal
knowledge and experience are becoming a highly valued information commodity.
At the same time, external, off-farm information is being reviewed by producers
through a filter of values, such as environmental impact, fossil fuel use, and
production/profit sustainability. We are beginning to measure external information
by the yardstick of specificity of application (Walters et al., 1989). As external,
off-farm information becomes incorporated and integrated with on-farm experiences
into internal knowledge, external knowledge is transformed and becomes a
"value-added commodity" for the specific farm and local conditions.

We conclude that closing the information cycle depends on a number of current
changes that are occurring in agriculture:

-- new methods of generating and disseminating information appear to be work-
ing as more farmers become involved with the process
-- research agendas that are developed with farmers and ranchers help to
address the most pertinent questions they face in management
-- researchers and extension specialists can recognize and document success
when results of their work are applied to solve current production constraints
-- producers are becoming more aware of the need for specific information that
can be applied to their own farm or ranch conditions
-- more farmers and ranchers are receptive to learning about the need for
unbiased comparisons and valid designs to test new alternatives










-- extension specialists are more aware of new communications technologies and
the need to incorporate these into their programs

As we project our plans into the future, it appears that participatory methods
with farmers and ranchers will become increasingly important. This is consistent
with the philosophy of farming systems research and extension, and program
planners can learn from available methods and recent experiences. Closing the
information cycle can help us to build cooperatively toward a more sustainable
agriculture.



REFERENCES

Dervin, R. 1989. Audience as listener and learner, teacher and confidante: the
sense making approach. In: Public Communication Campaigns, R. E. Rice and
C. K. Atkin, editors. Sage Publ. Co., Newbury Park, California. pp. 67-89.
Dickey, E. C., D. E. Eisenhauer, G. A. Wicks, and L. E. Lucas. 1989. Final
Report, Agricultural Energy Conservation Project to Conserve Energy, Soil and
Water by Expanding Conservation Tillage, Ecofallow, and Irrigation Manage-
ment Technology. Cooperative Extension Service, Univ. Nebraska, Lincoln,
Nebraska. 69 pp + appendices.
Dickey, E. C., P. Jasa, D. Shelton, D. Eisenhauer, R. Grisso, and A. Martin. 1989.
Area conservation tillage meetings. Conservation Tillage Proceedings No. 8,
Univ. Nebraska Cooperative Extension System, Lincoln, NE.
Francis, C. A. 1986. Dynamic integration of research and extension: igniting the
SPARC. Proc. Farming Systems Research/Extension Workshop, Kansas State
Univ., Manhattan, Kansas. October 5-8.
Franzluebbers, A. 1989. Project Report to Nebraska Energy Office (Unpublished
data).
Franzluebbers, A, C. Francis, P. E. Rzewnicki, R. Thompson, G. Lesoing, and R.
Elmore. 1988. Relative costs and efficiencies of on-farm versus on- station
research. Amer. Soc. Agron. Ann. Meetings, Anaheim, California, November
28, 6 pp. Agron. Abstr. p. 126.
Hildebrand, P. E., and F. Poey. 1985. On-farm Agronomic Trials in Farming
Systems Research and Extension. Lynne Reinner Publ. Co., Boulder, Colorado.
162 pp.
Jones, A. 1989. Progress Report on Neighbor-to-Neighbor Project, Coop. Extension
Service, Univ. Nebraska (unpublished).
King, J. W., and C. A. Francis. 1988. Back to the future: the power of communica-
tion and information. Plenary Address at Farming Systems Research/Extension
Conference, Univ. Arkansas, Fayetteville, October 11.










King, J. W., C. A. Francis, and J. G. Emal. 1989. Evolution in revolution: new
paradigms for agriculture and communication. Presented at "Future View", Sixth
General Assembly of World Future Society, Washington DC, July 16-20. 23 pp.
Lionberger, H. F., and P. H. Gwin. 1982. Communication strategies: a guide for
agricultural change agents. Interstate Press, Danville, Illinois.
Nafziger, E. D. 1984. Use of demonstration plots as extension tools. J. Agron.
Education 13:47-49.
Rhoades, R. E. 1984. Understanding small-scale farmers in developing countries:
sociocultural perspectives on agronomic farm trials. J. Agron. Education
13:64-68.
Rzewnicki, P. E., and S. K Rockwell. 1989. Farmer's perceptions of technology
transfer in agronomy. Agronomy Abstracts p. 28.
Rzewnicki, P. E., R. Thompson, G. W. Lesoing, R. W. Elmore, C. A. Francis, A. M.
Parkhurst, and R. S. Moomaw. 1988. On-farm experiment designs and
implications for locating research sites. Amer. J. Alternative Agric. 3:168-173.
Saeed, Mohammad, C. A. Francis, and J. F. Rajewski. 1984. Maturity effects on
genotype x environment interactions in grain sorghum. Agron. J. 76:55-58.
Walters, D. T., D. A. Mortensen, C. A. Francis, R. W. Elmore, and J. W. King.
1989. Specificity: the context of research for sustainability. J. Soil & Water
Conservation (submitted).