Farmers, Phosphorus and Water Quality, Part II: A Descriptive Report of Beliefs, Attitudes and Best Management Practices in the Maumee Watershed in Northwest Ohio

Abstract

The purpose of this study was to 1) better understand the prevalence of a variety of BMPs in the Maumee watershed, 2) identify why farmers choose to adopt certain BMPs, and 3) identify what motivates individual farmer willingness to adopt additional practices on their farm. This information may reveal what, if any, methods may be employed to increase BMP implementation, thereby ultimately improving water quality and protecting associated ecosystem services. Previous research has focused largely on socio-demographic predictors of adoption and economic motivations. To evaluate these complex decision-making processes, this survey incorporates a variety of behavioral and psychological motivators.

Full text

1
Elizabeth A. Burnett, Robyn S. Wilson, Brian Roe, Greg Howard, Elena Irwin, Wendong
Zhang, and Jay Martin.
2015
FARMERS, PHOSPHORUS AND
WATER QUALITY: PART II
A DESCRIPTIVE REPORT OF BELIEFS, ATTITUDES AND BEST
MANAGEMENT PRACTICES IN THE MAUMEE WATERSHED OF THE
WESTERN LAKE ERIE BASIN

2
Contact Information
Robyn S. Wilson, Associate Professor
School of Environment and Natural Resources
The Environmental and Social Sustainability Lab
The Ohio State University
2021 Coffey Rd
Columbus, OH 43210
(614) 247-6169 (phone)
(614) 292-7432 (fax)
wilson.1376@osu.edu
http://go.osu.edu/RobynWilson
Project Website: http://ohioseagrant.osu.edu/maumeebay/
Funding for this research was provided by the NSF Dynamics of Coupled Natural and Human
Systems Program (BCS-1114934)
Suggested Citation
Burnett, E.A., R. S. Wilson, B. Roe, G. Howard, E. Irwin, W. Zhang, and J. Martin. 2015.
Farmers, phosphorus and water quality: Part II. A descriptive report of beliefs, attitudes
and best management practices in the Maumee Watershed of the western Lake Erie
Basin. Columbus, OH: The Ohio State University, School of Environment & Natural
Resources.

3
Table of Contents
Suggested Citation ................................................................................................................................. 2
Executive Summary ................................................................................................................................... 4
Field Site .................................................................................................................................................. 6
Data Collection ....................................................................................................................................... 7
Survey Response ................................................................................................................................... 7
Analyses .................................................................................................................................................. 8
Farmers and Nutrient Loss: Individual Beliefs, Attitudes, and Perceptions ...................................... 9
Awareness of Water Quality and 4Rs ................................................................................................. 9
Nutrient Loss Beliefs ............................................................................................................................ 11
Perceived Control and Risk Regarding Nutrient Loss .................................................................... 13
Farm and Farmer Characteristics .......................................................................................................... 16
Gender and Age ................................................................................................................................... 16
Education and Experience .................................................................................................................. 16
Farm and Income ................................................................................................................................. 17
Risk Attitude .......................................................................................................................................... 17
Farmer Identity ...................................................................................................................................... 18
Behavior on Low Productivity Fields ................................................................................................. 19
Behavior on Average Productivity Fields .......................................................................................... 31
Behavior on High Productivity Fields ................................................................................................ 43
Summary of Field Comparisons ......................................................................................................... 55
References ................................................................................................................................................ 56
Appendix A: Preliminary differences between adopters and non-adopters .................................... 58
Adoption of Cover Crops ..................................................................................................................... 58
Adoption of Storm-Delay Broadcasting ............................................................................................. 60
Adoption of Seasonal-Delay Application .......................................................................................... 62
Adoption of Injection Application ........................................................................................................ 64

4
Executive Summary
Algal blooms have been a serious issue in Lake Erie since the 1960s. The blooms, which are
harmful to wildlife and humans (NOAA 2009), occur when phosphorus levels are high within the
lake. Although total phosphorus levels in Lake Erie decreased and stabilized during the 1980’s
and 1990’s due to both farmer best management practices and other policies (e.g., phosphorus
banned from detergents) (Pinto et al. 1986), data collected within the last decade have revealed
an increase in dissolved reactive phosphorus (DRP). While there is some uncertainty about the
current causes of this increase in DRP, experts are confident that the changes are likely due to
agricultural runoff during large rain events, particularly in the Maumee watershed. The
Phosphorus Task Force of Ohio recommends that farmers use best nutrient management
practices (BMPs) to reduce DRP loading into Lake Erie tributaries (Ohio EPA 2010).
Agricultural BMPs are meant to improve soil health (e.g., conservation tillage, cover cropping,
controlled traffic), increase nutrient management precision (e.g., soil testing, grid sampling,
comprehensive nutrient management planning), improve the filtration of surface and subsurface
runoff (e.g., filter strips, grass waterways, biofilters), and improve manure management (e.g.,
following Natural Resources Conservation Service (NRCS) guidelines). Adoption of a variety of
these practices can serve to curtail nutrient loss from agro-ecosystems, thereby decreasing the
overall impact of agriculture on water quality. Preliminary findings from our project indicate that
particular changes related to placement of fertilizer with the soil, avoiding application on frozen
or saturated ground, delaying application in light of a major rainfall event, and cover crops may
hold the most promise for decreasing DRP loss through field management strategies. See
Appendix A for correlational results and preliminary findings regarding what motivates adoption
of these particular practices, and how best to engage non-adopters.
Although many BMPs are known to be effective at reducing nutrient loss, their adoption is
largely voluntary in Ohio. The purpose of this study was to 1) better understand the prevalence
of a variety of BMPs in the Maumee watershed, 2) identify why farmers choose to adopt certain
BMPs, and 3) identify what motivates individual farmer willingness to adopt additional practices
on their farm. This information may reveal what, if any, methods may be employed to increase
BMP implementation, thereby ultimately improving water quality and protecting associated
ecosystem services. Previous research has focused largely on socio-demographic predictors of
adoption and economic motivations. To evaluate these complex decision-making processes,
this survey incorporates a variety of behavioral and psychological motivators.

5
The descriptive findings in this report are the result of a survey conducted in early 2014 among
row crop farmers living within the Maumee watershed of Lake Erie (a watershed in the Western
Lake Erie drainage basin). This report includes a description of the study area and survey
methods, as well as a summary of the survey findings with tables and figures.
Key findings in this report include:
1. The majority of farmers in the Maumee watershed perceive that the water quality of the
rivers, streams, and lakes near where they live is better than the water quality of Lake
Erie. A third of the farmers are not familiar with 4R Nutrient Stewardship.
2. While a minority of farmers agrees that taking additional steps to reduce nutrient loss on
their farms would be easy, a majority of farmers agree that they can engage in practices
that reduce nutrient loss on their farms. Most farmers have a strong sense of
responsibility to protect local water quality and to adopt BMPs that limit nutrient loss.
3. The majority of farmers believe that current practices on their own farms are sufficient to
minimize nutrient loss. About a quarter of farmers believe that other farmer’s practices
are insufficient to minimize nutrient loss, suggesting that many farmers feel that others in
their community should be doing more. In fact, nearly a third of farmers believed that
water quality issues in agriculture are the result of poor management among a small
number of farmers.
4. Farmers had a moderate perception of control, perceiving the most control over soil
erosion and the least control over phosphorus lost during heavy rainfall events. They
perceive relatively less control over subsurface drainage compared to surface runoff.
5. The majority of farmers believe that the impacts of nutrient loss will be most serious for
those on and around Lake Erie and for plants and animals, and the least serious for
his/her own family or community. However, when it comes to the likelihood of negative
impacts, farmers believed decreased crop yields and increased production costs were
more likely than decreased water quality and soil health.
6. There is great potential for increased adoption of most BMPs that may help to address
the current dissolved reactive phosphorus issues in Lake Erie. Adoption rates ranged
from a low of 13% for hiring a 4R certified applicator over an applicator without
certification, to a high of 57% for regular soil testing to inform management within the
rotation. Many of those who planned on adopting a particular practice next season were
new adopters (18-50%), meaning they had not yet adopted those practices on the

6
particular field. See Appendix A for a more detailed discussion of potential differences
between adopters and non-adopters of critical BMPs.
Study Methods
Field Site
The sample population for this survey consisted of corn and soybean farmers within the
Maumee River watershed (Figure 1). Similar to the larger watersheds of the Mississippi River
and Chesapeake Bay, the Maumee River watershed features a variety of environments
(agricultural fields and concentrated animal feeding operations, wetlands, and urban/industrial
settings), economies (extensive agriculture, manufacturing and chemical industries, and
transportation networks), and administrative settings (three states and 25 counties), as well as
extensive watershed impacts on downstream western Lake Erie (Kane et al. 2014).
Figure 1. The Maumee River and watershed spans 25 counties in 3 states and drains northeast from IN
to Lake Erie for a total of 16,000 km2 and watershed lengths of up to 170 km. (Map credit: Ohio
Department of Natural Resources, http://ohiodnr.com/tabid/23368/Default.aspx)

7
Data Collection
Researchers from the Colleges of Food, Agricultural and Environmental Sciences and Arts and
Sciences at The Ohio State University created three mail-back questionnaires used in this
study. Each of the three survey versions included a section of questions about field
management practices that were to be answered while the farmer considered a specific field
with low, average, or high crop productivity. Of the approximately 12,000 addresses in the 24
counties of the Maumee Watershed, a random sample (n = 2500 for each version) of corn and
soybean farmer addresses was purchased from a private sampling firm.
The surveys were conducted following Dillman’s Tailored Design method (Dillman 2000). In
February of 2014, an announcement letter was sent to the random sample of farmers informing
them that they would soon be receiving a survey in the mail. A cover letter and a survey booklet
with prepaid return postage were sent to all participants a week later. Included with this first
survey was a token incentive of a one dollar bill to increase response. In early March a reminder
postcard was sent to participants who did not return the survey. In late March an additional
mailing of the cover letters and survey booklets was sent out to those participants who had not
yet responded. In late April, a final reminder letter was sent to participants.
Survey Response
The three surveys were mailed out to a total of 7500 farmers (n = 2500 for each version). A total
of 3,937 surveys were initially returned. Of these, 599 (15.0%) were refusals and 43 (0.01%)
were undeliverable. Additionally, 60 (0.02%) were farmers who didn’t have operations in the
Maumee watershed. Therefore, a total of 3,234 surveys were initially included for potential
analysis, with an adjusted response rate of 43.12%. Of these, 438 (13.5%) were no longer
farming. These surveys were eliminated from analyses. An additional 32 surveys were taken out
for having an insufficient number of questions answered for analyses. In total, 2,764 surveys
were used in this analysis. The majority of these surveys (87.0%) were returned from the first
mailing, while 13% were returned from the second mailing. Independent samples t-tests
revealed that respondents who returned surveys from the first mailing were significantly older
(MD = 2.59, SED = .911, t(2596) = 2.84, p = .005) and had more experience with farming (MD =
2.88, SED = 1.25, t(2597) = 2.31, p = .021) than the respondents who returned surveys from the
second mailing. Further comparison of our sample data with the census data of the Maumee
watershed (USDA 2009) showed that our sample overrepresented farmers whose annual
income exceeded $50,000 and underrepresented farmers whose income was less than $50,000
(Table 1).

8
Table 1. Comparison of Annual Income between Census and Sample Data
Comparison of Census Income to Sample
Income
Census
Sample
Less than $50,000
38.0%
16.7%
$50,000-$99,999
17.1%
20.5%
$100,000-249,999
20.2%
27.8%
$250,000-$499,999
12.3%
15.9%
$500,000 or greater
12.4%
19.2%
Analyses
Data were analyzed using the Statistical Program for the Social Sciences (SPSS/PC+20).
Analysis involved descriptive statistics, frequency distributions, measures of central tendency
(mean, median, mode), valid percentages, Chi-square and ANOVA. Valid percentages were
gathered by eliminating missing responses from the variable being analyzed. All percentages
shown are valid percentages. The following sections of the report summarize the survey results.
Each section includes response frequencies and percentages for the relevant survey questions.
Results are illustrated using figures and tables. For additional relational analyses contact the
authors.

9
Farmers and Nutrient Loss: Individual Beliefs, Attitudes, and Perceptions
Awareness of Water Quality and 4Rs
To measure farmer awareness of water quality, respondents were asked to rate the overall
quality of the water near where they live as well as the quality of the water in Lake Erie (Figure
2). A third of respondents (31.2%) indicated that the water quality of Lake Erie was neither good
nor bad, however only a quarter of respondents (25.7%) reported this same quality of the water
near them. Overall, respondents believed the water quality of the water near them was better
than the water quality of Lake Erie.
Additionally, respondents were asked about their awareness of algae issues in Grand Lake St.
Mary’s and the western Lake Erie basin (Table 2). The majority of respondents (70.3%) reported
that they were moderately or very aware of the algae issues in Grand Lake St. Mary’s, whereas
about half of respondents (55.9%) reported the same about the western Lake Erie basin. A
similar percentage of respondents reported having no awareness of algae issues in Grand Lake
St. Mary’s and the western Lake Erie basin (14% and 16.5%, respectively).
In order to measure farmer awareness of the 4Rs of Nutrient Stewardship, respondents were
asked to indicate their familiarity with this concept, as well as the frequency with which they had
received information or participated in programming about 4R Nutrient Stewardship. Notably,
nearly a third of respondents (30.4%) indicated that they were not at all familiar with 4R Nutrient
Stewardship (Figure 3), while 40% of respondents said they had never been exposed to
information or programming about the 4Rs (Figure 4).
Figure 2. Respondent beliefs about the water quality of rivers, streams, and lakes nearby (n =
2730) as well as the water quality of Lake Erie (n = 2560) in percentages.
0
5
10
15
20
25
30
35
Very bad Neither bad nor good Very good
Valid Percentages
Farmer Beliefs About Water Quality
Water quality of rivers,
streams, and lakes near
where you live
Water quality of Lake Erie

10
Table 2. Respondent percentages for awareness of algae issues in Grand Lake St. Mary’s and
the western Lake Erie basin.
Awareness of the algae issues
in…
n
Not at all
aware
A little
aware
Moderately
aware
Very
aware
Grand Lake St. Mary's
2738
14.0
15.8
32.4
37.9
Western Lake Erie basin
2701
16.5
27.6
34.5
21.4
Figure 3. Respondent familiarity with the concept of 4R Nutrient Stewardship shown in
percentages (n = 2741).
Figure 4. Respondent exposure to 4R Nutrient Stewardship through receiving information or
participating in programs about the 4Rs, in percentages (n = 2743).
0
5
10
15
20
25
30
35
Not at all Slightly Moderately Very Extremely
Valid Percentages
Farmer Familiarity with 4R Nutrient
Stewardship
0
5
10
15
20
25
30
35
40
45
Never Rarely Sometimes Often Frequently
Valid Percentages
Exposure to 4R Information and
Programming

11
Nutrient Loss Beliefs
To evaluate farmer understanding of nutrient management, respondents were asked to agree or
disagree with a series of statements pertaining to efficacy and responsibility of taking action to
reduce nutrient runoff from their farms (Table 3).
Three statements in this section measured farmer’s sense of efficacy regarding nutrient loss.
Specifically, the questions attempted to assess farmer’s beliefs about their ability to reduce
nutrient loss on their farms. About a third of the respondents agreed that taking additional steps
to reduce nutrient loss on their farms would be easy, while nearly a quarter of respondents
disagreed with this statement. Nearly all respondents (79.7%) agreed that they could engage in
practices that limit nutrient loss on their farms, and over half (56.2%) agreed that they had the
ability to change their current practices to further limit nutrient loss on their farms. These
findings suggest that farmers, although able to take action to reduce nutrient loss on their farms,
perceive the adoption of additional steps to reduce nutrient loss to be a difficult task.
Three statements in this section measured farmer’s sense of responsibility regarding nutrient
runoff and adoption of nutrient practices. Over half of the respondents (59.3%) agreed that
farmers in northwest Ohio should be doing more to reduce nutrient runoff into waterways.
Interestingly, 87.3% of respondents agreed that it is their responsibility to adopt BMPs that limit
nutrient loss on their farms, while an even larger percent of respondents (90.8%) agreed that it
is their responsibility to help protect local water quality. These results indicate that farmers feel a
strong sense of responsibility to do more to reduce nutrient loss on their farms.
Two statements measured farmer’s sense of how sufficient their practices are related to other
farmers when it comes to minimizing nutrient loss. About a third of respondents (31.4%) agreed
that current practices on other farms were sufficient to minimize nutrient loss, whereas a
majority (69.8%) agreed that practices on their own farm were sufficient to minimize nutrient
loss. About a quarter of respondents (26.2%) disagreed that other farmer’s practices were
sufficient, suggesting that many farmers feel that they are doing enough on their own farms but
that others in their community should be doing more. A similar question in our previous report
(Wilson et al. 2013) revealed that 76.7% of respondents agreed that nutrient management
practices on their farms were sufficient to protect local water quality, suggesting that, since our
2012 survey, there has been a slight increase in farmers who no longer feel their practices are
sufficient to protect local water quality from nutrient runoff. The final statement measured
whether farmers believed water quality issues in agriculture are the result of poor management

12
among a small number of farmers. Responses to this statement were distributed fairly evenly,
with 36% disagreeing, 32% agreeing, and 32% staying neutral.
Table 3. Questions and response percentages for farmer beliefs regarding their sense of
efficacy and responsibility toward taking action to reduce nutrient loss.
Beliefs
n
Strongly
disagree
Disagree
Neither
disagree
nor
agree
Agree
Strongly
agree
*Taking additional steps to
reduce nutrient loss on my farm
would be easy
2738
2.3
21.9
38.8
34.4
2.7
*I can engage in practices that
limit nutrient loss on my farm
2727
0.4
2.2
17.9
68.0
11.7
*I have the ability to change my
practices to further limit nutrient
loss on my farm
2734
2.3
9.3
32.3
50.5
5.7
†Farmers in northwest Ohio
should be doing more to reduce
nutrient runoff into waterways
2730
1.0
5.2
34.5
46.6
12.7
†It is my responsibility to adopt
best management practices that
limit nutrient loss on my farm
2745
0.3
1.0
11.4
66.6
20.7
†It is my responsibility to help
protect local water quality
2746
0.1
0.9
8.2
64.9
25.9
Current nutrient management
practices on other farms in my
community are sufficient to
minimize nutrient loss
2732
4.5
26.2
37.8
28.9
2.5
Current nutrient management
practices on my farm are
sufficient to minimize nutrient
loss
2741
0.4
7.2
22.7
58.2
11.6
Water quality issues in
agriculture are the result of poor
management among a small
number of farmers
2734
8.2
28.2
31.6
25.6
6.4
*Statements pertaining to efficacy; †Statements pertaining to responsibility

13
Perceived Control and Risk Regarding Nutrient Loss
To measure perceived control over nutrient loss (Figure 5), five questions were asked, with
responses ranging on a scale from 0 (no control) to 6 (complete control). Overall, results
showed that farmers perceive a moderate amount of control over nutrient loss (M = 3.48, SD =
1.04, n = 2749). Farmers felt they had the most control over soil erosion on their farms and the
least control over phosphorus lost during heavy rainfall events. Additionally, they perceived
relatively less control over subsurface drainage as opposed to surface runoff.
Figure 5. Farmer perceived control over five different aspects of nutrient loss.
Several questions on the survey investigated respondent risk perception. Two of these
questions asked farmers to rate their level of concern about nutrient loss on their own farm and
the negative impacts of nutrient loss in western Lake Erie, choosing on a scale from 0 (not at all
concerned) to 6 (extremely concerned) (Figure 6). Levels 5 and 6 (i.e., high concern) made up
45.1% of responses for level of concern about nutrient loss occurring on your farm, whereas
34% of respondents marked these same levels for the negative impacts of nutrient loss in
western Lake Erie. These findings suggest that, although farmers have a moderate level of
concern for both of these issues, concern about nutrient loss on the farm is higher.
0
5
10
15
20
25
30
35
40
No control Complete control
Valid Percentages
Perceived Control Over Nutrient Loss
Control over soil erosion
on your farm (n = 2745)
Control over your farm's
impact on local water
quality (n = 2730)
Control over phosphorus
lost through surface
runoff (n = 2732)
Control over phosphorus
lost through subsurface
drainage (n = 2724)
Control over phosphorus
lost during heavy rainfall
events (n = 2731)

14
Figure 6. Respondent levels of concern regarding nutrient loss.
To evaluate farmer risk perception of the negative impacts of nutrient loss, respondents were
asked to rank the likelihood of agricultural nutrient loss leading to four outcomes: decreased
crop yield, decreased water quality, decreased soil health, and increased production costs
(Table 4). Overall, the majority of farmers (>80% for each) perceive that all categories are likely
negatively impacted by nutrient loss. The two strongest agreements were that nutrient loss
decreases crop yield (91.2%) and increases production costs (90.7%). Notably, nearly half of all
respondents (48.3%) marked that deceased water quality from nutrient loss was somewhat
likely, and only 9.1% of respondents marked that this impact was extremely likely.
Table 4. Respondent perception of the likelihood of negative impacts resulting from agricultural
nutrient loss, in valid percentages.
Factors that may be
impacted
n
Not at
all
likely
Somewhat
likely
Very
likely
Extremely
likely
Decreased crop yield
2747
8.7
42.0
38.0
11.2
Decreased water quality
2735
10.5
48.3
32.2
9.1
Decreased soil health
2740
12.7
39.4
36.9
11.0
Increased production costs
2742
9.3
33.3
41.7
15.7
The literature on risk perception indicates that overall subjective risk judgments are driven less
by likelihood estimation and more by the perceived severity of consequences. In particular,
those consequences that are seen as local and personal have a greater influence on perceived
risk (Tucker and Napier 1998; Lichtenberg and Zimmerman 1999; Liberman and Trope 2010;
0
5
10
15
20
25
30
35
Not at all concerned Extremely concerned
Valid Percentages
Farmer Concern About Nutrient Loss
Concern about nutrient loss
occuring on your farm (n = 2647)
Concern about negative impacts
of nutrient loss in western Lake
Erie (n = 2661)

15
Perrings and Hannon 2001; Spence et al. 2012). To better understand farmer risk perception of
nutrient loss, respondents were asked to rate the seriousness of impacts at a variety of scales
ranging from the personal (e.g., me and my family, my community) to the impersonal (e.g.,
plants and animals) (Table 5). Although most farmers agreed that the negative impacts on these
factors were at least slightly serious (>90% for each), trends showed that impacts on
communities on and around western Lake Erie as well as on plants and animals in Lake Erie
were more serious than impacts at a more personal or local scale.
Table 5. Farmer perception of the seriousness of agricultural nutrient loss impacts across a
range of personal to impersonal scales, in valid percentages.
Factors that may be impacted
n
Not at
all
serious
Slightly
serious
Serious
Moderately
serious
Extremely
serious
You and your family
2698
10.9
38.9
28.4
16.3
5.5
Your local community
2687
10.5
39.4
29.4
16.0
4.7
Communities on and around
western Lake Erie
2679
3.2
17.1
29.9
30.7
19.2
Plants and animals in local streams
2692
7.9
30.6
30.6
22.2
8.8
Plants and animals in Lake Erie
2691
3.7
20.6
26.9
29.5
19.2
In summary, farmers feel a strong sense of responsibility to protect local water quality and to do
more on their farms to reduce nutrient loss, although they perceive the adoption of additional
steps to reduce nutrient loss to be a difficult task. Many farmers feel that they are doing enough
on their own farms, but that others in their community should be doing more. In fact, a third of
respondents said that water quality issues in agriculture are the result of the poor management
among a small number of farmers, perhaps pointing to a frustration with a minority of the
farming community who choose not to take action to reduce nutrient loss. Respondents
perceived a moderate level of control over nutrient loss, but felt much less control over
phosphorus lost during heavy rain events. Although farmers believe that the impacts of nutrient
loss are most serious for communities on and around Lake Erie, as well as on plants and
animals in Lake Erie, their level of concern regarding nutrient loss impacts was higher for their
own farm than for Lake Erie. This personal, local concern was highest for the risks of decreased
crop yield and production costs. This suggests that, although the majority of farmers believe that
the impacts of nutrient loss will be most serious for those on and around Lake Erie, most have
greater concern for the risks to their own crops and finances.

16
Farm and Farmer Characteristics
Respondents were asked standard questions about gender, age, education, and basic farm
characteristics. This demographic information demonstrates what portion of the population is
best represented by the data presented here.
Gender and Age
The majority of respondents were male (97.9%) with a mean age of 58 (Table 6). Females
constituted 2.1% of respondents and had a mean age of 72. Although most farmers were in
their 50s, age ranged from 17 to 96 for men and 23 to 88 for women.
Table 6. Respondent gender and age (n = 2622).
Respondent Gender and Age
Gender
Valid %
Mean Age
Std. Deviation
Male
92.7
58.6
12.1
Female
7.3
71.6
66.2
Education and Experience
A small proportion (1.8%) of respondents did not finish high school, but a large proportion
(50.8%) of respondents did obtain a high school degree (Table 7). Nearly half of all respondents
(47.4%) studied in college. The mean number of years respondents had been farming was 37.8
years (SD = 13.9, n = 2596). Over half of the respondents reported that they were third
generation farmers (Table 8), and 28% of respondents said they were retired from an
occupation other than farming.
Table 7. Respondent education (n = 2611).
Respondent Education
Level of Education
Valid %
Some high school
1.8
High school degree
50.8
Some college
18.8
Associate's degree
10.8
Bachelor's degree
12.4
Graduate or professional degree
5.4

17
Table 8. Respondent farming generation (n = 2582).
Respondent Farming
Generation
Valid %
First generation farmer
10.4
Second generation farmer
22.2
Third generation farmer
67.4
Farm and Income
Approximately 30% of respondents manage livestock or poultry on their farms. Respondents
grew a mean of 211 acres of corn (SD = 348.7) and 236 acres of soybean (SD = 329.8) last
year. In the past year, a mean of 61.4% of total acreage (SD = 68.7) was planted in
conventional tillage, while a mean of 65.4% of total acreage (SD = 74.5) was planted in
conservation tillage, and a mean of 73% of total acreage (SD = 77.8) was planted in no-till. Only
7.6% of respondents have converted woodland, pasture, wasteland, fallow, or CRP land into
cropland since 2010.
Approximately 17% of respondents make less than $50,000 annually from their farm (Table 9).
About 20% of respondents are making $500,000 or greater annually from their farm. Over half
of the respondents receive off farm income from themselves (56.9%) and/or from their spouse
(54.1%). Of those respondents that do receive off farm income, 45.4% receive $10,000-49,999
annually, and 34% receive $50,000-99,999 annually.
Table 9. Annual gross sales from farm (n=2327).
Annual Gross Sales From Farm ($)
Valid %
< 50,000
16.7
50,000-99,999
20.5
100,000-249,999
27.8
250,000-499,999
15.9
> 500,000
19.2
Risk Attitude
To get an overall sense of respondent’s tolerance for risk in day-to-day life, respondents were
asked to rate their willingness to take risks on a scale from 0 (not willing to take risks) to 10
(very willing to take risks) for three separate items. These items included willingness to take
risks in general, with investments, and as a farmer (Table 10). Risk tolerance for respondents
was highest for occupational risk as a farmer (M = 5.41, SD = 2.40) and lowest for risk with

18
investments (M = 4.87, SD = 2.34). These items were combined to form an overall mean risk
attitude of 5.12 (SD = 2.15), or fairly moderate risk tolerance.
Table 10. This table shows farmer risk attitude. Risk was measured using a scale from 0 (not
willing to take risks or risk averse) to 10 (very willing to take risks or risk tolerant).
Risk Attitude
Mean
n
Std. Deviation
General risk
5.07
2696
2.27
Risk with investments
4.87
2697
2.34
Risk as a farmer
5.41
2702
2.404
Farmer Identity
The Good Farmer Identity Theory suggests that identities, or sets of values, may influence
farmer’s management decisions (Burton 2004; McGuire et al. 2013). We included five items
oriented toward a profit identity and seven items oriented toward a conservationist identity.
Respondents rated the importance of each item to their definition of a good farmer, on a scale
from 0 (not at all important) to 4 (very important), which allowed respondents to express their
values with high flexibility (e.g., respondents could mark values from both profit and
conservation identities as being of equal importance to their definition of a good farmer). The
mean was 1.73 (SD = 0.75) for profit identity and 2.99 (SD = 0.64) for conservation identity,
suggesting that respondents felt that conservation values (e.g., minimizing nutrient runoff into
waterways and maintaining or increasing soil organic matter) were more important in defining a
good farmer than were profit-oriented values (e.g., having the highest yields per acre and the
highest profit per acre).

19
Farmers and Nutrient Management: Field-Specific Behavior
The following section is broken into three categories: behavior on low productivity fields (0-150
bushels/acre of corn; 0-45 bushels/acre of soybeans), behavior on average productivity fields
(150-170 bushels/acre of corn; 45-50 bushels/acre of soybeans), and behavior on high
productivity fields (170 or greater bushels/acre of corn; 50 or greater bushels/acre of soybeans).
Respondents were asked a series of questions keeping a field in mind that belonged to one of
these three categories. The following sections are summaries of their responses, grouped into
sub-categories of basic field characteristics, recent management activities, and efficacy and
adoption of best management practices (BMPs).
Behavior on Low Productivity Fields
Basic field characteristics. The average size of the low productivity fields was 36.89 acres
(SD = 61.96), with a range from 0 to 1,100 acres (n = 943). The majority of respondents (94.6%)
reported that these fields were in an established rotation, with nearly half being in a corn and
bean rotation (Table 11). The mean yield expected for these fields was 135.36 bushels per acre
of corn (SD = 56.07; range: 0-200 bushels, n = 350), 62.79 bushels per acre of soybean (SD =
48.45; range: 0-336 bushels; n = 398), and a mean expected yield of 0.64 tons per acre of
forage (SD = 10.27; range: 0-300 tons; n = 930). The mean fair market rent for this field was
$126.94 per acre (SD = 89.18), with a range from $0 to $800. Over half of respondents (70.1%)
said their low productivity field had drainage tile installed, with the majority of these drainage
tiles being 30 feet spacing or greater (Table 12). Nearly half of respondents reported that their
low productivity field had a slope of 0-2% (Table 13), with dominant soil textures of clay and clay
loam (31.8% and 40.3%, respectively; Table 14). The majority of respondents (76.2%) reported
that their field was not classified by the USDA NRCS as highly erodible land (Table 15). The
mean distance of these fields from the nearest city of more than 10,000 people was 13.15 miles
(SD = 12.05), with a range of 0 to 100 miles.
Respondents were asked whether they were enrolled in any of four programs that provide
incentives for conservation practices (Figure 7). The programs included the Environmental
Quality Incentive Program (EQIP), Conservation Reserve Program (CRP), Conservation
Reserve Enhancement Program (CREP), and Conservation Security Program (CSP). Over a
third of respondents (37.4%) said that their field was enrolled in a conservation program, with
the most participation in CRP. Over half of respondents (65%) have their low productivity field

20
covered by a Federal Crop Insurance program. Of the 42.6% of respondents who do rent their
low productivity field, the majority of these farmers are the primary decision-makers regarding
nutrient management (Table 16). Of those who do rent their fields (359 respondents), 260
respondents reported that they rent for cash, whereas 106 respondents reported renting for a
share of crop.
Table 11. Established rotations on low productivity fields.
Established Rotation (n = 759)
Valid %
Frequency
Corn/beans
49.4
375
Corn/beans/wheat
41.6
316
Other with forage
2.2
17
Other
6.7
51
Table 12. Types of drainage used on low productivity fields.
Drainage Type (n = 936)
Frequency
<30ft tiles
60
30-49ft tiles
232
>50ft tiles
240
Drainage water management system
50
Table 13. Slopes of low productivity fields
Slope of Field (n = 846)
Valid %
Frequency
0-2%
43.7
370
2-5%
27.4
232
5-10%
8.7
74
>10%
2.2
19
I'm not sure
17.8
151
Table 14. Soil types of low productivity fields.
Soil Type (n = 827)
Valid %
Frequency
Clay
31.8
263
Clay loam
40.3
333
Silty loam
11.4
94
Sand
4.5
37
Sandy loam
12.1
100

21
Table 15. Classification of low productivity fields as highly erodible.
Highly Erodible Fields (n = 837)
Valid %
Frequency
No
76.2
638
Yes
11.4
95
I'm not sure
12.3
103
Figure 7. Conservation program enrollment for low productivity fields (n = 936).
Table. 16. Primary nutrient management decision-maker for those who rent their low
productivity fields.
Primary Decision-Maker (n = 350)
Valid %
Frequency
Me alone
82.9
290
Primarily me, with landlord input
10.3
36
Equally me and my landlord
2.0
7
Primarily my landlord, with my input
0.3
1
My landlord alone
0.6
2
Other
4.0
14
Recent management activities. Respondents were asked a series of questions about their
particular field that pertained to management activities within the last growing season, including
activities related to nutrient management such as tillage, soil testing, and fertilizer placement.
Nearly half of respondents (40.5%) do not till their low productivity field (Table 17). The most
popular crop grown in both 2012 and 2013 was soybean (47.1% and 48.1%, respectively),
0
10
20
30
40
50
60
70
80
90
100
EQIP CREP CRP CSP
Valid Percentages
Conservation Program Enrollment
Yes
No

22
followed by corn (42.7% and 42.2%, respectively; Tables 18 and 19). After their most recent
crop, only 16.6% of respondents planted cover crops on their low productivity field. Most
respondents (92.5%) said they use soil testing on this field to inform their nutrient management
decisions, and over half of these respondents test their field every three years (Table 20). Over
half of the respondents who test their soil (62.4%) had soil test results between 15 and 30 ppm
of phosphorus from their last soil test (Table 21).
The most popular times for applying fertilizer or manure on low productivity fields was in the fall,
spring pre-planting, and spring at planting, with the most popular method of application being
surface broadcast application (Table 22). A small proportion of respondents (12.5%) applied
manure on their low productivity field. Of the 106 respondents who applied manure, nearly half
applied manure sourced from dairy (Table 23). The quantity of manure applied on this field
varied greatly, with a median of 11,807 lbs per acre (SD = 30,557.62) and a range of 2 to
200,000 lbs per acre. About half of respondents (52.1%) have their phosphorus custom applied,
with the two most popular reasons being time management and to take advantage of variable
rate technology (Figure 8). Half of respondents (50.2%) applied a one-year application of
phosphorus on their most recent crop (Table 24). The mean rate of phosphorus applied on this
field was 42.81lbs per acre (SD = 159.48), with a mean price of $144.04 per ton. The mean rate
of nitrogen applied on this field was 55.77lbs per acre of nitrogen (SD = 97.97), with a mean
price of $90.86 per ton. The most popular form of phosphorus and nitrogen were MAP
(monoammonium phosphate) and UAN (urea-ammonium nitrate), respectively (Tables 25 and
26). The greatest number of passes across this low productivity field was 15 (Table 27), which
was during planting. The mean horsepower of the respondent’s largest tractor was 183.07 (SD
= 235.63), and 211.69 (SD = 722.16) for the respondent’s combine harvester. The mean
number of rows in the respondent’s planter was 27.51 (SD = 527.88). For this low productivity
field, nearly a third of respondents spent $30 per acre on herbicide, insecticide, and fungicide in
2013 (Table 28).

23
Table 17. Types of tillage for low productivity fields.
Type of Tillage (n = 834)
Valid %
Frequency
Conventional
21.3
178
Conservation
38.1
318
No-Till
40.5
338
Table 18. Crops grown on low productivity fields in 2012.
2012 Crop (n = 831)
Valid %
Frequency
Corn
42.7
355
Soybeans
47.1
391
Wheat
7.9
66
Other cash crop
2.3
19
Table 19. Crops grown on low productivity fields in 2013.
2013 Crop (n = 829)
Valid %
Frequency
Corn
42.2
350
Soybeans
48.1
399
Wheat
8.4
70
Other cash crop
1.2
10
Table 20. Frequency of soil testing on low productivity fields.
Soil Test Frequency (n = 777)
Valid %
Frequency
Every 2 years
24.2
188
Every 3 years
53.8
418
Every 4 years
13.1
102
Every 5 years
6.0
47
Other
2.8
22
Table 21. Soil test phosphorus results during last soil test on low productivity fields.
Soil Test Phosphorus (n = 542)
Valid %
Frequency
<15 ppm
24.5
133
15-30 ppm
62.4
338
>30 ppm
13.1
71

24
Table 22. Timing and method of nutrient application for respondents who applied it on their low
productivity field.
Nutrient Application
Timing (n = 1012)
Frequency
Fall
277
Winter
17
Spring pre-planting
238
Spring at planting
240
After planting
101
No fertilizer applied
139
Method (n = 964)
Incorporation with tillage
290
Band placement (planter, tool bar, strip
till)
188
Surface applied broadcast
347
No fertilizer applied
139
Table 23. Sources of manure for respondents who applied it on their low productivity field.
Manure Source (n = 106)
Frequency
Dairy
45
Swine
33
Poultry
21
Figure 8. Reasons for custom applying phosphorus on low productivity fields (n = 509).
0
50
100
150
200
250
Time
management Shifting risk Variable rate
technology Other
Frequency
Reasons for Custom Application of Phosphorus

25
Table 24. Most recent phosphorus application on low productivity fields.
Phosphorus Application (n = 839)
Valid %
Frequency
No application
21.8
183
1 year application
50.2
421
2 year application
23.2
195
3 year application
4.2
35
4 year application
0.6
5
Table 25. Phosphorus forms used for application on low productivity fields.
Phosphorus Forms Applied (n = 274)
Valid %
Frequency
MAP (monoammonium phosphate)
58.4
160
DAP (diammonium phosphate)
36.5
100
APP (ammonium polyphosphate)
5.1
14
Table 26. Nitrogen forms used for application on low productivity fields.
Nitrogen Forms Applied (n = 265)
Valid %
Frequency
Urea (carbamide)
26.4
70
UAN (urea-ammonium nitrate)
47.9
127
NH3 (ammonia)
25.7
68
Table 27. Mean number of passes on low productivity fields for management activities.
Passes Across Field
Passes for…
Mean
SD
n
Range
Preparation
0.92
1.01
934
0-10
Planting
0.90
0.62
934
0-15
Fertilizer application
0.78
0.72
934
0-8
Spraying
1.33
0.84
935
0-8

26
Table 28. Amount spent on herbicide, insecticide, and fungicide on low productivity fields in
2013.
Costs of Herbicide, Insecticide, and
Fungicide
Costs/acre
Valid %
Frequency
$10
3.6
25
$15
5.9
41
$20
14.3
100
$30
30.6
214
$40
17.9
125
$50
15.9
111
$60
7.4
52
$80
4.6
32
Efficacy and adoption of BMPs. Efficacy, or the ability to mitigate nutrient loss by taking
action, is an important aspect to consider when understanding farmer adoption of BMPs. If a
farmer believes that he or she has a problem with nutrient runoff, and believes that adopting
certain practices would curtail this runoff, the farmer may be more likely to adopt those
practices. Efficacy was measured for ten practices recommended by experts for reducing
nutrient loss (Ohio EPA, 2010), such as planting cover crops after a fall harvest and placing
fertilizer at least 2 to 3 inches below the soil surface (Table 29). Respondents were asked to
indicate the extent to which each practice reduced the amount of soluble phosphorus lost from
the farm field through surface runoff and/or subsurface drainage, with possible responses
ranging from 0 (not at all) to 4 (to a great extent). For each of the ten practices, the majority of
respondents indicated that the practice at least somewhat reduced the amount of soluble
phosphorus. Just over a third of respondents (34.4%) indicated that avoiding winter or frozen
ground surface application of phosphorus reduces the amount of phosphorus to a great extent.
Avoiding fall application of phosphorus had the lowest perceived efficacy, with 18.7% of
respondents believing it reduced phosphorus only a little, and 8.6% of respondents believing
that the practice didn’t reduce phosphorus at all. Interestingly, nearly half of respondents
(46.5%) indicated that requiring a 4R certification program for private applicators reduced
phosphorus runoff a good deal, while only 15.7% said it reduced phosphorus to a great extent.
Current adoption of BMPs was measured by asking respondents whether they had already
adopted each of the BMPs on their low productivity field (Table 30). Determining fertilizer rates
based on regular soil testing once within the rotation (or every three years) was the most

27
adopted practice (51.8% adoption), followed by avoiding winter or frozen ground surface
application of phosphorus. Respondents had the lowest adoption for hiring a 4R certified
applicator over an applicator without certification as well as managing field water levels with
drainage management systems.
To gain an understanding of farmer’s intentions to adopt these practices in the future,
respondents were asked to rate their willingness to adopt any practice for which they had not
yet adopted, with possible responses from 0 (will never adopt) to 4 (will definitely adopt). Similar
to the previous trends for practices that were already adopted, the highest willingness to adopt
was for avoiding winter or frozen ground surface application of phosphorus, followed by
determining fertilizer rates based on soil testing (Table 31). Respondents showed the least
amount of willingness to adopt hiring a 4R certified applicator over an applicator without
certification. To specify our understanding of farmer’s actual intentions, respondents were asked
to indicate whether they planned on using four different practices on their low productivity field in
the next growing season (Table 32). Out of the 936 total respondents, only 236 respondents
said they do not plan on using any of the four practices. The most popular practice planned for
next season was grid soil sampling with variable rate technology, which was followed by
restricting nutrient application so that phosphorus levels do not exceed 30 ppm. Interestingly,
many of those who planned on adopting a practice next season (22-44%) were new adopters,
meaning they had not yet adopted those practices on their field.

28
Table 29. Respondent perception of efficacy of ten BMPs, in valid percentages.
The following practices reduce
the amount of soluble
phosphorus…
n
Not at
all
A little
Somewhat
A good
deal
To a
great
extent
Grid soil sampling with variable rate
application
905
5.0
16.2
33.7
33.8
11.3
Planting cover crops after fall
harvest
914
5.4
10.6
25.8
41.2
17.0
Delaying broadcasting when the
forecast predicts a 50% or more
chance of at least 1 inch of total
rainfall in the next 12 hours
918
4.0
11.8
24.4
31.2
18.6
Managing field water levels with
drainage management systems
902
6.4
17.3
34.6
33.8
7.9
Avoiding winter or frozen ground
surface application of phosphorus
918
2.9
6.6
17.2
38.8
34.4
Avoiding fall application of
phosphorus
910
8.6
18.7
33.8
25.1
13.8
Determining rates based on regular
soil testing once within the rotation
(or every 3 years)
915
2.4
7.0
25.1
45.9
19.6
Placement of fertilizer at least 2-3
inches below the soil surface
913
3.5
10.5
26.4
43.9
15.7
Following soil test trends to
maintain the agronomic range for
phosphorus in the soil (15 to 30
ppm)
913
3.1
7.2
27.5
46.5
15.7
Requiring a 4R certification program
for private applicators
877
28.2
24.9
28.2
15.3
3.5

29
Table 30. Respondent adoption of the ten BMPs on their low productivity fields, in valid
percentages (n = 936).
BMPs
Adopted
Not
adopted
Grid soil sampling with variable rate application
34.9
65.1
Planting cover crops after fall harvest
16.8
83.2
Delaying broadcasting when the forecast predicts a 50% or
more chance of at least 1 inch of total rainfall in the next 12
hours
36.0
64.0
Managing field water levels with drainage management
systems
14.6
85.4
Avoiding winter or frozen ground surface application of
phosphorus
48.5
51.5
Avoiding fall application of phosphorus
30.2
69.8
Determining rates based on regular soil testing once within
the rotation (or every 3 years)
51.8
48.2
Placement of fertilizer at least 2-3 inches below the soil
surface
33.1
66.9
Following soil test trends to maintain the agronomic range
for phosphorus in the soil (15 to 30 ppm)
39.9
60.1
Hiring a 4R certified applicator, which adheres to
recommended management practices, over an applicator
without certification
14.5
85.5

30
Table 31. Respondent intentions to adopt those practices that they have not yet adopted on
their low productivity fields, in valid percentages.
BMPs
n
Will never
adopt
Unlikely
to adopt
Likely to
adopt
Will
definitely
adopt
Grid soil sampling with variable rate
application
497
10.5
45.3
39.8
4.4
Planting cover crops after fall harvest
671
7.0
41.4
44.3
7.3
Delaying broadcasting when the forecast
predicts a 50% or more chance of at least
1 inch of total rainfall in the next 12 hours
496
3.2
18.5
57.3
21.0
Managing field water levels with drainage
management systems
676
13.0
46.7
29.7
20.5
Avoiding winter or frozen ground surface
application of phosphorus
376
8.2
13.0
39.4
39.4
Avoiding fall application of phosphorus
539
6.7
36.5
38.4
18.4
Determining rates based on regular soil
testing once within the rotation (or every 3
years)
339
4.7
12.1
54.0
29.2
Placement of fertilizer at least 2-3 inches
below the soil surface
511
4.3
31.1
48.5
16.0
Following soil test trends to maintain the
agronomic range for phosphorus in the soil
(15 to 30 ppm)
450
3.1
12.0
58.4
26.4
Hiring a 4R certified applicator, which
adheres to recommended management
practices, over an applicator without
certification
667
15.1
39.9
33.9
11.1
Table 32. Respondent plans for adopting four practices on their low productivity field next
season.
Plans for Next Season (n = 936)
Frequency
% New*
Grid soil sampling with variable rate technology
311
21.9
Planting winter cover crops
198
43.9
Restricting nutrient application so that P levels do not exceed 15 ppm
132
42.4
Restricting nutrient application so that P levels do not exceed 30 ppm
207
40.6
I do not plan on using any of these practices next season
236
-
*Percent of respondents who had not yet adopted that practice but planned on adopting it next season

31
Behavior on Average Productivity Fields
Basic field characteristics. The average size of the average productivity fields was 43.98
acres (SD = 51.10), with a range from 0 to 730 acres (n = 908). The majority of respondents
(96.6%) reported that these fields were in an established rotation, with nearly half being in a
corn and bean rotation (Table 33). The mean yield expected for these fields was 150.27 bushels
per acre of corn (SD = 48.52; range: 0-220 bushels; n = 391), 66.80 bushels per acre of
soybeans (SD = 48.10; range: 0-220 bushels; n = 368), and a mean expected yield of 0.39 tons
per acre of forage (SD = 3.26; range: 0-60 tons; n = 904). The mean fair market rent for this field
was $137.38 per acre (SD = 91.75), with a range from $0 to $800. The majority of respondents
(84.3%) said their average productivity field had drainage tile installed, with the majority of these
drainage tiles being 30 feet spacing or greater (Table 34). Nearly half of respondents reported
that their average productivity field had a slope of 0-2% (Table 35), with dominant soil textures
of clay and clay loam (23.1% and 47.8%, respectively; Table 36). The majority of respondents
(80.4%) reported that their field was not classified by the USDA NRCS as highly erodible land
(Table 37). The mean distance of these fields from the nearest city of more than 10,000 people
was 13.02 miles (SD = 12.53), with a range of 0 to 200 miles.
Respondents were asked whether they were enrolled in any of four programs that provide
incentives for conservation practices (Figure 9). The programs included the Environmental
Quality Incentive Program (EQIP), Conservation Reserve Program (CRP), Conservation
Reserve Enhancement Program (CREP), and Conservation Security Program (CSP). Over a
third of respondents (39.4%) said that their field was enrolled in a conservation program, with
the most participation being in CRP. Over half of respondents (66%) have their average
productivity field covered by a Federal Crop Insurance program. Of the 32.3% of respondents
who do rent their average productivity field, the majority of these farmers are the primary
decision-makers regarding nutrient management (Table 38). Of those who do rent their fields
(266 respondents), 201 respondents reported that they rent for cash, whereas 78 respondents
reported renting for a share of crop.

32
Table 33. Established rotations on average productivity fields.
Established Rotation (n = 746)
Valid %
Frequency
Corn/beans
48.4
361
Corn/beans/wheat
45.4
339
Other with forage
2.3
17
Other
3.9
29
Table 34. Types of drainage used on average productivity fields.
Drainage Type (n = 906)
Frequency
<30ft tiles
78
30-49ft tiles
292
>50ft tiles
277
Drainage water management system
57
Table 35. Slopes of average productivity fields.
Slope of Field (n = 811)
Valid %
Frequency
0-2%
45.4
368
2-5%
29.2
237
5-10%
7.2
58
>10%
0.7
6
I'm not sure
17.5
142
Table 36. Soil types of average productivity fields.
Soil Type (n = 802)
Valid %
Frequency
Clay
23.1
185
Clay loam
47.8
383
Silty loam
16.1
129
Sand
1.5
12
Sandy loam
11.6
93
Table 37. Classification of average productivity fields as highly erodible.
Highly Erodible Fields (n = 810)
Valid %
Frequency
No
80.4
651
Yes
10.9
88
I'm not sure
8.6
70

33
Figure 9. Conservation program enrollment for average productivity fields (n = 906).
Table 38. Primary nutrient management decision-maker for those who rent their average
productivity field.
Primary Decision-Maker (n = 264)
Valid %
Frequency
Me alone
85.6
226
Primarily me, with landlord input
8.7
23
Equally me and my landlord
3.0
8
Primarily my landlord, with my input
1.5
4
My landlord alone
0.0
0
Other
1.1
3
Recent management activities. Respondents were asked a series of questions about their
particular average productivity field that pertained to management activities within the last
growing season, including activities related to nutrient management such as tillage, soil testing,
and fertilizer placement. A similar proportion of respondents use no-till or conservation tillage
practices on their average productivity field (35.4% and 37.8%, respectively; Table 39). Nearly
half of the respondents grew soybeans in 2012 (45.2%) and 2013 (45.7%), and a similar
proportion of respondents grew corn during these two years (40.4% and 48.6%, respectively;
Tables 40 and 41). After their most recent crop, only 17.6% of respondents planted cover crops
on their average productivity field. Most respondents (93.6%) said they use soil testing on this
field to inform their nutrient management decisions, and over half of these respondents test their
0
20
40
60
80
100
EQIP CREP CRP CSP
Valid Percentages
Conservation Program Enrollment
Yes
No

34
field every three years (Table 42). Over half of the respondents who test their soil (64%) had soil
test results between 15 and 30 ppm of phosphorus from their last soil test (Table 43).
The most popular times for applying fertilizer or manure on average productivity fields was in the
fall, spring pre-planting, and spring at planting, with the most popular method of application
being surface broadcast application (Table 44). A small proportion of respondents (16.5%)
applied manure on their most recent crop of their average productivity field. Of the 134
respondents who applied manure, over half used manure sourced from dairy (Table 45). The
quantity of manure applied on this field varied greatly, with a median of 4,000lbs per acre (SD =
7962.86) and a range of 3 to 50,000lbs per acre. The mean price paid for this manure was
$15.22 per pound (SD = 22.03), with a range of less than $1.00 per pound to $62 per pound.
About half of respondents (48.8%) have their phosphorus custom applied, with the two most
popular reasons being time management and to take advantage of variable rate technology
(Figure 10). Half of respondents (51.1%) applied a one-year application of phosphorus on their
most recent crop (Table 46). The mean rate of phosphorus applied on this field was 37.46lbs
per acre (SD = 68.17), with a mean price of $98.74 per ton. The mean rate of nitrogen applied
on this field was 62.74lbs per acre of nitrogen (SD = 97.97), with a mean price of $114.02 per
ton. The most popular form of phosphorus and nitrogen were MAP (monoammonium
phosphate) and UAN (urea-ammonium nitrate), respectively (Tables 47 and 48). The greatest
number of passes with equipment across this average productivity field was 50 (Table 49),
which was during planting. The mean horsepower of the respondent’s largest tractor was
173.15 (SD = 108.51), and 215.40 (SD = 747.38) for the respondent’s combine harvester. The
mean number of rows in the respondent’s planter was 8.96 (SD = 7.73). For this low productivity
field, nearly a third of respondents spent $30 per acre on herbicide, insecticide, and fungicide in
2013 (Table 50).
Table 39. Types of tillage for average productivity fields.
Type of Tillage (n = 802)
Valid %
Frequency
Conventional
26.8
215
Conservation
37.8
303
No-Till
35.4
284

35
Table 40. Crops grown on average productivity fields in 2012.
2012 Crop (n = 807)
Valid %
Frequency
Corn
40.4
326
Soybeans
45.2
365
Wheat
12.6
102
Other cash crop
1.7
14
Table 41. Crops grown on average productivity fields in 2013.
2013 Crop (n = 807)
Valid %
Frequency
Corn
48.6
392
Soybeans
45.7
369
Wheat
4.3
35
Other cash crop
1.4
11
Table 42. Frequency of soil testing on average productivity fields.
Soil Test Frequency (n = 768)
Valid %
Frequency
Every 2 years
23.3
179
Every 3 years
55.6
427
Every 4 years
13.5
104
Every 5 years
5.3
41
Other
2.2
17
Table 43. Soil test phosphorus results during last soil test on average productivity fields.
Soil Test Phosphorus (n = 528)
Valid %
Frequency
<15 ppm
23.7
125
15-30 ppm
64.0
338
>30 ppm
12.3
65

36
Table 44. Timing and method of nutrient application for respondents who applied it on their
average productivity field.
Nutrient Application
Timing (n = 906)
Frequency
Fall
275
Winter
39
Spring pre-planting
262
Spring at planting
253
After planting
113
No fertilizer applied
114
Method (n = 906)
Incorporation with tillage
284
Band placement (planter, tool bar, strip till)
215
Surface applied broadcast
358
No fertilizer applied
111
Table 45. Sources of manure for respondents who applied it on their average productivity field.
Manure Source (n = 134)
Frequency
Dairy
62
Swine
31
Poultry
27
Figure 10. Reasons for custom applying phosphorus on average productivity fields (n = 386).
0
50
100
150
200
250
Time
management Shifting risk Variable rate
technology Other
Reasons for Custom Applying Phosphorus

37
Table 46. Most recent phosphorus application on average productivity fields.
Phosphorus Application (n = 801)
Valid %
Frequency
No application
20.0
160
1 year application
51.1
409
2 year application
24.2
194
3 year application
3.9
31
4 year application
0.9
7
Table 47. Phosphorus forms used for application on average productivity fields.
Phosphorus Forms Applied (n = 283)
Valid %
Frequency
MAP (monoammonium phosphate)
50.2
142
DAP (diammonium phosphate)
42.8
121
APP (ammonium polyphosphate)
7.1
20
Table 48. Nitrogen forms used for application on average productivity fields.
Nitrogen Forms Applied (n = 301)
Valid %
Frequency
Urea (carbamide)
25.6
77
UAN (urea-ammonium nitrate)
39.5
119
NH3 (ammonia)
34.9
105
Table 49. Mean number of passes on average productivity fields for management activities.
Passes Across Field
Passes for…
Mean
SD
n
Range
Preparation
1.11
3.47
906
0-10
Planting
0.98
2.17
905
0-50
Fertilizer application
0.88
1.13
906
0-20
Spraying
1.36
1.09
905
0-16

38
Table 50. Amount spent on herbicide, insecticide, and fungicide on average productivity fields in
2013.
Costs of Herbicide, Insecticide, and
Fungicide
Costs/acre
Valid %
Frequency
$10
2.6
18
$15
6.0
42
$20
14.8
103
$30
28.6
199
$40
22.7
158
$50
12.8
89
$60
9.2
64
$80
3.2
22
Efficacy and adoption of BMPs. Efficacy, or the ability to mitigate nutrient loss by taking
action, is an important aspect to consider when understanding farmer adoption of BMPs. If a
farmer believes that he or she has a problem with nutrient runoff, and believes that adopting
certain practices would curtail this runoff, the farmer may be more likely to adopt those
practices. Efficacy was measured for ten practices recommended by experts for reducing
nutrient loss (Ohio EPA, 2010), such as planting cover crops after a fall harvest and placing
fertilizer at least 2 to 3 inches below the soil surface (Table 51). Respondents were asked to
indicate the extent to which each practice reduced the amount of soluble phosphorus lost from
the farm field through surface runoff and/or subsurface drainage, with possible responses
ranging from 0 (not at all) to 4 (to a great extent). For nine of the practices, the majority of
respondents (>70%) indicated that the practice at least somewhat reduced the amount of
soluble phosphorus, however only 47.9% indicated this level of efficacy for the tenth practice,
which was requiring a 4R certification program for private applicators. Just over a third of
respondents (35.2%) indicated that avoiding winter or frozen ground surface application of
phosphorus reduces the amount of phosphorus to a great extent. Requiring a 4R certification
program for private applicators had the lowest perceived efficacy, with nearly a quarter of
respondents (22.7%) believing it reduced phosphorus only a little and nearly a third of
respondents (29.4%) believing that the practice didn’t reduce phosphorus at all.
Current adoption of BMPs was measured by asking respondents whether they had already
adopted each of the BMPs on their average productivity field (Table 52). Determining fertilizer
rates based on regular soil testing once within the rotation (or every three years) was the most

39
adopted practice (52.3% adoption), followed by avoiding winter or frozen ground surface
application of phosphorus. Respondents had the lowest adoption for hiring a 4R certified
applicator over an applicator without certification (12.9% adoption) as well as planting cover
crops.
To gain an understanding of farmer’s intentions to adopt these practices in the future,
respondents were asked to rate their willingness to adopt any practice for which they had not
yet adopted, with possible responses from 0 (will never adopt) to 4 (will definitely adopt). Similar
to the previous trends for practices that were already adopted, the highest willingness to adopt
was for avoiding winter or frozen ground surface application of phosphorus (40.9% will definitely
adopt; Table 53). Respondents showed the least amount of willingness to adopt hiring a 4R
certified applicator over an applicator without certification. To specify our understanding of
farmer’s actual intentions, respondents were asked to indicate whether they planned on using
four different practices on their average productivity field in the next growing season (Table 54).
Out of the 906 total respondents, only 229 respondents said they do not plan on using any of
the four practices. The most popular practice planned for next season was grid soil sampling
with variable rate technology, which was followed by restricting nutrient application so that
phosphorus levels do not exceed 30 ppm. Interestingly, many of those who planned on adopting
a practice next season (18-39%) were new adopters, meaning they had not yet adopted those
practices on their field.

40
Table 51. Respondent perception of efficacy of ten BMPs, in valid percentages.
The following practices reduce the
amount of soluble phosphorus…
n
Not at
all
A little
Somewhat
A good
deal
To a
great
extent
Grid soil sampling with variable rate
application
882
4.6
15.1
33.0
37.2
10.1
Planting cover crops after fall harvest
885
4.0
12.2
28.4
39.7
15.8
Delaying broadcasting when the
forecast predicts a 50% or more
chance of at least 1 inch of total
rainfall in the next 12 hours
889
3.1
8.9
26.0
43.8
18.2
Managing field water levels with
drainage management systems
888
5.4
15.7
36.0
34.1
8.8
Avoiding winter or frozen ground
surface application of phosphorus
887
2.4
6.7
14.5
41.3
35.2
Avoiding fall application of
phosphorus
887
7.9
18.0
28.4
30.8
14.9
Determining rates based on regular
soil testing once within the rotation
(or every 3 years)
889
1.3
8.8
21.7
46.3
21.8
Placement of fertilizer at least 2-3
inches below the soil surface
888
3.0
9.7
29.7
41.9
15.7
Following soil test trends to maintain
the agronomic range for phosphorus
in the soil (15 to 30 ppm)
884
1.8
7.9
28.2
46.5
15.6
Requiring a 4R certification program
for private applicators
863
29.4
22.7
29.2
13.6
5.1

41
Table 52. Respondent adoption of the ten BMPs on their average productivity fields, in valid
percentages (n = 906).
BMPs
Adopted
Not
adopted
Grid soil sampling with variable rate application
34.4
65.6
Planting cover crops after fall harvest
16.9
83.1
Delaying broadcasting when the forecast predicts a 50% or
more chance of at least 1 inch of total rainfall in the next 12
hours
31.7
68.3
Managing field water levels with drainage management
systems
17.8
82.2
Avoiding winter or frozen ground surface application of
phosphorus
47.1
52.9
Avoiding fall application of phosphorus
29.6
70.4
Determining rates based on regular soil testing once within
the rotation (or every 3 years)
52.3
47.7
Placement of fertilizer at least 2-3 inches below the soil
surface
34.1
65.9
Following soil test trends to maintain the agronomic range
for phosphorus in the soil (15 to 30 ppm)
40.7
59.3
Hiring a 4R certified applicator, which adheres to
recommended management practices, over an applicator
without certification
12.9
87.1

42
Table 53. Respondent intentions to adopt those practices that they have not yet adopted on
their average productivity fields, in valid percentages.
BMPs
n
Will never
adopt
Unlikely
to adopt
Likely to
adopt
Will
definitely
adopt
Grid soil sampling with variable rate
application
490
4.7
43.9
44.1
7.3
Planting cover crops after fall harvest
640
5.6
46.4
42.7
5.3
Delaying broadcasting when the forecast
predicts a 50% or more chance of at least
1 inch of total rainfall in the next 12 hours
512
3.5
21.7
53.9
20.9
Managing field water levels with drainage
management systems
635
10.4
47.9
34.2
7.6
Avoiding winter or frozen ground surface
application of phosphorus
374
5.6
13.1
40.4
40.9
Avoiding fall application of phosphorus
531
7.7
37.7
33.9
20.7
Determining rates based on regular soil
testing once within the rotation (or every 3
years)
330
2.4
15.5
54.8
27.3
Placement of fertilizer at least 2-3 inches
below the soil surface
484
4.1
32.9
48.6
14.5
Following soil test trends to maintain the
agronomic range for phosphorus in the soil
(15 to 30 ppm)
424
3.1
13.0
56.1
27.8
Hiring a 4R certified applicator, which
adheres to recommended management
practices, over an applicator without
certification
667
13.5
40.9
34.2
11.4
Table 54. Respondent plans for adopting four practices on their average productivity field next
season.
Plans for Next Season (n = 906)
Frequency
% New
Grid soil sampling with variable rate technology
297
18.6
Planting winter cover crops
189
39.4
Restricting nutrient application so that P levels do not exceed 15 ppm
115
39.4
Restricting nutrient application so that P levels do not exceed 30 ppm
200
34.8
I do not plan on using any of these practices next season
229
-
*Percent of respondents who had not yet adopted that practice but planned on adopting it next season

43
Behavior on High Productivity Fields
Basic field characteristics. The average size of the high productivity fields was 52.27 acres
(SD = 62.63), with a range from 0 to 1,000 (n = 924). The majority of respondents (95%)
reported that these fields were in an established rotation, with over half being in corn and beans
(Table 55). The mean yield expected for these fields was 161.08 bushels per acre of corn (SD =
51.47; range: 0-250 bushels; n = 533), 83.73 bushels per acre of soybeans (SD = 58.16; range:
0-220 bushels; n = 290), and a mean expected yield of 0.37 tons per acre of forage (SD = 3.39;
range: 0-50 tons; n = 913). The mean fair market rent for this field was $161.73 per acre (SD =
99.35), with a range from $0 to $700. The majority of respondents (88.2%) said their high
productivity field had drainage tile installed, with the majority of these drainage tiles being 30
feet spacing or greater (Table 56). About half of respondents reported that their high productivity
field had a slope of 0-2% (Table 57), with dominant soil textures of clay loam and silty loam
(45.3% and 20.7%, respectively; Table 58). The majority of respondents (86%) reported that
their field was not classified by the USDA NRCS as highly erodible land (Table 59). The mean
distance of these fields from the nearest city of more than 10,000 people was 12.97 miles (SD =
12.05), with a range of 0 to 150 miles.
Respondents were asked whether they were enrolled in any of four programs that provide
incentives for conservation practices (Figure 11). The programs included the Environmental
Quality Incentive Program (EQIP), Conservation Reserve Program (CRP), Conservation
Reserve Enhancement Program (CREP), and Conservation Security Program (CSP). About a
third of respondents (35.5%) said that their field was enrolled in a conservation program, with
the most participation being in CRP. Over half of respondents (67%) have their high productivity
field covered by a Federal Crop Insurance program. Of the 29% of respondents who do rent
their high productivity field, the majority of these farmers are the primary decision-makers
regarding nutrient management (Table 60). Of those who do rent their fields (252 respondents),
188 respondents reported that they rent for cash, whereas 86 respondents reported renting for a
share of crop.

44
Table 55. Established rotations on high productivity fields.
Established Rotation (n = 786)
Valid %
Frequency
Corn/beans
56.1
441
Corn/beans/wheat
38.3
301
Other with forage
1.7
13
Other
3.9
31
Table 56. Types of drainage used on high productivity fields.
Drainage Type (n = 906)
Frequency
<30ft tiles
84
30-49ft tiles
364
>50ft tiles
268
Drainage water management system
39
Table 57. Slopes of high productivity fields.
Slope of Field (n = 860)
Valid %
Frequency
0-2%
50.1
431
2-5%
26.0
224
5-10%
5.3
46
>10%
1.6
14
I'm not sure
16.9
145
Table 58. Soil types of high productivity fields.
Soil Type (n = 841)
Valid %
Frequency
Clay
17.0
143
Clay loam
45.3
381
Silty loam
20.7
174
Sand
1.3
11
Sandy loam
15.7
132
Table 59. Classification of high productivity fields as highly erodible.
Highly Erodible Fields (n = 863)
Valid %
Frequency
No
86.0
742
Yes
7.3
63
I'm not sure
6.7
58

45
Figure 11. Conservation program enrollment for high productivity fields (n = 922).
Table 60. Primary nutrient management decision-maker for those who rent their high
productivity field.
Primary Decision-Maker (n = 247)
Valid %
Frequency
Me alone
77.3
191
Primarily me, with landlord input
13.8
34
Equally me and my landlord
4.0
10
Primarily my landlord, with my input
1.2
3
My landlord alone
0.4
1
Other
3.2
8
Recent management activities. Respondents were asked a series of questions about their
particular high productivity field that pertained to management activities within the last growing
season, including activities related to nutrient management such as tillage, soil testing, and
fertilizer placement. Over a third of respondents (39.2%) use conservation tillage practices on
their high productivity field (Table 61). Nearly half of the respondents grew soybeans in 2012
(49.8%) and over half grew corn in 2013 (61.5%; Tables 62 and 63). After their most recent
crop, only 11.9% of respondents planted cover crops on their high productivity field. Most
respondents (93.2%) said they use soil testing on this field to inform their nutrient management
decisions, and over half of these respondents test their field every three years (Table 64). Over
half of the respondents who test their soil (64.7%) had soil test results between 15 and 30 ppm
of phosphorus from their last soil test (Table 65).
0
20
40
60
80
100
EQIP CREP CRP CSP
Valid Percentages
Conservation Program Enrollment
Yes
No

46
The most popular times for applying fertilizer or manure on high productivity fields was in the
fall, spring pre-planting, and spring at planting, with the most popular method of application
being incorporated with tillage (Table 66). A small proportion of respondents (19.4%) applied
manure on their most recent crop of their high productivity field. Of the 169 respondents who
applied manure, over half used manure sourced from dairy (Table 67). The quantity of manure
applied on this field varied greatly, with a median of 4,000lbs per acre (SD = 12,244.11) and a
range of 2 to 80,000lbs per acre. The mean price paid for this manure was $37.36 per pound
(SD = 54.85), with a range of less than $1.00 per pound to $200 per pound. The mean price
paid for this manure was $11.04 per pound (SD = 26.56), with a range of less than $1.00 per
pound to $125 per pound. About half of respondents (52.1%) have their phosphorus custom
applied, with the two most popular reasons being time management and to take advantage of
variable rate technology (Figure 12). Half of respondents (51.8%) applied a one-year application
of phosphorus on their most recent crop (Table 68). The mean rate of phosphorus applied on
this field was 66.11lbs per acre (SD = 444.61), with a mean price of $123.28 per ton. The mean
rate of nitrogen applied on this field was 97.34lbs per acre of nitrogen (SD = 430.14), with a
mean price of $272.63 per ton. The most popular form of phosphorus and nitrogen were MAP
(monoammonium phosphate) and UAN (urea-ammonium nitrate), respectively (Tables 69 and
70). The greatest number of passes with equipment across this high productivity field was 200
(Table 71), which was for preparation. The mean horsepower of the respondent’s largest tractor
was 182.69 (SD = 104.78), and 203.11 horsepower (SD = 543.60) for the respondent’s combine
harvester. The mean number of rows in the respondent’s planter was 18.56 (SD = 270.23). For
this high productivity field, nearly a third of respondents spent $30 per acre on herbicide,
insecticide, and fungicide in 2013 (Table 72).
Table 61. Types of tillage for high productivity fields.
Type of Tillage (n = 865)
Valid %
Frequency
Conventional
31.3
271
Conservation
39.2
339
No-Till
29.5
255

47
Table 62. Crops grown on high productivity fields in 2012.
2012 Crop (n = 859)
Valid %
Frequency
Corn
35.4
304
Soybeans
49.8
428
Wheat
12.1
104
Other cash crop
2.7
23
Table 63. Crops grown on high productivity fields in 2013.
2013 Crop (n = 859)
Valid %
Frequency
Corn
61.5
528
Soybeans
33.8
290
Wheat
3.8
33
Other cash crop
0.9
8
Table 64. Frequency of soil testing on high productivity fields.
Soil Test Frequency (n = 809)
Valid %
Frequency
Every 2 years
25.0
202
Every 3 years
53.5
433
Every 4 years
12.6
102
Every 5 years
6.3
51
Other
2.6
21
Table 65. Soil test phosphorus results during last soil test on high productivity fields.
Soil Test Phosphorus (n = 580)
Valid %
Frequency
<15 ppm
17.9
104
15-30 ppm
64.7
375
>30 ppm
17.4
101

48
Table 66. Timing and application of nutrients on high productivity fields.
Nutrient Application
Timing (n = 922)
Frequency
Fall
331
Winter
28
Spring pre-planting
257
Spring at planting
316
After planting
143
No fertilizer applied
97
Method (n = 922)
Incorporation with tillage
358
Band placement (planter, tool bar, strip till)
258
Surface applied broadcast
348
No fertilizer applied
100
Table 67. Sources of manure for respondents who applied it on their high productivity field.
Manure Source (n = 169)
Frequency
Dairy
80
Swine
35
Poultry
35
Figure 12. Reasons for custom applying phosphorus on high productivity fields (n = 438).
0
50
100
150
200
250
300
Time
management Shifting risk Variable rate
technology Other
Reasons for Custom Applying Phosphorus

49
Table 68. Most recent phosphorus application on high productivity fields.
Phosphorus Application (n = 853)
Valid %
Frequency
No application
17.8
151
1 year application
51.8
442
2 year application
24.9
212
3 year application
4.1
35
4 year application
1.5
13
Table 69. Phosphorus forms used for application on high productivity fields.
Phosphorus Forms Applied (n = 306)
Valid %
Frequency
MAP (monoammonium phosphate)
52.3
160
DAP (diammonium phosphate)
43.1
132
APP (ammonium polyphosphate)
4.6
14
Table 70. Nitrogen forms used for application on high productivity fields.
Nitrogen Forms Applied (n = 365)
Valid %
Frequency
Urea (carbamide)
18.9
69
UAN (urea-ammonium nitrate)
42.7
156
NH3 (ammonia)
38.4
140
Table 71. Mean number of passes on high productivity fields for management activities.
Passes Across Field
Passes for…
Mean
SD
n
Range
Preparation
1.38
6.67
918
0-200
Planting
1.07
2.85
919
0-75
Fertilizer application
1.06
2.49
919
0-70
Spraying
1.38
1.24
920
0-25

50
Table 72. Amount spent on herbicide, insecticide, and fungicide on high productivity fields in
2013.
Costs of Herbicide, Insecticide, and
Fungicide
Costs/acre
Valid %
Frequency
$10
2.5
19
$15
6.3
47
$20
13.5
101
$30
27.9
209
$40
22.2
166
$50
13.1
98
$60
10.2
76
$80
4.3
32
Efficacy and adoption of BMPs. Efficacy, or the ability to mitigate nutrient loss by taking
action, is an important aspect to consider when understanding farmer adoption of BMPs. If a
farmer believes that he or she has a problem with nutrient runoff, and believes that adopting
certain practices would curtail this runoff, the farmer may be more likely to adopt those
practices. Efficacy was measured for ten practices recommended by experts for reducing
nutrient loss (Ohio EPA, 2010), such as planting cover crops after a fall harvest and placing
fertilizer at least 2 to 3 inches below the soil surface (Table 73). Respondents were asked to
indicate the extent to which each practice reduced the amount of soluble phosphorus lost from
the farm field through surface runoff and/or subsurface drainage, with possible responses
ranging from 0 (not at all) to 4 (to a great extent). For nine of the practices, the majority of
respondents (>70%) indicated that the practice at least somewhat reduced the amount of
soluble phosphorus, however only 49.1% indicated this same level of efficacy for the tenth
practice, which was requiring a 4R certification program for private applicators. Just over a third
of respondents (34.3%) indicated that avoiding winter or frozen ground surface application of
phosphorus reduces the amount of phosphorus to a great extent. Requiring a 4R certification
program for private applicators had the lowest perceived efficacy, with a quarter of respondents
(25.1%) believing it reduced phosphorus only a little and a similar amount (25.7%) believing that
the practice didn’t reduce phosphorus at all.
Current adoption of BMPs was measured by asking respondents whether they had already
adopted each of the BMPs on their high productivity field (Table 74). Determining fertilizer rates
based on regular soil testing once within the rotation (or every three years) was the most

51
adopted practice (56.6% adoption), followed by avoiding winter or frozen ground surface
application of phosphorus. Respondents had the lowest adoption for hiring a 4R certified
applicator over an applicator without certification (13.9% adoption) as well as planting cover
crops.
To gain an understanding of farmer’s intentions to adopt these practices in the future,
respondents were asked to rate their willingness to adopt any practice for which they had not
yet adopted, with possible responses from 0 (will never adopt) to 4 (will definitely adopt). Similar
to the previous trends for practices that were already adopted, the highest willingness to adopt
was for avoiding winter or frozen ground surface application of phosphorus (35.3% will definitely
adopt; Table 75). Respondents showed the least amount of willingness to adopt hiring a 4R
certified applicator over an applicator without certification. To specify our understanding of
farmer’s actual intentions, respondents were asked to indicate whether they planned on using
four different practices on their low productivity field in the next growing season (Table 76). Out
of the 922 total respondents, only 216 respondents said they do not plan on using any of the
four practices. The most popular practice planned for next season was grid soil sampling with
variable rate technology, which was followed by restricting nutrient application so that
phosphorus levels do not exceed 30 ppm. Interestingly, many of those who planned on adopting
a practice next season (20-50%) were new adopters, meaning they had not yet adopted those
practices on their field.

52
Table 73. Respondent perception of efficacy of ten BMPs on their high productivity fields, in
valid percentages.
The following practices reduce the
amount of soluble phosphorus…
n
Not at
all
A little
Somewhat
A good
deal
To a
great
extent
Grid soil sampling with variable rate
application
895
4.7
17.5
31.1
36.3
10.4
Planting cover crops after fall harvest
902
3.8
12.0
26.9
41.7
15.6
Delaying broadcasting when the
forecast predicts a 50% or more
chance of at least 1 inch of total
rainfall in the next 12 hours
899
2.3
8.7
26.4
42.8
19.8
Managing field water levels with
drainage management systems
893
5.3
16.7
36.6
34.5
6.9
Avoiding winter or frozen ground
surface application of phosphorus
900
2.3
7.6
16.1
39.7
34.3
Avoiding fall application of
phosphorus
897
6.7
19.5
32.1
29.5
12.2
Determining rates based on regular
soil testing once within the rotation
(or every 3 years)
899
2.2
6.8
23.1
45.1
22.8
Placement of fertilizer at least 2-3
inches below the soil surface
898
2.8
12.6
28.6
40.9
15.1
Following soil test trends to maintain
the agronomic range for phosphorus
in the soil (15 to 30 ppm)
892
1.6
8.4
26.1
48.0
15.9
Requiring a 4R certification program
for private applicators
867
25.7
25.1
28.8
16.0
4.3

53
Table 74. Respondent adoption of the ten BMPs on their high productivity fields, in valid
percentages (n = 922).
BMPs
Adopted
Not
adopted
Grid soil sampling with variable rate application
38.7
61.3
Planting cover crops after fall harvest
14.9
85.1
Delaying broadcasting when the forecast predicts a 50% or
more chance of at least 1 inch of total rainfall in the next 12
hours
37.9
62.1
Managing field water levels with drainage management
systems
22.0
78.0
Avoiding winter or frozen ground surface application of
phosphorus
50.9
49.1
Avoiding fall application of phosphorus
29.1
70.9
Determining rates based on regular soil testing once within
the rotation (or every 3 years)
56.6
43.4
Placement of fertilizer at least 2-3 inches below the soil
surface
36.4
63.6
Following soil test trends to maintain the agronomic range
for phosphorus in the soil (15 to 30 ppm)
44.7
55.3
Hiring a 4R certified applicator, which adheres to
recommended management practices, over an applicator
without certification
13.9
86.1

54
Table 75. Respondent intentions to adopt those practices that they have not yet adopted on
their high productivity fields, in valid percentages.
BMPs
n
Will never
adopt
Unlikely
to adopt
Likely to
adopt
Will
definitely
adopt
Grid soil sampling with variable rate
application
484
7.4
46.1
40.9
5.6
Planting cover crops after fall harvest
707
4.5
43.8
46.4
5.2
Delaying broadcasting when the forecast
predicts a 50% or more chance of at least 1
inch of total rainfall in the next 12 hours
501
2.8
18.2
59.1
20.0
Managing field water levels with drainage
management systems
635
9.6
51.7
30.9
7.9
Avoiding winter or frozen ground surface
application of phosphorus
377
4.8
16.4
43.5
35.3
Avoiding fall application of phosphorus
572
7.7
39.0
39.2
14.2
Determining rates based on regular soil
testing once within the rotation (or every 3
years)
320
3.4
16.3
48.1
32.2
Placement of fertilizer at least 2-3 inches
below the soil surface
504
4.2
31.3
46.0
18.5
Following soil test trends to maintain the
agronomic range for phosphorus in the soil
(15 to 30 ppm)
426
2.8
16.4
57.3
23.5
Hiring a 4R certified applicator, which
adheres to recommended management
practices, over an applicator without
certification
676
18.2
35.9
35.7
10.2
Table 76. Respondent plans for adopting four practices on their high productivity field next
season.
Plans for Next Season (n = 922)
Frequency
% New
Grid soil sampling with variable rate technology
342
20.6
Planting winter cover crops
187
49.0
Restricting nutrient application so that P levels do not exceed 15 ppm
131
40.2
Restricting nutrient application so that P levels do not exceed 30 ppm
190
30.4
I do not plan on using any of these practices next season
216
-
*Percent of respondents who had not yet adopted that practice but planned on adopting it next season

55
Summary of Field Comparisons
Further analyses comparing the three different types of fields revealed significant differences for
several farm and farmer characteristics including yield as well as the number of passes on the
field for fertilizer application, with both factors significantly increasing from low to high
productivity
1
. Willingness and adoption differed between field types for two of the ten nutrient
management practices examined in this study. Specifically, willingness of farmers to manage
field water levels with drainage management systems was significantly higher on their high
productivity fields compared to low productivity fields. Willingness of farmers to adopt storm-
delay broadcasting (i.e., delaying broadcasting when the forecast predicts a 50% or more
chance of at least 1 inch of total rainfall in the next 12 hours) was similar on low and high
productivity fields, which were significantly higher than farmer’s willingness to adopt this practice
on average productivity fields
2
,
3
. For more detailed information regarding adoption of select
nutrient management practices, please refer to the Appendix.
1
Significant differences for yield and number of passes on the field were found using a one-way ANOVA,
F(2,2756) = 22.2, p <.0005 and F(2,2755) = 6.92, p = .001, respectively. Chi Square analysis revealed
numerous other differences including tillage type, use of cover crops, use of manure, slope, drainage,
erodible classification, soil texture, rent, as well as timing and method of nutrient application.
2
Significant differences for willingness to adopt drainage management and storm-delay broadcasting
were found using a one-way ANOVA, F(2,2373) = 4.05, p = .017 and F(2,2343) = 5.77, p = .003,
respectively.
3
The format of our survey assumed that all farmers had at least one acre in each of the low, average,
and high productivity categories; however, simple crosstabs revealed that 50% of our respondents had
zero acres in our assigned category of field productivity. Therefore, we cannot guarantee that the field
specified by the respondent fell into our assigned category, which may bias these results.

56
References
Ajzen, I. 1985. From intentions to actions: a theory of planned behavior. In J. Kuhl & J. Beckmann
(Eds.), Action-control: from cognition to behavior pg. 11-39. Heidelberg: Springer.
Burton, R.J. 2004. Seeing through the ‘Good Farmer’s’ eyes: towards developing an understanding
of the social symbolic value of ‘productivist’ behavior. Sociologia Ruralis 44: 195-215.
Camboni, S. M., & Napier, T. L. 1993. Factors affecting use of conservation farming practices in
east central Ohio. Agriculture, Ecosystems and Environment, 45, 79-94.
Carpenter, S., Caraco, N., Correll, D., Howarth, R., Sharpley, A., & Smith, V. 1998. Nonpoint
pollution of surface waters with phosphorus and nitrogen. Ecological Applications, 8(3), 559-
568.
Dillman, D. 2000. Mail and internet surveys: The tailored design method. New York: Wiley.
Douglas, M. and A.B. Wildavsky. 1982. Risk and Culture: An Essay on the Selection of Technical
and Environmental Dangers. Berkeley, CA: Univ. of California Press.
Ferry, J. 2011. Phosphorus management in the agroecosystem: A mental models analysis of
knowledge and perceived risk. M.S. Thesis. The Ohio State University: Columbus, OH
http://etd.ohiolink.edu/send-pdf.cgi/Ferry%20Joshua%20David.pdf?osu1300980591
Kane, D., J. Conroy, R. Richards, D. Baker, and D. Culver. 2014. Re-eutrophication of Lake Erie:
correlations between tributary nutrient loads and phytoplankton biomass. Journal of Great
Lakes Research 40(3): 496-501.
Liberman, N. and Trope, Y. 2010. Construal-level theory of psychological distance. American
Psychological Association 117(2): 440-463.
Lichtenberg, E., and Zimmerman, R. 1999. Adverse health experiences, environmental attitudes,
and pesticide usage behavior of farm operators. Risk Analysis 19(2): 283-294.
McGuire, J., Morton, L., and Cast, A. 2013. Reconstructing the good farmer identity: shifts in farmer
identities and farm management practices to improve water quality. Agriculture and Human
Values 30: 57-69.
National Oceanographic and Atmospheric Administration. 2009. Center of Excellence for Great
Lakes and Human Health. Harmful algal bloom event response.
www.glerl.noaa.gov/res/Centers/HABS/sampling_data.html.
Perrings, C. and Hannon, B. 2001. An introduction to spatial discounting. Journal of Regional
Science 41(1): 957-972.
Pinto, J.V., T.C. Young and L.M. McIlroy. 1986. Great Lakes water quality improvement: The
strategy of phosphorus discharge control is evaluated. Environmental Science & Technology
20(8): 752-759.

57
Rayner, S. 1992. “Cultural Theory and Risk Analysis,” in S. Krimsky and D. Goldin, eds., Social
Theories of Risk, p. 83. Westport, CT: Praeger.
Tucker, M., and T. Napier. 1998. Percepitons of risk associated with use of farm chemicals:
implications for conservation initiatives. Environmental Management 22(4): 575-587.
Spence, A., Poortinga, W., and Pigeon, N. 2012. The psychological distance of climate change.
Risk Analysis 32(6): 957-972.
US Department of Agriculture. 2009, 2007. United States Census of Agriculture [summary statistics
on confidential microdata obtained from special tabulations by USDA - National Agricultural
Statistics Service]
Wilson, R.S., L. Burnett, T. Ritter, B. Roe and G. Howard. 2013. Farmers, phosphorus and water
quality: A descriptive report of beliefs, attitudes and practices in the Maumee Watershed of
northwest Ohio. The Ohio State University, School of Environment & Natural Resources.
.

58
Appendix A: Preliminary differences between adopters and non-adopters
The following section summarizes preliminary group differences for adoption of four of the
nutrient management practices investigated in this report (i.e., planting cover crops, storm-delay
broadcasting, seasonal-delay application, and injection application) and suggests methods for
effective messaging based on these findings.
Adoption of Cover Crops
Exploratory Analyses
4
. Results from our survey of soybean and corn farmers in the Maumee
watershed show that 16.2% of farmers plant cover crops after the fall harvest. Of those who
have not planted cover crops (on their assigned low, average, or high productivity field), 44.5%
said they were likely to plant cover crops in the future, while 43.9% said they were unlikely to
plant them in the future. Nearly a quarter of farmers (20.8%) plan on planting winter cover crops
next season, and of this group of farmers, just under half (about 46%) had not yet planted on
their field, meaning they were ‘new’ adopters of planting cover crops.
Proportional to non-adopters, cover crop adopters were more likely to use no-till on their
respective field, and more often had drainage tile installed proportional to non-adopters.
Adopters were also more likely to use soil testing to inform their nutrient management decisions
and have their field covered by the Federal Crop Insurance program. Compared to non-
adopters, adopters were significantly younger, had higher gross farm income, more owned
acres, and more rented acres. A further exploration of the differences between non-adopters,
current adopters, and ‘new’ adopters (i.e., those who had not yet planted cover crops on their
situational field, but planned on doing it next season) revealed that new adopters had
significantly less education than current adopters, as well as significantly less owned acres than
both current and non-adopters.
In terms of intentions to adopt, significant differences were found for age, education, annual
gross income, and total owned acres between those unwilling to adopt, willing to adopt, and
already adopting (Table A1). Those unwilling to adopt cover crops were significantly older than
both the other groups. Farmers who had already adopted cover crops had significantly more
education than either of the groups who had not yet adopted cover crops, and education did not
differ between unwilling and willing farmers. Similarly, those who had adopted cover crops
owned more of their acres than either of the groups who had not yet adopted, but total owned
acres did not differ significantly between unwilling and willing farmers. Most importantly, annual
gross income differed significantly across all three groups, with adopters of cover crops having
the highest income and unwilling farmers having the lowest income.
These three groups (i.e., unwilling to adopt, willing to adopt, and already adopted) were also
compared to investigate differences regarding various psychological factors (e.g., perceived
efficacy, responsibility, etc.). Compared to unwilling and willing adopters, those farmers who had
already adopted cover crops had a significantly higher sense of efficacy about nutrient loss (i.e.,
4
Based on Chi-square tests and one-way ANOVAs

59
a sense that they have the ability to reduce nutrient loss on their farms), a higher sense of
responsibility to take care of local water quality, and were more willing to take risks. Additionally,
they had a higher sense of control over various types of nutrient loss occurring on their farm.
Those who had already adopted cover crops rated conservation-oriented values (e.g.,
minimizes soil erosion and nutrient runoff, maintains soil organic matter, etc.) as being more
important than those unwilling or willing to adopt, meaning that this group related strongly to a
conservation-oriented identity. Differences between groups were not significant for profit-
oriented identity, suggesting that profit-oriented values (e.g., highest yields per acre, use of
latest seed and chemical technology, etc.) have no significant influence on a farmer’s
willingness to adopt cover crops. Trends between groups were the same for each psychological
variable, with the unwilling having the lowest values and actual adopters having the highest.
Table A1. Summary of variables significantly influencing adoption of cover crops
VARIABLES
UNWILLING TO ADOPT
WILLING TO ADOPT
ADOPTED
Age
Older
Younger
Education
Less
More
Income
Low
Moderate
High
Total Owned Acres
Less
More
Perceived Control
Low
Moderate
High
Risk Attitude
Risk Averse
Risk Tolerant
Risk Taking
Perceived Efficacy
Low
Moderate
High
Perceived Responsibility
Low
Moderate
High
Conservation Values
Weak
Moderate
Strong
Summary. In general, cover crop adopters tend to be younger, more educated, have higher
income, and own and rent more acres. The “new” adopters, those who were about to plant
cover crops for the first time after the upcoming harvest (the 2014 harvest), tended to be less
educated and owned less acreage. Those who were relatively unwilling to adopt cover crops
were older with less education and lower overall income. It appears that cover crop adopters are
more conservation-oriented in their thinking about farm management, although they were as
concerned about profit and production as those not using cover crops. Similarly, adopters had a
higher sense of responsibility to protect local water quality. Interestingly, adopters also had a
greater belief in their ability to reduce nutrient loss on their farms as a well a higher perceived
level of control over various types of nutrient loss on their farm. It is possible that these higher
degrees of efficacy and control are a result of having practices like cover crops in place;
however those who hadn’t yet adopted cover crops but who were willing to adopt them also had
higher degrees of efficacy and control compared to those who were unwilling to adopt them.
Finally, given the tendency for cover crops to be perceived as risky (due to timing, weather,
potential interference with planting, etc.), adopters tended to be more willing to take risks on
their farm, perhaps increasing their willingness to try out a new practice like cover crops.

60
Adoption of Storm-Delay Broadcasting
Exploratory Analyses
5
. Results from our survey of soybean and corn farmers in the Maumee
watershed show that 35.2% of farmers delay broadcasting nutrients on their fields when the
forecast predicts a 50% or more chance of at least 1 inch of total rainfall in the next 12 hours
(heretofore referred to as storm-delay broadcasting). Of those who have not adopted storm-
delay broadcasting (on their assigned low, average, or high productivity field), 77.3% said they
were likely to adopt this practice in the future, while 22.7% said they were unlikely to adopt.
Proportional to non-adopters, adopters of storm-delay broadcasting were more likely to have
their field in an established rotation and to have planted cover crops after their most recent crop.
Adopters of storm-delay broadcasting were more likely to apply fertilizer or manure in the fall, in
spring at planting, and after planting. A higher proportion of adopters incorporated their nutrients
with tillage and applied using band placement. Compared to non-adopters, adopters also had
proportionally less fields classified as highly erodible land by the USDA NRCS. More adopters
used soil testing on their field to inform their nutrient management decisions and had their field
covered by a Federal Crop Insurance program than non-adopters. Lastly, adopters of this
practice test their soils more frequently than non-adopters.
In terms of intentions to adopt, significant differences between those unwilling to adopt, willing to
adopt, and already adopting were found for age, annual gross income, education, and owned
and rented acreage (Table A2). Adopters of storm-delay broadcasting were significantly
younger, earned a higher income, and had completed more education than non-adopters.
Adopters also had more total owned and rented acreage than farmers who were willing to
adopt, however, their acreage did not significantly differ from farmers who were unwilling.
These three groups (i.e., unwilling to adopt, willing to adopt, and already adopted) were also
compared in order to investigate differences regarding various psychological factors (e.g.,
perceived efficacy, responsibility, etc.). Farmers who were unwilling to adopt storm-delay
broadcasting were less willing to take risks and had a lower sense of efficacy about nutrient loss
(i.e., lower belief in the ability to reduce nutrient loss on their farms) than willing farmers and
farmers who had already adopted this practice. Importantly, adopters of storm-delay
broadcasting had a higher sense of responsibility to take care of local water quality as well as a
higher sense of control over various types of nutrient loss occurring on their farm than both
unwilling and willing farmers. Similarly, we found that those who had adopted this practice rated
conservation-oriented values (e.g., minimizes soil erosion and nutrient runoff, maintains soil
organic matter, etc.) as being more important than those unwilling or willing to adopt, meaning
that this group related strongly to a conservation-oriented farmer identity. Responsibility, control,
and conservation identity significantly differed between each group, with unwilling adopters
having the lowest values and actual adopters having the highest values. We did observe a
difference between groups for profit-oriented identity, which suggests that profit-oriented values
(e.g., highest yields per acre, highest profit per acre, etc.) may be more important to farmers
who are willing to adopt this practice than to farmers who are not willing to adopt it.
5
Based on Chi-square tests and one-way ANOVAs

61
Summary. In general, adopters of storm-delay broadcasting tended to be younger, have higher
gross farm income, more education, and more total and rented acreage than those who haven’t
adopted this practice. These factors did not differ significantly between farmers who were willing
versus unwilling to adopt the practice in the future. Adopters of storm-delay broadcasting are
more conservation-oriented in their thinking about farm management, although they were as
concerned about profit and production as those not using this practice. Farmers who were
willing to adopt this practice, however, placed more importance on production values than
farmers who were unwilling to adopt this practice
6
, perhaps pointing to a belief that delaying
broadcasting in light of a major storm event would benefit the farm’s bottom line (an important
motivator for those considering adoption of the practice). Adopters had a higher sense of
responsibility to protect local water quality, and this sense increased significantly between
unwilling farmers, willing farmers, and adopters. Interestingly, adopters also had a greater belief
in their ability to reduce nutrient loss on their farms and higher perceived control over nutrient
loss on the farm. Although it is possible that these higher degrees of efficacy and control are a
result of using practices like storm-delay broadcasting, farmers who were willing to adopt this
practice had similar levels of efficacy to adopters, suggesting that this belief precedes adoption,
and may in fact be a critical motivator. Finally, given the unpredictability of weather, willing
farmers and adopters tended to be more willing to take risks on their farm, perhaps increasing
their willingness to adopt storm-delay broadcasting and risk further delays in production.
Table A2. Summary of variables significantly influencing adoption of storm-delay broadcasting
VARIABLES
UNWILLING TO ADOPT
WILLING TO ADOPT
ADOPTED
Age
Older
Younger
Income
Lower
Higher
Education
Less
More
Total Rented Acres
N/A*
Less
More
Total Owned Acres
N/A*
Less
More
Perceived Control
Low
Moderate
High
Risk Attitude
Risk Averse
Risk Taking
Efficacy
Lower
Higher
Responsibility
Low
Moderate
High
Production Values
Weak
Moderate†
N/A*
Conservation Values
Weak
Moderate
Strong
* N/A means that group did not differ significantly from the other two groups.
† The mean score for production values (M = 1.73, SD = .754) was lower than the mean for
conservation values (M = 3.00, SD = .636) on a 5-point scale.
6
The mean score for production values (M = 1.73, SD = .754) was lower than the mean for conservation
values (M = 3.00, SD = .636) on a 5-point scale. A basic comparison of the distributions for these values
showed higher importance placed on conservation values than production values.

62
Adoption of Seasonal-Delay Application
Exploratory Analyses
7
. Results from our survey of soybean and corn farmers in the Maumee
watershed show that 51.4% of farmers avoid winter or frozen ground surface application of
phosphorus to their fields (heretofore referred to as seasonal-delay application). Of those who
have not yet adopted seasonal-delay application (on their assigned low, average, or high
productivity field), 79.6% said they were likely to adopt this practice in the future, while 20.4%
said they were unlikely to adopt it in the future.
Proportional to non-adopters, adopters of seasonal-delay application were more likely to use
conservation tillage and have an established rotation. Adopters also grew proportionally more
corn and soybeans than non-adopters in 2013. A higher proportion of adopters used manure on
their most recent crop and planted cover crops compared to non-adopters. Additionally, a higher
proportion of adopters had their fields enrolled in conservation programs and had their fields
covered by crop insurance. More adopters used soil testing to inform their nutrient management
decisions than non-adopters, and they did this soil testing more frequently than non-adopters.
Compared to non-adopters, proportionally more adopters used a two year phosphorus
application on their most recent crop, although most of them used a one-year application.
Lastly, a higher proportion of adopters used the phosphorus form MAP (monoammonium
phosphate) than non-adopters.
In terms of intentions to adopt, significant differences between those who were unwilling to
adopt seasonal-delay application, willing to adopt this practice, and farmers who had already
adopted this practice were found for age, annual gross income, education, and total owned and
rented acres (Table A3). Farmers who had adopted seasonal-delay application were
significantly younger, had higher gross farm income, more education, and more total owned and
rented acreage than farmers who were willing or unwilling to adopt this practice. Importantly,
education was the only significant difference in farm and farmer characteristics between farmers
who were willing to adopt and farmers who were unwilling to adopt this practice, with willing
farmers having more education than unwilling farmers.
These three groups (i.e., unwilling to adopt, willing to adopt, and already adopted) were also
compared in order to investigate differences regarding various psychological factors (e.g.,
perceived efficacy, responsibility, etc.). Farmers who had adopted seasonal-delay application on
their chosen field were significantly more willing to take risks compared to those unwilling to
adopt, however they did not differ significantly from those willing to adopt this practice. Farmers
who were unwilling to adopt seasonal-delay application had a significantly lower sense of
efficacy about nutrient loss (i.e., lesser belief in his/her ability to reduce nutrient loss on the
farm) than willing farmers as well as those who had already adopted this practice. Perceived
control over various types of nutrient loss occurring on their farm and sense of responsibility to
take care of local water quality differed significantly between all three groups, with adopters
having the highest sense of responsibility and control and unwilling farmers having the lowest
sense of responsibility and control. Those who had already adopted seasonal-delay application
were also found to rate conservation-oriented values (e.g., minimizes soil erosion and nutrient
7
Based on Chi-square tests and one-way ANOVAs

63
runoff, maintains soil organic matter, etc.) as being more important than those unwilling or
willing to adopt, meaning that this group related strongly to a conservation-oriented identity.
Differences between groups were not significant for profit-oriented identity, suggesting that
profit-oriented values (e.g., highest yields per acre, use of latest seed and chemical technology,
etc.) have no significant influence on a farmer’s willingness to adopt seasonal-delay application.
Trends between groups were the same for each psychological variable, with unwilling adopters
having the lowest values and actual adopters having the highest values.
Summary. In general, adopters of seasonal-delay application tend to be younger, have higher
income, more education, and own and rent more acres than those who haven’t yet adopted this
practice. These factors did not differ significantly between willing and unwilling farmers. Among
farm or farmer characteristics, we found that only education differed significantly between
unwilling and willing to adopt farmers, suggesting that these two groups may be very similar to
each other for this particular practice. It appears that adopters of seasonal-delay application
(and those who are willing to adopt this practice) are more conservation-oriented in their
thinking about farm management, although they were as concerned about profit and production
as those unwilling to use this practice. Similarly, adopters had a higher sense of responsibility to
protect local water quality and perceived a greater level of control over various types of nutrient
loss occurring on their farms. These senses of responsibility and control increased significantly
between unwilling farmers, willing farmers, and adopters, suggesting that these beliefs exist
prior to actual adoption of the practice. Interestingly, farmers who were unwilling to adopt
seasonal-delay application had a weaker belief in their ability to reduce nutrient loss on their
farms than both willing farmers and current adopters. Finally, given the seasonal restrictions of
this practice, adopters tended to be more willing to take risks on their farm than unwilling
farmers, perhaps increasing their willingness to shift to seasonal-delay application.
Table A3. Summary of variables significantly influencing farmer adoption of seasonal-delay
application of nutrients
VARIABLES
UNWILLING TO ADOPT
WILLING TO ADOPT
ADOPTED
Age
Older
Younger
Income
Lower
Higher
Education
Little
Moderate
High
Total Owned Acres
Less
More
Total Rented Acres
Less
More
Perceived Control
Low
Moderate
High
Risk Attitude
Risk Averse
N/A*
Risk Taking
Efficacy
Low
High
Responsibility
Low
Moderate
High
Conservation Values
Weak
Moderate
Strong
*N/A means that group did not differ significantly from the other two groups

64
Adoption of Injection Application
Exploratory Analyses
8
. Results from our survey of soybean and corn farmers in the Maumee
watershed show that 34.5% of farmers place fertilizer at least 2-3 inches below the soil surface
(heretofore referred to as injection application). Of those who have not adopted injection
application (on their assigned low, average, or high productivity field), 64% said they were likely
to inject in the future, while 36% said they were unlikely to inject in the future.
Adopters of injection application more often had drainage tile installed proportional to non-
adopters. Most adopters used soil testing to inform their nutrient management decisions, had
their field covered by the Federal Crop Insurance program, and had their field in an established
rotation. Adopters of injection application more often grew corn in 2013 proportional to non-
adopters. Compared to non-adopters, a higher proportion of adopters did their fertilizer
application in the spring at planting as well as after planting.
In terms of intentions to adopt, significant differences were found between farmers who were
unwilling to adopt injection application, willing to adopt this practice, and those who had already
adopted this practice for farming experience, education, annual gross income, total rented
acres, and total owned acres (Table A4). Those unwilling to adopt injection application had
significantly less experience farming than farmers who were willing to adopt this practice as well
as farmers who had already adopted this practice. Farmers who were willing to adopt this
practice had significantly less education than those unwilling to adopt and those who had
already adopted injection application. No significant differences in education were observed
between unwilling farmers and farmers who had already adopted the practice. Both unwilling
and willing farmers had significantly lower annual gross income than those who already injected
nutrients on their field. Farmers willing to adopt injection application had significantly less total
owned acres than those farmers who were unwilling or who had already adopted this practice.
Similarly, willing farmers had significantly less total rented acres than those farmers who had
already adopted injection application, but their acreage did not differ significantly compared to
farmers who were unwilling to adopt.
These three groups (i.e., unwilling to adopt, willing to adopt, and already adopted) were also
compared in order to investigate differences regarding various psychological factors (e.g.,
perceived efficacy, responsibility, etc.). Significant differences between the groups were found
for five psychological factors. Compared to unwilling and willing adopters, those farmers who
had already adopted injection application were more willing to take risks and had significantly
higher perceived control over various types of nutrient loss occurring on their farm. These
farmers also had a significantly higher sense of efficacy about nutrient loss (i.e., a sense that
they have the ability to reduce nutrient loss on their farms) compared to those farmers who were
unwilling to adopt injection application, however, perceived efficacy did not differ significantly
between those who had adopted and those who were willing to adopt this practice. Importantly,
adopters of injection application had a higher sense of responsibility to take care of local water
quality than either unwilling or willing farmers. Those who had already adopted injection
application also rated conservation-oriented values (e.g., minimizes soil erosion and nutrient
8
Based on Chi-square tests and one-way ANOVAs

65
runoff, maintains soil organic matter, etc.) as being more important than those unwilling or
willing to adopt, meaning that this group related strongly to a conservation-oriented identity.
Differences between groups were not significant for production-oriented identity, suggesting that
profit-oriented values (e.g., highest yields per acre, highest profit per acre, etc.) have no
significant influence on a farmer’s willingness to inject apply. Trends between groups were the
same for each psychological variable, with unwilling adopters having the lowest values and
actual adopters having the highest values.
Summary. In general, adopters of injection application tend to have more farming experience,
higher gross farm income, and own and rent more acres. Those who were relatively unwilling to
adopt this practice were similar to adopters in education and owned acreage, but had less
farming experience and a lower income. Those who were relatively willing to adopt this practice
had less rented and owned acreage, less income, and less education than adopters and
unwilling farmers, but they had more farming experience compared to unwilling farmers. It
appears that adopters of injection application are more conservation-oriented in their thinking
about farm management, although they were as concerned about profit and production as those
who did inject apply. Similarly, adopters had a higher sense of responsibility to protect local
water quality. Adopters also had a greater belief in their ability to reduce nutrient loss on their
farms as well as a higher perceived level of control over various types of nutrient loss on their
farm. It is possible that these higher degrees of efficacy and control are a result of using
practices like injection application; however, those who hadn’t yet adopted this practice but who
were willing to adopt it also had higher degrees of efficacy compared to those who were
unwilling to adopt them. Finally, adopters tended to be more willing to take risks on their farm,
perhaps increasing their willingness to try out a new practice like injection application.
Table A4. Summary of variables that significantly influence adoption of injection application
VARIABLES
UNWILLING TO ADOPT
WILLING TO ADOPT
ADOPTED
Farming Experience
Less
More
Education
More
Less
More
Income
Lower
Higher
Total Rented Acres
N/A*
Less
More
Total Owned Acres
More
Less
More
Perceived Control
Lower
Higher
Risk Attitude
Risk Averse
Risk Taking
Efficacy
Lower
Higher
Responsibility
Low
Moderate
High
Conservation Values
Weak
Moderate
Strong
*N/A means that group did not differ significantly from the other two groups

66
Final Conclusion and Recommendations Across Practices. In order to encourage the next
phase of adopters to implement these nutrient management practices, it is important to
emphasize the economic benefits of each practice in terms of productivity of the landscape, as
those less likely to adopt are not as motivated by conservation objectives. Those less likely to
adopt also have a greater tendency to avoid risky endeavors, making it critical that the
perceived risks associated with each practice be minimized. This can be done through
messaging by emphasizing the benefits, which, due to the typical inverse relationship between
perceived risk and benefit, may then decrease the perceived risks associated with implementing
each practice. For example, if issues of timing are a concern, then one should emphasize the
soil health benefits of cover crops or the economic benefits of delayed application, as well as
address how the issues associated with timing can directly be addressed. Providing tailored on-
farm (or even field-level) support can help farmers to navigate the potential risks and increase
the probability of successful implementation. Such support may also strengthen the perceived
efficacy or ability of farmers with fewer resources to successfully implement new practices,
which would then increase the likelihood of adoption given the importance of perceived efficacy
for many of the non-adopters or less willing farmers, and the tendency for these individuals to
also have less resources at their disposal.